Welcome to Superdesk Publisher documentation¶

The next-generation publishing platform for newsrooms

Superdesk Publisher is a lightweight open source renderer for news articles and other content delivered via an API feed. The code is released under the GNU Affero General Public Licence, version 3.

Publisher is designed to work with the Superdesk newsroom management system from Sourcefabric, but it can also be adapted to work with any compatible API. Publisher is a lightweight PHP 7 renderer for HTTP-pushed content in both HTML/CSS/JavaScript and PWA templates, and it runs on a standard web server or in a Docker container. A PostgreSQL database is also required.

The presentation of articles is taken care of by a flexible, device-responsive themes system, which can be customised to suit your publications.

This documentation includes text and code examples from the Symfony and Sylius projects, released under the Creative Commons BY-SA 3.0 licence. Pull requests to improve the documentation are welcome.

Manual¶

Superdesk Publisher: 360° digital output on an industrial scale

Publisher is the latest tool made for the Superdesk newsroom suite. It publishes multimedia content across multiple outputs, from websites to apps to outdoor displays to social media, and allows simultaneous, centralised monitoring and management of all of your assets.

Publisher is designed to complement Superdesk, which already powers content creation, production, distribution and curation at media businesses such as global news agencies and print/online newspapers. Superdesk is a structured-data polyglot and converses in formats as diverse as the legacy ANPA 1312 all the way to media-rich NewsML G2. But Publisher can also be extended to handle content created in third-party systems, even in exotic, custom formats.

We encourage any media outfit dealing with multiple outputs, platforms and channels to investigate the power of Publisher. Whether you author and produce content in Superdesk or not, Publisher gives you the real-time multi-tenancy overview you need to ensure that it performs optimally, everywhere.

Introduction¶

Why choose Publisher?¶

If your organisation already creates and produces content in Superdesk, Publisher is built to work with it natively. If you are not using Superdesk, but your back-end system or systems are still fit for purpose and your need is to manage a portfolio of digital assets (from multiple websites to apps to social feeds), Superdesk Publisher can be integrated with your legacy tools until their deprecation and replacement.

What is Publisher’s focus?¶

Efficient content delivery to multiple digital channels.
Rapid development of new website and digital layouts, independently from back-end systems to avoid disrupting editorial workflows.
All the latest embeds and custom widgets for your web pages.
Full commercial support from the upstream development and implementation teams.

How is Publisher structured?¶

This graphic shows how Publisher is configured.

Data: Symfony uses Postgres and Doctrine.
Basic frameworks: Publisher is based on Symfony and the Symfony CMF.
Publisher: The rest is Publisher

Notable features¶

Multitenancy
Built-in support for reverse proxy caching (Varnish, Nginx, internal system as a fallback)
Twig or PWA theme support

Website architecture¶

A basic Twig theme must have the following structure:

ExampleTheme/               <=== Theme starts here
    views/                  <=== Views directory
        article.html.twig
        base.html.twig
        category.html.twig
        home.html.twig
    translations/           <=== Translations directory
        messages.en.xlf
        messages.de.xlf
    public/                 <=== Assets directory
        css/
        js/
        images/
    theme.json              <=== Theme configuration

Getting started¶

The Superdesk Publisher themes system is built on fast, flexible and easy-to-use Twig templates.

Installing Superdesk Publisher¶

This guide describes how to install Superdesk Publisher (refered to as Publisher) on an Ubuntu 18.04 server using Nginx web server. This guide was verified as accurate and tested using Superdesk Publisher 2.0.3.

Installing Publisher Prerequisites¶

See the Publisher Requirements to read more about specific requirements.

Before starting, make sure your Ubuntu server has the latest packages available by running the commands:

sudo apt update

sudo apt upgrade

Install PHP-FPM 7.3 and Extensions¶

Add the ondrej/php repository, which has the PHP 7.3 package and required extensions:

#!/bin/bash

sudo apt install software-properties-common
sudo add-apt-repository ppa:ondrej/php
sudo apt update

Install PHP 7.3 and required extensions:

sudo apt-get install -y php7.3-fpm php7.3-pgsql php7.3-gd php7.3-xml php7.3-intl php7.3-zip php7.3-mbstring php7.3-curl php7.3-bcmath

Configure PHP-FPM 7.3 by running the command:

cd /etc/php/7.3/fpm/pool.d/ &&
sudo curl -s -O https://gist.githubusercontent.com/takeit/2ee16ee50878eeab01a7ca11b69dec10/raw/e9eda2801ac3657495374fcb846c2ff101a3e070/www.conf &&
sudo service php7.3-fpm restart

Install PostgreSQL¶

Install PostgreSQL:

1	sudo apt-get install postgresql postgresql-contrib -y

The default PostgreSQL user is postgres with no password set.

Install Memcached¶

Install Memcached:

1	sudo apt-get install -y memcached

Install the Memcached PHP extension:

1	sudo apt-get install -y php7.3-memcached

Install ElasticSearch¶

ElasticSearch v5.6 will be used. Run the following command to install ES:

curl -L -O https://artifacts.elastic.co/downloads/elasticsearch/elasticsearch-5.6.0.deb &&
sudo dpkg -i elasticsearch-5.6.0.deb && sudo apt-get -y update &&
sudo apt-get -y install --no-install-recommends openjdk-8-jre-headless &&
sudo systemctl enable elasticsearch && sudo systemctl restart elasticsearch

The ElasticSearch should be running on port 9200. You can run the following command to verify this:

1	curl -s "http://localhost:9200"

If you get no response in the console after running that command, use this command to check for error messages:

1	systemctl status elasticsearch

Install and Configure Nginx Server¶

Install Nginx:

1	sudo apt-get -y install nginx

Configure Nginx site-enabled by editing the file /etc/nginx/sites-enabled/default. Paste in the following configuration:

server {
        server_name example.com
        listen 80 default;
        root /var/www/publisher/public;

        location / {
                try_files $uri /index.php$is_args$args;
        }

        location ~ ^/index\.php(/|$) {
                fastcgi_pass 127.0.0.1:9000;
                fastcgi_split_path_info ^(.+\.php)(/.*)$;
                include fastcgi_params;

                fastcgi_param SCRIPT_FILENAME $realpath_root$fastcgi_script_name;
                fastcgi_param DOCUMENT_ROOT $realpath_root;

                internal;
        }

        location ~ \.php$ {
                return 404;
        }

        error_log /var/log/nginx/project_error.log;
        access_log /var/log/nginx/project_access.log;
}

Restart the Nginx service:

1	sudo service nginx restart

Install RabbitMQ Server¶

Install RabbitMQ:

1	sudo apt install -y rabbitmq-server

Install the AMQP PHP extension:

1	sudo apt-get install -y php7.3-amqp

Install Supervisor¶

Install Supervisor:

1	sudo apt-get install -y supervisor

Installing Publisher¶

Clone the source code from the Publisher repository on GitHub, then install dependencies and configure the Publisher server.

Clone the Publisher Repository¶

The default directory where the Publisher source code will be downloaded can be /var/www/publisher and all console commands need to be executed inside that directory starting from now on.

Run the clone command in your terminal:

cd /var/www/ && sudo git clone https://github.com/superdesk/web-publisher.git publisher && cd publisher

All commands must be run in the /var/www/publisher directory from now on.

Install Publisher Dependencies¶

Install Composer:

1	sudo apt-get install composer -y

Install Publisher’s dependencies (which can be found in composer.json) using the following command:

1	composer install

Create and Populate the Database¶

Create a new terminal session and log into the postgres user:

1	su - postgres

Create a new user ‘root’ as a superuser to match Publisher’s default database connection configuration:

1	createuser -s -d root

Next, find the location of PostgreSQL’s pg_hba.conf file:

psql -t -P format=unaligned -c 'show hba_file';

In this guide we are using version 10 of PostgreSQL, so our pg_hba.conf is located at /etc/postgresql/10/main/pg_hba.conf. Edit this file and change the local connections authentication method from peer or md5 to trust.

Danger

Changing this setting to trust will allow anyone, even remote, to be able to log into the database as any user without authentication. You will learn how to secure PostgreSQL in Configure and secure your Publisher server.

Now, reload the pg_hba.conf file:

psql -t -P format=unaligned -c 'select pg_reload_conf()';

Exit the postgres user session:

exit

Create the database using Doctrine:

1	php bin/console doctrine:database:create

Populate the database schema:

1	php bin/console doctrine:migrations:migrate

If you’re not installing Publisher for a production environment and want to quickly add test data, populate the database with test data using the following command:

1	php bin/console doctrine:fixtures:load

or

1	php -d memory_limit=-1 bin/console doctrine:fixtures:load

Generate the SSH keys to properly use the authentication (readers)¶

Generate the SSH keys:

#!/bin/bash

mkdir -p config/jwt
openssl genrsa -out config/jwt/private.pem -aes256 4096
openssl rsa -pubout -in config/jwt/private.pem -out config/jwt/public.pem

In case first openssl command forces you to input password use following to get the private key decrypted:

#!/bin/bash

openssl rsa -in config/jwt/private.pem -out config/jwt/private2.pem
mv config/jwt/private.pem config/jwt/private.pem-back
mv config/jwt/private2.pem config/jwt/private.pem

Create an Organization and Tenant¶

Create a new organization:

1	php bin/console swp:organization:create Acme

An organization code will be output. Take note of it.

Create a new tenant under your organization:

1	php bin/console swp:tenant:create <organization_code> example.com AcmeTenant

Take note of the tenant code output by this command.

Install the Demo Theme¶

Install and activate the demo theme. Replace 123abc with your tenant code:

php bin/console swp:theme:install 123abc src/SWP/Bundle/FixturesBundle/Resources/themes/DefaultTheme/ -f -p -a

Install the theme assets (images, stylesheets, JavaScript, etc.):

1	php bin/console sylius:theme:assets:install

Configure Supervisor¶

Supervisor is used to automatically start services that Publisher depends on.

The program configuration files for Supervisor programs are founds in the /etc/supervisor/conf.d directory, normally with one program per file and a .conf extension. We prepared ready-to-use configuration files for Publisher consumers. You can find them in etc/scripts/supervisor directory.

Copy them to the Supervisor configs directory:

1	cp -r etc/scripts/supervisor/. /etc/supervisor/conf.d

Then, reload Supervisor:

1	systemctl reload supervisor

Bind websocket queue to websocket exchange:

#!/bin/bash

sudo rabbitmq-plugins enable rabbitmq_management
wget http://127.0.0.1:15672/cli/rabbitmqadmin
chmod +x rabbitmqadmin
sudo mv rabbitmqadmin /etc/rabbitmq
/etc/rabbitmq/rabbitmqadmin --vhost=/ declare binding source="swp_websocket_exchange" destination="swp_websocket"

Start the web server:

1	php bin/console server:start

Use your web browser to navigate to your Publisher instance, using the domain you specified earlier when creating a new tenant. You should now see the home page for your tenant!

Next Steps¶

How to Install and Configure Superdesk Publisher with Superdesk¶

This guide describes the functions of Superdesk Publisher and Superdesk, along with the required steps to run both applications concurrently in a production environment on two different servers. (However, both applications can also work on a single machine.)

This guide assumes that you have:

A working instance of Superdesk
A working instance of Superdesk Publisher

For this guide, we will assume your Superdesk server runs on superdesk.example.com and your Publisher server runs on example.com.

Install the Superdesk Publisher Component in Superdesk¶

Login to the server where Superdesk is installed.

Superdesk Publisher Component is a JavaScript component that is a separate dependency and can be included in Superdesk to manage your Superdesk Publisher application.

To add this component as a dependency edit /opt/superdesk/client/package.json and include the Superdesk Publisher repository:

"dependencies": {
        ....,
        "superdesk-publisher": "superdesk/superdesk-publisher#2.0"
}

You can replace master with whichever branch you require.

Next, run the following command to install the package.json dependencies:

1	npm install

Update Your Superdesk Configuration File¶

Inside /opt/superdesk/client directory on your server open the superdesk.config.js and integrate the following code with your existing configuration:

apps: [
        ....,
        'superdesk-publisher'
],
importApps: [
        ....,
        'superdesk-publisher'
]
publisher: {
        protocol: 'http',                /* http or https */
        tenant: '',                      /* tenant - semantically subdomain, '' is allowed */
        domain: 'example.com',           /* domain name for the publisher */
        base: 'api/v2',                  /* api base path */

        wsProtocol: 'wss',                /* ws or wss (websocket); if unspecified or '' defaults to 'wss' */
        wsDomain: 'WebSocketDomain.com',  /* domain name (usually domain as above) */
                                                                          /* e.g.: example.com, abc.example.com */
                                                                          /* tenant, as above, is NOT used for websocket */
        wsPath: '/ws',                    /* path to websocket root dir */
        wsPort: '8080'                    /* if not specified: defaults to 443 for wss, 80 for ws */
},

Finally, rebuild the front-end by running this command:

1	grunt build

Update Your Publisher Configuration File¶

Login to the server where Publisher is installed.

Edit (or create, if it doesn’t already exist) the file /var/www/publisher/.env.local and add the following:

SUPERDESK_SERVERS='["superdesk.example.com"]'
CORS_ALLOW_ORIGIN=http://superdesk.example.com

Note

For CORS_ALLOW_ORIGIN, it is important to include the protocol your Superdesk server uses (http or https) and to not have a trailing slash after your domain name (e.g. your value should not be http://superdesk.example.com/).

Clear the Publisher server cache for the change to take effect:

1	php bin/console cache:clear

That’s it! Now, when you log in to Superdesk in the left hamburger menu, you will see the Publisher Settings menu item available:

Configure Subscriber to Publish Content from Superdesk to Superdesk Publisher¶

You can read more about this in the official Superdesk Publisher documentation.

Configuring Publisher¶

Publisher integrates into Superdesk simply by adding new option Publisher Settings to its main left-hand sidebar navigation.

This option, when chosen, opens Publisher configuration which allows configuring one or more websites. Setting a website actually means defining routes, creating navigation menus (whose menu items are linked to these routes), and creating content lists.

Detailed explanation of website management steps can be found in chapter Admin interface

Managing websites and routes¶

When Publisher is installed, it is integrated into Superdesk and expects output channels (or, in other words - websites) to be set - as already mentioned in Configuring Publisher.

The main concepts are:

there can be more than one website that you can configure and feed from Superdesk
each website is configured firstly by its routes; routes can be of type collection, content and custom.

Route of type collection is expected to get articles attached to it - think of it as a kind of category page (Business, or Politics, or simply News). When configuring this route, you need to also specify an article template name - the one that will be used to show articles attached to that route.

Route of type content is the destination - it holds the content! Either only one article is attached to it, or no articles at all! So it can be either a static article page (About us), or a special template (contact form, confirmation page, or simply a route that doesn’t directly hold attached articles, like ‘home’ route for example, or Trending articles, which would use a special template to show that kind of content).

Routes of type custom are dynamically generated and thus have more technical aspect in essence. They are used to define, for example, author profile routes - /author/john-smith, /author/sarrah-staffwriter etc. The first part is static (‘author’), while the second part is dynamic and based on an author slug in this case. Written in json format, a custom route definition looks like this:

Here a theme developer would needs to get the author slug value from the context and construct the url dynamically using it.

Navigation management¶

Navigations are menus that you can use on your websites. The advantage of creating them here (and not building the navigation menu only in the theme) is that they can later be managed via the Superdesk Publisher interface (meaning menu items added, removed and reordered) even by website editors, and not just theme developers.

Only after configuring Routes, we can proceed to configure Navigation. That’s because navigation is built on menu items, and menu items are partly defined by routes.

However, it doesn’t have to be the case, as menu items can also have custom URIs. But as SEO algorithms sort-of penalize navigations with external links, it should be used wisely (ie. to make a link to a specific custom route).

The elements needed for menu item definition are:

Name
Label - the value that is shown on the front end for that menu item in the navigation menu
Parent - useful when building nested, drop-down menus
Route - one of previously defined routes
Uri - automatically filled in when route is selected

Configure image formats¶

In Superdesk Settings -> Vocabularies -> Image Crop sizes it is possible to define all needed image sizes.

If you are configuring one website, depending on the layout design it can require one image size for lead story on front, another one for second-level stories, thumbnail image for small teasers, big article page image, and finally image size optimized for Facebook sharing for example.

So all these sizes should be crops of originally uploaded image that journalist or editor are adding to the story as feature image (one that represents the story in teasers).

If you have more websites powered by one Superdesk instance, then all of the needed crops are defined here.

Even though original image gets cropped automatically, this process can be overviewed and best point of interest and area of interest are customizable.

Templates System¶

The Superdesk Publisher templates system has its own git repository, at: https://github.com/SuperdeskWebPublisher/templates-system

Rendering pages with Twig¶

Superdesk Publisher uses Twig templating engine to render website HTML. Twig is modern, flexible, extensible and secure templating system, and has great documentation, as well as active support community at Stack Overflow.

This is how Twig code looks like:

{% for user in users %}
* {{ user.name }}
{% else %}
   No users have been found.
{% endfor %}

If you are creating completely new theme for your Publisher project, or going to modify some of the existing demo themes, you can follow this handy guide.

Generally, if starting from scratch, we advise you to develop your HTML/CSS/JS first with some dummy content, and once it’s ready, you can proceed with translating this markup into twig templates.

We have developed three demo themes which can serve as a refference for quick start (more about it here)

Superdesk Publisher demo theme, located at /src/SWP/Bundle/FixturesBundle/Resources/themes/DefaultTheme inside your Publisher instance (this theme is distributed as part of Publisher package)
The Modern Times theme, whose Git repo is here: https://github.com/SuperdeskWebPublisher/theme-dailyNews
Magazine theme, whose Git repo is here: https://github.com/SuperdeskWebPublisher/theme-magazine

Creating a page templates¶

Page template can render content in a straight-forward way: take article delivered to Publisher from Superdesk (or from any other source) and put its elements in markup of your choice (article title in <h1>, wrap main body in block-level element, etc).

Creating custom templates¶

Setting error pages from theme¶

Publisher provides simple default templates for error pages. You can find them in app/Resources/TwigBundle/views/Exception/ directory.

To override these templates from theme you need to create TwigBundle/views/Exception/ directory in your theme, and put there new error pages files.

Example Structure:

ThemeName/
└─ TwigBundle/
   └─ views/
      └─ Exception/
         ├─ error404.html.twig
         ├─ error403.html.twig
         ├─ error500.html.twig
         ├─ error.html.twig      # All other HTML errors

Testing error pages during theme development¶

You can use URLs like

http://wepublisher.dev/app_dev.php/_error/404
http://wepublisher.dev/app_dev.php/_error/403
http://wepublisher.dev/app_dev.php/_error/500
http://wepublisher.dev/app_dev.php/_error/501 # error.html.twig will be loaded

to preview the error page for a given status code as HTML.

Properties¶

Templates features¶

Custom Twig tags¶

Gimme allows you to fetch the Meta object you need in any place of your template. It supports single Meta objects (with gimme ) and collections of Meta objects (with gimmelist).

gimme¶

The tag gimme has one required parameter and one optional parameter:

(required) Meta object type (and name of variable available inside block), for example: article

(optional) Keword with and parameters for Meta Loader, for example: { param: "value" }

{% gimme article %}
    {# article Meta will be available under "article" variable inside block #}
    {{ article.title }}
{% endgimme %}

Meta Loaders sometimes require special parameters - like the article number, language of the article, user id, etc..

{% gimme article with { articleNumber: 1 } %}
    {# Meta Loader will use provided parameters to load article Meta #}
    {{ article.title }}
{% endgimme %}

gimmelist¶

The gimmelist tag has two required parameters and two optional parameters:

(required) Name of variable available inside block: article

(required) Keyword from and type of requested Metas in collection: from articles with filters passed to Meta Loader as extra parameters (start, limit, order)

(optional) Keyword with and parameters for Meta Loader, for example: with {foo: 'bar', param1: 'value1'}

(optional) Keyword without and parameters for Meta Loader, for example: without {source: 'AAP'}

(optional) Keyword if and expression used for results filtering

(optional) Keyword ignoreContext and optional array of selected meta to be ignored

Example of the required parameters:

{% gimmelist article from articles %}
    {{ article.title }}
{% endgimmelist %}

Example with ignoring selected context parameters:

{% gimmelist article from articles ignoreContext ['route', 'article'] %}
...

Example with ignoring whole context

{% gimmelist article from articles ignoreContext [] %}
...

Or even without empty array

{% gimmelist article from articles ignoreContext %}
...

Example with filtering articles by metadata:

{% gimmelist article from articles with {metadata: {byline: "Karen Ruhiger", located: "Sydney"}} %}
    {{ article.title }}
{% endgimmelist %}

The above example will list all articles by metadata which contain byline equals to Karen Ruhiger AND located equals to Sydney.

To list articles by authors you can also do:

{% gimmelist article from articles with {author: ["Karen Ruhiger", "Doe"]} %}
    {{ article.title }}
    Author(s): {% for author in article.authors %}<img src="{{ url(author.avatar) }}" />{{ author.name }} ({{ author.role }}) {{ author.biography }} - {{ author.jobTitle.name }},{% endfor %}
{% endgimmelist %}

It will then list all articles written by Karen Ruhiger AND Doe.

To list articles from the Forbes source but without an AAP source you can also do:

{% gimmelist article from articles with {source: ["Forbes"]} without {source: ["AAP"]} %}
    {% for source in article.sources %} {{ source.name }} {% endfor %}
{% endgimmelist %}

It will then list all articles with source Forbes and without AAP.

Listing article’s custom fields:

{% gimmelist article from articles %}
    {{ article.title }}
    {{ article.extra['my-custom-field'] }}
{% endgimmelist %}

Example with usage of all parameters:

{% gimmelist article from articles|start(0)|limit(10)|order('id', 'desc')
    with {foo: 'bar', param1: 'value1'}
    contextIgnore ['route', 'article']
    if article.title == "New Article 1"
%}
    {{ article.title }}
{% endgimmelist %}

How to work with `gimmelist` pagination?¶

gimmelist is based on Twig for tag, like in Twig there is loop variable available. In addition to default loop properties there is also totalLength. It’s filled by loader with number of total elements in storage which are matching criteria. Thanks to this addition we can build real pagination.

TemplateEngine Bundle provides simple default pagination template file: pagination.html.twig.

Note

You can override that template with SWPTemplatesSystemBundle/views/pagination.html.twig file in Your theme. Or You can use own file used for pagination rendering.

Here is commented example of pagination:

{# Setup list and pagination parameters #}
{% set itemsPerPage, currentPage = 1, app.request.get('page', 1) %}
{% set start = (currentPage / itemsPerPage) - 1 %}

{# List all articles from route '/news' and limit them to `itemsPerPage` value starting from `start` value #}
{% gimmelist article from articles|start(start)|limit(itemsPerPage) with {'route': '/news'} %}

    <li><a href="{{ url(article) }}">{{ article.title }} </a></li>

    {# Render pagination only at end of list #}
    {% if loop.last  %}
        {#
            Use provided by default pagination template

            Parameters:
            * currentFilters (array) : associative array that contains the current route-arguments
            * currentPage (int) : the current page you are in
            * paginationPath (Meta|string) : the route name (or supported by router Meta object) to use for links
            * lastPage (int) : represents the total number of existing pages
            * showAlwaysFirstAndLast (bool) : Always show first and last link (just disabled)
        #}
        {% include '@SWPTemplatesSystem/pagination.html.twig' with {
            currentFilters: {}|merge(app.request.query.all()),
            currentPage: currentPage,
            paginationPath: gimme.route,
            lastPage: (loop.totalLength/itemsPerPage)|round(1, 'ceil'),
            showAlwaysFirstAndLast: true
        } only %}
    {% endif %}
{% endgimmelist %}

For referrence, see original pagination.html.twig template (if you want to customize it and use instead of default one):

{#
  Source: http://dev.dbl-a.com/symfony-2-0/symfony2-and-twig-pagination/
  Updated by: Simon Schick <simonsimcity@gmail.com>

  Parameters:
    * currentFilters (array) : associative array that contains the current route-arguments
    * currentPage (int) : the current page you are in
    * paginationPath (string) : the route name to use for links
    * showAlwaysFirstAndLast (bool) : Always show first and last link (just disabled)
    * lastPage (int) : represents the total number of existing pages
#}
{% spaceless %}
    {% if lastPage > 1 %}

        {# the number of first and last pages to be displayed #}
        {% set extremePagesLimit = 3 %}

        {# the number of pages that are displayed around the active page #}
        {% set nearbyPagesLimit = 2 %}

        <nav class="pagination">
            <div class="numbers">
                <ul>
                    {% if currentPage > 1 %}
                        <li><a href="{{ path(paginationPath, currentFilters|merge({page: currentPage-1})) }}">Previous</a></li>

                        {% for i in range(1, extremePagesLimit) if ( i < currentPage - nearbyPagesLimit ) %}
                            <li><a href="{{ path(paginationPath, currentFilters|merge({page: i})) }}">{{ i }}</a></li>
                        {% endfor %}

                        {% if extremePagesLimit + 1 < currentPage - nearbyPagesLimit %}
                            <span class="sep-dots">...</span>
                        {% endif %}

                        {% for i in range(currentPage-nearbyPagesLimit, currentPage-1) if ( i > 0 ) %}
                            <li><a href="{{ path(paginationPath, currentFilters|merge({page: i})) }}">{{ i }}</a></li>
                        {% endfor %}
                    {% elseif showAlwaysFirstAndLast %}
                        <span class="disabled">Previous</span>
                    {% endif %}

                    <li class="current"><a href="{{ path(paginationPath, currentFilters|merge({ page: currentPage })) }}">{{ currentPage }}</a></li>

                    {% if currentPage < lastPage %}
                        {% for i in range(currentPage+1, currentPage + nearbyPagesLimit) if ( i <= lastPage ) %}
                            <li><a href="{{ path(paginationPath, currentFilters|merge({page: i})) }}">{{ i }}</a></li>
                        {% endfor %}

                        {% if  (lastPage - extremePagesLimit) > (currentPage + nearbyPagesLimit) %}
                            <li><span class="sep-dots">...</span></li>
                        {% endif %}

                        {% for i in range(lastPage - extremePagesLimit+1, lastPage) if ( i > currentPage + nearbyPagesLimit ) %}
                            <li><a href="{{ path(paginationPath, currentFilters|merge({page: i})) }}">{{ i }}</a></li>
                        {% endfor %}

                        <li><a href="{{ path(paginationPath, currentFilters|merge({page: currentPage+1})) }}">Next</a></li>
                    {% elseif showAlwaysFirstAndLast %}
                        <li><span class="disabled">Next</span></li>
                    {% endif %}
                </ul>
            </div>
        </nav>
    {% endif %}
{% endspaceless %}

How to work with Meta objects¶

On the template level, every variable in Context and fetched by gimme and gimmelist is a representation of Meta objects.

dump

{{ dump(article) }}

print

{{ article }}

If the meta configuration has the to_string property then the value of this property will be printed, otherwise it will be represented as JSON.

access property

{{ article.title }}
{{ article['title']}}

generate url

{{ url(article) }}    // absolute url
{{ path(article) }}   // relative path

Here’s an example using gimmelist:

{% gimmelist article from articles %}
    <li><a href="{{ url(article) }}">{{ article.title }} </a></li>
{% endgimmelist %}

Stringy twig extensions¶

We have extended the twig syntax, adding a number of functions for working with strings from a php library. A list of the functions together with a description of each, and of how they are to be invoked in PHP can be found here: https://github.com/danielstjules/Stringy#instance-methods

To call one of these functions in twig, if it returns a boolean, it is available as a twig function. So, for example, the function contains() can be called like this in twig:

{% set string_var = 'contains' %}
{% if contains(string_var, 'tain') %}string_var{% endif %} // will render contains

Any php function which returns a string is available in twig as a filter. So, for example, the function between() can be called like this in twig:

{% set string_var = 'Beginning' %}
{{ string_var|between('Be', 'ning') }} // will render gin

And the function camelize(), which doesn’t require any parameters, can simply be called like this:

{% set string_var = 'Beginning' %}
{{ string_var|camelize }} // will render bEGINNING

Redirects¶

We provide two functions which can be used for redirects:

redirect¶

{# redirect(route, code, parameters) #}
{{ redirect('homepage', 301, [] }}

route param can be string with ruute name or route meta object (article meta object will work also).
code is a redirection HTTP code (301 by default)
parameters - if route requires any it can be provided here

notFound¶

{{ notFound('Error message visible for reader' }}

notFound function will redirect user to 404 error page with provided message (it’s usefull when in your custom route, loader can’t find requested data).

Handling Articles¶

Listing Articles¶

Publisher have concept of Meta Loaders - one of built in loaders covers articles.

The articles loader parameters:

(optional) key route - id or name or array of id’s used for loading meta (if omitted then current route is used).

{% gimmelist article from articles %} <!-- It will use route from context -->
    <img src="{{ url(article) }}" />
{% endgimmelist %}

{% gimmelist article from articles with {'route': 1} %} <!-- route id -->
    <img src="{{ url(article) }}" />
{% endgimmelist %}

{% gimmelist article from articles with {'route': '/news'} %} <!-- route staticPrefix -->
    <img src="{{ url(article) }}" />
{% endgimmelist %}

{% gimmelist article from articles with {'route': ['/news', '/sport/*']} %} <!-- route staticPrefix -->
    <img src="{{ url(article) }}" />
{% endgimmelist %}

Note

'/sport/*' syntax will load articles from main route ('/sport') and all of it 1st level children (eg. '/sport/football').

{% gimmelist article from articles with {'route': [1, 2, 3]} %} <!-- array with routes id -->
    <img src="{{ url(article) }}" />
{% endgimmelist %}

(optional) key metadata - It matches article’s metadata, and you can use all metadata fields that are defined for the article, i.e.: language, located etc.

{% gimmelist article from articles with {'metadata':{'language':'en'}} %}
    <img src="{{ url(article) }}" />
{% endgimmelist %}

(optional) key keywords - It matches article’s keywords,

{% gimmelist article from articles with {'keywords':['keyword1', 'keyword2']} %}
    <img src="{{ url(article) }}" />
{% endgimmelist %}

Filtering out selected articles (useful when you want to exclude articles listed already in content list)

{% gimmelist article from articles without {article:[1,2]} %} <!-- pass articles ids (collected before) -->
    <img src="{{ url(article) }}" />
{% endgimmelist %}

{% gimmelist article from articles without {article:[gimme.article]} %} <!-- pass articles meta objects -->
    <img src="{{ url(article) }}" />
{% endgimmelist %}

Ordering by article comments count (set by external system)

{% gimmelist article from articles|order('commentsCount', 'desc') %}
    <img src="{{ url(article) }}" />
{% endgimmelist %}

Fetching first article url (when it’s changed after slug or route change)¶

{% gimme article with {slug: "test-article"} %}
    <a href="{{ original_url(article) }}">{{ article.title }}</a>
{% endgimme %}

Note

original_url function will always return valid article url. If article url was changed after publication, then it will return it’s original (first) value.

Handling Article Media¶

Listing Article Media¶

Publisher have concept of Meta Loaders - one of built in loaders covers article media.

Article Media¶

The articleMedia loader have one optional parameter:

(optional) key article - article Meta instance used for loading meta (if omitted then one available in context is used).

Simple usage:

{% gimmelist media from articleMedia %} <!-- It will use article from context -->
    <img src="{{ url(media) }}" />
{% endgimmelist %}

With optional parameter:

{% gimmelist media from articleMedia with {'article': gimme.article} %}
    <img src="{{ url(media) }}" />
{% endgimmelist %}

Note

Media Meta is handled by default by url and uri functions. It will return url for original image or file.

Image Renditions¶

If provided article media is an Image then it can have custom renditions. You can loop through renditions and display them.

Usage:

{% gimmelist media from articleMedia with {'article': gimme.article} %}
    {% if media.renditions is iterable %}
        {% for rendition in media.renditions %}
            <img src="{{ url(rendition) }}" style="width: {{ rendition.width }}px; height: {{ rendition.height }}px;" />
        {% endfor %}
    {% endif %}
{% endgimmelist %}

Get selected rendition only:

{% gimmelist media from articleMedia with {'article': gimme.article} %}
    {% gimme rendition with { 'name': '16-9', 'fallback': 'original' } %}
        <img src="{{ url(rendition) }}" style="width: {{ rendition.width }}px; height: {{ rendition.height }}px;" />
    {% endgimme %}
{% endgimmelist %}

Note

‘original’ is default feedback value for single rendition loader.

Feature Media¶

If Item comes with featuremedia association then Article will have this media set as featureMedia.

Usage:

{% if gimme.article.featureMedia.renditions is iterable %}
    {% for rendition in gimme.article.featureMedia.renditions %}
        <img src="{{ url(rendition) }}" style="width: {{ rendition.width }}px; height: {{ rendition.height }}px;" />
    {% endfor %}
{% endif %}

Or get selected rendition:

{% gimme rendition with { 'media': gimme.article.featureMedia, 'name': '16-9', 'fallback': 'original' } %}
    <img src="{{ url(rendition) }}" style="width: {{ rendition.width }}px; height: {{ rendition.height }}px;" />
{% endgimme %}

Handling Article Authors¶

Listing a Single Article’s Author¶

Usage:

<ul>
{% gimme author with { id: 1 } %}
    <li>{{ author.name }}</li> <!-- Author's name -->
    <li>{{ author.role }}</li> <!-- Author's name -->
    <li>{{ author.biography }}</li> <!-- Author's biography -->
    <li>{{ author.jobTitle.name }}</li> <!-- Author's job title name -->
    <li>{{ author.jobTitle.qcode }}</li> <!-- Author's job title code -->
    {% if author.avatar %}<li> <img src="{{ url(author.avatar) }}"><li>{% endif %} <!-- Author's avatar url. Check first if it's not null - author can be without avatar. -->
    <li>{{ author.facebook }}</li> <!-- Author's Facebook -->
    <li>{{ author.instagram }}</li> <!-- Author's Instagram -->
    <li>{{ author.twitter }}</li> <!-- Author's Twitter -->
{% endgimme %}
</ul>

Parameters:

{% gimme author with { id: 1 } %} {{ author.name }} {% endgimmelist %} - select author by it's id.

{% gimme author with { name: "Tom" } %} {{ author.name }} {% endgimmelist %} - select author by it's name.

Listing a collection of Article’s Authors¶

Usage:

<ul>
{% gimmelist author from authors with { role: ["writer"] } without {role: ["subeditor"]} %}
    <li>{{ author.name }}</li> <!-- Author's name -->
    <li>{{ author.role }}</li> <!-- Author's role -->
    <li>{{ author.biography }}</li> <!-- Author's biography -->
    <li>{{ author.jobTitle.name }}</li> <!-- Author's job title name -->
    <li>{{ author.jobTitle.qcode }}</li> <!-- Author's job title code -->
    {% if author.avatar %}<li> <img src="{{ url(author.avatar) }}"><li>{% endif %} <!-- Author's avatar url. Check first if it's not null - author can be without avatar. -->
    <li>{{ author.facebook }}</li> <!-- Author's Facebook -->
    <li>{{ author.instagram }}</li> <!-- Author's Instagram -->
    <li>{{ author.twitter }}</li> <!-- Author's Twitter -->
{% endgimmelist %}
</ul>

The above twig code, will render the list of articles where author’s role is writer and is not subeditor.

Filter authors by author’s name:

{% gimmelist author from authors with { name: ["Tom"] } %}
    {{ author.name }}
{% endgimmelist %}

Filter authors by author’s slug (automatically created from name. Example: John Doe -> john-doe):

{% gimmelist author from authors with { slug: ["john-doe"] } %}
    {{ author.name }}
{% endgimmelist %}

Filter authors by author’s name and role:

{% gimmelist author from authors with { role: ["Writer"], name: ["Tom"] } %}
    {{ author.name }}
{% endgimmelist %}

Filter authors by job title:

{% gimmelist author from authors with {jobtitle: {name: "quality check"}} %}
    {{ author.name }}
{% endgimmelist %}

{% gimmelist author from authors with {jobtitle: {qcode: "123"}} %}
    {{ author.name }}
{% endgimmelist %}

Handling Article Slideshows¶

Listing a Single Article’s Slideshow¶

Usage:

{% gimme slideshow with { name: "slideshow1" } %}
    {{ slideshow.code }} <!-- Slideshow's code -->
    {{ slideshow.createdAt|date('Y-m-d hh:mm') }} <!-- Slideshow's created at datetime -->
    {{ slideshow.updatedAt|date('Y-m-d hh:mm') }} <!-- Slideshow's updated at datetime-->
{% endgimme %}

or

{% gimme slideshow with { name: "slideshow1", article: gimme.article } %}
    {{ slideshow.code }} <!-- Slideshow's code -->
    {{ slideshow.createdAt|date('Y-m-d hh:mm') }} <!-- Slideshow's created at datetime -->
    {{ slideshow.updatedAt|date('Y-m-d hh:mm') }} <!-- Slideshow's updated at datetime-->
{% endgimme %}

Parameters:

{% gimme slideshow with { name: "slideshow1", article: gimme.article } %} {{ slideshow.code }} {% endgimme %} - select slideshow by it's code/name and current article.

If the article parameter is not provided, the slideshow will be loaded for the current article that is set in the context.

Listing a collection of Article’s Slideshows¶

Usage:

{% gimmelist slideshow from slideshows with { article: gimme.article } %}
    {{ slideshow.code }} <!-- Slideshow's code -->
    {{ slideshow.createdAt|date('Y-m-d hh:mm') }} <!-- Slideshow's created at datetime -->
    {{ slideshow.updatedAt|date('Y-m-d hh:mm') }} <!-- Slideshow's updated at datetime-->
{% endgimmelist %}

The above twig code will render the list of articles slideshows for the current article set in context.

Listing all Article’s Slideshows Items¶

Usage:

{% gimmelist slideshowItem from slideshowItems with { article: gimme.article } %}
    {% gimme rendition with {'media': slideshowItem.articleMedia, 'name': '770x515', 'fallback': 'original' } %}
        <img src="{{ url(rendition) }}" />
    {% endgimme %}
{% endgimmelist %}

The above twig code will render the list of articles slideshows for the current article set in context.

Or if there are audio, video, image files in slideshow:

{% gimmelist slideshow from slideshows with { article: gimme.article } %}
    <h2>{{ slideshow.code }}</h2>
    {% gimmelist slideshowItem from slideshowItems with { article: gimme.article, slideshow: slideshow } %}

        {% if slideshowItem.articleMedia.mimetype starts with 'image' %}
            {% gimme rendition with {'media': slideshowItem.articleMedia, 'name': '770x515', 'fallback': 'original' } %}
                <img src="{{ url(rendition) }}" />
            {% endgimme %}
        {% elseif slideshowItem.articleMedia.mimetype starts with 'audio' %}
            <audio src="{{ url(slideshowItem.articleMedia) }}" controls>
                <a href="{{ url(slideshowItem.articleMedia) }}">Download song</a>
            </audio>
        {% elseif slideshowItem.articleMedia.mimetype starts with 'video' %}
            <video src="{{ url(slideshowItem.articleMedia) }}" controls>
                <a href="{{ url(slideshowItem.articleMedia) }}">Download video</a>
            </video>
        {% endif %}

    {% endgimmelist %}
{% endgimmelist %}

Listing all Article’s Slideshows and its Items¶

Usage:

{% gimmelist slideshow from slideshows with { article: gimme.article } %}
    {{ slideshow.code }} <!-- Slideshow's code -->
    <!-- Slideshow items -->
    {% gimmelist slideshowItem from slideshowItems with { article: gimme.article, slideshow: slideshow } %}
        {% gimme rendition with {'media': slideshowItem.articleMedia, 'name': '770x515', 'fallback': 'original' } %}
            <img src="{{ url(rendition) }}" />
        {% endgimme %}
    {% endgimmelist %}
    {{ slideshow.createdAt|date('Y-m-d hh:mm') }} <!-- Slideshow's created at datetime -->
    {{ slideshow.updatedAt|date('Y-m-d hh:mm') }} <!-- Slideshow's updated at datetime-->
{% endgimmelist %}

The article parameter in gimmelist is optional. If not provided, it will load slideshows for current article.

Listing a Single Slideshow and its Items by Name¶

Usage:

{% gimmelist slideshow from slideshows with { article: gimme.article, name: "slideshow1" } %}
    {{ slideshow.code }} <!-- Slideshow's code -->
    <!-- Slideshow items -->
    {% gimmelist slideshowItem from slideshowItems with { article: gimme.article, slideshow: slideshow } %}
        {% gimme rendition with {'media': slideshowItem.articleMedia, 'name': '770x515', 'fallback': 'original' } %}
            <img src="{{ url(rendition) }}" />
        {% endgimme %}
    {% endgimmelist %}
    {{ slideshow.createdAt|date('Y-m-d hh:mm') }} <!-- Slideshow's created at datetime -->
    {{ slideshow.updatedAt|date('Y-m-d hh:mm') }} <!-- Slideshow's updated at datetime-->
{% endgimmelist %}

The article parameter in gimmelist is optional. If not provided, it will load slideshows for current article.

Handling Routes¶

Listing a Single Route¶

Usage:

<ul>
{% gimme route with { parent: 5, slug: 'test-route', name: 'Test Route'} %}
    <li>{{ route.name }}</li> <!-- Route's name -->
    <li>{{ route.slug }}</li> <!-- Route's slug -->
    <li>{{ url(route) }}</li> <!-- Route's url -->
{% endgimme %}
</ul>

parent - an id of parent route
slug - route’s slug
name - route’s name

Listing a Routes Collection¶

Usage:

<ul>
{% gimmelist route from routes %}
    <li>{{ route.name }}
{% endgimme %}
</ul>

<ul>
{% gimmelist route from routes with {parent: 5} %} <!-- possible values for parent: (int) 5, (string) 'Test Route', (meta) gimme.route -->
    <li>{{ route.name }}
{% endgimmelist %}
</ul>

Handling Content List Items¶

Listing Content List Items¶

Note

Content List can store many different content types (articles, events, packages).

Content List¶

Usage:

<ul>
{% gimme contentList with { contentListName: "List1" } %}
    <li>{{ contentList.name }}</li> <!-- Content list name -->
    <li>{{ contentList.description }}</li> <!-- Content list description -->
    <li>{{ contentList.type }}</li> <!-- Content list type) -->
{% endgimme %}
</ul>

Parameters:

{% gimme contentList with { contentListId: 1 } %} - select list by it's id.

{% gimme contentList with { contentListName: "List Name" } %} - select list by it's name.

Content List Items¶

Usage:

<ul>
{% gimmelist item from contentListItems with { contentListName: "List1" } %}
    <li>{{ item.content.title }}</li> <!-- Article title -->
    <li>{{ item.position }}</li> <!-- Item position in list -->
    <li>{{ item.sticky ? "pinned" : "not pinned" }}</li> <!-- Checks if item is sticky (positioned on top of list) -->
{% endgimmelist %}
</ul>

or

{% gimme contentList with { contentListName: "List1"} %}
    {% cache 'top-articles' {gen: contentList} %}
    <ul>
    {% gimmelist item from contentListItems with { contentList: contentList } %}
        <li>{{ item.content.title }}</li> <!-- Article title -->
        <li>{{ item.position }}</li> <!-- Item position in list -->
        <li>{{ item.sticky ? "pinned" : "not pinned" }}</li> <!-- Checks if item is sticky (positioned on top of list) -->
    {% endgimmelist %}
    </ul>
    {% endcache %}
{% endgimme %}

Note

Passing previously fetched contentList (for cache key generation needs) is good for performance.

Parameters:

{% gimmelist item from contentListItems with { contentListId: 1 } %} - select list by it's Id.

{% gimmelist item from contentListItems with { contentListId: 1, sticky: true } %} - filter by sticky value.

{% gimmelist item from contentListItems with { contentListId: 1, sticky: true } without {content: [1]} %} - exclude article with id 1 (in case of articles as items).

Keywords¶

List a Single Keyword¶

Usage:

<ul>
{% gimme keyword with { slug: 'big-city' } %}
    <li>{{ keyword.name }}</li> <!-- Keyword's name -->
    <li>{{ keyword.slug }}</li> <!-- Keyword's slug -->
{% endgimme %}
</ul>

Template Caching¶

For now we support just template block caching with the cache block.

The Cache block is simple, and accepts only two parameters: cache key and strategy object (with strategy key and value).

Note

Cache blocks can be nested:

{% cache 'v1' {time: 900} %}
    {% for item in items %}
        {% cache 'v1' {gen: item} %}
            {# ... #}
        {% endcache %}
    {% endfor %}
{% endcache %}

The annotation can also be an expression:

{% set version = 42 %}
{% cache 'hello_v' ~ version {time: 300} %}
    Hello {{ name }}!
{% endcache %}

There is no need to invalidate keys - the system will clear unused cache entries automatically.

Strategies¶

There are two available cache strategies: lifetime and generational.

With lifetime as a strategy key you need to provide time with a value in seconds.

{# delegate to lifetime strategy #}
{% cache 'v1/summary' {time: 300} %}
    {# heavy lifting template stuff here, include/render other partials etc #}
{% endcache %}

With generational as a strategy key you need to provide gen with object or array as the value.

{# delegate to generational strategy #}
{% cache 'v1/summary' {gen: gimme.article} %}
    {# heavy lifting template stuff here, include/render other partials etc #}
{% endcache %}

Note

You can pass Meta object to generational strategy and it will be used for key generation. If Meta value have created_at or updated_at then those properties will be used, otherwise key will be generated only from object id.

Content list blocks caching¶

It’s important to always use generational strategy for content lists (and it items) caching. Publisher will update cache key generated with it every time when items on list are added/removed/reordered or when list criteria are updated or even when article used by list will be unpublished or updated.

{% cache 'frontPageManualList' {gen: contentList} %}
    {# get and render list items here #}
{% endcache %}

How to implement Search using ElasticSearch?¶

To make use of features (e.g. full-text search) provided by ElasticSearch and it’s extension/plugin ElasticSearchBundle created specifically for Superdesk Publisher, you need to do the following steps.

Create a new Route¶

You can create a new route using admin interface as described in Create new route with extension (example: feed/sitemap.rss) section. In this example the created route under which we will place search will be named: search.

Create a template file¶

Example search template which loads search and its results when filtered by criteria:

# ../view/search.html.twig
<form name="filter" method="get">
    <input type="search" id="filter_search" name="q">
</form>

{% set itemsPerPage, currentPage = 8, app.request.get('page', 1) %}
{% set start = ((currentPage - 1) * itemsPerPage) %}

{% gimmelist article from searchResults|limit(app.request.get('limit', 10))|order(app.request.get('field', 'publishedAt'), app.request.get('direction', 'desc')) with {
    term: app.request.get('q', ''),
    page: app.request.get('page', 1),
    routes: app.request.get('route', []),
    term: app.request.get('q', ''),
    publishedBefore: app.request.get('publishedBefore'),
    publishedAfter: app.request.get('publishedAfter'),
    publishedAt: app.request.get('publishedAt'),
    sources: app.request.get('source', []),
    authors: app.request.get('author', []),
    statuses: app.request.get('status', []),
    metadata: app.request.get('metadata', []),
    keywords: app.request.get('keywords', []),
} %}
    <h4>{{ article.title }}</h4>
    <p>{{ article.lead }}</p>

{% if loop.last  %}
    {% include '_tpl/pagination.html.twig' with {
    currentFilters: {}|merge(app.request.query.all()),
    currentPage: currentPage,
    paginationPath: gimme.route,
    lastPage: (loop.totalLength/itemsPerPage)|round(0, 'ceil')
    } only %}

    Showing {{ searchResults|length }} out of {{ loop.totalLength }} articles.
{% endif %}
{% endgimmelist %}

<a href="search?route[]=50">Business</a>
<a href="search?route[]=49">Politics</a>

Alternatively, to built-in order function, you can also use sort: app.request.get('sort', []), parameter to sort by different fields and directions which needs to be passed directly to the with statement.

Available search criteria:¶

Based on the above template.

Criteria name	Description	Example	Format
sort	Sorting	sort[publishedAt]=desc	sort[<field>]=<direction>
page	Pagination	page=1	page=<page_number>
limit	Items per page	limit=10	limit=<limit>
routes	An array of routes ids	route[]=10&route[]=12	route[]=<routeId>&route[]=<routeId>
q	Search query	q=Lorem ipsum	q=<search_term>
publishedBefore	Published before date time	publishedBefore=1996-10-15T00:00:00	publishedBefore=<datetime>
publishedAfter	Published before date time	publishedBefore=1996-10-15T00:00:00	publishedAfter=<datetime>
sources	Sources of articles	source[]=APP&source[]=NTB	source[]=<source>&source[]=<source>
authors	An array of authors	author[]=Joe&author[]=Doe	author[]=<auth1>&author[]=<auth2>
statuses	An array of statues	status[]=new&status[]=published	status[]=new&status[]=published
metadata	An array metadata	metadata[located]=Sydney	metadata[<field>]=<value>
keywords	An array of keywords	keywords[]=Joe&keywords[]=Doe	keywords[]=<key1>&keywords[]=<key2>

Tips¶

Templates inheritance¶

Default template name for route and articles in Publisher is article.html.twig.

Inheritance overview:

> article.html.twig
    > Route custom template
        > Article custom template

If route is collection type then it can have declared two default templates:

default_template used for rendering Route content (eg. /sport).

default_articles_template used for rendering content attached to this route (eg. /sport/usain-bolt-fastest-man-in-theo-world).

Note

When route default_template property is set but not default_articles_template, then Web Publisher will load all articles attached to this route with template chosen in default_template (not with article.html.twig).

If content have assigned custom template then it will always override other already set templates.

How to change the Route/Article template name?¶

You can change default template name values for article and route either in routes management, or with API calls (providing it on resource create or resource update calls).

Example resource update calls:

article:¶

curl -X "PATCH" -d "article[template_name]=new_template_name.html.twig" -H "Content-type:\ application/x-www-form-urlencoded" /api/v1/content/articles/features

route:¶

curl -X "PATCH" -d "route[template_name]=custom_route_template.html.twig" -H "Content-type:\ application/x-www-form-urlencoded" /api/v1/content/routes/news

Themes¶

About themes and multitenancy¶

Themes provide templates for the customization of the appearance and functionality of your public-facing websites. Themes can also be translated to support multiple written languages and regional dialects.

There are two Superdesk Web Publisher themes systems; default one is built on top of fast, flexible and easy to use Twig templates. Alternativelly, PWA Next.js (React/Node.js) Superdesk Publisher renderer can be used.

Twig - Publisher’s default theme system¶

By default, themes are located under the app/themes directory. A basic theme must have the following structure:

ExampleTheme/               <=== Theme starts here
    views/                  <=== Views directory
        home.html.twig
    translations/           <=== Translations directory
        messages.en.xlf
        messages.de.xlf
    screenshots/                <=== Theme screenshots
        front.jpg
    public/                 <=== Assets directory
        css/
        js/
        images/
    theme.json              <=== Theme configuration

Publisher does not support the option of Sylius theme structure to have bundle resources nested inside a theme.

Progressive Web App (PWA) Themes¶

PWA theme, on the other hand, is built as Hybrid app - one React app on both server and client side. It is built on modern and highly optimised code which ensures lightning fast performance.

Our PWA solution is Server Side Generated (SSG, not SSR - server side rendered) and Client Side Rendered (CSR, React) - on build, app renders pages to HTML and JSON. It refreshes these files during runtime on defined schedule. The end users ALWAYS get a static file - either HTML (on initial load) or JSON (when navigating between pages), with data needed to render given page on client side.

Beside standard front - section - article page functionality, and tag - author - search pages, default Publisher’s PWA theme also includes:

Responsiveness - fits any form factor: desktop, mobile, tablet, or whatever is next. It makes a project available to more people on more devices with wildly varying operating systems, browser capabilities, system APIs, and screen sizes. It ensures that websites work on any device that can access the web, regardless of a browser.
app-like experiences which users enjoy using. Also, it allow users to add the app to their home screen. With the option to install websites, users are offered the ability to install PWA and easily access it on their home screens without the hassle of an app store.
integration of Web Vitals recording into Google Analytics (that way one gets real data from users about page speed and other measurements that can be then visualised in Analytics using custom dashboard or online app. (More about web vitals)
Publisher Analytics: app reports views back to publisher endpoint
Static/Dynamic sitemaps and sitemap-news 0.9
Installable as an app on mobiles and even on desktop Chrome
Possibility of offline usage, thanks to service workers and manifest.json
AMP support out of the box
Re-engagement - PWAs feature Push Notifications is used for promotions and specials, as those updates can be displayed to the users even if they don’t have the PWA installed or a browser tab open to the website.
Sentry integration
User Login/Register

Superdesk Publisher can serve many websites from one instance, and each website tenant can have multiple themes. The active theme for the tenant can be selected with a local settings file, or by API.

If only one tenant is used, which is the default, the theme should be placed under the app/themes/default directory, (e.g. app/themes/default/ExampleTheme).

If there were another tenant configured, for example client1, the files for one of this tenant’s themes could be placed under the app/themes/client1/ExampleTheme directory.

Superdesk Publisher’s default theme system Twig provides an easy way to create device-specific templates. This means you only need to put the elements in a particular template which are going to be used on the target device.

The supported device types are: desktop, phone, tablet, plain.

A theme with device-specific templates could be structured like this:

ExampleTheme/                   <=== Theme starts here
    phone                       <=== Views used on phones
        views/                  <=== Views directory
            home.html.twig
    tablet                      <=== Views used on tablets
        views/                  <=== Views directory
            home.html.twig
    views/                      <=== Default templates directory
        home.html.twig
    translations/               <=== Translations directory
        messages.en.xlf
        messages.de.xlf
    screenshots/                <=== Theme screenshots
        front.jpg
    public/                     <=== Assets directory
        css/
        js/
        images/
    theme.json                  <=== Theme configuration

Note

If a device is not recognized by the Publisher, it will fall back to the desktop type. If there is no desktop directory with the required template file, the locator will try to load the template from the root level views directory.

More details about theme structure and configuration can be found in the Sylius Theme Bundle documentation.

Create or install a theme¶

To install theme assets you need to run swp:theme:install command.

The swp:theme:install command installs your custom theme for given tenant:

    bin/console swp:theme:install <tenant> <theme_dir>

You need specify the directory (theme_dir) argument to install
theme from any directory:

    bin/console swp:theme:install <tenant> /dir/to/theme

Once executed, it will create directory app/themes/<tenant>
where <tenant> is the tenant code you typed in the first argument.

To force an action, you need to add an option: --force:

    bin/console swp:theme:install <tenant> <theme_dir> --force

To activate this theme in tenant, you need to add and option --activate:
    bin/console swp:theme:install <tenant> <theme_dir> --activate

If option --processGeneratedData will be passed theme installator will
generate declared in theme config elements like: routes, articles, menus and content lists

Work with theme assets¶

To install theme assets you need to run sylius:theme:assets:install command.

Theme assets (JavaScript, CSS etc. files) should be placed inside the theme directory. There are few ways of reading theme assets in your Twig templates. The below how-to describes where to place the assets, how to install it and use it.

Load assets from the theme’s public directory (`app/themes/<theme-name>/public`)¶

Put the example.css asset file inside <theme-name>/public/css/ directory.
Install assets by running command: php bin/console sylius:theme:assets:install.
Make use of the asset file in twig templates:

<!-- loads test.css file directly /public/css/ in theme directory -->
<link rel="stylesheet" href="{{ asset('theme/css/example.css') }}" />

Load assets from the public `web` directory¶

Put the example.css asset file directly inside web directory.
Make use of the asset file in twig templates:

<!-- loads asset file directly from `web` dir (`web/example.css`) -->
<link rel="stylesheet" href="{{ asset('example.css') }}" />

Generate simple links for current theme assets¶

If You need to get link to asset from outside of twig template then you can use this url:

/public/{filePath}

ex. <link rel="stylesheet" href="/public/css/example.css" />

Where {filePath} is path for your file from public directory inside theme.

Load Service Worker files (from domain root level)¶

If You want to use service worker or manifest file (it must be placed in root level) then you can use this url:

/{fileName}.{fileExtension}

ex. <link rel="manifest" href="/manifest.json">

Where {fileName} can be only sw or manifest.

Load bundles’ assets¶

Install Symfony assets by running command: php bin/console assets:install.
Make use of the asset file in twig templates:

<!-- loads bundle's asset file from bundles dir -->
<link rel="stylesheet" href="{{ asset('bundles/framework/css/body.css') }}" />

Override bundles’ assets from the theme¶

There is a possibility to override bundle specific assets. For example, you have AcmeDemoBundle registered in your project. Let’s assume there is a body.css file placed inside this bundle (Resources/public/css/body.css). To override body.css file from your theme, you need to place your new body.css file inside app/themes/<theme-name>/AcmeDemoBundle/public directory:

Put the body.css asset file inside app/themes/<theme-name>/AcmeDemoBundle/public directory.
Install assets by running command: php bin/console sylius:theme:assets:install.
Make use of the asset file in twig templates:

<link rel="stylesheet" href="{{ asset('theme/acmedemo/css/body.css') }}" />

Note

theme prefix in {{ asset('theme/css/example.css') }} indicates that the asset refers to current theme.

Translations¶

The Symfony Translation component supports a variety of file formats for translation files, but in accordance with best practices suggested in the Symfony documentation, the XLIFF file format is preferred. JMSTranslationBundle has been added to the project to facilitate the creation and updating of such files.

The use of abstract keys such as index.welcome.title is preferred, with an accompanying description desc in English to inform a translator what needs to be translated. This description could simply be the English text which is to be displayed, but additional information about context could be provided to help a translator.

Abstract keys are used for two main reasons:

Translation messages are mostly written by developers, and changes might be necessitated later. These changes would then result in changes for all supported languages instead of only for the source language, and some translations might be lost in the process.

Some words in English are spelled differently in other languages, depending on their meaning, so providing context is important.

Here is an example of the preferred syntax in twig templates:

{{ 'index.welcome.title'|trans|desc('Welcome to Default Theme!') }}

Translation labels added to Twig and php files can be extracted and added to XLIFF files using a console command bin/console translation:extract. This command can be used to create or update a XLIFF file in the locale en for the DefaultTheme of the FixturesBundle:

bin/console translation:extract en --dir=./src/SWP/Bundle/FixturesBundle/Resources/themes/DefaultTheme/ --output-dir=./src/SWP/Bundle/FixturesBundle/Resources/themes/DefaultTheme/translations

This will create or update a XLIFF file in English called messages.en.xlf, which can be used with a translation tool.

AMP HTML Integration¶

Google AMP HTML integration comes with Superdesk Publisher out of the box. This integration gives you a lot of features provided by Google. To name a few: fast loading time and accessibility via Google engines etc. There is no need to install any dependencies, all you need to do is to create AMP HTML compatible theme or use the default one provided by us.

Default AMP HTML theme is bundled in our main Demo Theme and can be installed using php bin/console swp:theme:install command.

You could also copy it to your own main theme and adjust it in a way you wish.

Note

See Setting up a demo theme section for more details on how to install demo theme.

How to create AMP HTML theme?¶

You can find more info about it in AMP HTML official documentation.

Where to upload AMP HTML theme?¶

Publisher expects to load AMP HTML theme from main theme directory which is app/themes/<tenant_code>/<theme_name>. AMP HTML theme should be placed in app/themes/<tenant_code>/<theme_name>/amp/amp-theme folder. index.html.twig is the starting template for that theme. If that template doesn’t exist, theme won’t be loaded. Once the theme is placed in a proper directory, it will be automatically loaded.

To test if the theme has been loaded properly you can access your article at e.g.: https://example.com/news/my-articles?amp.

Linking AMP page and non-AMP page¶

To add a link to AMP page from article template in the form of <link> tags (which is required by AMP HTML integration for discovery and distribution), you can use amp Twig filter:

{# app/themes/<tenant_code>/<theme_name>/views/article.html.twig #}
<link rel="amphtml" href="{{ url(gimme.article)|amp }}"> {# https://example.com/news/my-articles?amp #}

And from AMP page:

{# app/themes/<tenant_code>/<theme_name>/amp/amp-theme/index.html.twig #}
<link rel="canonical" href="{{ url(gimme.article) }}"> {# https://example.com/news/my-articles #}

Theme Settings¶

Theme settings are defined in the theme.json configuration file which should be in every theme directory.

An example of theme.json file with defined settings will look like:

{
    "name": "swp/default-theme",
    "title": "Default Theme",
    "description": "Superdesk Publisher default theme",
    "authors": [
        {
            "name": "Sourcefabric z.ú.",
            "email": "contact@sourcefabric.org",
            "homepage": "https://www.sourcefabric.org",
            "role": "Organization"
        }
    ],
    "settings": {
        "primary_font_family": {
            "label": "Primary Font Family",
            "value": "Roboto",
            "type": "string",
            "help": "The primary font",
            "options": [
                {"value": "Roboto", "label": "Roboto"},
                {"value": "Lato", "label": "Lato"},
                {"value": "Oswald", "label": "Oswald"}
            ]
        },
        "secondary_font_family": {
            "value": "Roboto",
            "type": "string",
            "options": [
                {"value": "Roboto", "label": "Roboto"},
                {"value": "Lato", "label": "Lato"},
                {"value": "Oswald", "label": "Oswald"}
            ]
        },
        "body_font_size": {
            "label": "Body Font Size",
            "value": 14,
            "type": "integer",
            "options": [
                {"value": 14, "label": "14px"},
                {"value": 16, "label": "16px"},
                {"value": 18, "label": "18px"}
            ]
        }
    }
}

In the settings property of the JSON file you will find the default theme’s settings.

Each setting can be overridden through the Publisher settings interface or by the API (for this, see /settings/ API endpoint for more details in the /api/doc route in your Superdesk Publisher instance).

Read more about settings in the Settings chapter to find out more.

Every setting is a JSON object which may contain the following properties:

label - Setting’s label, will be visible in API when defined,
value - Setting’s value, will be visible in API when defined,
type - Setting’s type, either it’s string, integer, boolean or array.
help - Settins’s helper text.
options - an array of optional values that can be used to implement select box.

Read more about theme structure in the Themes chapter.

How to display the current theme settings in templates?¶

{# app/themes/<tenant_code>/<theme_name>/views/index.html.twig #}
{{ themeSetting('primary_font_family') }} # will print "Roboto"

In development environment, if the theme’s setting doesn’t exists an exception will be thrown with a proper message that it does not exist. In production environment no exception will be thrown, the page will render normally.

How to work with theme settings in the GUI¶

In Publisher’s Website management, after selecting your desired tenant (if there is more than one), the last ‘tab’ in horizontal navigation will be ‘Theme customization’. This is a graphical representation of theme.json - all the fields and default settings you set there, are visible on this screen, and can be updated.

Based on these custom values, there are dialogues to manage up to three logos. It is meant to be be used for quick adjustments of example themes (header and footer logos, or maybe logo on inner pages if it is somehow different than the main one), but these custom upload files can be incorporated into the site in other ways; for example they can be used for graphical announcements / banners that are changed by site editors from time to time.

How to display current theme’s settings using an API?¶

Theme settings can be accessed by calling an /theme/settings/ API endpoint using GET method.

How to update current theme settings using an API?¶

To update theme settings using an API, a PATCH request must be submitted to the /settings/ endpoint with the JSON payload:

{
    "settings": {
        "name": "primary_font_family",
        "value": "custom font"
    }
}

How to restore current theme settings using an API?¶

There is an option to restore the current theme settings to the default ones, defined in the theme.json file.

This can be done using the API and calling a /settings/revert/{scope} endpoint using the POST method. The scope parameter should be set to theme in order to restore settings for the current theme.

Example themes¶

Superdesk Publisher demo theme¶

Theme location: /src/SWP/Bundle/FixturesBundle/Resources/themes/DefaultTheme

Publisher Mag ships within Superdesk Publisher Release, thus can be considered the default and most basic Publisher theme. It serves purpose of showing most common features of the software, such as listing articles, showing article elements and full content, working with menus etc.

To create richer user experience, 3rd-party services can be incorporated. In Publisher Mag theme we showcase it with Disqus article comments.

The Modern Times theme¶

Theme repo: https://github.com/SuperdeskWebPublisher/theme-dailyNews

The Modern Times theme is fresh, fully responsive, highly adaptible Superdesk Publisher theme built primarily to serve those media operations with high daily news production. It offers editors flexibility in ways they can present sets of news (category or tag based; manually curated, fully- or semi-automated content lists; and more).

In this theme we also showcase how 3rd-party services can be incorporated for reacher user experience (Open weather data integration for any weather station in the world, Disqus article comments, Playbuzz voting poll, Google Custom Search Engine).

Magazine theme¶

Theme repo: https://github.com/SuperdeskWebPublisher/theme-magazine

Magazine theme is fresh, fully responsive, simple and ultra fast Superdesk Publisher theme built to serve those media operations that are not primarily focused on daily content production, but on fewer stories per day/week that have longer time span. Naturally this applies to traditional weekly, be-weekly or even monthly type of magazines from the print world.

Magazine theme features customizable menu, html and content list widgets which enable live-site editing from frontend.

To create richer user experience, 3rd-party services can be incorporated. In Magazine theme we showcase it with Disqus article comments.

PWA theme¶

PWA theme is built as Hybrid app - one React app on both server and client side. It is built on modern and highly optimised code which ensures lightning fast performance.

Our PWA solution is Server Side Generated (SSG, not SSR - server side rendered) and Client Side Rendered (CSR, React) - on build, app renders pages to HTML and JSON. It refreshes these files during runtime on defined schedule. The end users ALWAYS get a static file - either HTML (on initial load) or JSON (when navigating between pages), with data needed to render given page on client side.

Beside standard front - section - article page functionality, and tag - author - search pages, default Publisher’s PWA theme also includes:

Responsiveness - fits any form factor: desktop, mobile, tablet, or whatever is next. It makes a project available to more people on more devices with wildly varying operating systems, browser capabilities, system APIs, and screen sizes. It ensures that websites work on any device that can access the web, regardless of a browser.
app-like experiences which users enjoy using. Also, it allow users to add the app to their home screen. With the option to install websites, users are offered the ability to install PWA and easily access it on their home screens without the hassle of an app store.
integration of Web Vitals recording into Google Analytics (that way one gets real data from users about page speed and other measurements that can be then visualised in Analytics using custom dashboard or online app. (More about web vitals)
Publisher Analytics: app reports views back to publisher endpoint
Static/Dynamic sitemaps and sitemap-news 0.9
Installable as an app on mobiles and even on desktop Chrome
Possibility of offline usage, thanks to service workers and manifest.json
AMP support out of the box
Re-engagement - PWAs feature Push Notifications is used for promotions and specials, as those updates can be displayed to the users even if they don’t have the PWA installed or a browser tab open to the website.
Sentry integration
User Login/Register

Editorial tools¶

Content Lists¶

What playlists are to music, content lists are to articles - customizable sets of content organized and ordered in the way that suits your output channel best. For example, when you access a Publisher website homepage as a visitor, chances are that the teasers for articles you see are curated by website editors using Content lists. The order of articles can be easily updated as news arrives, new articles are added, some others removed etc.

Manual Content Lists¶

With manual content lists, articles need to be drag-and-dropped into the list manually. This gives complete freedom to editors to put which stories where. A list can be (and should be) limited in length, so when a new article is added on top, the last one drops out.

Automatic Content Lists¶

If a list doesn’t need to be fully customizable, or if it can be programmed with certain rules to collect appropriate articles, then the automatic content list steps in.

For example, a block on the webiste shows the most recent articles from the Sport section, where the metadata location is set to ‘Europe’ and author is not John Smith; in this situation, obviously editors don’t need to put in manual labour, but can rather use an automated set of rules to add articles to the automatic list.

Built in criteria:

route - an array of route ids, e.g. [1,5]
author - an array of authors, e.g. [“Test Persona”,”Doe”]
publishedBefore - date string, articles published before this date will be added to the list,

e.g. date: “2021-01-20”. (date format must be YYYY-MM-DD)

publishedAfter - date string, articles published after that date will be added to the list, format is the same as in the publishedBefore case.
publishedAt - date string, when defined articles matching this publish date are added to the list when published, the format is the same as in case of publishedBefore and publishedAfter
metadata - metadata is the dropdown field where filtering can be done by any of an article’s metadata: Categories, Genre, Urgency, Priority, Keywords, and any custom vocabulary that is set as a field in a content profile in Superdesk

All criteria can be combined together with that the result it will add articles to the list (on publish) depending on your needs.

Admin Interface¶

Publisher Settings¶

Publisher is built to work with any News API feed and thus it can be used as a frontend solution for, say, Wordpress-created content, a custom CMS solution, or just a feed that you subscribe to with any provider.

But if you use Publisher along with Superdesk, then you have a Web Site Management GUI that integrates seamlessly into the Superdesk environment. From there, you can manage one or many websites (or, more generally, output channels) with as many tenants as you want.

Website Management¶

In our example (screenshot 1) you can see several websites configured. As you can see, each site is represented with its name (which links to the front end) and list of available settings.

Options to manage or delete websites are shown after clicking on the three-dots icon.

Next screenshot (2) shows initial step in managing a single website - set its name and (sub)domain, language(s) and other options.

Definition of site routes is the next step (3). Routes can be of type collection, content or custom - the first is a category-like route to which articles are later attached, second is an end - with specific content assigned, while the third is for creating dynamic routes based on author name, tag or other dynamic value.

As you can see, the route definition consists of name, type (collection, content or custom), eventual parent route, template used to show this route, and article template to open articles attached to this route (if route is of type collection). There is also a switch called ‘Paywall secured’ which can be used to indicate locked content (and thus open it only to, say, logged-in users).

The third step in managing a website is to define its navigation. In other words, this option can be used to make navigation menus for a header, footer, sidebar or wherever. If created this way, the menu can be later also managed through LiveSite editing (of course, menus can also be defined in templates but then they are not dynamic in a way that can be managed by website editors).

Next option is to control site redirects (4), which can be route to route, or custom url redirections.

In navigations (5), it is possible to create dynamic site menus, so they can be changed/adjusted by site editors (and not technical people exclusively).

Navigation menu consists of menu items that can be either route-based (Politics, Business etc), or leading to a custom url. Each of these menu items is defined by name, label, parent, route (pre-defined in previous step) and/or uri.

Webhooks

Webhooks are HTTP callbacks (with associated URLs) defined by the user. Some events trigger them. Superdesk Publisher uses webhooks to notify external applications when an event happens in Publisher. Examples of these events include: when an editor publishes/unpublishes/updates an article; when a website administrator creates a new route or creates a menu.

The next step in site management offers possibilites to choose or change themes (8).

You can either upload your custom theme or choose one of available themes from the list. The current theme is highlighted in green.

If a selected theme supports Theme Settings you can also customise your theme with GUI.

Publishing rules¶

In Publisher, it is possible to automate content publishing, by setting rules that, when triggered, puts article online (thus publishing it to the tenant, and on the predefined route).

Triggering mechanism for publishing rules can be simple (single expression) or complex (multiple expressions - in which case they behave in the AND logical manner).

There two levels of rules - organization rules and tenant rules.

Organization rule is top-level rule which is meant to catch artilces on tenant level. There is even the switch _catch all_ which will publish all incoming content to specified tenant (keep in mind, though, that it is yet not enough for an article to get online! for that, route needs to be assigned too).

Tenant rule can be created only after defining organization rule for that tenant. It is logical - first we need to dirfect incoming article by organization rule, then specify under which conditions it gets assigned to specific route. For example, if article’s metadata ‘category’ is set to ‘Business’, then publish that article to the route ‘business’.

Publisher dashboard¶

This screen is the starting point for working with content that arrives in Publisher (from Superdesk, some other CMS or even from an ingest source).

By default, the content list manager is shown. Here, every site that is configured (through Publisher settings) is represented by the list of its content lists.

Content lists¶

Publisher offers the option to create manual or automatic content lists.

With automatic content lists, it is necessary to set up rules which will be used to fill the list with articles. It can be a single rule or a combination of several rules - route, author, date-based options, or one of many metadata values that are part of an article (for example: topic, category etc)

Setting criteria for automated content list

As you can see, setting automated list criteria is a straightforward process where you can define as many rules as you need.

When working with manual content lists, editors need to drag and drop stories from the right pane (showing all published articles) to the left pane. The order of articles here is the order in which they will appear on the front page.

Although the newest articles are on top of the listing, it is also possible to narrow down the number of articles shown in the right pane by using filtering options (top left button in that pane), or simply typing a keyword.

Output control¶

Output control displays content that comes to the Publisher instance, and is divided into two listings - incoming and published articles.

The Incoming list shows articles that have arrived in Publisher, but are not yet visible online. In order to get them online, they need to be assigned to a tenant, and route on which the article will be attached.

Also, it is possible to filter the list by keyword or by criteria.

The Published list is for manipulations with published content. There, one can change the route to which an article is attached, unpublish an article, publish it to additional tenants etc. This list can also be filtered to narrow down the search.

Error log¶

As the name suggests, this screen is for listing potential errors that appear during the process of publishing and manipulating articles. It is meant to be informative for the tech part of the team.

Cookbooks¶

Templates Cookbooks¶

Users registration and login¶

Registration¶

User can be registered in Publisher with REST API /{version}/users/register/ POST request.

curl -X POST 'http://webpublisher.dev/api/v1/users/' -H 'Origin: http://webpublisher.dev' -H 'Content-Type: application/x-www-form-urlencoded' -H 'Accept: */*' -H 'Connection: keep-alive' -H 'DNT: 1' --data '_format=json&user_registration%5Bemail%5D=pawel.mikolajczuk%40sourcefabric.org&user_registration%5Busername%5D=pawel.mikolajczuk&user_registration%5BplainPassword%5D%5Bfirst%5D=superStronP%40SSword&user_registration%5BplainPassword%5D%5Bsecond%5D=superStronP%40SSword' --compressed

After that user will get email message with link for account confirmation. Link will redirect him to /register/confirmed page.

Customize sender email address:¶

By default email will be sent from 'contact@{tenant domain} - example: contact@example.com. You can override it by customizing registration_from_email.confirmation setting.

Customize confirmation email template:¶

Default template used for confirmation is @FOSUser/Registration/email.txt.twig You can override it by customizing registration_from_email.confirmation setting.

Customize account confirmation page template:¶

After clicking on conformation link (from email) user will be redirected to /register/confirmed url. To render this page publisher by default use '@FOSUser/Registration/confirmed.html.twig.

You can override it in Your theme (with creating FOSUser/Registration/confirmed.html.twig file in your theme.

Note

Read more about settings in SettingsBundle.

Login¶

Publisher don’t provide single page for login action, instead that we made sure that login can be placed in any template (or even widget) of You choice. Only hardcoded url’s for security are /security/login_check (used for user authentication - you need to send your login form data there) and /security/logout (used for logging out user).

Example login form:¶

{% if app.session.has('_security.last_error') %}
    {# show error message after unsuccessful login attempt #}
    {% set error = app.session.get('_security.last_error') %}
    <div>{{ error.messageKey|trans(error.messageData, 'security') }}</div>
{% endif %}
<form action="{{ path('security_login_check') }}" method="POST">
    <input type="text" name="_username" value="{{ app.session.get('_security.last_username') }}" />
    <input type="password" name="_password" value="" />
    <input type="hidden" name="_login_success_path" value="{{ url('homepage') }}">
    <input type="hidden" name="_login_failure_path" value="{{ url('homepage') }}">
    <input type="submit" value="Login" />
</form>

Parameters explanation:

_username - User username parameter (You can use {{ app.session.get('_security.last_username') }} as a default value - it will be filled with previously entered username when login will fail and user will be redirected to login form again).

_password - User password

_login_success_path - url used for redirection after successful login

_login_failure_path - url used for redirection after unsuccessful login (use {{ url(gimme.page) }} to generate url for current page)

Check if users is logged in:¶

{% if app.user %}
    Hey {{ app.user.username }}. <a href="{{ url('security_logout') }}">Logout</a>.
{% else %}
    {# Show login form #}
{% endif %}

Password Reset¶

By default, there are a few routes exposed for password reset functionality (provided by FOSUserBundle).

When going to route /resetting/request (route name is fos_user_resetting_request) the default password reset form will be rendered.

You can override this template in your theme by creating FOSUserBundle/views/Resetting/request.html.twig file in your theme.

All templates related to password reset functionality which can be overridden can be found here.

Create new route with extension (example: `feed/sitemap.rss`)¶

Very common case for newspaper website is providing some custom rss feeds. To accomplish that with Publisher you need to create new route (ex: feed/sitemap) and attach selected template to it.

After hitting this route you will see output generated by template but you will find out that Content Type of this route response is set to text/html - and it can create issues for some parsers.

To fix that problem you just need to rename your route to one with extension at end. For example feed/sitemap.rss. Publisher will automatically recognize route extension and will set proper Content-Type header - in this case: application/rss+xml.

This feature can be used for generating custom xml files, or even JavaScript files (with extension .js).

Create new custom route¶

Our use case will be article authors pages. As we don’t want to have special route for every author - we need to create one witch will fit all our needs.

We need url like that: publisher.dev/authors/{authorSlug}. Author slug is a parameter passed to template and will be used by us for loading desired author. Example filled route: publisher.dev/authors/john-doe.

We can create route like that with API:

POST /api/v1/content/routes/

{
    "name": "Authors",
    "slug": "authors",
    "type": "custom",
    "templateName": "author.html.twig",
    "variablePattern": "/{authorSlug}",
    "requirements": [
        {
            "key": "authorSlug",
            "value": "[a-zA-Z\\-_]+"
        }
    ]
}

Important parts:

type custom: says to publisher that we want to set variablePattern and requirements manually.
variablePattern - string with set parameters placeholders added to end of route.
requirements = array of objects with regular expression for provided parameters.

Now in template author.html.twig we can load author by slug provided in url.

{% gimme author with { slug: app.request.attributes.get('authorSlug') } %}
    Author name: {{ author.name }}
{% endgimme %}

And done - you have custom route with option to pass parameters in url and use them later in template.

Most popular (read) articles list¶

Article meta have property articleStatistics and inside it You can find pageViewsNumber. To simplify syntax for ordering we created pageViews alias.

Here is example how to order articles by their page views number:

{% gimmelist article from articles|order('pageViews', 'desc') %} <!-- ordering by page views -->
    <a href="{{ url(article) }}">{{ article.title }}</a>
{% endgimmelist %}

There is also option to get most popular (ordered by page views) articles from date range.

For example this is how You can list yesterday most popular articles:

{% gimmelist article from articles|order('pageViews', 'desc')|dateRange('now', '-1 day') %}
    <a href="{{ url(article) }}">{{ article.title }}</a>
{% endgimmelist %}

Filter dateRange takes two parameters compatible with PHP strtotime syntax (http://php.net/manual/en/function.strtotime.php):

start date (<= equal or in past ) - time will be reset to 23:59:59

end date (>= equal or in feature) - time will be reset to 00:00:00

So |dateRange('now', '-1 day') will filter all page views from whole day today and whole yesterday (from midnight)

To activate article page views counting you need to call short twig function.

{{ countPageView(gimme.article) }}

It will print <script> tag with some generated by Publisher javascript code inside.

Elements generated on theme installation¶

To provide better experience for end users after theme installation Publisher can create some default content used by theme. Thanks to this feature theme developer can be sure that all of his article or category templates will be visible without complicated configuration by end user.

Generated elements can be declared in theme.json config file under generatedData key. Example:

{
 "name": "my/custom-theme",
 "generatedData": { ... }
}

Theme generators supports now those elements: routes, menus and contentLists. All elements have this same properties as are supported by API requests, plus few extra like:

in routes: numberOfArticles - number of fake articles generated and attached to route

in menus: children - array of child menus attached to parent one

Example routes block¶

"generatedData": {
    "routes": [
        {
            "name": "Politics",                             # required
            "slug": "politics",                             # optional
            "type": "collection",                           # required
            "templateName": "category.html.twig",           # optional
            "articlesTemplateName": "article.html.twig",    # optional
            "numberOfArticles": 1                           # optional (number of articles generated and attached to route)
        },
    ...

Example menus block¶

"generatedData": {
    "menus": [
        {
            "name": "mainNavigation",               # required
            "label": "Main Navigation",             # optional
            "children": [                           # optional (array of child menus attached to parent one)
                {
                    "name": "home",                 # required
                    "label": "Home",                # optional
                    "uri": "/"
                }
            ]
        },
        {
            "name": "footerPrim",                   # required
            "label": "Footer Navigation",           # optional
            "children": [                           # optional (array of child menus attached to parent one)
                {
                    "name": "politics",             # required
                    "label": "Politics",            # optional
                    "route": "Politics"             # optional (route name - can be one defined in this config)
                },
    ...

Example contentLists block¶

"generatedData": {
    "contentLists": [
        {
            "name": "Example automatic list",                       # required
            "type": "automatic",                                    # required
            "description": "New list",                              # required
            "limit": 5,                                             # optional
            "cacheLifeTime": 30,                                    # optional
            "filters": "{\"metadata\":{\"located\":\"Porto\"}}"     # optional
        }
    ...

Technical Cookbooks¶

Installing Composer¶

Note

This section is based on Symfony2 documentation.

Composer is the package manager used by modern PHP applications. Use Composer to manage dependencies in your Symfony applications and to install Symfony Components in your PHP projects.

It’s recommended to install Composer globally in your system as explained in the following sections.

Install Composer on Linux and Mac OS X¶

To install Composer on Linux or Mac OS X, execute the following two commands:

curl -sS https://getcomposer.org/installer | php
sudo mv composer.phar /usr/local/bin/composer

Note

If you don’t have curl installed, you can also just download the installer file manually at https://getcomposer.org/installer and then run:

php installer
sudo mv composer.phar /usr/local/bin/composer

Install Composer on Windows¶

Download the installer from getcomposer.org/download, execute it and follow the instructions.

Learn more¶

Read the Composer documentation to learn more about its usage and features.

Updating database schema after customizing model’s mapping¶

If you modified one of the model’s mapping file or you added completely new model with mapping, you will need to update the database schema.

There are two ways to do it.

via direct database schema update:

1	$ php bin/console doctrine:schema:update --force

via migrations:

We recommend to update the schema using migrations so you can easily rollback and/or deploy new changes to the database without any issues.

$ php bin/console doctrine:migrations:diff
$ php bin/console doctrine:migrations:migrate

Tip

Read more about the database modifications and migrations in the Symfony documentation here.

Meta Loaders¶

Meta Loaders are services injected into the SWP\TemplatesSystemBundle\Gimme\Loader\ChainLoader class and are used for loading specific types of Meta objects.

Every Meta Loader must implement the SWP\TemplatesSystemBundle\Gimme\Loader\LoaderInterface interface.

Required methods:

load
isSupported

Example Meta Loader

<?php

namespace SWP\Component\TemplatesSystem\Gimme\Loader;

use SWP\Component\TemplatesSystem\Gimme\Factory\MetaFactory;
use SWP\Component\TemplatesSystem\Gimme\Meta\Meta;
use Symfony\Component\Yaml\Parser;

class ArticleLoader implements LoaderInterface
{
    /**
     * @var string
     */
    protected $rootDir;

    /**
     * @var MetaFactory
     */
    protected $metaFactory;

    /**
     * @param string $rootDir path to application root directory
     */
    public function __construct($rootDir, MetaFactory $metaFactory)
    {
        $this->rootDir = $rootDir;
        $this->metaFactory = $metaFactory;
    }

    /**
     * Load meta object by provided type and parameters.
     *
     * @MetaLoaderDoc(
     *     description="Article Meta Loader provide simple way to test Loader, it will be removed when real loaders will be merged.",
     *     parameters={}
     * )
     *
     * @param string $type              object type
     * @param array  $parameters        parameters needed to load required object type
     * @param array  $withoutParameters parameters used to exclude items from result
     * @param int    $responseType      response type: single meta (LoaderInterface::SINGLE) or collection of metas (LoaderInterface::COLLECTION)
     *
     * @return Meta|array false if meta cannot be loaded, a Meta instance otherwise
     */
    public function load($type, array $parameters = null, $withoutParameters = [], $responseType = LoaderInterface::SINGLE)
    {
        if (!is_readable($this->rootDir.'/Resources/meta/article.yml')) {
            throw new \InvalidArgumentException('Configuration file is not readable for parser');
        }
        $yaml = new Parser();
        $configuration = (array) $yaml->parse(file_get_contents($this->rootDir.'/Resources/meta/article.yml'));

        if ($responseType === LoaderInterface::SINGLE) {
            return $this->metaFactory->create([
                'title' => 'New article',
                'keywords' => 'lorem, ipsum, dolor, sit, amet',
                'don\'t expose it' => 'this should be not exposed',
            ], $configuration);
        } elseif ($responseType === LoaderInterface::COLLECTION) {
            return [
                $this->metaFactory->create([
                    'title' => 'New article 1',
                    'keywords' => 'lorem, ipsum, dolor, sit, amet',
                    'don\'t expose it' => 'this should be not exposed',
                ], $configuration),
                $this->metaFactory->create([
                    'title' => 'New article 2',
                    'keywords' => 'lorem, ipsum, dolor, sit, amet',
                    'don\'t expose it' => 'this should be not exposed',
                ], $configuration),
            ];
        }
    }

    /**
     * Checks if Loader supports provided type.
     *
     * @param string $type
     *
     * @return bool
     */
    public function isSupported($type)
    {
        return in_array($type, ['articles', 'article']);
    }
}

Content Lists¶

Automatic Content Lists¶

Pinned articles¶

This feature gives you a way of exposing more important articles on top of the list.

In automatic content lists you have the possibility to pin/unpin articles from the list. If you decide to pin one of the article from the list, it will always show up on top of the list, no matter how many new articles will be added to that list.

You can pin as many articles as you want.

Pinned article can be unpinned too. Once it’s done, the unpinned article will be removed from the top of the list and will remain in the list on the corresponding position.

By default, articles in Automatic Content List are ordered by “sticky” flag (desc): and item creation date (desc).

How articles are being added to Automatic Content Lists¶

When a new list is created and relevant criteria/filters are set for that list, articles to be published will automatically be added to the list but only when they meet the list’s criteria.

If the list’s criteria are not defined, or if the articles do not match the list’s criteria, articles will not be added to the list.

Here is an example to demonstrate this:

Given that your list’s criteria/filters are set to match “Sports” route so when you publish an article which is assigned to “Sports” route, it will be automatically assigned to the list.

Built in criteria:

route - an array of route ids, e.g. [1,5]
author - an array of authors, e.g. [“Test Persona”,”Doe”]
publishedBefore - date string, articles published before this date will be added to the list, e.g. date: “2017-01-20”. (date format must be YYYY-MM-DD)
publishedAfter - date string, articles published after that date will be added to the list, format is the same as in the publishedBefore case.
publishedAt - date string, when defined articles matching this publish dates will be added to the list when published, format is the same as in case of publishedBefore and publishedAfter
metadata - metadata field is json string, e.g. {"metadata":{"language":"en"}}. It matches article’s metadata, and you can use all metadata fields that are defined for the article, i.e.: language, located etc.

All criteria can be combined together which in the result it will add articles to the list (on publish) depending on your needs.

Article Preview¶

Article preview is based on user roles. Every article which is not published yet can be previewed by users with special roles assigned to them. This role is named ROLE_ARTICLE_PREVIEW.

If a user has ROLE_ARTICLE_PREVIEW role assigned, he/she can preview article using url: domain.com/preview/article/<routeId>/<article-slug>/?auth_token=<token>.

Where <routeId> is route identifier on which you want to preview given article by it’s slug (<article-slug> parameter).

Important here is to provide token, in order to be authorized to preview an article.

Tip

See API Authentication section for more details on how to obtain user token.

For example, if you created an article that has a slug test-article and this article is assigned to news route which id is 5, it will be available for preview under /preview/article/5/test-article?auth_token=uty56392323== url but only when the user has ROLE_ARTICLE_PREVIEW role assigned. In other cases 403 error will be thrown.

If you are building JavaScript app and you want to preview article, the preview url of an article can be taken and loaded in an iframe for preview.

User roles eligible for article preview:¶

Role
`ROLE_EDITOR`
`ROLE_INTERNAL_API`
`ROLE_ADMIN`
`ROLE_SUPER_ADMIN`

Rules¶

Rules are a way to define the business logic inside the system. They define how that business logic should be organized. A good example is a rule defining an article auto-publish. The organization rule can be configured to forward the content received from Superdesk to one of the defined tenants. Then tenant’s rule can be configured to automatically publish an article under specific route, so if the content is received in Superdesk Publisher, it can be automatically forwarded to the defined tenants (it won’t be published by default) and will be published under the specific route defined by the tenant’s rule.

There are two types of rules which are defined in Superdesk Publisher: - organization rules - applied to the level of organization - tenant rules - applied to the level of tenant

Auto-publish content based on rules¶

1. Adding an organization rule to push content to the desired tenants¶

Let’s assume there is a single organization created which contains a single tenant with code 123abc.

The next step is to create an organization rule by making a POST request to the /api/v1/organization/rules/ API endpoint with the JSON body:

{
   "name":"Test rule",
   "description":"Test rule description",
   "priority":1,
   "expression":"package.getLocated() matches \"/Sydney/\"",
   "configuration":[
      {
         "key":"destinations",
         "value":[
            {
               "tenant":"123abc"
            }
         ]
      }
   ]
}

In the JSON above we define that if the content which comes from Superdesk has a field located and it matches the value of Sydney, then push the content to the tenant with the code equal to 123abc.

On the tenant level, a new article will be created based on the pushed content which won’t be published by default. Right now, this article can be manually published or route can be assigned to it manually etc.

In order to publish it automatically, read below.

2. Adding a tenant rule to autopublish content under specific route¶

The next step is to create a rules on the level of tenant to automatically publish the article under specific route.

In order to do that, a POST request must be made to the /api/v1/rules/ API endpoint with the JSON body:

{
   "name":"Test tenant rule",
   "description":"Test tenant rule description",
   "priority":1,
   "expression":"article.getMetadataByKey(\"located\") matches \"/Sydney/\"",
   "configuration":[
      {
         "key":"route",
         "value":6
      },
      {
         "key":"published",
         "value":true
      }
   ]
}

The above JSON string defines, if an article’s metadata field called located equals to Sydney then assign route with id 6 to the article and automatically publish it.

3. Adding a tenant rule to publish content to Facebook Instant Articles¶

Article can be also published to Facebook Instant Articles. To do that, create a new tenant rule by making a POST request to the /api/v1/rules/ API endpoint with the JSON body:

{
   "name":"Test tenant rule",
   "description":"Test tenant rule description",
   "priority":1,
   "expression":"article.getMetadataByKey(\"located\") matches \"/Sydney/\"",
   "configuration":[
      {
         "key":"fbia",
         "value":true
      }
   ]
}

Note the fbia key in the configuration property is set to true.

If the content will be pushed to the tenant, the content will be also submitted to the Facebook Instant Articles.

Read more about Facebook Instant Articles in this section.

4. Adding a tenant rule to make an article paywall-secured¶

Articles can be marked as paywall-secured so an access can be restricted to such articles. To do that, create a new tenant rule by making a POST request to the /api/v1/rules/ API endpoint with the JSON body:

{
   "name":"Make articles paywall-secured",
   "description":"Marks articles as paywall-secured.",
   "priority":1,
   "expression":"article.getMetadataByKey(\"located\") matches \"/Sydney/\"",
   "configuration":[
      {
         "key":"paywallSecured",
         "value":true
      }
   ]
}

Note the paywallSecured key in the configuration property is set to true.

If the content will be pushed to the tenant and will match given expression, the “paywall-secured” flag will be set to true.

Read more about Paywall in this section.

Evaluation of the rules which match given package/item¶

Based on the package, there is a possibility to evaluate rules that match given package’s/item’s metadata. If the package/item in NINJS format will be passed to the /api/v1/organization/rules/evaluate as a request’s payload, then the (organization and/or tenant) rules that match that package’s/item’s metadata will be returned.

Usage¶

What is Output Channel Adapter?¶

Output Channel Adapter is a service which helps to communicate with the external system, for example, Wordpress. Thanks to the concept of adapters it is possible to exchange data between 3rd party services. It is possible to send the data from Publisher to an external system and also get the data from that system.

The Output Channel Adapters are strictly connected to the concept of Output Channels.

It is possible to choose the type of the external system (e.g. WordPress) if a new tenant is created.

It means that the content which is sent to Superdesk Publisher will not only be stored in the Publisher’s storage, but it can also be transmitted to an external system.

In this case, thanks to the output channels, you can send content wherever you want.

Superdesk Publisher acts as a hub where you can control where the content goes.

Adding New Adapter to work with Output Channels¶

A new adapter must implement SWP\Bundle\CoreBundle\Adapter\AdapterInterface interface.

Add a new const (CUSTOM_TYPE) to the SWP\Component\OutputChannel\Model\OutputChannelInterface interface.
Create a custom class:

<?php

// CustomAdapter.php
declare(strict_types=1);

namespace SWP\Bundle\CoreBundle\Adapter;

use GuzzleHttp\ClientInterface;
use SWP\Bundle\CoreBundle\Model\ArticleInterface;
use SWP\Bundle\CoreBundle\Model\OutputChannelInterface;

final class CustomAdapter implements AdapterInterface
{
    /**
     * @var ClientInterface
     */
    private $client;

    /**
     * WordpressAdapter constructor.
     *
     * @param ClientInterface $client
     */
    public function __construct(ClientInterface $client)
    {
        $this->client = $client;
    }

    /**
     * {@inheritdoc}
     */
    public function send(OutputChannelInterface $outputChannel, ArticleInterface $article): void
    {
        $url = $outputChannel->getConfig()['url'];

        $this->client->post($url, [
            'headers' => ['Content-Type' => 'application/json'],
            'body' => $article->getBody(),
            'timeout' => 5,
        ]);
    }

    /**
     * {@inheritdoc}
     */
    public function supports(OutputChannelInterface $outputChannel): bool
    {
        return OutputChannelInterface::TYPE_CUSTOM === $outputChannel->getType();
    }
}

Create Custom Adapter configuration form type

This form type will define which fields we can specify for the adapter. It can be credentials to connect to the 3rd party service etc.

<?php

declare(strict_types=1);

namespace SWP\Bundle\OutputChannelBundle\Form\Type;

use Symfony\Component\Form\AbstractType;
use Symfony\Component\Form\Extension\Core\Type\TextType;
use Symfony\Component\Form\FormBuilderInterface;
use Symfony\Component\Validator\Constraints\NotBlank;
use Symfony\Component\Validator\Constraints\Url;

final class CustomOutputChannelConfigType extends AbstractType
{
    /**
     * {@inheritdoc}
     */
    public function buildForm(FormBuilderInterface $builder, array $options): void
    {
        $builder
            ->add('url', TextType::class, [
                'constraints' => [
                    new NotBlank(),
                    new Url(),
                ],
            ])
            ->add('key', TextType::class, [
                'constraints' => [
                    new NotBlank(),
                ],
            ])
            ->add('secret', TextType::class, [
                'constraints' => [
                    new NotBlank(),
                ],
            ])
        ;
    }
}

Include a new type and form type in the OutputChannelType

<?php

declare(strict_types=1);

namespace SWP\Bundle\OutputChannelBundle\Form\Type;

use SWP\Bundle\CoreBundle\Model\OutputChannel;
use SWP\Component\OutputChannel\Model\OutputChannelInterface;
use Symfony\Component\Form\AbstractType;
use Symfony\Component\Form\Extension\Core\Type\ChoiceType;
use Symfony\Component\Form\FormBuilderInterface;
use Symfony\Component\Form\FormEvent;
use Symfony\Component\Form\FormEvents;
use Symfony\Component\Form\FormInterface;
use Symfony\Component\OptionsResolver\OptionsResolver;

final class OutputChannelType extends AbstractType
{
    /**
     * {@inheritdoc}
     */
    public function buildForm(FormBuilderInterface $builder, array $options): void
    {
        $builder
            ->add('type', ChoiceType::class, [
                'choices' => [
                    'Wordpress' => OutputChannelInterface::TYPE_WORDPRESS,
                    'Custom' => OutputChannelInterface::TYPE_CUSTOM,
                ],
            ])
        ;

        $formModifier = function (FormInterface $form, ?string $type) {
            if (OutputChannelInterface::TYPE_WORDPRESS === $type) {
                $form->add('config', WordpressOutputChannelConfigType::class);
            }

            if (OutputChannelInterface::TYPE_CUSTOM === $type) {
                $form->add('config', CustomOutputChannelConfigType::class);
            }
        };

        $builder->addEventListener(
            FormEvents::POST_SET_DATA,
            function (FormEvent $event) use ($formModifier) {
                $data = $event->getData();
                if (null !== $event->getData()) {
                    $formModifier($event->getForm(), $data->getType());
                }
            }
        );

        $builder->get('type')->addEventListener(
            FormEvents::POST_SUBMIT,
            function (FormEvent $event) use ($formModifier) {
                $type = $event->getForm()->getData();

                $formModifier($event->getForm()->getParent(), $type);
            }
        );
    }

    /**
     * {@inheritdoc}
     */
    public function configureOptions(OptionsResolver $resolver): void
    {
        $resolver->setDefaults([
            'csrf_protection' => false,
            'data_class' => OutputChannel::class,
        ]);
    }

    /**
     * {@inheritdoc}
     */
    public function getBlockPrefix(): string
    {
        return 'swp_output_channel';
    }
}

Register your new adapter

Your new adapter must be registered so it can be detected by the system and used by the Publisher. It can be done by taggin a service with swp.output_channel_adapter tag.

services:
    # ..
    SWP\Bundle\CoreBundle\Adapter\CustomAdapter:
        public: true
        arguments:
            - '@GuzzleHttp\Client'
        tags:
            - { name: swp.output_channel_adapter, alias: custom_adapter }

Create a new tenant with output channel which will use the newly created adapter

Now, when you want to create a new tenant, it will be possible to choose your output channel type and define the configuration which will use the newly created CustomAdapter.

curl -X POST \
  http://example.com/api/v1/tenants/ \
  -H 'Authorization: key' \
  -H 'Cache-Control: no-cache' \
  -H 'Content-Type: application/json' \
  -d '      {
        "tenant": {
          "domainName": "example.com",
          "name": "Custom tenant",
          "subdomain": "custom",
          "outputChannel": {
            "type": "custom",
            "config": {
              "url": "https://api.custom.com",
              "key": "private key",
              "secret": "secret"
            }
          }
        }
      }'

Using Wordpress Adapter¶

Usage:

// example.php
// ..

use SWP\Bundle\CoreBundle\Adapter\WordpressAdapter;
use SWP\Bundle\CoreBundle\Model\Article;
use SWP\Component\OutputChannel\Model\OutputChannel;
// ..

$article = new Article();

$guzzle = new GuzzleHttp\Client();
// ...

$wordpressAdapter = new WordpressAdapter($guzzle);
$outputChannel = new OutputChannel();
$outputChannel->setType('wordpress');
// ...

if ($adapter->supports($outputChannel)) {
    $adapter->send($outputChannel, $article);
    // ...
}

Using Composite Output Channel Adapter¶

The Composite Output Channel Adapter service loops for each of the registered adapter, checks if adapter supports given output channel and executes appropriate adapter functions.

Usage:

<?php
 // example.php
 // ..

 use SWP\Bundle\CoreBundle\Adapter\CompositeOutputChannelAdapter;
 use SWP\Bundle\CoreBundle\Adapter\WordpressAdapter;
 use SWP\Bundle\CoreBundle\Model\Article;
 use SWP\Component\OutputChannel\Model\OutputChannel;
 // ..

 $article = new Article();
 $guzzle = new GuzzleHttp\Client();
 // ...

 $wordpressAdapter = new WordpressAdapter($guzzle);
 // ...

 $compositeAdapter = new CompositeOutputChannelAdapter();
 $compositeAdapter->addAdapter($wordpressAdapter);
 // ...
 $outputChannel = new OutputChannel();
 $outputChannel->setType('wordpress');
 // ...

 $compisiteAdapter->send($outputChannel, $article);
 // ...

WebSocket Communication¶

There is a WebSocket server where the push notifications can be sent to the connected clients. These push notifications are used to refresh the views or in other words, keeps everything synchronized.

In the background, the WebSocket server is using AMQP queue (`WAMP sub-protocol and PubSub patterns<http://socketo.me/docs/wamp>`_) and from there it sends everything to clients. There is no communication from client to server, all changes are handled via API. For example, if the new content is pushed to Publisher, it is immediately sent to all the clients, meaning that the new content has been delivered.

Authentication¶

To keep the WebSocket server protected from unauthorized persons there is an authentication mechanism implemented which is based on tokens. Only clients with a valid token can access the WebSocket server. It means if you want to connect to the WebSocket server you have to authenticate via API first. This is the standard token which can be also used to access API. Read more about ../internal_api/authentication. The obtained token must be placed as a query parameter inside the WS url: ws://127.0.0.1:8080?token=<token>, where <token> is the proper token.

How it works?¶

A client must connect to the WebSocket server and subscribe to the specific topic. In this case it is package_created.

If the new content will be sent to the Publisher, we will automatically receive info from the WebSocket server about newly delivered package/content. Based on that info we can refresh or update existing view.

The default WebSocket server port is 8080 and host 127.0.0.1.

<script language="javascript" type="text/javascript" src="https://cdn.rawgit.com/cboden/fcae978cfc016d506639c5241f94e772/raw/e974ce895df527c83b8e010124a034cfcf6c9f4b/autobahn.js"></script>
<script>
    var conn = new ab.Session('ws://127.0.0.1:8080?token=12345',
        function() {
            conn.subscribe('package_created', function(topic, data) {
                // This is where you would add the new article to the DOM (beyond the scope of this tutorial)
                console.log('New article published to "' + topic + '" : ' + data.title);
            });
        },
        function() {
            console.warn('WebSocket connection closed');
        },
        {'skipSubprotocolCheck': true}
    );
</script>

Paywall¶

Paywall only retrieves the subscriptions data from an external Subscriptions System. In order to buy subscriptions, you have to directly interact with the 3rd party Subscriptions System if you want to create new subscriptions. For example, a simple JS application can be written where users can buy an access to the website.

How it works¶

Your app (via API) or web front-end (via Twig) requests the subscriptions data.
The Publisher passes the request parameters to the adapter.
Adapter calls external Subscriptions System.
Once the connection is initialized and the adapter has made a request to Subscriptions System, the Subscriptions System communicates that information and returns a list of subscriptions or a single subscription.
Publisher then saves the retrieved subscriptions in the Memcached and caches it for 24 hours, by default. If the request parameters did not change, each request to get subscriptions will be executed to Memcached server.
The Publisher gets the subscription(s) from the Memcached.

How to configure subscriptions cache lifetime¶

By default, the retrieved subscriptions are cached for 24 hours (86400 seconds). This can be changed by setting env(SUBSCRIPTIONS_CACHE_LIFETIME) env var or parameter value to different number of seconds.

How to render user subscriptions in Twig¶

To render user subscriptions:

{% if app.user %}
    Hey {{ app.user.username }}. <a href="{{ url('security_logout') }}">Logout</a>.
    {% gimmelist subscription from subscriptions with { user: app.user, articleId: 10, routeId: 20 } %}
        {{ subscription.id }} # subscription's id, this is set in the Paywall Adapter
        {{ subscription.code }} # subscription's code, this is set in the Paywall Adapter
        {{ subscription.active }} # is subscription active, this is set in the Paywall Adapter
        {{ subscription.type }} # type of the subscription, this is set in the Paywall Adapter
        {{ subscription.details.articleId }} # an array with more details about the subscription, this is set in the Paywall Adapter
        {{ subscription.updatedAt|date('Y-m-d') }}
        {{ subscription.createdAt|date('Y-m-d') }}
    {% endgimmelist %}
{% endif %}

To render a single user subscription by article id:

{% if app.user %}
    Hey {{ app.user.username }}. <a href="{{ url('security_logout') }}">Logout</a>.
    {% gimme subscription with { user: app.user, articleId: 10 } %}
        {{ subscription.id }}
        # ...
    {% endgimme %}
{% endif %}

To render a single user subscription by article id and name:

{% if app.user %}
    Hey {{ app.user.username }}. <a href="{{ url('security_logout') }}">Logout</a>.
    {% gimme subscription with { user: app.user, articleId: 10, name: "premium_content" } %}
        {{ subscription.id }}
        {{ subscription.details.name }}
        # ...
    {% endgimme %}
{% endif %}

How to check if used is “paywall-secured”¶

Route and the article objects can be marked as “paywall-secured”. This can be done via Routes API and Articles API by setting the value of paywallSecured property to true.

To check if the article or route is “paywall-secured” do:

Articles:

{% gimmelist article from articles %}
    {% if article.paywallSecured %}
        # render content of the article
    {% else %}
        # need to buy an access to read this article
    {% endif %}
    # ...
{% endgimmelist %}

Routes:

{% gimmelist route from routes %}
    {% if route.paywallSecured %}
        # render articles under this route
    {% else %}
        # need to buy an access to read this section
    {% endif %}
    # ...
{% endgimme %}

How to mark articles as paywall-secured using rules¶

How to mark articles as paywall-secured by manually publishing packages¶

You can also directly publish a package and mark articles as “paywall-secured” by making a POST request to /api/v1/packages/<package_id>/publish/ API endpoint with body:

{
    "publish":{
      "destinations":[
        {
          "tenant":"123abc",
          "route":6,
          "fbia":false,
          "published":true,
          "paywallSecured":true
        }
      ]
    }
}

How to override paywall-secured option using publish destination¶

If there is a rule configured that marks all the articles matching given expression as “paywall-secured”, you can use publish destinations to override existing publish workflow for specific packages on specific tenants.

To do this, make a POST request to /api/v1/organization/destinations/ API endpoint with body:

{
  "publish_destination":{
    "tenant":"123abc",
    "route":5,
    "fbia":false,
    "published":true,
    "paywallSecured":false,
    "packageGuid": "urn:newsml:sd-master.test.superdesk.org:2022-09-19T09:26:52.402693:f0d01867-e91e-487e-9a50-b638b78fc4bc"
  }
}

The following destination will be processed when package will be published. The package will be published to tenant with code 123abc, route with id 5 and won`t be marked as “paywall-secured” even if there is a rule marking it as paywall-secured.

GeoIP¶

GeoIP feature allows to restrict access to the specific articles based on the geolocation metadata.

How it works?¶

When this feature is enabled, the reader’s IP address is read on the visited article page. Then, the IP address is checked in the GeoIP2 database to check which country or state it comes from. The GeoIP database has to be downloaded first. If the article place metadata matches the IP country or state, access to the articles is denied.

The GeoIP features are disabled by default. To enable the GeoIP features you have to set the GEO_IP_ENABLED environment variable to true (GEO_IP_ENABLED=true) in your .env.local file.

Before enabling this feature, the GeoIP2 database must be downloaded:

1	php bin/console swp:geoip:db:update

Executing the above command will download the GeoIP2 database to the cache dir. From this directory, the Publisher will read the GeoIP data.

Calls to the GeoIP database are cached by default.

Performance¶

To increase the performance it’s recommended to install native PHP extension for GeoIP2.

MaxMind provides an optional C extension that is a drop-in replacement for MaxMindDbReader. This will speed up the location lookups for GeoIp2 PHP provider enormously and is recommended for high traffic instances.

Installing libmaxminddb¶

The PHP extension requires the C library libmaxminddb for reading MaxmindDB (GeoIP2) files.

Read how to install this library

Installing PHP Extension¶

Download the https://github.com/maxmind/MaxMind-DB-Reader-php repository and in the top-level directory execute:

cd ext
phpize
./configure
make
sudo make install

Then add extension="maxminddb.so" into your php.ini config file.

The reads from the GeoIP2 database will be automatically faster.

Editors Cookbooks¶

Facebook Instant Articles Integration¶

Superdesk Publisher has built-in integration with Facebook Instant Articles. This cookbook describes all the steps needed for proper configuration.

Step 1. Register Facebook Page and Application in Publisher¶

Note

As at the moment of writing this documentation there is no UI in Superdesk for this feature, needed actions will be described with CURL direct API calls.

Note

Any Publisher API request requires authentication. Read more about this here: API authentication

Instant Articles are strongly connected with a Facebook Page. To start you need to enable that feature in Your Facebook Page settings. Once this is set up, call our API to register that page in Publisher.

A Facebook Page can be registered in Publisher with a REST API /api/{version}/facebook/pages/ POST request.

Required parameters:

pageId - Unique ID of your Facebook Page

name - Facebook Page Name

curl -X POST 'http://webpublisher.dev/api/v1/facebook/pages/' -H 'Origin: http://webpublisher.dev' -H 'Content-Type: application/x-www-form-urlencoded' -H 'Accept: */*' -H 'Connection: keep-alive' -H 'DNT: 1' -d "facebook_page[pageId]=1234567890987654321&facebook_page[name]=Test Page" --compressed

The next step is registering the Facebook Application (You need to create it first on the Facebook Platform). The application is used for retrieving never expired access token - it will be used by Publisher in Facebook API calls.

Facebook Application can be registered in Publisher with a REST API /api/{version}/facebook/applications/ POST request.

Required parameters:

appId - Unique ID of your Facebook Application

appSecret - Generated by Facebook Application secret

curl -X POST 'http://webpublisher.dev/api/v1/facebook/applications/' -H 'Origin: http://webpublisher.dev' -H 'Content-Type: application/x-www-form-urlencoded' -H 'Accept: */*' -H 'Connection: keep-alive' -H 'DNT: 1' -d "facebook_application[appId]=1234567890987654321&facebook_application[appSecret]=superS3cretSecretFromFacebook" --compressed

Step 2. Facebook Page/Application authentication¶

Assuming that in your database you have Application with id 123456789 and Page with id 987654321 (and both it exists on Facebook platform), You need to call this url (route: swp_fbia_authorize): /facebook/instantarticles/authorize/123456789/987654321

In response You will be redirected to Facebook where You will need allow for all required permissions.

After that Facebook will redirect You again to application where (in background - provided by Facebook code will be exchanged for access token and that access) you will get JSON response with pageId and accessToken (never expiring access token).

Step 3. Create Content List (bucket)¶

In most cases you wouldn’t want to push everything to Instant Articles. Publisher allows you to define rules for articles selected for Instant Articles publication. This solution is based on Content Lists. Content lists allow you to define custom criteria and apply them to every published article - if an article matches the criteria it’s added to that Content List and automatically published to Instant Articles.

Content Lists can be created in Publisher with a REST API /api/{version}/content/lists/ POST request.

Required parameters:

name - Content List name

type - Content List type, in this case it must be “bucket”

expression - (optional) Expression used for testing published articles eg.: article.getPriority() > 4

curl -X POST 'http://webpublisher.dev/api/{version}/content/lists/' -H 'Origin: http://webpublisher.dev' -H 'Content-Type: application/x-www-form-urlencoded' -H 'Accept: */*' -H 'Connection: keep-alive' -H 'DNT: 1' -d "content_list[name]=Facebook Instant Articles&content_list[type]=bucket" --compressed

This list don’t have expression parameter defined so it will catch all published articles.

Step 3. Create Feed¶

Feeds are used to connect Facebook Pages and Content Lists. With them, you can send selected articles to different Facebook Pages.

Feeds can be created in Publisher with a REST API /api/{version}/facebook/instantarticles/feed/ POST request.

Required parameters:

contentBucket - Content List id

facebookPage - Facebook Page id (from publisher)

mode - Instant Article publishing mode: 0 (devlopment) or 1 (production)

curl -X POST 'http://webpublisher.dev/api/{version}/content/lists/' -H 'Origin: http://webpublisher.dev' -H 'Content-Type: application/x-www-form-urlencoded' -H 'Accept: */*' -H 'Connection: keep-alive' -H 'DNT: 1' -d "facebook_instant_articles_feed[contentBucket]=1&facebook_instant_articles_feed[facebookPage]=1&&facebook_instant_articles_feed[mode]=0" --compressed

Step 4. Create Instant Article template¶

An Instant Article is created from parsed template file. The look and feel of Instant Articles can be controlled by template files in the theme. That file must be located here: views\platforms\facebook_instant_article.html.twig. Publisher autmatically attaches current article meta like in regular page template (remember that there is gimme.route set in this case).

Minimal code for Instant Article templates needs to look like this:

<!doctype html>
<html lang="en" prefix="op: http://media.facebook.com/op#">
    <head>
        <meta charset="utf-8">
        <link rel="canonical" href="{{ url(gimme.article) }}">
    </head>
    <body>
        <header>
            <h1>{{ gimme.article.title }}</h1>
            <time class="op-published" datetime="{{ gimme.article.publishedAt|date("Y-m-d\\Th:i:s\\Z", false) }}">{{ gimme.article.publishedAt|date("Y-m-d h:i ") }}</time>

        </header>
        <p>{{ gimme.article.lead }}</p>
        {{ gimme.article.body|raw }}
    </body>
</html>

The HTML code of the article body (gimme.article.body) will be parsed by Transformer. Transformer will try to match html elements to Instant Articles tags (for example images). If it does not recognize some elements they will be removed. You can preview how your template works with currently published articles here: /facebook/instantarticles/preview/{articleId}.

Note

Preview url is available only in Publisher Development mode. To send that article to FBIA library from preview add ?listId={content bucket Id} to url.

Step 5. Publish Your articles to Facebook Instant Articles¶

After all previous steps - publishing should happen automatically just after publishing article matching Content List criteria.

Implementators Cookbooks¶

Secure content push to Publisher¶

Content can be pushed to Publisher via HTTP requests from any system. But it’s important to store and publish only requests from approved by us sources.

To verify incoming requests we use special header: x-superdesk-signature. Value of this header have format like that: sha1={token}.

token is a result of HMAC (keyed-Hash Message Authentication Code) function. It’s created from request content and secret token value with sha1 algorithm applied on it.

secret token can be defined in Organization (when created or updated). Example command:

php bin/console swp:organization:update OrganizationName --secretToken secret_token

Organization secret token is not visible in any API.

If token is set in organization then Publisher will reject all requests without x-superdesk-signature header or with wrong value in it.

[Migration] Redirect articles from previous url’s to publisher¶

Almost always after content migration you need to make sure that all previous links to articles works ok with new system. In many publishing systems articles are identified by article/post numbers or some special codes. In Publisher we use unique combination of route and article slug’s. Because of that it’s impossible to redirect requests only with server redirects.

Publisher provide solution for this case. In short: based on package external data (imported from external system) we localize articles and return redirect responses with link to new article location.

External data¶

With special API endpoint (secured with secret code) You can associate pair’s of keys and values with imported article (package).

Example of setting external data (from our behat feature):

When I add "Content-Type" header equal to "application/json"
And I add "x-publisher-signature" header equal to "sha1=0dcd1953d72dda47f4a4acedfd638a3c58def7bc"
And I send a "PUT" request to "/api/v1/packages/extra/test-news-article" with body:
"""
{
  "articleNumber": "123456",
  "some other key": "some other value"
}
"""

Then the response status code should be 200

Redirect url¶

In our case server is responsible for composing redirect (supported with regular expressions) url containing article identifier from original url and pushed as external data to package.

For example for article with url like that: /en/sport/123456/mundial-winner we need to push indetifier (123456) as a external data to /api/v1/packages/extra/mundial-winner.

After that server can rediret our url to this one: /redirecting/extra/articleNumber/123456. Publisher in response will return redirect response (with code 301) to new article location.

Setup Wordpress as a Publisher Output Channel¶

Output channel allows you to use tenant as a bridge for publishing content in external systems. Thanks to it content pushed to Publisher can be automatically published also in Wordpress or other even internal systems.

We assume that Publisher is installed and running. So now we need Wordpress instance.

Download: wget https://wordpress.org/latest.zip
Unpack: unzip latest.zip
Download router.php file: cd wordpress && wget https://gist.githubusercontent.com/ginfuru/1dfd9a054f27d268e9e3f445896150f5/raw/9f5a4c71e9bd6592e113914e64f7c36c31c5a1ad/router.php
Run Wordpress with built in php server: php -S wordpress.test:8080 router.php
Install Wordpress
Install plugin “Application Passwords” (author: George Stephanis). Dont forget to activate it ;)

7. Go to http://wordpress.test:8080/wp-admin/profile.php adn create new Application Password - it’s on bottom of page. 7. Encode your password with echo -n "admin:8e7M k22B znze mLVF 3vmc i4Vc" | base64 (run this in terminal) 9. Copy generated password (we will use it later). 10. Create new tenant for wordpress:

{
  "tenant": {
    "domainName": "yourpublisherdomain.com",
    "name": "Wordpress Output Channel",
    "subdomain": "validvhostsubdomain",
    "outputChannel": {
      "type": "wordpress",
      "config": {
        "url": "http://wordpress.test:8080",
        "authorization_key": "Basic YWRtaW46OGU3TSBrMjJCIHpuemUgbUxWRiAzdm1jIGk0VmM="
      }
    }
  }
}

In authorization_key we type “Basic ” and generated before password.

From now all content published to created tenant will be published, unpublished and updated in Wordpress instance.

Setup oAuth login with auth0.com¶

Publisher from version 2.0 have support for authentication via oAuth protocol. Here is an example of configuration with auth0.com

Let’s assume that you have publisher instance under: https://www.publisher.wip url.

First create fee account on auth0.com and create new Application of type Regular Web Applications. As technology choose PHP.

Now let’s configure required env variables in Publisher. Go to settings tab in your application page and look for variables defined bellow.

Note

IMPORTANT: Add this url https://www.publisher.wip/connect/oauth/check to Allowed Callback URLs field. And click Save changes at the bottom of settings page.

In file .env.local set those variables:

EXTERNAL_OAUTH_CLIENT_ID=<value of Client ID>

EXTERNAL_OAUTH_CLIENT_SECRET=<value of Client Secret>

EXTERNAL_OAUTH_BASE_URL=<value of Domain (with https://)

Now go to https://www.publisher.wip/connect/oauth.

And it’s done. After redirect to auth0, logging with selected provider - You will be redirected back to publisher as an authenticated user.

Reference¶

In this chapter you can explore complete reference of things that exist in Publisher, list of functionality, how to use certain functions etc. This reference is meant to be quick reminder and sum up of everythnig that is already described in other chapters.

System Requirements¶

Latest requirements can be found in project README.md file

Twig Extensions¶

Publisher provides its own Twig functions and filters in addition to the standard set of Twig functions which you can use in website templates.

Functions¶

Bundles¶

Fixtures Bundle¶

Overview¶

The Fixtures Bundle helps developers to create fixtures or fake data that can be used for development and/or testing purposes.

It relies on the following 3rd party libraries:

DoctrineFixturesBundle (gives possibility to load data fixtures programmatically into the Doctrine ORM or ODM)

fzaninotto/Faker (generates fake data for you)

nelmio/alice (gives you a few essential tools to make it very easy to generate complex data with constraints in a readable and easy to edit way)

It also provides the possibility to set up a ready-to-use demo theme, for development.

How to use fixtures¶

The following chapter describes how to make use of the Fixtures Bundle features.

Creating a simple PHP fixture class¶

Fixtures should be created inside SWP\FixturesBundle\DataFixtures\<db_driver> directory and by convention, should be called like: LoadPagesData, LoadArticlesData, LoadUsersData etc.

Replace db_driver either with ORM for the Doctrine ORM or PHPCR for the Doctrine PHPCR ODM.

Example Fixture class:

<?php
namespace SWP\FixturesBundle\DataFixtures\ORM;

use Doctrine\Common\DataFixtures\FixtureInterface;
use Doctrine\Common\Persistence\ObjectManager;
use SWP\Bundle\FixturesBundle\AbstractFixture;

class LoadPagesData extends AbstractFixture implements FixtureInterface
{
    /**
     * {@inheritdoc}
     */
    public function load(ObjectManager $manager)
    {
        $env = $this->getEnvironment();
        $this->loadFixtures(
            '@SWPFixturesBundle/Resources/fixtures/ORM/'.$env.'/page.yml',
            $manager
        );
    }
}

Each fixture class extends AbstractFixture class and implements FixtureInterface. This way we can use the load method, inside which we can create some objects using PHP and/or make use of nelmio/alice and fzaninotto/Faker by loading fixtures in YAML format, as shown in the example above.

Creating a simple Alice fixture (YAML format)¶

For more details on how to create Alice fixtures, please see the Alice documentation as a reference.

Example Alice fixture:

SWP\CoreBundle\Entity\Page:
    page1:
        name: "About Us"
        type: 1
        slug: "about-us"
        templateName: "static.html.twig"
        contentPath: "/swp/content/about-us"
    page2:
        name: "Features"
        type: 1
        slug: "features"
        templateName: "features.html.twig"
        contentPath: "/swp/content/features"
    page3:
        name: "Get Involved" # we can also use faker library formatters like: <paragraph(20)> etc
        type: 1
        slug: "get-involved"
        templateName: "involved.html.twig"
        contentPath: "/swp/content/get-involved"

The above configuration states that we want to persist into the database three objects of type SWP\CoreBundle\Entity\Page. We can use the faker formatters where, for example, <paragraph(20)> is one of the fzaninotto/Faker formatters, which tells Alice to generate 20 paragraphs filled with fake data.

By convention, Alice YAML files should be placed inside Resources/fixtures/<db_driver>/<environment>, where <environment> is the current environment name (dev, test).

For instance, having the Resources/fixtures/ORM/test/page.yml Alice fixture, we will be able to persist fake data defined in the YAML file into the database (using Doctrine ORM driver), only when the test environment is set or defined differently in SWP\FixturesBundle\DataFixtures\ORM\LoadPagesData.php.

There is a lot of flexibility on how to define fixtures, so it’s up to the developer how to create them.

Loading all fixtures¶

Note: Remember to update your database schema before loading fixtures! To do this, run in a console:

1	php bin/console doctrine:schema:update --force

Once you have your fixtures defined, we can simply load them. To do that you must execute console commands.

To load Doctrine ORM fixtures:

1	php bin/console doctrine:fixtures:load --append

See php bin/console doctrine:fixtures:load --help for more details.

After executing the commands above, your database will be filled with the fake data, which can be used by themes.

Setting up a demo theme¶

To make it easier to start with the Web Publisher, we created a simple demo theme. To set this theme as an active one, you need to execute the following command in a console:

php bin/console swp:theme:install 123abc src/SWP/Bundle/FixturesBundle/Resources/themes/DefaultTheme/ -f -p

This command will install default theme for the default tenant which was already created by loading fixtures (see above).

See php bin/console swp:theme:install --help for more details.

MultiTenancyBundle¶

This bundle provides the tools to build multi-tenant architecture for your PHP applications.

The MultiTenancy Component, which is used by this bundle, provides a generic interfaces to create different implementations of multi-tenancy in PHP applications.

The idea of this bundle is to have the ability to create multiple websites (tenants) within many organizations.

So far, we aim to support two persistence backends:

Doctrine ORM
Doctrine PHPCR ODM

Note

This documentation describes installation and configuration for Doctrine PHPCR ODM at the moment.

Features:

Allows to create many organizations
Allows to create many websites (tenants) within a single organization
Organization can have multiple websites assigned
Each website have parent organization
Allows to create default organization and default website (tenant)

E.g. The Vox Media organization can have multiple websites: The Verge. Polygon, Eater etc. Each website has it’s own content.

Prerequisites¶

This version of the bundle requires Symfony >= 2.6 and PHP version >=7.0.

Installation¶

Install the Bundle with Composer¶

In your project directory, execute the following command to download the latest stable version of the MultiTenancyBundle:

1	composer require swp/multi-tenancy-bundle

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

Note

By default Jackalope Doctrine DBAL is required for PHPCR ODM in this bundle. See Choosing a PHPCR Implementation for alternatives.

Enable the bundle and its dependencies (DoctrinePHPCRBundle, DoctrineBundle) by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(
            // ...

            new Doctrine\Bundle\PHPCRBundle\DoctrinePHPCRBundle(),
            new Doctrine\Bundle\DoctrineBundle\DoctrineBundle(),
            new SWP\MultiTenancyBundle\SWPMultiTenancyBundle(),
        );

        // ...
    }

    // ...
}

Note

All dependencies will be installed automatically. You will just need to configure the respective bundles.

Configure the SWPMultiTenancyBundle¶

Let’s enable PHPCR persistence backend.

YAML

# app/config/config.yml
swp_multi_tenancy:
    persistence:
        phpcr:
            # if true, PHPCR is enabled in the service container
            enabled: true

XML

<!-- app/config/config.xml -->
<?xml version="1.0" encoding="UTF-8" ?>
<swp_multi_tenancy>
    <persistence>
        <phpcr>
            <!-- if true, PHPCR is enabled in the service container -->
            <enabled>true</enabled>
        </phpcr>
    </persistence>
</swp_multi_tenancy>

Tip

See CMF RoutingBundle Integration.

Note

See Configuration Reference for more details.

DoctrinePHPCRBundle Configuration¶

See how to set PHPCR Session Configuration.

Add the domain parameter¶

Add the following parameter to your parameters file, so the current tenant can be resolved and matched against the configured domain.

# app/config/parameters.yml
domain: example.com

Update your database schema¶

Note

This step assumes you have already configured and created the database.

Execute the following commands in the console:

php bin/console doctrine:schema:update --force
php bin/console doctrine:phpcr:repository:init
php bin/console swp:organization:create --default
php bin/console swp:tenant:create --default
php bin/console doctrine:phpcr:repository:init

That’s it, the bundle is configured properly now!

Services¶

TenantContext¶

The TenantContext service allows you to manage the currently used tenant. Its getTenant method gets the current tenant by resolving its subdomain from the request.

For example, if the host name is: subdomain.example.com the TenantContext will first resolve the subdomain from the host provided in the parameters file domain, and then it will try to find the object of instance TenantContextInterface in the storage. When found, it will return the tenant object.

You can also set the current tenant in the context, so whenever you request the current tenant from the context it will return you the object you set. The setTenant method is used to set the tenant. It accepts as the first parameter an object of type TenantContextInterface.

TenantAwareRouter¶

Note

This service requires the CMF Routing Bundle to be installed and configured.

The TenantAwareRouter generates tenant-aware routes. It extends DynamicRouter from the CMF Routing Bundle.

In some cases you may need to generate a statically configured route. Let’s say we have a path defined in PHPCR: /swp/<organization_code>/<tenant_code>/routes/articles/features. If you want to generate a route for the current tenant in a Twig template, you could use the following code:

<a href="{{ path('/routes/articles/features') }}">Features</a>

The TenantAwareRouter will resolve the current tenant from the host name and will internally create a route /swp/<organization_code>/<tenant_code>/routes/articles/features where swp is the root path defined in the bundle configuration, <tenant_code> is the current tenant’s unique code, and routes is the configured route_basepaths.

The result will be:

<a href="/articles/features">Features</a>

You can also generate the route by content path:

<a href="{{ path(null, {'content_id': '/content/articles/features'}) }}">Features</a>

If the content is stored under the path /swp/<organization_code>/<tenant_code>/content/articles/features in the PHPCR tree, the router will search for the route for that content and will return the route associated with it. In this case, the associated route is /swp/<organization_code>/<tenant_code>/routes/articles/features so it will generate the same route: /articles/features as in the example above.

Note

We do not recommend hard-coding the route name in the template because if the route is removed, the page will break.

See CMF RoutingBundle Integration on how to enable and make use of this router.

PrefixCandidates¶

Note

This service requires the CMF Routing Bundle to be installed and configured.

This service extends Symfony CMF RoutingBundle PrefixCandidates service, to set tenant-aware prefixes. Prefixes are used to generate tenant-aware routes. Prefixes are built from the configured root path, which by default is /swp and from route_basepaths which you can set in the configuration file.

See the Full Default Configuration reference for more details.

PHPCR ODM Repository Initializer¶

PHPCRBasePathsInitializer¶

Note

This service requires DoctrinePHPCRBundle to be installed and configured.

The Initializer is the PHPCR equivalent of the ORM schema tools. PHPCRBasePathsInitializer creates base paths in the content repository based on tenants and organizations, configures and registers PHPCR node types. It is disabled by default, but can be enabled in the configuration when using PHPCR ODM persistence backend.

You can execute this initializer, together with the generic one, by running the following command:

1	php bin/console doctrine:phpcr:repository:init

Running this command will trigger the generic initializer which is provided by the DoctrinePHPCRBundle. The generic initializer will be fired before this one, and will create the root base path in the content repository.

See CMF RoutingBundle Integration on how to enable this initializer.

Repositories¶

TenantRepository¶

This repository allows you to fetch a single tenant by its subdomain name and all available tenants from the Doctrine ORM storage. It extends EntityRepository from Doctrine.

This service implements TenantRepositoryInterface and it has three methods:

findBySubdomain($subdomain) - Finds the tenant by subdomain. $subdomain is the subdomain of string type.
findByCode($code) - Finds the tenant by code. $code is the unique code of string type.
findAvailableTenants() - Finds all available tenants. Returns an array of all tenants.

SQL Query Filters¶

TenantableFilter¶

This filter adds the where clause to the select queries, to make sure the query will be executed for the current tenant. If the tenant exists in the context and the tenant id is 1, it will add WHERE tenant_id = 1 to every select query. This way, we always make sure we get the data for the current tenant.

In order to make use of the filter every class needs to implement TenantAwareInterface which indicates that it should be associated with the specific tenant.

It extends Doctrine\ORM\Query\Filter\SQLFilter.

When PHPCR ODM persistence backend is enabled it will rely on tenant’s unique code instead of the tenant id. In this case, if the tenant exists in the context and the tenant code is 123abc, it will add WHERE tenant_id = 123abc to every select query.

Event Listeners¶

TenantableListener¶

This event listener runs on every kernel request (kernel.request event). If the tenant is set in the TenantContext it enables Doctrine ORM Query TenantableFilter, otherwise it doesn’t do anything. Its responsibility is to ensure that every SQL select query will be tenant-aware (tenant_id will be added in the query).

TenantSubscriber¶

This subscriber subscribes to every Doctrine ORM prePersist event, when persisting the data. It makes sure that the persisted object (which needs to implement TenantAwareInterface) will be associated with the current tenant when saving the object.

Console Commands¶

This section describes console commands available in this bundle.

Create a new organization¶

To make use of this bundle, you need to first create the default organization. You may also need to create some other, custom organization if needed.

Note

This command persists organizations in database depending on your enabled persistence backend. If the PHPCR backend is enabled it will store tenants in PHPCR tree.

To create the default organization, execute the following console command:

1	php bin/console swp:organization:create --default

To create a custom organization which will be disabled by default, use the command:

1	php bin/console swp:organization:create --disabled

To create a custom organization, execute the following console command:

1	php bin/console swp:organization:create

Run php bin/console swp:organization:create --help to see more details of how to use this command.

List available organizations¶

This command list all available organizations.

Usage:

1	php bin/console swp:organization:list

Run php bin/console swp:organization:list --help to see more details of how to use this command.

Create a new tenant¶

To make use of this bundle, you need to first create the default tenant. You may also need to create some other, custom tenants.

Note

This command persists tenants in database depending on your enabled persistence backend. If the PHPCR backend is enabled it will store tenants in PHPCR tree.

To create the default tenant, execute the following console command:

1	php bin/console swp:tenant:create --default

Note

When creating default tenant the command requires you to have the default organization created.

To create a custom tenant which will be disabled by default, use the command:

1	php bin/console swp:tenant:create --disabled

To create a custom tenant, execute the following console command:

1	php bin/console swp:tenant:create

You will need to specify organization unique code so tenant can be assigned to the organization.

Run php bin/console swp:tenant:create --help to see more details of how to use this command.

Twig Extension¶

This bundle provides Twig global variables in your project. See below for more details.

Global variables¶

tenant - provides data about the current website/tenant. This variable is an object of type TenantInterface

Usage:

{{ tenant.name }}
{{ tenant.subdomain }}
{{ tenant.organization.name }} # get tenant's organization name
{# ... #}

Configuration Reference¶

The SWPMultiTenancyBundle can be configured under the swp_multi_tenancy key in your configuration file. This section describes the whole bundle’s configuration.

Full Default Configuration¶

# app/config/config.yml
swp_multi_tenancy:
    use_orm_listeners: false
    persistence:
        phpcr:
            enabled: true
            basepath: "/swp"
            route_basepaths: ["routes"]
            content_basepath: "content"
            menu_baseapth: "menu"
            media_baseapth: "media"
            tenant_aware_router_class: SWP\MultiTenancyBundle\Routing\TenantAwareRouter
            classes:
                tenant:
                    model: SWP\Component\MultiTenancy\Model\Tenant
                    repository: SWP\Bundle\MultiTenancyBundle\Doctrine\PHPCR\TenantRepository
                    factory: SWP\Component\MultiTenancy\Factory\TenantFactory
                    object_manager_name: ~
                organization:
                    model: SWP\Component\MultiTenancy\Model\Organization
                    repository: SWP\Bundle\MultiTenancyBundle\Doctrine\PHPCR\OrganizationRepository
                    factory: SWP\Bundle\MultiTenancyBundle\Factory\OrganizationFactory
                    object_manager_name: ~
        orm:
            enabled: true
            classes:
                tenant:
                    model: SWP\Component\MultiTenancy\Model\Tenant
                    repository: SWP\Bundle\MultiTenancyBundle\Doctrine\ORM\TenantRepository
                    factory: SWP\Component\MultiTenancy\Factory\TenantFactory
                    object_manager_name: ~
                organization:
                    model: SWP\Component\MultiTenancy\Model\Organization
                    repository: SWP\Bundle\MultiTenancyBundle\Doctrine\ORM\OrganizationRepository
                    factory: SWP\Bundle\MultiTenancyBundle\Factory\OrganizationFactory
                    object_manager_name: ~

persistence¶

use_orm_listeners¶

type: Boolean default: false

Use this setting to activate the TenantableListener and TenantableSubscriber. This will enable tenantable SQL extension and will make sure your Doctrine ORM entities are tenant aware. See Event Listeners for more details.

persistence¶

phpcr¶

# app/config/config.yml
swp_multi_tenancy:
    # ..
    persistence:
        phpcr:
            enabled: true
            basepath: "/swp"
            route_basepaths: ["routes"]
            content_basepath: "content"
            menu_baseapth: "menu"
            media_baseapth: "media"
            tenant_aware_router_class: SWP\MultiTenancyBundle\Routing\TenantAwareRouter

enabled¶

type: boolean default: false

If true, PHPCR is enabled in the service container.

PHPCR can be enabled by multiple ways such as:

phpcr: ~ # use default configuration
# or
phpcr: true # straight way
# or
phpcr:
    route_basepaths: ... # or any other option under 'phpcr'

basepath¶

type: string default: /swp

The basepath for documents in the PHPCR tree.

route_basepaths¶

type: array default: ['routes']

A set of paths where routes should be located in the PHPCR tree.

content_basepath¶

type: string default: content

The basepath for content objects in the PHPCR tree. This information is used to offer the correct subtree to select content documents.

media_basepath¶

type: string default: media

The basepath for media objects in the PHPCR tree. This information is used to offer the correct subtree to select media documents.

tenant_aware_router_class¶

type: string default: SWP\MultiTenancyBundle\Routing\TenantAwareRouter

The TenantAwareRouter service’s fully qualified class name to use.

classes¶

# app/config/config.yml
swp_multi_tenancy:
    # ..
    persistence:
        phpcr:
            # ..
            classes:
                tenant:
                    model: SWP\Component\MultiTenancy\Model\Tenant
                    repository: SWP\Bundle\MultiTenancyBundle\Doctrine\PHPCR\TenantRepository
                    factory: SWP\Component\MultiTenancy\Factory\TenantFactory
                    object_manager_name: ~
                organization:
                    model: SWP\Component\MultiTenancy\Model\Organization
                    repository: SWP\Bundle\MultiTenancyBundle\Doctrine\PHPCR\OrganizationRepository
                    factory: SWP\Bundle\MultiTenancyBundle\Factory\OrganizationFactory
                    object_manager_name: ~

tenant.model¶

type: string default: SWP\Component\MultiTenancy\Model\Tenant

The FQCN of the Tenant model class which is of type TenantInterface.

tenant.factory¶

type: string default: SWP\Component\MultiTenancy\Factory\TenantFactory

The FQCN of the Tenant Factory class.

tenant.repository¶

type: string default: SWP\Bundle\MultiTenancyBundle\Doctrine\PHPCR\TenantRepository

The FQCN of the Tenant Repository class.

tenant.object_manager_name¶

type: string default: null

The name of the object manager. If set to null it defaults to default. If PHPCR ODM persistence backend is enabled it will register swp.object_manager.tenant service which is an alias for “doctrine_phpcr.odm.default_document_manager”.

organization.model¶

type: string default: SWP\Component\MultiTenancy\Model\Organization

The FQCN of the Organization model class which is of type OrganizationInterface.

organization.factory¶

type: string default: SWP\Bundle\MultiTenancyBundle\Factory\OrganizationFactory

The FQCN of the Organization Factory class.

organization.repository¶

type: string default: SWP\Bundle\MultiTenancyBundle\Doctrine\PHPCR\OrganizationRepository

The FQCN of the Organization Repository class.

organization.object_manager_name¶

type: string default: null

The name of the object manager. If set to null it defaults to default. If PHPCR ODM persistence backend is enabled it will register swp.object_manager.organization service which is an alias for doctrine_phpcr.odm.default_document_manager.

orm¶

# app/config/config.yml
swp_multi_tenancy:
    # ..
    persistence:
        orm:
            enabled: true

enabled¶

type: boolean default: false

If true, ORM is enabled in the service container.

ORM can be enabled by multiple ways such as:

orm: ~ # use default configuration
# or
orm: true # straight way
# or
orm:
    enabled: true ... # or any other option under 'orm'

classes¶

# app/config/config.yml
swp_multi_tenancy:
    # ..
    persistence:
        orm:
            # ..
            classes:
                tenant:
                    model: SWP\Component\MultiTenancy\Model\Tenant
                    repository: SWP\Bundle\MultiTenancyBundle\Doctrine\ORM\TenantRepository
                    factory: SWP\Component\MultiTenancy\Factory\TenantFactory
                    object_manager_name: ~
                organization:
                    model: SWP\Component\MultiTenancy\Model\Organization
                    repository: SWP\Bundle\MultiTenancyBundle\Doctrine\ORM\OrganizationRepository
                    factory: SWP\Bundle\MultiTenancyBundle\Factory\OrganizationFactory
                    object_manager_name: ~

tenant.model¶

type: string default: SWP\Component\MultiTenancy\Model\Tenant

The FQCN of the Tenant model class which is of type TenantInterface.

tenant.factory¶

type: string default: SWP\Component\MultiTenancy\Factory\TenantFactory

The FQCN of the Tenant Factory class.

tenant.repository¶

type: string default: SWP\Bundle\MultiTenancyBundle\Doctrine\ORM\TenantRepository

The FQCN of the Tenant Repository class.

tenant.object_manager_name¶

type: string default: null

The name of the object manager. If set to null it defaults to default. If Doctrine ORM persistence backend is enabled it will register swp.object_manager.tenant service which is an alias for doctrine.orm.default_entity_manager.

organization.model¶

type: string default: SWP\Component\MultiTenancy\Model\Organization

The FQCN of the Organization model class which is of type OrganizationInterface.

organization.factory¶

type: string default: SWP\Bundle\MultiTenancyBundle\Factory\OrganizationFactory

The FQCN of the Organization Factory class.

organization.repository¶

type: string default: SWP\Bundle\MultiTenancyBundle\Doctrine\ORM\OrganizationRepository

The FQCN of the Organization Repository class.

organization.object_manager_name¶

type: string default: null

The name of the object manager. If set to null it defaults to default. If Doctrine ORM persistence backend is enabled it will register swp.object_manager.organization service which is an alias for doctrine.orm.default_entity_manager.

Tutorials¶

CMF RoutingBundle Integration¶

The SWPMultiTenancyBundle can be integrated with the CMF RoutingBundle. This section describes how to integrate the CMF Routing Bundle when using PHPCR ODM or ORM as persistence backends.

Note

If you don’t have CMF RoutingBundle installed, see the documentation on how to install and configure it.

Doctrine PHPCR ODM integration¶

Enable PHPCR persistence backend¶

Make sure the PHPCR persistence backend is enabled in CMF RoutingBundle.

You need to enable the PHPCR as a persistence backend for the SWPMultiTenancyBundle and fully integrate this bundle with the CMF RoutingBundle. Add the following lines to the configuration file:

# app/config/config.yml
swp_multi_tenancy:
    persistence:
        phpcr:
            # if true, PHPCR is enabled in the service container
            enabled: true
            # route base paths under which routes will be stored
            route_basepaths: ["routes"]
            # PHPCR content base path under which content will be stored
            content_basepath: "content"

Once the enabled property is set to true, PHPCRBasePathsInitializer, TenantAwareRouter and PrefixCandidates will be available in the application.

Enable TenantAwareRouter¶

To register the TenantAwareRouter service in the CMF RoutingBundle, add the following lines to your configuration file:

cmf_routing:
    chain:
        routers_by_id:
            # other routers
            # TenantAwareRouter with the priority of 150
            swp_multi_tenancy.tenant_aware_router: 150

The RoutingBundle example configuration can be found here:

cmf_routing:
    chain:
        routers_by_id:
            # default Symfony Router
            router.default: 200
            # TenantAwareRouter
            swp_multi_tenancy.tenant_aware_router: 150
            # CMF Dynamic Router
            cmf_routing.dynamic_router: 100
    dynamic:
        route_collection_limit: 100
        persistence:
            phpcr:
                enabled: true

Note

Please see the documentation of the CMF RoutingBundle for more details.

Doctrine ORM integration¶

Not implemented yet.

Creating a custom Tenant class¶

In this tutorial you will learn how to create custom Tenant class. For example, you want: to store info about tenant’s theme name and you want to make use of it in your project.

Note

This tutorial covers creating a custom Tenant class for PHPCR ODM.

This new class must implement TenantInterface which is provided by The MultiTenancy Component, or you can extend the default Tenant class, which is also part of the MultiTenancy Component.

Create an interface first which will require to implement theme name behaviour.

<?php

namespasce Acme\AppBundle\Document;

use SWP\Component\MultiTenancy\Model\TenantInterface as BaseTenantInterface;

interface ThemeAwareTenantInterface extends BaseTenantInterface
{
    /**
     * @return string
     */
    public function getThemeName();

    /**
     * @param string $themeName
     */
    public function setThemeName($themeName);
}

Let’s create a new Tenant class now:

<?php

namespasce Acme\AppBundle\Document;

use Acme\AppBundle\Document\ThemeAwareTenantInterface;
use SWP\Component\MultiTenancy\Model\Tenant as BaseTenant;

class Tenant extends BaseTenant implements ThemeAwareTenantInterface
{
    /**
     * @var string
     */
    protected $themeName;

    /**
     * {@inheritdoc}
     */
    public function getThemeName()
    {
        return $this->themeName;
    }

    /**
     * {@inheritdoc}
     */
    public function setThemeName($themeName)
    {
        $this->themeName = $themeName;
    }
}

Create a mapping file for your newly created document:

# src/Acme/AppBundle/Resources/config/doctrine/Document.Tenant.phpcr.yml

Acme\AppBundle\Document\Tenant:
referenceable: true
fields:
    themeName:
        type: string
        nullable: true

Once your class is created, you can now put its FQCN into the MultiTenancy bundle’s configuration:

# app/config/config.yml
swp_multi_tenancy:
    persistence:
        phpcr:
            enabled: true
            # ..
            classes:
                tenant:
                    model: Acme\AppBundle\Document\Tenant

From now on your custom class will be used and you will be able to make use of the $themeName property in your app.

Tip

See Configuration Reference for more configuration details.

That’s it, you can now refer to Acme\AppBundle\Document\Tenant to manage tenants in the PHPCR tree.

StorageBundle¶

This bundle provides tools to build a persistence-agnostic storage layer.

The Storage Component, which is used by this bundle, provides generic interfaces to create different types of repositories, factories, and drivers which can be used for various storages. So far, this bundle supports:

Doctrine ORM
Doctrine PHPCR ODM

By default this bundle uses the Doctrine ORM persistence backend. If you would like to make use of, for example, Doctrine PHPCR, you would need to install and configure DoctrinePHPCRBundle.

Prerequisites¶

This version of the bundle requires Symfony >= 2.8 and PHP version >=5.6.

Installation¶

Install the Bundle with Composer¶

In your project directory execute the following command to download the latest stable version:

1	composer require swp/storage-bundle

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

Enable the bundle and its dependency (DoctrineBundle) by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(
            // ...

            new Doctrine\Bundle\DoctrineBundle\DoctrineBundle(),
            new SWP\Bundle\StorageBundle\SWPStorageBundle(),
        );

        // ...
    }

    // ...
}

Note

All dependencies will be installed automatically. You will just need to configure the respective bundles if needed.

That’s it, the bundle is configured properly now!

Usage¶

How to Provide Model Classes for several Doctrine Implementations¶

When building a bundle that could be used not only with Doctrine ORM but also the CouchDB ODM, MongoDB ODM or PHPCR ODM, you should still only write one model class. The Doctrine bundles provide a compiler pass to register the mappings for your model classes. This bundle helps you easily register these mappings.

Note

See Symfony documentation for more details.

Let’s say you created a new bundle called ContentBundle and you want to register your model classes mappings. You could do that in a normal way by defining it in the configuration file as follows:

# app/config/config.yml

doctrine:
    # ..
    orm:
        entity_managers:
            default:
                # ..
                auto_mapping: false
                mappings:
                    AcmeContentBundle:
                        type: yml
                        prefix: Acme\ContentBundle\Model
                        dir: Resources/config/doctrine

If you want to have a flexible way to register your model classes mappings both for Doctrine ORM and PHPCR ODM using a single model class, extend your class with SWP\Bundle\StorageBundle\DependencyInjection\Bundle\Bundle from StorageBundle.

Note

SWPStorageBundle is able to load mappings in XML, YAML and annotation formats. By default YAML mappings files are loaded. You can change this by setting the $mappingFormat property to: protected $mappingFormat = BundleInterface::MAPPING_XML; (see SWP\Component\Storage\Bundle\BundleInterface) if you wish to load mapping files in XML format.

For example, to extend the AcmeContentBundle, you could use the code:

// src/Acme/ContentBundle/AcmeContentBundle.php
namespace Acme\ContentBundle\AcmeContentBundle;

use SWP\Bundle\StorageBundle\DependencyInjection\Bundle\Bundle;
use SWP\Bundle\StorageBundle\Drivers;
use Symfony\Component\DependencyInjection\ContainerBuilder;

class AcmeContentBundle extends Bundle
{
    /**
     * {@inheritdoc}
     */
    public function getSupportedDrivers()
    {
        return [
            Drivers::DRIVER_DOCTRINE_ORM,
            Drivers::DRIVER_DOCTRINE_PHPCR_ODM,
        ];
    }

    /**
     * {@inheritdoc}
     */
    public function getModelClassNamespace()
    {
        return 'Acme\\ContentBundle\\Model';
    }
}

Your bundle can now support multiple drivers. According to the example above, your Article model class namespace is specified in the getModelClassNamespace() method. Two drivers are configured: PHPCR and ORM. In this case, you can create a model class for PHPCR and ORM and extend the default Acme\ContentBundle\Model\Article class. You would then have different implementations for PHPCR and ORM using the same model class:

Acme\ContentBundle\ODM\PHPCR\Article should extend Acme\ContentBundle\Model\Article

Acme\ContentBundle\ORM\Article should extend Acme\ContentBundle\Model\Article

All that you need to do now is to place the mapping files for each model classes. In this case the Acme\ContentBundle\ODM\PHPCR\Article class mapping should be placed inside the Resources/config/doctrine-phpcr directory. The mappings for ORM classes should be placed inside the Resources/config/doctrine-orm directory.

Note

A reference to the directories where mapping files should be placed for model classes is generated automatically, based on the supported driver. In the case of PHPCR it will be Resources/config/doctrine-phpcr and in the case of Doctrine ORM it will be Resources/config/doctrine-orm. These directories should be created manually if they don’t exist already.

The getSupportedDrivers defines supported drivers by the ContentBundle e.g. PHPCR ODM, MongoDB ODM etc. You should use the SWP\Bundle\StorageBundle\Drivers class to specify the supported drivers, as shown in the example above.

The Drivers class provides drivers’ constants:

namespace SWP\Bundle\StorageBundle;

class Drivers
{
    const DRIVER_DOCTRINE_ORM = 'orm';
    const DRIVER_DOCTRINE_MONGODB_ODM = 'mongodb';
    const DRIVER_DOCTRINE_PHPCR_ODM = 'phpcr';
}

How to automatically register Services required by the configured Storage Driver¶

This bundle enables you to register required services on the basis of the configured storage driver, to standardize the definitions of registered services.

By default, this bundle registers:

repository services
factory services
object manager services
parameters

Based on the provided model class it will register the default factory and repository, where you will be able to create a new object based on the provided model class name, adding or removing objects from the repository.

Set Configuration for your Bundle¶

Let’s start from your bundle configuration, where you will need to specify the default configuration. In this example let’s assume you already have a bundle called ContentBundle and you want to have working services to manage your resources.

The default Configuration class would look something like this:

<?php

namespace Acme\ContentBundle\DependencyInjection;

use Symfony\Component\Config\Definition\Builder\TreeBuilder;
use Symfony\Component\Config\Definition\ConfigurationInterface;

class Configuration implements ConfigurationInterface
{
    /**
     * {@inheritdoc}
     */
    public function getConfigTreeBuilder()
    {
        $treeBuilder = new TreeBuilder();
        $treeBuilder->root('acme_content');

        return $treeBuilder;
    }
}

Now, let’s add the configuration for the Acme\ContentBundle\ODM\PHPCR\Article model class, for the PHPCR driver:

<?php

namespace Acme\ContentBundle\DependencyInjection;
// ..

use Acme\ContentBundle\ODM\PHPCR\Article;

class Configuration implements ConfigurationInterface
{
    /**
     * {@inheritdoc}
     */
    public function getConfigTreeBuilder()
    {
        $treeBuilder = new TreeBuilder();
        $treeBuilder->root('acme_content')
            ->children()
                ->arrayNode('persistence')
                    ->addDefaultsIfNotSet()
                    ->children()
                        ->arrayNode('phpcr')
                            ->addDefaultsIfNotSet()
                            ->canBeEnabled()
                            ->children()
                                ->arrayNode('classes')
                                    ->addDefaultsIfNotSet()
                                    ->children()
                                        ->arrayNode('article')
                                            ->addDefaultsIfNotSet()
                                            ->children()
                                                ->scalarNode('model')->cannotBeEmpty()->defaultValue(Article::class)->end()
                                                ->scalarNode('repository')->defaultValue(null)->end()
                                                ->scalarNode('factory')->defaultValue(null)->end()
                                                ->scalarNode('object_manager_name')->defaultValue(null)->end()
                                            ->end()
                                        ->end()
                                    ->end()
                                ->end()
                            ->end()
                        ->end() // phpcr
                    ->end()
                ->end()
            ->end();

        return $treeBuilder;
    }
}

Note

The repository, factory and object_manager_name nodes are configured to use null as the default value. It means that the default factory, repository and object manager services will be registered in the container.

Register configured classes in your Extension class¶

Now that you have the configuration defined, it is time to register those classes using the Extension class in your bundle. By default, this class is generated inside the DependencyInjection folder in every Symfony Bundle.

In this ContentBundle example it will be located under the namespace Acme\ContentBundle\DependencyInjection. The fully qualified class name will be Acme\ContentBundle\DependencyInjection\AcmeContentExtension.

You need to extend this class by the SWP\Bundle\StorageBundle\DependencyInjection\Extension\Extension class, which will give you access to register configured classes needed by the storage. The registerStorage method will do the whole magic for you. See the code below:

<?php

namespace Acme\ContentBundle\DependencyInjection;

// ..
use SWP\Bundle\StorageBundle\Drivers;
use SWP\Bundle\StorageBundle\DependencyInjection\Extension\Extension;
use Symfony\Component\DependencyInjection\ContainerBuilder;
use Symfony\Component\Config\FileLocator;
use Symfony\Component\DependencyInjection\Loader;

class AcmeContentExtension extends Extension
{
    /**
     * {@inheritdoc}
     */
    public function load(array $configs, ContainerBuilder $container)
    {
        $config = $this->processConfiguration(new Configuration(), $configs);
        $loader = new Loader\YamlFileLoader($container, new FileLocator(__DIR__.'/../Resources/config'));
        $loader->load('services.yml');

        if ($config['persistence']['phpcr']['enabled']) {
            $this->registerStorage(Drivers::DRIVER_DOCTRINE_PHPCR_ODM, $config['persistence']['phpcr'], $container);
        }
    }
}

If the PHPCR persistence backend is enabled, it will register the following services in the container:

Service ID	Class name
swp.factory.article	SWP\Bundle\StorageBundle\Factory\Factory
swp.repository.article.class \| Acme\ContentBundle\PHPCR\Article
swp.repository.article	SWP\Bundle\StorageBundle\Doctrine\ODM\PHPCR\DocumentRepository

together with all parameters:

Parameter Name	Value
swp.factory.article.class	SWP\Bundle\StorageBundle\Factory\Factory
swp.model.article.class	Acme\ContentBundle\PHPCR\Article
swp.repository.article.class	SWP\Bundle\StorageBundle\Doctrine\ODM\PHPCR\DocumentRepository

If your configuration supports Doctrine ORM instead of PHPCR, the default service definitions would be:

Service ID	Class name
swp.factory.article	SWP\Bundle\StorageBundle\Factory\Factory
swp.object_manager.article	alias for “doctrine.orm.default_entity_manager”
swp.repository.article	SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository

And all parameters in the container would look like:

Parameter Name	Value
swp.factory.article.class	SWP\Bundle\StorageBundle\Factory\Factory
swp.model.article.class	Acme\ContentBundle\ORM\Article
swp.repository.article.class	SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository

You could then access parameters from the container, as visible below:

<?php
//..
$className = $container->getParameter('swp.model.article.class');
var_dump($className); // will return Acme\ContentBundle\PHPCR\Article

Now, register all classes in the configuration file:

# app/config/config.yml
swp_content:
    persistence:
        phpcr: true

The above configuration is equivalent to:

# app/config/config.yml
swp_content:
    persistence:
        phpcr:
            enabled: true
            classes:
                article:
                    model: Acme\ContentBundle\ODM\PHPCR\Article
                    factory: ~
                    repository: ~
                    object_manager_name: ~

How to create and use custom repository service for your model¶

For some use cases you would need to implement your own methods in the repository, like findOneBySlug() or findAllArticles(). It’s very easy!

You need to create your custom implementation for the repository. In this example you will create a custom repository for the Article model class and Doctrine PHPCR persistence backend.

Firstly, you need to create your custom repository interface. Let’s name it ArticleRepositoryInterface and extend it by the SWP\Component\Storage\Repository\RepositoryInterface interface:

<?php

namespace Acme\ContentBundle\PHPCR;

use Acme\ContentBundle\Model\ArticleInterface;
use SWP\Component\Storage\Repository\RepositoryInterface;

interface ArticleRepositoryInterface extends RepositoryInterface
{
    /**
     * Find one article by slug.
     *
     * @param string $slug
     *
     * @return ArticleInterface
     */
    public function findOneBySlug($slug);

    /**
     * Find all articles.
     *
     * @return mixed
     */
    public function findAllArticles();
}

Secondly, you need to create your custom repository class. Let’s name it ArticleRepository and implement the ArticleRepositoryInterface interface:

<?php

namespace Acme\ContentBundle\PHPCR;

use Acme\ContentBundle\Model\ArticleRepositoryInterface;
use SWP\Bundle\StorageBundle\Doctrine\ODM\PHPCR\DocumentRepository;

class ArticleRepository extends DocumentRepository implements ArticleRepositoryInterface
{
    /**
     * {@inheritdoc}
     */
    public function findOneBySlug($slug)
    {
        return $this->findOneBy(['slug' => $slug]);
    }

    /**
     * {@inheritdoc}
     */
    public function findAllArticles()
    {
        return $this->createQueryBuilder('o')->getQuery();
    }
}

Note

If you want to create a custom repository for the Doctrine ORM persistence backend, you need to extend your custom repository class by the SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository class.

The last step is to add your custom repository to the configuration file:

# app/config/config.yml
swp_content:
    persistence:
        phpcr:
            enabled: true
            classes:
                article:
                    model: Acme\ContentBundle\ODM\PHPCR\Article
                    factory: ~
                    repository: Acme\ContentBundle\PHPCR\ArticleRepository
                    object_manager_name: ~

Note

Alternatively, you could add it directly in your Configuration class.

Note

You can change repository class by simply changing your bundle configuration, without needing to change the code.

How to create and use custom factory service for your model¶

You may need to have a different way of creating objects than the default way of doing it. Imagine you need to create an Article object with the route assigned by default.

Note

In this example you will create a custom factory for your Article object and Doctrine PHPCR persistence backend.

Let’s create a custom interface for your factory. Extend your custom class by the SWP\Component\Storage\Factory\FactoryInterface class:

<?php

namespace Acme\ContentBundle\Factory;

use SWP\Bundle\ContentBundle\Model\ArticleInterface;
use SWP\Component\Bridge\Model\PackageInterface;
use SWP\Component\Storage\Factory\FactoryInterface;

interface ArticleFactoryInterface extends FactoryInterface
{
    /**
     * Create a new object with route.
     *
     * @param string $route
     *
     * @return ArticleInterface
     */
    public function createWithRoute($route);
}

Create the custom Article factory class:

<?php

namespace Acme\ContentBundle\Factory;

use SWP\Component\Storage\Factory\FactoryInterface;

class ArticleFactory implements ArticleFactoryInterface
{
    /**
     * @var FactoryInterface
     */
    private $baseFactory;

    /**
     * ArticleFactory constructor.
     *
     * @param FactoryInterface $baseFactory
     */
    public function __construct(FactoryInterface $baseFactory)
    {
        $this->baseFactory = $baseFactory;
    }

    /**
     * {@inheritdoc}
     */
    public function create()
    {
        return $this->baseFactory->create();
    }

    /**
     * {@inheritdoc}
     */
    public function createWithRoute($route)
    {
        $article = $this->create();
        // ..
        $article->setRoute($route);

        return $article;
    }
}

Create a compiler pass to override the default Article factory class with your custom factory on container compilation:

<?php

namespace Acme\ContentBundle\DependencyInjection\Compiler;

use SWP\Component\Storage\Factory\Factory;
use Symfony\Component\DependencyInjection\ContainerBuilder;
use Symfony\Component\DependencyInjection\Compiler\CompilerPassInterface;
use Symfony\Component\DependencyInjection\Definition;
use Symfony\Component\DependencyInjection\Parameter;

class RegisterArticleFactoryPass implements CompilerPassInterface
{
    /**
     * {@inheritdoc}
     */
    public function process(ContainerBuilder $container)
    {
        if (!$container->hasDefinition('swp.factory.article')) {
            return;
        }

        $baseDefinition = new Definition(
            Factory::class,
            [
                new Parameter('swp.model.article.class'),
            ]
        );

        $articleFactoryDefinition = new Definition(
            $container->getParameter('swp.factory.article.class'),
            [
                $baseDefinition,
            ]
        );

        $container->setDefinition('swp.factory.article', $articleFactoryDefinition);
    }
}

Don’t forget to register your new compiler pass in your Bundle class (AcmeContentBundle):

<?php

use Acme\ContentBundle\DependencyInjection\Compiler\RegisterArticleFactoryPass;
// ..

/**
 * {@inheritdoc}
 */
public function build(ContainerBuilder $container)
{
    parent::build($container);
    $container->addCompilerPass(new RegisterArticleFactoryPass());
}

The last thing required to make use of your new factory service is to add it to the configuration file, under the factory node:

# app/config/config.yml
swp_content:
    persistence:
        phpcr:
            enabled: true
            classes:
                article:
                    model: Acme\ContentBundle\ODM\PHPCR\Article
                    factory: Acme\ContentBundle\Factory\ArticleFactory
                    repository: ~
                    object_manager_name: ~

Note

Alternatively, you could add it directly in your Configuration class.

You would then be able to use the factory like so:

$article = $this->get('swp.factory.article')->createWithRoute('some-route');
// or create flat object
$article = $this->get('swp.factory.article')->create();

Note

You can change factory class by simply changing your bundle configuration, without needing to change the code.

Configuring object manager for your model¶

As you can see, there is the object_manager_name option in the Configuration class, which is the default Object Manager (Contract for a Doctrine persistence layer) name.

In the case of Doctrine ORM it’s doctrine.orm.default_entity_manager, in PHPCR it’s doctrine_phpcr.odm.default_document_manager.

If you set this option to be, for example, test the doctrine.orm.test_entity_manager object manager service’s id will be used. Of course this new test document, in the case of PHPCR, should be first configured in the Doctrine PHPCR Bundle as described in the bundle documentation on multiple document managers. For Doctrine ORM it should be configured as shown in the Doctrine ORM Bundle documentation on multiple entity managers.

The possibility of defining a default Object Manager for a Doctrine persistence layer, and making use of it in the registered repositories and factories in your Bundle, is very useful in case you are using different databases or even different sets of entities.

Note

Factories and repositories are defined as a services in Symfony container to have better flexibility of use.

Resolve target entities¶

This chapter is strictly related to How to Define Relationships with Abstract Classes and Interfaces so please read it first.

This functionality allows you to define relationships between different entities without making them hard dependencies. All you need to do is to define interface node in your bundle’s Configuration class.

See example below:

<?php

namespace Acme\Bundle\CoreBundle\DependencyInjection;
// ..

use Acme\Component\MultiTenancy\Model\Tenant;
use Acme\Component\MultiTenancy\Model\TenantInterface;
use Acme\Component\MultiTenancy\Model\Organization;
use Acme\Component\MultiTenancy\Model\OrganizationInterface;

class Configuration implements ConfigurationInterface
{
    /**
     * {@inheritdoc}
     */
    public function getConfigTreeBuilder()
    {
        $treeBuilder = new TreeBuilder();
        $treeBuilder->root('acme_core')
            ->children()
                ->arrayNode('persistence')
                    ->addDefaultsIfNotSet()
                    ->children()
                        ->arrayNode('phpcr')
                            ->addDefaultsIfNotSet()
                            ->canBeEnabled()
                            ->children()
                                ->arrayNode('classes')
                                    ->addDefaultsIfNotSet()
                                    ->children()
                                        ->arrayNode('tenant')
                                            ->addDefaultsIfNotSet()
                                            ->children()
                                                ->scalarNode('model')->cannotBeEmpty()->defaultValue(Tenant::class)->end()
                                                ->scalarNode('interface')->cannotBeEmpty()->defaultValue(TenantInterface::class)->end()
                                                ->scalarNode('repository')->defaultValue(null)->end()
                                                ->scalarNode('factory')->defaultValue(null)->end()
                                                ->scalarNode('object_manager_name')->defaultValue(null)->end()
                                            ->end()
                                        ->end()
                                        ->arrayNode('organization')
                                            ->addDefaultsIfNotSet()
                                            ->children()
                                                ->scalarNode('model')->cannotBeEmpty()->defaultValue(Organization::class)->end()
                                                ->scalarNode('interface')->cannotBeEmpty()->defaultValue(OrganizationInterface::class)->end()
                                                ->scalarNode('repository')->defaultValue(null)->end()
                                                ->scalarNode('factory')->defaultValue(null)->end()
                                                ->scalarNode('object_manager_name')->defaultValue(null)->end()
                                            ->end()
                                        ->end()
                                    ->end()
                                ->end()
                            ->end()
                        ->end() // phpcr
                    ->end()
                ->end()
            ->end();

        return $treeBuilder;
    }
}

In this case you will be able to specify your interface for your model via config file:

# app/config/config.yml
swp_content:
    persistence:
        phpcr:
            enabled: true
            classes:
                tenant:
                    model: Acme\Bundle\CoreBundle\Model\Tenant # extends default Acme\Component\MultiTenancy\Model\Tenant class
                    interface: ~
                    # ..
                organization:
                    model: Acme\Bundle\CoreBundle\Model\Organization # extends default Acme\Component\MultiTenancy\Model\Organization class
                    interface: ~
                    # ..

Now, no matter which model (implementing for example Acme\Component\MultiTenancy\Model\OrganizationInterface): you will use in your bundle’s configuration above, the interface will be automatically resolved to defined entity and will be used by your mapping file without a need to change any extra code or configuration setup.

The above is equivalent to if the Tenant has a relation to Organization and vice versa.

# app/config/config.yml
doctrine:
    # ...
    orm:
        # ...
        resolve_target_entities:
            Acme\Component\MultiTenancy\Model\OrganizationInterface: Acme\Bundle\CoreBundle\Model\Organization
            Acme\Component\MultiTenancy\Model\TenantInterface: Acme\Bundle\CoreBundle\Model\Tenant

In this example above every time you will want to change your model inside your bundle’s configuration you would also need to care about the Doctrine config as the specified entity will not change automatically to a new one which was defined in bundle’s config.

Inheritance Mapping¶

By default every entity inside bundle should be mapped as Mapped superclass. This bundle helps you manage and simplify inheritance mapping in case you want to use default mapping or extend it. In this case the following applies:

If you do not configure your custom class, the default mapped superclasses become entites.
Otherwise they become mapped superclasses and move the conflicting mappings (these which you cannot normally configure on mapped superclass) to your class mapping. For example, you do not need anymore to map Organization -> Tenants inside your custom class, it is copied transparently from the bundle.
It also works on all levels, so you can cleanly override the core bundle models! If you configure other class than core one, your entity will be used and the core model will remain mapped superclass.

Note

This feature and its description has been ported from Sylius project. See related issue.

BridgeBundle¶

This bundle provides tools which help you to integrate Superdesk data with Superdesk Web Publisher.

The Bridge Component, which is used by this bundle, provides a generic interface to create different type of validators, data transformers, and models, which in turn help to pull and process data from Superdesk.

Prerequisites¶

This version of the bundle requires Symfony >= 2.8 and PHP version >=5.6.

Installation¶

Install the Bundle with Composer¶

In your project directory execute the following command to download the latest stable version:

composer require swp/bridge-bundle jms/serializer-bundle swp/jms-serializer-bridge

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

Enable the bundle by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(
            // ...
            new JMS\SerializerBundle\JMSSerializerBundle(),
            new SWP\Bundle\BridgeBundle\SWPBridgeBundle()
        );

        // ...
    }

    // ...
}

Note

All dependencies will be installed automatically. You will just need to configure the respective bundles if needed.

That’s it, the bundle is configured properly now!

Usage¶

Using Validator Chain Service¶

The Validator Chain service is used to register all validators with a tag validator.http_push_validator.

Usage:

// ...
use Symfony\Component\HttpFoundation\Response;

public function indexAction()
{
    $value = 'some value';
    $result = $this->get('swp_bridge.http_push.validator_chain')->isValid($value);

    return new Response($result);
}

How to Create and Register Custom Validators¶

Validators are used to process incoming request content and validate it against the specific schema. Read more about it in the Bridge component documentation (in the Usage section).

Creating the Validator Class¶

A new Validator has to implement the SWP\Component\Bridge\Validator\ValidatorInterface and SWP\Component\Bridge\Validator\ValidatorOptionsInterface interfaces.

CustomValidator class example:

<?php

namespace Acme\DemoBundle\Validator;

use SWP\Component\Bridge\Validator\ValidatorInterface;
use SWP\Component\Bridge\Validator\ValidatorOptionsInterface

final class CustomValidator implement ValidatorInterface, ValidatorOptionsInterface
{
    /**
     * @var string
     */
    private $schema = 'custom schema';

    /**
     * {@inheritdoc}
     */
    public function isValid($data)
    {
        // custom validation here
    }

    /**
     * {@inheritdoc}
     */
    public function getSchema()
    {
        return $this->schema;
    }

    /**
     * {@inheritdoc}
     */
    public function getFormat()
    {
        return 'custom';
    }
}

Configuring the Validator¶

To register your new validator, simply add a definition to your services file and tag it with a special name: validator.http_push_validator:

# Resources/config/services.yml
acme_my_custom_validator:
    class: 'Acme\DemoBundle\Validator\CustomValidator'
    tags:
        - { name: validator.http_push_validator, alias: http_push.custom }

Note

You can use the SWP\Component\Bridge\Validator\JsonValidator abstract class if you wish to create custom JSON validator.

Using Transformer Chain Service¶

Transformer Chain service is used to register all transformers with a tag transformer.http_push_transformer.

Usage:

// ...

public function indexAction()
{
    $value = 'some value';
    $result = $this->get('swp_bridge.http_push.transformer_chain')->transform($value);
    $result = $this->get('swp_bridge.http_push.transformer_chain')->reverseTransform($value);
}

How to Create and Register Custom Data Transformers¶

Data transformers are used to transform one value/object into another. Read more about it in the Bridge component documentation (in the Usage section).

Creating the Data Transformer Class¶

To create a new Data Transformer, your new class should implement the SWP\Component\Bridge\Transformer\DataTransformerInterface interface.

CustomValidator class example:

<?php

namespace Acme\DemoBundle\Transformer;

use Acme\DemoBundle\Model\Custom;
use SWP\Component\Bridge\Exception\MethodNotSupportedException;
use SWP\Component\Bridge\Exception\TransformationFailedException;
use SWP\Component\Bridge\Validator\ValidatorInterface;
use SWP\Component\Common\Serializer\SerializerInterface;

final class JsonToObjectTransformer implements DataTransformerInterface
{
    /**
     * @var SerializerInterface
     */
    private $serializer;

    /**
     * @var ValidatorInterface
     */
    private $validatorChain;

    /**
     * JsonToPackageTransformer constructor.
     *
     * @param SerializerInterface $serializer
     * @param ValidatorInterface  $validatorChain
     */
    public function __construct(SerializerInterface $serializer, ValidatorInterface $validatorChain)
    {
        $this->serializer = $serializer;
        $this->validatorChain = $validatorChain;
    }

    /**
     * {@inheritdoc}
     */
    public function transform($json)
    {
        if (!$this->validatorChain->isValid($json)) {
            throw new TransformationFailedException('None of the chained validators were able to validate the data!');
        }

        return $this->serializer->deserialize($json, Custom::class, 'json');
    }

    /**
     * {@inheritdoc}
     */
    public function reverseTransform($value)
    {
        throw new MethodNotSupportedException('reverseTransform');
    }
}

Configuring the Data Transformer¶

To register your new Data Transformer, simply add a definition to your services file and tag it with a special name: transformer.http_push_transformer:

# Resources/config/services.yml
acme_my_custom_transformer:
    class: 'Acme\DemoBundle\Transformer\CustomTransformer'
    arguments:
        - '@swp.serializer'
        - '@swp_bridge.http_push.validator_chain'
    tags:
        - { name: transformer.http_push_transformer, alias: transformer.json_to_object }

Enabling a separate Monolog channel for Validators¶

It is possible to enable a separate Monolog channel to which all validators related logs will be forwarded. An example log entry might be logged when the incoming payload can not be validated properly.. You could have then a separate log file for which will be usually saved under the directory app/logs/ in your application and will be named, for example: swp_validators_<env>.log. By default, a separate channel is disabled. You can enable it by adding an extra channel in your Monolog settings (in one of your configuration files):

# app/config/config.yml
monolog:
    handlers:
        swp_validators:
            level:    debug
            type:     stream
            path:     '%kernel.logs_dir%/swp_validators_%kernel.environment%.log'
            channels: swp_validators

For more details see the Monolog documentation.

ContentBundle¶

This bundle provides functionality for defining the content of a site managed with Superdesk Web Publisher.

Usage¶

Rules to assign routes and/or templates to articles¶

With the API, it is possible to create and manage rules in order to assign a route and/or a template to content received in a package from a provider based on the metadata in that package.

The rules themselves are to be written in Symfony’s expression language, documentation for which can be found here: http://symfony.com/doc/current/components/expression_language/syntax.html

The article document generated from the package can be referenced directly in the rule. So, here is an example of a rule:

'article.getMetadataByKey("var_name") matches "/regexExp/"'
# ..
'article.getMetadataByKey("urgency") > 1'
# or
'article.getMetadataByKey("lead") matches "/Text/"'
# etc.

A priority can also be assigned to the rule. This is simply an integer. The rules are ordered by their priority (the greater the value, the higher the priority) before searching for the first one which matches.

The route to be assigned is identified by its id, for example:

1	'articles/features'

If a template name parameter (templateName) is given with the rule, this template will be assigned to the article instead of the one in the route.

Every rule is automatically processed when the content is pushed to api/v1/content/push API endpoint. The SWP\Bundle\ContentBundle\EventListener\ProcessArticleRulesSubscriber subscriber subscribes to SWP\Bundle\ContentBundle\ArticleEvents::PRE_CREATE event and runs the processing when needed.

Pushing Content to Web Publisher¶

Content Push API Endpoint¶

Content Push API Endpoint

Method	URL
POST	/api/v1/content/push

Resource details

Response format	JSON
Authentication	No

Example Request

1	POST https://<tenant_name>.domain.com/api/v1/content/push

Example Response

{
    "status": "OK"
}

Response status code is 201.

Which content formats are allowed to be pushed?¶

The Superdesk Web Publisher can receive content of different formats, by default the IPTC’s ninjs format is supported.

ninjs - standardizes the representation of news in JSON - a lightweight, easy-to-parse, data interchange format.

In Superdesk Web Publisher we use an extension of the standard IPTC’s ninjs format (to validate incoming request’s content), which is extended by some additional fields:

uri was replaced by guid: uri should be the resource identifier on the web but since the item was not published yet it can’t be determined at this point
added priority field
added service field
added slugline field
added keywords field
associations dictionary may contain entire items like in this ninjs example: http://dev.iptc.org/ninjs-Examples-3

By default, Superdesk Web Publisher is meant to work with our in-house content creation software called Superdesk which uses the above extension of ninjs format to store the data. That’s why in Web Publisher the incoming content is being validated in the same format.

In the future we could also support other IPTC formats, like: NewsML-G2, NITF etc. So you would be able to push content to Web Publisher in format that fits you best.

Superdesk ninjs extension schema used by Web Publisher:

{
    "$schema": "http://json-schema.org/draft-03/schema#",
    "id" : "http://www.iptc.org/std/ninjs/ninjs-schema_1.1.json#",
    "type" : "object",
    "title" : "IPTC ninjs - News in JSON - version 1.1 (approved, 2014-03-12) / document revision of 2014-11-15: geometry_* moved under place",
    "description" : "A news item as JSON object -- copyright 2014 IPTC - International Press Telecommunications Council - www.iptc.org - This document is published under the Creative Commons Attribution 3.0 license, see  http://creativecommons.org/licenses/by/3.0/  $$comment: as of 2014-03-13 ",
    "additionalProperties" : false,
    "patternProperties" : {
        "^description_[a-zA-Z0-9_]+" : {
            "description" : "A free-form textual description of the content of the item. (The string appended to description_ in the property name should reflect the format of the text)",
            "type" : "string"
        },
        "^body_[a-zA-Z0-9_]+" : {
            "description" : "The textual content of the news object. (The string appended to body_ in the property name should reflect the format of the text)",
            "type" : "string"
        }
    },
    "properties" : {
        "guid" : {
            "description" : "The identifier for this news object",
            "type" : "string",
            "format" : "guid",
            "required" : true
        },
        "type" : {
            "description" : "The generic news type of this news object",
            "type" : "string",
            "enum" : ["text", "audio", "video", "picture", "graphic", "composite"]
        },
        "slugline" : {
            "description" : "The slugline",
            "type" : "string",
            "required" : true
        },
        "mimetype" : {
            "description" : "A MIME type which applies to this news object",
            "type" : "string"
        },
        "representationtype" : {
            "description" : "Indicates how complete this representation of a news item is",
            "type" : "string",
            "enum" : ["complete", "incomplete"]
        },
        "profile" : {
            "description" : "An identifier for the kind of content of this news object",
            "type" : "string"
        },
        "version" : {
            "description" : "The version of the news object which is identified by the uri property",
            "type" : "string"
        },
        "versioncreated" : {
            "description" : "The date and time when this version of the news object was created",
            "type" : "string",
            "format" : "date-time"
        },
        "embargoed" : {
            "description" : "The date and time before which all versions of the news object are embargoed. If absent, this object is not embargoed.",
            "type" : "string",
            "format" : "date-time"
        },
        "pubstatus" : {
            "description" : "The publishing status of the news object, its value is *usable* by default.",
            "type" : "string",
            "enum" : ["usable", "withheld", "canceled"]
        },
        "urgency" : {
            "description" : "The editorial urgency of the content from 1 to 9. 1 represents the highest urgency, 9 the lowest.",
            "type" : "number"
        },
        "priority" : {
            "description" : "The editorial priority of the content from 1 to 9. 1 represents the highest priority, 9 the lowest.",
            "type" : "number"
        },
        "copyrightholder" : {
            "description" : "The person or organisation claiming the intellectual property for the content.",
            "type" : "string"
        },
        "copyrightnotice" : {
            "description" : "Any necessary copyright notice for claiming the intellectual property for the content.",
            "type" : "string"
        },
        "usageterms" : {
            "description" : "A natural-language statement about the usage terms pertaining to the content.",
            "type" : "string"
        },
        "language" : {
            "description" : "The human language used by the content. The value should follow IETF BCP47",
            "type" : "string"
        },
        "service" : {
            "description" : "A service e.g. World Photos, UK News etc.",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of a service",
                        "type" : "string"
                    },
                    "code" : {
                        "description": "The code for the service in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "person" : {
            "description" : "An individual human being",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of a person",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the person",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the person",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the person in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "organisation" : {
            "description" : "An administrative and functional structure which may act as as a business, as a political party or not-for-profit party",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of the organisation",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the organisation",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the organisation",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the organisation in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    },
                    "symbols" : {
                        "description" : "Symbols used for a finanical instrument linked to the organisation at a specific market place",
                        "type" : "array",
                        "items" : {
                            "type" : "object",
                            "additionalProperties" : false,
                            "properties" : {
                                "ticker" : {
                                    "description" : "Ticker symbol used for the financial instrument",
                                    "type": "string"
                                },
                                "exchange" : {
                                    "description" : "Identifier for the marketplace which uses the ticker symbols of the ticker property",
                                    "type" : "string"
                                }
                            }
                        }
                    }
                }
            }
        },
        "place" : {
            "description" : "A named location",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "patternProperties" : {
                    "^geometry_[a-zA-Z0-9_]+" : {
                        "description" : "An object holding geo data of this place. Could be of any relevant geo data JSON object definition.",
                        "type" : "object"
                    }
                },
                "properties" : {
                    "name" : {
                        "description" : "The name of the place",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the place",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the place",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "qcode" : {
                        "description": "The code for the place in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    },
                    "state" : {
                        "description" : "The state for the place",
                        "type" : "string"
                    },
                    "group" : {
                        "description" : "The place group",
                        "type" : "string"
                    },
                    "name" : {
                        "description" : "The place name",
                        "type" : "string"
                    },
                    "country" : {
                        "description" : "The country name",
                        "type" : "string"
                    },
                    "world_region" : {
                        "description" : "The world region",
                        "type" : "string"
                    }
                }
            }
        },
        "subject" : {
            "description" : "A concept with a relationship to the content",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of the subject",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the subject",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the subject",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the subject in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "event" : {
            "description" : "Something which happens in a planned or unplanned manner",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of the event",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the event",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the event",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the event in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "object" : {
            "description" : "Something material, excluding persons",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of the object",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the object",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the object",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the object in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "byline" : {
            "description" : "The name(s) of the creator(s) of the content",
            "type" : "string"
        },
        "headline" : {
            "description" : "A brief and snappy introduction to the content, designed to catch the reader's attention",
            "type" : "string"
        },
        "located" : {
            "description" : "The name of the location from which the content originates.",
            "type" : "string"
        },
        "keywords": {
            "description" : "Content keywords",
            "type" : "array"
        },
        "renditions" : {
            "description" : "Wrapper for different renditions of non-textual content of the news object",
            "type" : "object",
            "additionalProperties" : false,
            "patternProperties" : {
                "^[a-zA-Z0-9]+" : {
                    "description" : "A specific rendition of a non-textual content of the news object.",
                    "type" : "object",
                    "additionalProperties" : false,
                    "properties" : {
                        "href" : {
                            "description" : "The URL for accessing the rendition as a resource",
                            "type" : "string",
                            "format" : "uri"
                        },
                        "mimetype" : {
                            "description" : "A MIME type which applies to the rendition",
                            "type" : "string"
                        },
                        "title" : {
                            "description" : "A title for the link to the rendition resource",
                            "type" : "string"
                        },
                        "height" : {
                            "description" : "For still and moving images: the height of the display area measured in pixels",
                            "type" : "number"
                        },
                        "width" : {
                            "description" : "For still and moving images: the width of the display area measured in pixels",
                            "type" : "number"
                        },
                        "sizeinbytes" : {
                            "description" : "The size of the rendition resource in bytes",
                            "type" : "number"
                        }
                    }
                }
            }
        },
        "associations" : {
            "description" : "Content of news objects which are associated with this news object.",
            "type" : "object",
            "additionalProperties" : false,
            "patternProperties" : {
                "^[a-zA-Z0-9]+" :  { "$ref": "http://www.iptc.org/std/ninjs/ninjs-schema_1.0.json#" }
            }
        }
    }
}

How Content is Pushed to Web Publisher?¶

Making a POST request to api/v1/content/push API endpoint, you can push whatever content you want in request’s payload. See Content Push API Endpoint section.

The below diagram shows the data flow from the moment of their receipt by the content/push API endpoint until it’s saved as a resulting Article object in the persistence backend.

As you can see the request is being sent first, then the request’s content is validated using Web Publisher validators (see validators Usage section for more info about validators).

Once the content passes the validation (i.e. submitted content is for example in ninjs format), a respective transformer/parser (see Using Transformer Chain Service for more details) transforms (according to submitted content) the incoming data to format which is understandable by Web Publisher, i.e. Package and Item objects which are reflecting the submitted content.

Once this is done, the converted request’s content is being persisted in Web Publisher persistence backend as a representation of Package and Item objects.

The last step is converting already persisted Package and Item objects to Article object which is used by Web Publisher internally and on which every operation is being made.

How the article’s slug is being generated?¶

The article’s slug is being generated from the slugline field, if it is not empty, else the headline property’s value (according to ninjs IPTC format) is used to populate the article`s slug.

Warning

If the slugline property in incoming data is missing or is empty, the article’s slug will be generated from the headline, which means if you would want to change the headline and submit content again, a new article will be created instead as the article’s slug will be generated from the headline field.

What happens when I want to change article’s title?¶

You can change existing article’s title in the content that you are sending to Web Publisher. Let’s say we have a simple text item or package in ninjs format, as it is defined according to Which content formats are allowed to be pushed?.

Once the item/package headline is changed and the whole content is pushed to Web Publisher again, the article’s title will be updated automatically.

What happens when I want to change article’s slug?¶

If an article already exists and you want to change the article’s slug, the content which you used to create the article for the first time should be re-sent with modified slugline property. Once you change the slugline property’s value and submit it again to Web Publisher, a new article will be created.

How do I auto-publish an article?¶

In some cases, you will need to publish an article automatically, without additional action. In this bundle a special logic has been implemented which is responsible for the auto publishing articles, which is based on the rules. You can read more about rules in sections: Rules to assign routes and/or templates to articles, RuleBundle - Usage section.

All you need to do in order to auto-publish your articles, you need to first add a rule. If the article will match the rule, it will be auto published.

Create a new rule:

$ curl 'http://localhost/api/v1/rules/' -H 'Content-Type: application/x-www-form-urlencoded' --data 'rule%5Bpriority%5D=1&rule%5Bexpression%5D=article.getMetadataByKey(%22located%22)+matches+%22%2FSydney%2F%22&rule%5Bconfiguration%5D%5B0%5D%5Bkey%5D=published&rule%5Bconfiguration%5D%5B0%5D%5Bvalue%5D=true' --compressed

Submitted rule’s expression:

1	article.getMetadataByKey("located") matches "/Sydney/"

Submitted rule’s configuration:

rule[configuration][0][key]: published
rule[configuration][0][value]: true

It means that if the above rule’s expression matches any article, it will apply the configuration to it - in this case it will publish article.

RuleBundle¶

This bundle provides a simple business rules engine for Symfony applications.

The Rule Component, which is used by this bundle, provides a generic interface to create different type of rule applicators, models etc., which in turn help to create powerful business rules engine.

It means you can create your own rules and apply them to whatever objects you need:

# Get the special price if
user.getGroup() in ['good_customers', 'collaborator']

# Promote article to the homepage when
article.commentCount > 100 and article.category not in ["misc"]

# Send an alert when
product.stock < 15

Prerequisites¶

This version of the bundle requires Symfony >= 2.8 and PHP version >=5.6.

Installation¶

Install the Bundle with Composer¶

In your project directory execute the following command to download the latest stable version:

1	composer require swp/rule-bundle

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

Enable the bundle by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(
            new Burgov\Bundle\KeyValueFormBundle\BurgovKeyValueFormBundle(),
            new SWP\Bundle\RuleBundle\SWPStorageBundle()
            // ...
            new SWP\Bundle\RuleBundle\SWPRuleBundle()
        );

        // ...
    }

    // ...
}

Note

All dependencies will be installed automatically. You will just need to configure the respective bundles if needed.

Configure the bundle¶

YAML

# app/config/config.yml
swp_rule:
    persistence:
        orm:
            # if true, ORM is enabled as a persistence backend
            enabled: true

Note

By default this bundle supports only Doctrine ORM as a persistence backend.

Update Database Schema¶

Run the following command:

1	$ php bin/console doctrine:schema:update --force

That’s it, the bundle is configured properly now!

Usage¶

How to Create and Register Custom Rule Applicator¶

Rule applicators are used to apply given rule’s configuration to an object. Read more about it in the Rule component documentation (in the Usage section).

Creating the Rule Applicator Class¶

A new Rule Applicator has to implement the SWP\Component\Rule\Applicator\RuleApplicatorInterface interface.

ArticleRuleApplicator class example:

<?php

namespace Acme\DemoBundle\Applicator;
// ..
use Psr\Log\LoggerInterface;
use SWP\Bundle\ContentBundle\Model\ArticleInterface;
use SWP\Bundle\ContentBundle\Provider\RouteProviderInterface;
use SWP\Component\Rule\Applicator\RuleApplicatorInterface;
use SWP\Component\Rule\Model\RuleSubjectInterface;
use SWP\Component\Rule\Model\RuleInterface;
use Symfony\Component\OptionsResolver\OptionsResolver;

final class ArticleRuleApplicator implements RuleApplicatorInterface
{
    /**
     * @var RouteProviderInterface
     */
    private $routeProvider;

    /**
     * @var LoggerInterface
     */
    private $logger;

    /**
     * @var array
     */
    private $supportedKeys = ['route', 'templateName'];

    /**
     * ArticleRuleApplicator constructor.
     *
     * @param RouteProviderInterface $routeProvider
     * @param LoggerInterface        $logger
     */
    public function __construct(RouteProviderInterface $routeProvider, LoggerInterface $logger)
    {
        $this->routeProvider = $routeProvider;
        $this->logger = $logger;
    }

    /**
     * {@inheritdoc}
     */
    public function apply(RuleInterface $rule, RuleSubjectInterface $subject)
    {
        $configuration = $this->validateRuleConfiguration($rule->getConfiguration());

        if (!$this->isAllowedType($subject) || empty($configuration)) {
            return;
        }

        /* @var ArticleInterface $subject */
        if (isset($configuration[$this->supportedKeys[0]])) {
            $route = $this->routeProvider->getOneById($configuration[$this->supportedKeys[0]]);

            if (null === $route) {
                $this->logger->warning('Route not found! Make sure the rule defines an existing route!');

                return;
            }

            $subject->setRoute($route);
        }

        $subject->setTemplateName($configuration[$this->supportedKeys[1]]);

        $this->logger->info(sprintf(
            'Configuration: "%s" for "%s" rule has been applied!',
            json_encode($configuration),
            $rule->getExpression()
        ));
    }

    /**
     * {@inheritdoc}
     */
    public function isSupported(RuleSubjectInterface $subject)
    {
        return $subject instanceof ArticleInterface && 'article' === $subject->getSubjectType();
    }

    private function validateRuleConfiguration(array $configuration)
    {
        $resolver = new OptionsResolver();
        $this->configureOptions($resolver);

        try {
            return $resolver->resolve($configuration);
        } catch (\Exception $e) {
            $this->logger->warning($e->getMessage());
        }

        return [];
    }

    private function configureOptions(OptionsResolver $resolver)
    {
        $resolver->setDefaults([$this->supportedKeys[1] => null]);
        $resolver->setDefined($this->supportedKeys[0]);
    }

    private function isAllowedType(RuleSubjectInterface $subject)
    {
        if (!$subject instanceof ArticleInterface) {
            $this->logger->warning(sprintf(
                '"%s" is not supported by "%s" rule applicator!',
                is_object($subject) ? get_class($subject) : gettype($subject),
                get_class($this)
            ));

            return false;
        }

        return true;
    }
}

Configuring the Rule Applicator¶

To register your new rule applicator, simply add a definition to your services file and tag it with a special name: applicator.rule_applicator, it will be automatically added to the chain of rule applicators:

# Resources/config/services.yml
acme_my_custom_rule_applicator:
    class: 'Acme\DemoBundle\Applicator\ArticleRuleApplicator'
    arguments:
        - '@swp.provider.route'
        - '@logger'
    tags:
        - { name: applicator.rule_applicator }

How to Create and Enable Custom Rule Entity¶

In some cases you would want to extend the default Rule model to add some extra properties etc. To do this you need to create a custom class which extends the default one.

<?php
 // ..
 namespace Acme\DemoBundle\Entity;

 use SWP\Component\Rule\Model\Rule as BaseRule;

 class Rule extends BaseRule
 {
     protected $something;

     public function getSomething()
     {
         return $this->something;
     }

     public function setSomething($something)
     {
         $this->something = $something;
     }
 }

Add class’s mapping file:

# Acme\DemoBundle\Resources\config\doctrine\Rule.orm.yml
Acme\DemoBundle\Entity\Rule:
    type: entity
    table: custom_rule
    fields:
        something:
            type: string

The newly created class needs to be now added to the bundle’s configuration:

# app/config/config.yml
swp_rule:
    persistence:
        orm:
            # ..
            classes:
                rule:
                    model: Acme\DemoBundle\Entity\Rule

That’s it, a newly created class will be used instead.

Note

You could also provide your own implementation for Rule Factory and Rule Repository. To find out more about it check How to automatically register Services required by the configured Storage Driver

How rules are processed?¶

You can create Event Subscriber which can listen on whatever event is defined. If the event is dispatched, the subscriber should run Rule Processor which will process all rules.

Example subscriber:

<?php

namespace SWP\Bundle\ContentBundle\EventListener;

use SWP\Bundle\ContentBundle\ArticleEvents;
use SWP\Bundle\ContentBundle\Event\ArticleEvent;
use SWP\Component\Rule\Processor\RuleProcessorInterface;
use Symfony\Component\EventDispatcher\EventSubscriberInterface;

class ProcessArticleRulesSubscriber implements EventSubscriberInterface
{
    /**
     * @var RuleProcessorInterface
     */
    private $ruleProcessor;

    /**
     * ProcessArticleRulesSubscriber constructor.
     *
     * @param RuleProcessorInterface $ruleProcessor
     */
    public function __construct(RuleProcessorInterface $ruleProcessor)
    {
        $this->ruleProcessor = $ruleProcessor;
    }

    /**
     * {@inheritdoc}
     */
    public static function getSubscribedEvents()
    {
        return [
            ArticleEvents::PRE_CREATE => 'processRules',
        ];
    }

    /**
     * @param ArticleEvent $event
     */
    public function processRules(ArticleEvent $event)
    {
        $this->ruleProcessor->process($event->getArticle());
    }
}

Enabling separate Monolog channel¶

It is possible to enable a separate Monolog channel to which all Rule Bundle related logs will be forwarded. An example log entry might be logged when the rule can not be evaluated properly etc. You could have then a separate log file for (which will log everything related to that bundle) which will be saved under the directory app/logs/ in your application and will be named, for example: swp_rule_<env>.log. By default, a separate channel is not enabled. You can enable it by adding an extra channel in your Monolog settings (in one of your configuration files):

# app/config/config.yml
monolog:
    handlers:
        swp_rule:
            level:    debug
            type:     stream
            path:     '%kernel.logs_dir%/swp_rule_%kernel.environment%.log'
            channels: swp_rule

For more details see the Monolog documentation.

Models¶

Rule¶

Rule is to check if your “rule aware” objects are allowed to be processed and if some rule’s configuration can be applied to it.

A rule is configured using the configuration attribute which is an array serialized into database. Your custom Rule Applicator should define which configuration key-value pair should be applied. For example, you could configure the route key to define which route should be applied to an object if the given rule evaluates to true. You could also apply templateName or any other keys.

See Usage section for more details.

RuleSubjectInterface¶

To make use of the Rule bundle and allow to apply rule to an object, the entity must be “rule aware”, it means that “subject” class needs to implement SWP\Component\Rule\Model\RuleSubjectInterface interface.

If you make your custom entity rule aware, the Rule Processor will automatically process all rules for given object.

By implementing SWP\Component\Rule\Model\RuleSubjectInterface interface, your object will have to define the following method:

getSubjectType() - should return the name of current object, for example: article.

Rule Evaluator is using this method to evaluate rule on an object.

If the getSubjectType() returns article the rule expression should be related to this object using article prefix. For example: article.getSomething('something') > 1.

Configuration Reference¶

The SWPRuleBundle can be configured under the swp_rule key in your configuration file. This section describes the whole bundle’s configuration.

Full Default Configuration¶

# app/config/config.yml
swp_rule:
    persistence:
        orm:
            enabled: true
            classes:
                rule:
                    model: SWP\Component\Rule\Model\Rule
                    repository: SWP\Bundle\RuleBundle\Doctrine\ORM\RuleRepository
                    factory: SWP\Bundle\StorageBundle\Factory\Factory
                    object_manager_name: ~

persistence¶

persistence¶

orm¶

# app/config/config.yml
swp_rule:
    # ..
    persistence:
        orm:
            enabled: true

enabled¶

type: boolean default: false

If true, ORM is enabled in the service container.

ORM can be enabled by multiple ways such as:

orm: ~ # use default configuration
# or
orm: true # straight way
# or
orm:
    enabled: true ... # or any other option under 'orm'

classes¶

# app/config/config.yml
swp_rule:
    # ..
    persistence:
        orm:
            # ..
            classes:
                rule:
                    model: SWP\Component\Rule\Model\Rule
                    repository: SWP\Bundle\RuleBundle\Doctrine\ORM\RuleRepository
                    factory: SWP\Bundle\StorageBundle\Factory\Factory
                    object_manager_name: ~

type: string default: null

The name of the object manager. If set to null it defaults to default. If Doctrine ORM persistence backend is enabled it will register swp.object_manager.rule service which is an alias for doctrine.orm.default_entity_manager.

rule.model¶

type: string default: SWP\Component\Rule\Model\Rule

The FQCN of the Rule model class which is of type SWP\Component\Rule\Model\RuleInterface.

rule.factory¶

type: string default: SWP\Bundle\StorageBundle\Factory\Factory

The FQCN of the Rule Factory class.

rule.repository¶

type: string default: SWP\Bundle\RuleBundle\Doctrine\ORM\RuleRepository

The FQCN of the Rule Repository class.

rule.object_manager_name¶

type: string default: null

The name of the object manager. If set to null it defaults to default. If Doctrine ORM persistence backend is enabled it will register swp.object_manager.rule service which is an alias for doctrine.orm.default_entity_manager.

ContentListBundle¶

This bundle gives you an ability to create powerful content lists where lists’ items can be of any type. For example, imagine a list which contains a lot of articles which you can add, reorder, drag and drop etc.

The Content List Component, which is used by this bundle, provides classes to build powerful: and flexible content lists.

Prerequisites¶

This version of the bundle requires Symfony >= 2.8 and PHP version >=7.0

Installation¶

Install the Bundle with Composer¶

In your project directory execute the following command to download the latest stable version:

1	composer require swp/content-list-bundle

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

Enable the bundle by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(
            new SWP\Bundle\ContentListBundle\SWPStorageBundle(),
            new Stof\DoctrineExtensionsBundle\StofDoctrineExtensionsBundle(),
            // ...
            new SWP\Bundle\ContentListBundle\SWPContentListBundle(),
        );

        // ...
    }

    // ...
}

Note

All dependencies will be installed automatically. You will just need to configure the respective bundles if needed.

Configure the bundle¶

YAML

# app/config/config.yml
swp_content_list:
    persistence:
        orm:
            # if true, ORM is enabled as a persistence backend
            enabled: true

Note

By default this bundle supports only Doctrine ORM as a persistence backend.

Note

If this bundle is used together with ContentBundle, configuration will be automatically pre-pended and enabled, so there is no need to configure it in your config file.

Configure Doctrine extensions which are used by this bundle:

# app/config/config.yml
stof_doctrine_extensions:
    orm:
        default:
            timestampable: true
            softdeleteable: true
            loggable: true

Using your custom list item content class:

YAML

# app/config/config.yml
swp_content_list:
    persistence:
        orm:
            # if true, ORM is enabled as a persistence backend
            enabled: true
            classes:
                # ..
                list_content:
                    model: Acme\MyBundle\Entity\Post

Note

Acme\MyBundle\Entity\Post must implement SWP\Component\ContentList\Model\ListContentInterface interface.

Update Database Schema¶

Run the following command:

1	$ php bin/console doctrine:schema:update --force

That’s it, the bundle is configured properly now!

Usage¶

Creating new content lists¶

Here is an example on how to create a new content list:

// src/AppBundle/Controller/MyController.php

use SWP\Component\ContentList\Model\ContentListInterface;
// ...
public function createAction()
{
    $repository = $this->container->get('swp.repository.content_list');

    /* @var ContentListInterface $contentList */
    $contentList = $this->get('swp.factory.content_list')->create();
    $contentList->setName('my content list');
    $contentList->setDescription('description');
    $contentList->setLimit(10);
    $contentList->setType(ContentListInterface::TYPE_AUTOMATIC);
    // ...

    $repository->add($contentList);

    // ...
}

Adding new items to content list¶

// src/AppBundle/Controller/MyController.php

use SWP\Component\ContentList\Model\ContentListInterface;
use SWP\Component\ContentList\Model\ContentListItemInterface;
use Acme\AppBundle\Entity\Article;
// ...
public function createAction()
{
    $repository = $this->container->get('swp.repository.content_list');

    /* @var ContentListInterface $contentList */
    $contentList = $this->get('swp.factory.content_list')->create();
    $contentList->setName('my content list');
    $contentList->setDescription('description');
    $contentList->setLimit(10);
    $contentList->setType(ContentListInterface::TYPE_AUTOMATIC);
    // ...

    /* @var ContentListItemInterface $contentListItem */
    $contentListItem = $this->get('swp.factory.content_list_item')->create();
    $contentListItem->setPosition(6);
    $contentListItem->setContent(new Article());
    $contentListItem->setSticky(true);
    $contentList->addItem($contentListItem);

    $repository->add($contentList);

    // ...
}

Note

Article class must implement SWP\Component\ContentList\Model\ListContentInterface.

Deleting content lists¶

// src/AppBundle/Controller/MyController.php

use SWP\Component\ContentList\Model\ContentListInterface;
use Acme\AppBundle\Entity\Article;
// ...
public function deleteAction($id)
{
    $repository = $this->container->get('swp.repository.content_list');

    /* @var ContentListInterface $contentList */
    $contentList = $repository->findOneBy(['id' => $id]);
    // ...

    $repository->remove($contentList);

    // ...
}

Deleting content lists items¶

// src/AppBundle/Controller/MyController.php

use SWP\Component\ContentList\Model\ContentListItemInterface;
use Acme\AppBundle\Entity\Article;
// ...
public function deleteAction($id)
{
    $repository = $this->container->get('swp.repository.content_list_item');

    /* @var ContentListItemInterface $contentListItem */
    $contentListItem = $repository->findOneBy(['id' => $id]);
    // ...

    $repository->remove($contentListItem);

    // ...
}

Forms¶

Content list type selector¶

If you want to use content list type selector inside your custom form you can do it by adding SWP\Bundle\ContentListBundle\Form\Type\ContentListTypeSelectorType form field type to your form:

namespace Acme\AppBundle\Form\Type;

use Symfony\Component\Form\AbstractType;
use Symfony\Component\Form\Extension\Core\Type\TextType;
use Symfony\Component\Form\FormBuilderInterface;
use Symfony\Component\Validator\Constraints\NotBlank;


class MyListType extends AbstractType
{
    public function buildForm(FormBuilderInterface $builder, array $options)
    {
        $builder->add('name', TextType::class, [
            'constraints' => [
                new NotBlank(),
            ],
            'description' => 'List name',
        ])
        ->add('type', ContentListTypeSelectorType::class, [
            'constraints' => [
                new NotBlank(),
            ],
            'description' => 'List type',
        ])
    }
}

Alternatively, you could also extend from the default SWP\Bundle\ContentListBundle\Form\Type\ContentListType class if you would only add more fields on top of the existing form.

Note

For more details on how to register custom factory, repository, object manager, forms using custom classes see SWPStorageBundle Usage section.

Getting content lists from repository¶

To get single or all content lists from the repository you can use default Doctrine ORM SWP\Bundle\ContentListBundle\Doctrine\ORM\ContentListRepository repository. It has the same methods as Doctrine ORM EntityRepository, but it contains an extra method to get content lists by its type:

findByType(string $type): array - it gets many content lists by its type, type can be either: automatic or manual.

// src/AppBundle/Controller/MyController.php

use SWP\Component\ContentList\Model\ContentListInterface;
// ...
public function getAction()
{
    $repository = $this->container->get('swp.repository.content_list');

    $lists = $repository->findByType(ContentListInterface::TYPE_AUTOMATIC);
    var_dump($lists);die;
    // ...
}

Note

This repository is automatically registered as a service for you and can be accessible under service id: swp.repository.content_list in Symfony container.

Getting content lists items from repository¶

To get content list items you can use default repository which is registered as a service under the swp.repository.content_list_item key in Symfony container. It extends default Doctrine ORM EntityRepository.

Models¶

ContentList¶

ContentList model is a main class which defines default list properties. This includes list’s items, name, description and more:

list can be limited to display certain number of items
list can have cache life time defined in seconds (which is useful, for example, if you want to cache list for some time when rendering it on frontend)
list can be of one of the types:
- automatic
- manual

Automatic list is meant to be created manually but the items in that list should not be draggable and droppable. It just a flat list that you can add items and simply render list with it’s items. Whatever content you want to place in this list you should be able to do it. An example can be that if some part of your business logic is able to decide where the article should go, if it matches some criteria, you can use that logic and add an article to the list automatically - this list will be then called automatic list.

As in the case of Automatic lists, the Manual list is meant to be created manually but you should be able to add, remove, drag and drop, sort items in this list manually by simply linking items to lists.

ContentListItem¶

ContentListItem model represents an item which can be placed inside content list. It has a reference to content list, position at which it should be placed inside the list and a content. Content can be of any type, but it should implement SWP\Component\ContentList\Model\ListContentInterface.

ListContentInterface¶

This interface should be implemented by your class if you want objects of that type to be a content of the list. For example, if you have Article class in your project and you want to make objects of this type to be a content of list item, it needs to implement SWP\Component\ContentList\Model\ListContentInterface.

Configuration Reference¶

The SWPContentListBundle can be configured under the swp_content_list key in your configuration file. This section describes the whole bundle’s configuration.

Full Default Configuration¶

# app/config/config.yml
swp_content_list:
    persistence:
        orm:
            enabled: true
            classes:
                content_list:
                    model: SWP\Component\ContentList\Model\ContentList
                    interface: SWP\Component\ContentList\Model\ContentListInterface
                    repository: SWP\Bundle\ContentListBundle\Doctrine\ORM\ContentListRepository
                    factory: SWP\Bundle\StorageBundle\Factory\Factory
                    object_manager_name: ~
                content_list_item:
                    model: SWP\Component\ContentList\Model\ContentListItem
                    interface: SWP\Component\ContentList\Model\ContentListItemInterface
                    repository: SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository
                    factory: SWP\Bundle\StorageBundle\Factory\Factory
                    object_manager_name: ~
                list_content:
                    model: ~
                    interface: SWP\Component\ContentList\Model\ListContentInterface
                    repository: SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository
                    factory: SWP\Bundle\StorageBundle\Factory\Factory
                    object_manager_name: ~

persistence¶

persistence¶

orm¶

# app/config/config.yml
swp_content_list:
    # ..
    persistence:
        orm:
            enabled: true

enabled¶

type: boolean default: false

If true, ORM is enabled in the service container.

ORM can be enabled by multiple ways such as:

orm: ~ # use default configuration
# or
orm: true # straight way
# or
orm:
    enabled: true ... # or any other option under 'orm'

classes¶

# app/config/config.yml
swp_content_list:
    # ..
    persistence:
        orm:
            # ..
            classes:
                content_list:
                    model: SWP\Component\ContentList\Model\ContentList
                    interface: SWP\Component\ContentList\Model\ContentListInterface
                    repository: SWP\Bundle\ContentListBundle\Doctrine\ORM\ContentListRepository
                    factory: SWP\Bundle\StorageBundle\Factory\Factory
                    object_manager_name: ~
                content_list_item:
                    model: SWP\Component\ContentList\Model\ContentListItem
                    interface: SWP\Component\ContentList\Model\ContentListItemInterface
                    repository: SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository
                    factory: SWP\Bundle\StorageBundle\Factory\Factory
                    object_manager_name: ~
                list_content:
                    model: ~
                    interface: SWP\Component\ContentList\Model\ListContentInterface
                    repository: SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository
                    factory: SWP\Bundle\StorageBundle\Factory\Factory
                    object_manager_name: ~

type: string default: null

The name of the object manager. If set to null it defaults to default. If Doctrine ORM persistence backend is enabled it will register swp.object_manager.content_list service which is an alias for doctrine.orm.default_entity_manager.

content_list.model¶

type: string default: SWP\Component\ContentList\Model\ContentList

The FQCN of the ContentList model class which is of type SWP\Component\ContentList\Model\ContentListInterface.

content_list.interface¶

type: string default: SWP\Component\ContentList\Model\ContentListInterface

The FQCN of your custom interface which is used by your model class.

content_list.factory¶

type: string default: SWP\Bundle\StorageBundle\Factory\Factory

The FQCN of the ContentList Factory class.

content_list.repository¶

type: string default: SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository

The FQCN of the ContentList Repository class.

content_list.object_manager_name¶

type: string default: null

The name of the object manager. If set to null it defaults to default. If Doctrine ORM persistence backend is enabled it will register swp.object_manager.content_list service which is an alias for doctrine.orm.default_entity_manager.

content_list_item.model¶

type: string default: SWP\Component\ContentList\Model\ContentListItem

The FQCN of the ContentListItem model class which is of type SWP\Component\ContentList\Model\ContentListItemInterface.

content_list_item.interface¶

type: string default: SWP\Component\ContentList\Model\ContentListItemInterface

The FQCN of your custom interface which is used by your model class.

content_list_item.factory¶

type: string default: SWP\Bundle\StorageBundle\Factory\Factory

The FQCN of the ContentListItem Factory class.

content_list_item.repository¶

type: string default: SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository

The FQCN of the ContentListItem Repository class.

content_list_item.object_manager_name¶

type: string default: null

The name of the object manager. If set to null it defaults to default. If Doctrine ORM persistence backend is enabled it will register swp.object_manager.content_list_item service which is an alias for doctrine.orm.default_entity_manager.

list_content.model¶

type: string default: null

The FQCN of the model class which must be of type SWP\Component\ContentList\Model\ContentListInterface. This is the content of the list item. You can use your custom classes here so for example, ACME\DemoBundle\Entity\Post could be your content.

list_content.interface¶

type: string default: SWP\Component\ContentList\Model\ListContentInterface

The FQCN of your custom interface which is used by your model class.

list_content.factory¶

type: string default: SWP\Bundle\StorageBundle\Factory\Factory

The FQCN of the List Item’s content Factory class.

list_content.repository¶

type: string default: SWP\Bundle\StorageBundle\Doctrine\ORM\EntityRepository

The FQCN of the List Item’s content Repository class.

list_content.object_manager_name¶

type: string default: null

The name of the object manager. If set to null it defaults to default. If Doctrine ORM persistence backend is enabled it will register swp.object_manager.content_list service which is an alias for doctrine.orm.default_entity_manager.

FacebookInstantArticlesBundle¶

This bundle integrates Facebook Instant Articles PHP SDK with Symfony application.

Authentication¶

To work with Facebook Instant Articles REST API you need to have valid access token. Bundle provides controller for authentication url generation and handling authorization callback requests from Facebook.

Authorization procedure requires providing Facebook Application and Page (token will be generated for that combination) ID’s. Bundle provide entities for both (Application and Page) and will look for matching rows in database.

Note

Authentication controller checks if provided page and application are in your storage (database). But bundle doesn’t provide controllers for adding them (there are only pre-configured factories and repositories) - You need to implement it manually in your application.

Authentication flow¶

Assuming that in your database you have Application with id 123456789 and Page with id 987654321 (and both it exists on Facebook platform), You need to call this url (route: swp_fbia_authorize): /facebook/instantarticles/authorize/123456789/987654321

In response You will be redirected to Facebook where You will need allow for all required permissions.

After that Facebook will redirect You again to application where (in background - provided by Facebook code will be exchanged for access token and that access) you will get JSON response with pageId and accessToken (never expiring access token).

Article Parsing¶

Instant Article view¶

For Instant Article rendering this template file is used /platforms/facebook_instant_article.html.twig. Basic version is provided by Publisher but you can (should) override it with your theme.

To control how exactly look pushed to Facebook Instant Articles API article you can use preview url.

/facebook/instantarticles/preview/{articleId} - shows how article template was converted into FBIA article. Parser is removing all not recognized tags. Read more about allowed rules here: https://developers.facebook.com/docs/instant-articles/sdk/transformer.

By default we use standard SDK rules: https://github.com/facebook/facebook-instant-articles-sdk-php/blob/master/src/Facebook/InstantArticles/Parser/instant-articles-rules.json

SettingsBundle¶

This bundle provides tools to define settings and save user changes.

Installation¶

Install the Bundle with Composer¶

In your project directory execute the following command to download the latest stable version:

1	composer require swp/settings-bundle

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

Enable the bundle and its dependencies by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(
            // ...

            new SWP\Bundle\SettingsBundle\SWPSettingsBundle(),
        );

        // ...
    }

    // ...
}

Enable the bundle in application configuration (config.yml):

swp_settings:
    persistence:
        orm:
            enabled: true

Note

All dependencies will be installed automatically. You will just need to configure the respective bundles if needed. To see all required dependencies - check bundle composer.json file.

That’s it, the bundle is configured properly now!

Settings definitions¶

In your application or bundle configuration add settings definitions:

swp_settings:
    settings:
        registration_confirmation.template:
            value: "example value"
            scope: user
            type: string

Minimal definition looks like that:

swp_settings:
    settings:
        registration_confirmation.template: ~

Note

Default values:

value: null
scope: global (possible options: global, user)
type: string (possible options: string, array)

Settings scopes¶

Scope defines level for custom changes. If setting have scope user then every user will have his own value for this setting.

Settings value types¶

Setting value can be string or array (it will be saved as json).

Example with array as value:

parameters:
    array_value:
        a: 1
        b: 2

swp_settings:
    settings:
        custom_setting_1:
            value: "%array_value%"
        custom_setting_2:
            value: '{"a":1, "b": 2}'

Usage¶

REST API¶

List settings¶

You can list all settings (with values loaded for scope) by API.

GET /api/{version}/settings/

Update settings¶

You can update settings with API call:

curl -X "PATCH" -d "settings[name]=setting_name&settings[value]=setting_value" -H "Content-type:\ application/x-www-form-urlencoded" /api/v1/settings

In code¶

Get all settings¶

$settingsManager = $this->get('swp_settings.manager.settings');
$settings = $settingsManager->all();

Get single setting¶

$settingsManager = $this->get('swp_settings.manager.settings');
$setting = $settingsManager->get('setting_name');

Set setting value¶

$settingsManager = $this->get('swp_settings.manager.settings');
$setting = $settingsManager->set('setting_name', 'setting value');

For details check SWP\Bundle\SettingsBundle\Manager\SettingsManager class.

Scope context¶

Hint

Scope defines level for custom changes. If setting have scope user then every user will have his own value for this setting.

ScopeContext class defines available scopes (in getScopes() method). Every setting must have scope and can have setting owner. Scope Context collects owners for defined scopes from current application state.

...
$scopeContext = new \SWP\Bundle\SettingsBundle\Context\ScopeContext();
...

// Set user in scope
$scopeContext->setScopeOwner(ScopeContextInterface::SCOPE_USER, $user);

Note

Owner object set to scope context must implement SettingsOwnerInterface. Scope owner allows for system to fill settings with correct custom set values kept in storage.

Bundle already register event subscriber responsible for setting currently logged user in scope context - SWP\Bundle\SettingsBundle\EventSubscriber\ScopeContextSubscriber.

WebhookBundle¶

This bundle provides tools to build a webhook’s system in Symfony application.

This bundle uses the Doctrine ORM persistence backend.

Installation¶

Install the Bundle with Composer¶

In your project directory execute the following command to download the latest stable version:

1	composer require swp/webhook-bundle

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

This bundle require StorageBundle to be installed and configured. Enable the bundle by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(

            new SWP\Bundle\StorageBundle\SWPStorageBundle()
            // ...
            new SWP\Bundle\WebhookBundle\SWPWebhookBundle(),
        );

        // ...
    }

    // ...
}

Note

All dependencies will be installed automatically.

That’s it, the bundle is configured properly now!

Usage¶

Webhooks Management¶

Bundle provides abstract Controller class for CRUD webhook actions. It cane be used in Your own controllers.

Sending Webhooks¶

Bundle allows to create new webhooks, and provides repository for searching them by event name. Sending events need to be implemented in end application. We recommend to do that with dispatcher and event listeners/subscribers.

Example implementation can be found in SWP\Bundle\CoreBundle\EventSubscriber\WebhookEventsSubscriber class (in Superdesk Publisher project).

OutputChannelBundle¶

This bundle provides tools to build output channels abstraction in Symfony application.

This bundle uses the Doctrine ORM persistence backend.

Installation¶

Install the Bundle with Composer¶

In your project directory execute the following command to download the latest stable version:

1	composer require swp/output-channel-bundle

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

This bundle requires StorageBundle to be installed and configured. Enable the bundle by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(

            new SWP\Bundle\StorageBundle\SWPStorageBundle()
            // ...
            new SWP\Bundle\OutputChannelBundle\SWPOutputChannelBundle(),
        );

        // ...
    }

    // ...
}

Note

All dependencies will be installed automatically.

That’s it, the bundle is configured properly now!

Usage¶

This bundle provide the JMS Serializer mapping files, Doctrine ORM mapping files, form types and the basic configuration.

OutputChannelType¶

This bundle comes with the OutputChannelType to build forms using Symfony.

WordpressOutputChannelConfigType¶

This bundle contains the WordpressOutputChannelConfigType form type and defines the configuration for the Wordpress output channel.

Paywall Bundle¶

This bundle provides tools to build paywall solution in Symfony application.

Installation¶

Install the Bundle with Composer¶

In your project directory execute the following command to download the latest stable version:

1	composer require swp/paywall-bundle

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

Enable the bundle by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(
            // ...
            new SWP\Bundle\PaywallBundle\SWPPaywallBundle(),
        );

        // ...
    }

    // ...
}

Import config to into your config.yml file:

imports:
    - { resource: "@SWPPaywallBundle/Resources/config/app/config.yml" }

Note

All dependencies will be installed automatically.

That’s it, the bundle is configured properly now!

Usage¶

This bundle provide the services which help to interact with PaymentsHub. It also allows you to create your custom implementations to interact with different/custom subscriptions systems.

How to add a new adapter¶

Adapters are used to retrieve the subscriptions data from the external subscription system. It is possible to implement your custom adapter and use it to fetch the subscriptions data from the 3rd party subscription system.

Create your custom adapter class which uses GuzzleHttp client

// src/AcmeBundle/Adapter/CustomAdapter.php

namespace AcmeBundle\Adapter;

use GuzzleHttp\ClientInterface;
use GuzzleHttp\Exception\RequestException;
use Psr\Http\Message\ResponseInterface;
use SWP\Component\Paywall\Factory\SubscriptionFactoryInterface;
use SWP\Component\Paywall\Model\SubscriberInterface;
use SWP\Component\Paywall\Model\SubscriptionInterface;

// ...
final class CustomAdapter implements PaywallAdapterInterface
{
    public function __construct(array $config, SubscriptionFactoryInterface $subscriptionFactory, ClientInterface $client)
    {
        $this->config = $config;
        $this->subscriptionFactory = $subscriptionFactory;
        $this->client = $client;
    }

    public function getSubscriptions(SubscriberInterface $subscriber, array $filters = []): array
    {
        // custom logic here to get subscriptions
        // ...

        $subscription = $this->subscriptionFactory->create();

        // ...
    }

    public function getSubscription(SubscriberInterface $subscriber, array $filters = []): ?SubscriptionInterface
    {
        // custom logic here to get a single subscription
        // ...
    }

    // ...
}

Register your adapter as a service

# services.yml
AcmeBundle\Adapter\CustomAdapter:
    arguments:
        -
            serverUrl: "%env(resolve:PAYWALL_SERVER_URL)%"
            credentials:
                username: "%env(resolve:PAYWALL_SERVER_USERNAME)%"
                password: "%env(resolve:PAYWALL_SERVER_PASSWORD)%"
        - '@SWP\Component\Paywall\Factory\SubscriptionFactory'
        - '@GuzzleHttp\Client'

Enabled newly created adapter in bundle’s config

# config.yml
swp_paywall:
    adapter: AcmeBundle\Adapter\CustomAdapter

SeoBundle¶

This bundle provides tools for SEO metadata in Symfony application.

This bundle uses the Doctrine ORM persistence backend.

Installation¶

Install the Bundle with Composer¶

In your project directory execute the following command to download the latest stable version:

1	composer require swp/seo-bundle

This command requires you to have Composer installed globally. If it’s not installed globally, download the .phar file locally as explained in Composer documentation.

Enable the bundle and its dependencies¶

This bundle requires StorageBundle to be installed and configured. Enable the bundle by adding the following lines in the app/AppKernel.php file:

// app/AppKernel.php

// ...
class AppKernel extends Kernel
{
    public function registerBundles()
    {
        $bundles = array(

            new SWP\Bundle\StorageBundle\SWPStorageBundle()
            // ...
            new SWP\Bundle\SeoBundle\SWPSeoBundle(),
        );

        // ...
    }

    // ...
}

Note

All dependencies will be installed automatically.

That’s it, the bundle is configured properly now!

Developer Guide¶

Contributing¶

Note

This section is based on Symfony2 documentation.

Contributing Code¶

Reporting a Bug¶

Whenever you find a bug, we kindly ask you to report it. It helps us make a better Superdesk Publisher.

Caution

If you think you’ve found a security issue, please use the special procedure instead.

Before submitting a bug:

Double-check the official documentation to see if you’re not misusing the project;
Ask for assistance on StackOverflow or on the official bug tracker if you’re not sure if your issue is really a bug.

If your problem definitely looks like a bug, report it using the bug tracker and follow some basic rules:

Use the title field to clearly describe the issue;
Describe the steps needed to reproduce the bug with short code examples (providing a unit test that illustrates the bug is best);
Give as much detail as possible about your environment (OS, PHP version, Superdesk Publisher version, enabled extensions, …);
(optional) Attach a patch.

Submitting a Patch¶

Patches are the best way to provide a bug fix or to propose enhancements to Superdesk Publisher.

Step 1: Setup your Environment¶

Install the Software Stack¶

Before working on Superdesk Publisher, setup a friendly environment with the following software:

Git;
PHP version 5.5.9 or above;
PHPUnit 4.2 or above.
PostgreSQL

Configure Git¶

Set up your user information with your real name and a working email address:

git config --global user.name "Your Name"
git config --global user.email you@example.com

Tip

If you are new to Git, you are highly recommended to read the excellent and free ProGit book.

Tip

If your IDE creates configuration files inside the project’s directory, you can use a global .gitignore file (for all projects) or a .git/info/exclude file (per project) to ignore them. See GitHub’s documentation.

Tip

Windows users: when installing Git, the installer will ask what to do with line endings, and suggests replacing all LF with CRLF. This is the wrong setting if you wish to contribute to Superdesk Publisher! Selecting the as-is method is your best choice, as Git will convert your line feeds to the ones in the repository. If you have already installed Git, you can check the value of this setting by typing:

1	git config core.autocrlf

This will return either “false”, “input” or “true”; “true” and “false” being the wrong values. Change it to “input” by typing:

1	git config --global core.autocrlf input

Replace –global by –local if you want to set it only for the active repository

Get the Superdesk Publisher Source Code¶

Get the Superdesk Publisher source code:

Create a GitHub account and sign in;
Fork the Superdesk Publisher repository (click on the “Fork” button);
After the “forking action” has completed, clone your fork locally (this will create a web-publisher directory):

1	git clone git@github.com:USERNAME/web-publisher.git

Add the upstream repository as a remote:

cd web-publisher
git remote add upstream git://github.com/superdesk/web-publisher.git

Check that the current Tests Pass¶

Now that Superdesk Publisher is installed, check that all tests pass for your environment as explained in the dedicated document.

Step 2: Work on your Patch¶

The License¶

Before you start, you must know that all the patches you are going to submit must be released under the GNU AGPLv3 license, unless explicitly specified in your commits.

Create a Topic Branch¶

Each time you want to work on a patch for a bug or an enhancement, create a topic branch:

1	git checkout -b BRANCH_NAME master

Tip

Use a descriptive name for your branch, containing the ticket number from the bug tracker.

The above checkout commands automatically switch the code to the newly created branch (check the branch you are working on with git branch).

Work on your Patch¶

Work on the code as much as you want and commit as much as you want; but keep in mind the following:

Read about the Superdesk Publisher conventions and follow the coding standards (use git diff --check to check for trailing spaces – also read the tip below);
Add unit tests to prove that the bug is fixed or that the new feature actually works;
Try hard to not break backward compatibility (if you must do so, try to provide a compatibility layer to support the old way) – patches that break backward compatibility have less chance to be merged;
Do atomic and logically separate commits (use the power of git rebase to have a clean and logical history);
Never fix coding standards in some existing code as it makes the code review more difficult;
Write good commit messages (see the tip below).

Tip

When submitting pull requests, StyleCI checks your code for common typos and verifies that you are using the PHP coding standards as defined in PSR-1 and PSR-2.

A status is posted below the pull request description with a summary of any problems it detects or any Travis CI build failures.

Tip

A good commit message is composed of a summary (the first line), optionally followed by a blank line and a more detailed description. The summary should start with the Component you are working on in square brackets ([MultiTenancy], [MultiTenancyBundle], …). Use a verb (fixed ..., added ..., …) to start the summary and don’t add a period at the end.

Prepare your Patch for Submission¶

When your patch is not about a bug fix (when you add a new feature or change an existing one for instance), it must also include the following:

An explanation of the changes in the relevant CHANGELOG file(s) (the [BC BREAK] or the [DEPRECATION] prefix must be used when relevant);
An explanation on how to upgrade an existing application in the relevant UPGRADE file(s) if the changes break backward compatibility or if you deprecate something that will ultimately break backward compatibility.

Step 3: Submit your Patch¶

Whenever you feel that your patch is ready for submission, follow the following steps.

Rebase your Patch¶

Before submitting your patch, update your branch (needed if it takes you a while to finish your changes):

git checkout master
git fetch upstream
git merge upstream/master
git checkout BRANCH_NAME
git rebase master

When doing the rebase command, you might have to fix merge conflicts. git status will show you the unmerged files. Resolve all the conflicts, then continue the rebase:

git add ... # add resolved files
git rebase --continue

Check that all tests still pass and push your branch remotely:

1	git push --force origin BRANCH_NAME

Make a Pull Request¶

You can now make a pull request on the superdesk/web-publisher GitHub repository.

To ease the core team work, always include the modified components in your pull request message, like in:

[MultiTenancy] fixed something
[Common] [MultiTenancy] [MultiTenancyBundle] added something

The pull request description must include the following checklist at the top to ensure that contributions may be reviewed without needless feedback loops and that your contributions can be included into Superdesk Publisher as quickly as possible:

| Q             | A
| ------------- | ---
| Bug fix?      | [yes|no]
| New feature?  | [yes|no]
| BC breaks?    | [yes|no]
| Deprecations? | [yes|no]
| Tests pass?   | [yes|no]
| Fixed tickets | [comma separated list of tickets fixed by the PR]
| License       | AGPLv3

An example submission could now look as follows:

| Q             | A
| ------------- | ---
| Bug fix?      | no
| New feature?  | no
| BC breaks?    | no
| Deprecations? | no
| Tests pass?   | yes
| Fixed tickets | #12, #43
| License       | AGPLv3

The whole table must be included (do not remove lines that you think are not relevant). For simple typos, minor changes in the PHPDocs, or changes in translation files, use the shorter version of the check-list:

| Q             | A
| ------------- | ---
| Fixed tickets | [comma separated list of tickets fixed by the PR]
| License       | GPLv3

Some answers to the questions trigger some more requirements:

If you answer yes to “Bug fix?”, check if the bug is already listed in the Superdesk Publisher bug tracker and reference it/them in “Fixed tickets”;
If you answer yes to “New feature?”, you must submit a pull request to the documentation and reference it under the “Doc PR” section;
If you answer yes to “BC breaks?”, the patch must contain updates to the relevant CHANGELOG and UPGRADE files;
If you answer yes to “Deprecations?”, the patch must contain updates to the relevant CHANGELOG and UPGRADE files;
If you answer no to “Tests pass”, you must add an item to a todo-list with the actions that must be done to fix the tests;
If the “license” is not as AGPLv3 here, please don’t submit the pull request as it won’t be accepted anyway.

If some of the previous requirements are not met, create a todo-list and add relevant items:

- [ ] fix the tests as they have not been updated yet
- [ ] submit changes to the documentation
- [ ] document the BC breaks

Caution

When submitting pull requests which require some documentation changes, please also update the documentation where appropriate, as it is kept in the same repository (documentation dir)

If the code is not finished yet because you don’t have time to finish it or because you want early feedback on your work, add an item to the todo-list:

- [ ] finish the code
- [ ] gather feedback for my changes

As long as you have items in the todo-list, please prefix the pull request title with “[WIP]”.

In the pull request description, give as much detail as possible about your changes (don’t hesitate to give code examples to illustrate your points). If your pull request is about adding a new feature or modifying an existing one, explain the rationale for the changes. The pull request description helps the code review and it serves as a reference when the code is merged (the pull request description and all its associated comments are part of the merge commit message).

Rework your Patch¶

Based on the feedback on the pull request, you might need to rework your patch. Before re-submitting the patch, rebase with upstream/master, don’t merge; and force push to the origin:

git rebase -f upstream/master
git push --force origin BRANCH_NAME

Note

When doing a push --force, always specify the branch name explicitly to avoid messing with other branches in the repo (--force tells Git that you really want to mess with things, so do it carefully).

If moderators asked you to “squash” your commits, this means you will need to convert many commits to one commit.

Security Issues¶

This document explains how security issues affecting code in the main superdesk/web-publisher Git repository are handled by the Superdesk Publisher core team.

Reporting a Security Issue¶

If you think that you have found a security issue in Superdesk Publisher, don’t use the bug tracker and don’t publish it publicly. Instead, all security issues must be sent to security [at] superdesk.org. Emails sent to this address are forwarded to the Superdesk Publisher core-team private mailing list.

Resolving Process¶

For each report, we first try to confirm the vulnerability. When it is confirmed, the core team works on a solution following these steps:

Send an acknowledgement to the reporter;
Work on a patch;
Write a security announcement for the official Superdesk Publisher blog about the vulnerability. This post should contain the following information:
- a title that always include the “Security release” string;
- a description of the vulnerability;
- the affected versions;
- the possible exploits;
- how to patch/upgrade/workaround affected applications;
- credits.
Send the patch and the announcement to the reporter for review;
Apply the patch to all maintained versions of Superdesk Publisher;
Package new releases for all affected versions;
Publish the post on the official Superdesk Publisher blog

Note

Releases that include security issues should not be made on a Saturday or Sunday, except if the vulnerability has been publicly posted.

Note

While we are working on a patch, please do not reveal the issue publicly.

Running Superdesk Publisher Tests¶

The Superdesk Publisher project uses a third-party service which automatically runs tests for any submitted patch. If the new code breaks any test, the pull request will show an error message with a link to the full error details.

In any case, it’s a good practice to run tests locally before submitting a patch for inclusion, to check that you have not broken anything.

Before Running the Tests¶

To run the Superdesk Publisher test suite, install the external dependencies used during the tests, such as Doctrine, Twig and Monolog. To do so, install Composer and execute the following:

1	composer install

Note

For unit tests we use PHPSpec, for functional tests PHPUnit and Behat for integration.

Running the PHPUnit Tests¶

Then, run the test suite from the Superdesk Publisher root directory with the following command:

1	bin/phpunit -c app/

The output should display OK. If not, read the reported errors to figure out what’s going on and if the tests are broken because of the new code.

Tip

The entire Superdesk Publisher suite can take up to several minutes to complete. If you want to test a single component/bundle, type its path after the phpunit command, e.g.:

1	bin/phpunit src/SWP/Bundle/MultiTenancyBundle/

Tip

On Windows, install the ConEmu, ANSICON or Mintty free applications to see coloured test results.

Running the PHPSpec specs¶

Note

This section is based on Sylius documentation.

PHPSpec is a PHP toolset to drive emergent design by specification. It is not really a testing tool, but a design instrument, which helps structuring the objects and how they work together.

The Superdesk Publisher approach is to always describe the behaviour of the next object you are about to implement.

As an example, we’ll write a service, which sets the current tenant in the context. To initialize a new spec, use the desc command.

We just need to tell PHPSpec we will be working on the TenantContext class.

bin/phpspec desc "SWP\Component\MultiTenancy\Context\TenantContext"
Specification for TenantContext created in spec.

What have we just done? PHPSpec has created the spec for us. You can navigate to the spec folder and see the spec there:

<?php

namespace spec\SWP\Component\MultiTenancy\Context;

use PhpSpec\ObjectBehavior;
use Prophecy\Argument;

class TenantContextSpec extends ObjectBehavior
{
    function it_is_initializable()
    {
        $this->shouldHaveType('SWP\Component\MultiTenancy\Context\TenantContext');
    }
}

The object behaviour is made of examples. Examples are encased in public methods, started with it_ or its_.

PHPSpec searches for such methods in your specification to run. Why underscores for example names? just_because_its_much_easier_to_read than someLongCamelCasingLikeThat.

Now, let’s write the first example, which will set the current tenant:

<?php

namespace spec\SWP\Component\MultiTenancy\Context;

use PhpSpec\ObjectBehavior;
use SWP\Component\MultiTenancy\Model\TenantInterface;

class TenantContextSpec extends ObjectBehavior
{
    function it_is_initializable()
    {
        $this->shouldHaveType('SWP\Component\MultiTenancy\Context\TenantContext');
    }

    function it_should_set_tenant(TenantInterface $tenant)
    {
        $tenant->getId()->willReturn(1);
        $tenant->getSubdomain()->willReturn('example1');
        $tenant->getName()->willReturn('example1');

        $this->setTenant($tenant)->shouldBeNull();
    }
}

The example looks clear and simple, the TenantContext service should obtain the tenant id, name, subdomain and call the method to set the tenant.

Try running the example by using the following command:

bin/phpspec run

> spec\SWP\Component\MultiTenancy\Context\TenantContext

  ✘ it should set tenant
      Class TenantContext does not exists.

         Do you want me to create it for you? [Y/n]

Once the class is created and you run the command again, PHPSpec will ask if it should create the method as well. Start implementing the initial version of the TenantContext.

<?php

namespace SWP\Component\MultiTenancy\Context;

use SWP\Component\MultiTenancy\Model\TenantInterface;

/**
 * Class TenantContext.
 */
class TenantContext implements TenantContextInterface
{
    /**
     * @var TenantInterface
     */
    protected $tenant;

    /**
     * {@inheritdoc}
     */
    public function setTenant(TenantInterface $tenant)
    {
        $this->tenant = $tenant;
    }
}

Done! If you run PHPSpec again, you should see the following output:

bin/phpspec run

> spec\SWP\Component\MultiTenancy\Context\TenantContext

  ✔ it should set tenant

1 examples (1 passed)
123ms

This example is greatly simplified, in order to illustrate how we work. More examples might cover errors, API exceptions and other edge-cases.

A few tips & rules to follow when working with PHPSpec & Superdesk Publisher:

RED is good, add or fix the code to make it green;
RED-GREEN-REFACTOR is our rule;
All specs must pass;
When writing examples, describe the behaviour of the object in the present tense;
Omit the public keyword;
Use underscores (_) in the examples;
Use type hinting to mock and stub classes;
If your specification is getting too complex, the design is wrong. Try decoupling a bit more;
If you cannot describe something easily, probably you should not be doing it that way;
shouldBeCalled or willReturn, never together, except for builders;
Use constants in assumptions but strings in expected results;

Coding Standards¶

When contributing code to Superdesk Publisher, you must follow its coding standards. To make a long story short, here is the golden rule: Imitate the existing Superdesk Publisher code. Most open-source Bundles and libraries used by Superdesk Publisher also follow the same guidelines, and you should too.

Remember that the main advantage of standards is that every piece of code looks and feels familiar, it’s not about this or that being more readable.

Superdesk Publisher follows the standards defined in the PSR-0, PSR-1, PSR-2 and PSR-4 documents.

Since a picture - or some code - is worth a thousand words, here’s a short example containing most features described below:

<?php

/**
 * This file is part of the Superdesk Publisher package.
 *
 * Copyright 2016 Sourcefabric z.ú. and contributors.
 *
 * For the full copyright and license information, please see the
 * AUTHORS and LICENSE files distributed with this source code.
 *
 * @copyright 2016 Sourcefabric z.ú.
 * @license http://www.superdesk.org/license
 */

namespace Acme;

/**
 * Coding standards demonstration.
 */
class FooBar
{
    const SOME_CONST = 42;

    /**
     * @var string
     */
    private $fooBar;

    /**
     * @param string $dummy Some argument description
     */
    public function __construct($dummy)
    {
        $this->fooBar = $this->transformText($dummy);
    }

    /**
     * @return string
     *
     * @deprecated
     */
    public function someDeprecatedMethod()
    {
        @trigger_error(sprintf('The %s() method is deprecated since version 1.0 and will be removed in 2.0. Use Acme\Baz::someMethod() instead.', __METHOD__), E_USER_DEPRECATED);

        return Baz::someMethod();
    }

    /**
     * Transforms the input given as first argument.
     *
     * @param bool|string $dummy   Some argument description
     * @param array       $options An options collection to be used within the transformation
     *
     * @return string|null The transformed input
     *
     * @throws \RuntimeException When an invalid option is provided
     */
    private function transformText($dummy, array $options = array())
    {
        $defaultOptions = array(
            'some_default' => 'values',
            'another_default' => 'more values',
        );

        foreach ($options as $option) {
            if (!in_array($option, $defaultOptions)) {
                throw new \RuntimeException(sprintf('Unrecognized option "%s"', $option));
            }
        }

        $mergedOptions = array_merge(
            $defaultOptions,
            $options
        );

        if (true === $dummy) {
            return;
        }

        if ('string' === $dummy) {
            if ('values' === $mergedOptions['some_default']) {
                return substr($dummy, 0, 5);
            }

            return ucwords($dummy);
        }
    }

    /**
     * Performs some basic check for a given value.
     *
     * @param mixed $value     Some value to check against
     * @param bool  $theSwitch Some switch to control the method's flow
     *
     * @return bool|null The resultant check if $theSwitch isn't false, null otherwise
     */
    private function reverseBoolean($value = null, $theSwitch = false)
    {
        if (!$theSwitch) {
            return;
        }

        return !$value;
    }
}

Structure¶

Add a single space after each comma delimiter;
Add a single space around binary operators (==, &&, …), with the exception of the concatenation (.) operator;
Place unary operators (!, --, …) adjacent to the affected variable;
Always use identical comparison unless you need type juggling;
Use Yoda conditions when checking a variable against an expression to avoid an accidental assignment inside the condition statement (this applies to ==, !=, ===, and !==);
Add a comma after each array item in a multi-line array, even after the last one;
Add a blank line before return statements, unless the return is alone inside a statement-group (like an if statement);
Use just return; instead of return null; when a function must return void early;
Use braces to indicate control structure body regardless of the number of statements it contains;
Define one class per file - this does not apply to private helper classes that are not intended to be instantiated from the outside and thus are not concerned by the PSR-0 and PSR-4 autoload standards;
Declare class properties before methods;
Declare public methods first, then protected ones and finally private ones. The exceptions to this rule are the class constructor and the setUp and tearDown methods of PHPUnit tests, which should always be the first methods to increase readability;
Use parentheses when instantiating classes regardless of the number of arguments the constructor has;
Exception and error message strings should be concatenated using sprintf.
Calls to trigger_error with type E_USER_DEPRECATED should be switched to opt-in via @ operator. Read more at Deprecations;

Naming Conventions¶

Use camelCase, not underscores, for variable, function and method names, arguments;
Use underscores for option names and parameter names;
Use namespaces for all classes;
Prefix abstract classes with Abstract.
Suffix interfaces with Interface;
Suffix traits with Trait;
Suffix exceptions with Exception;
Use alphanumeric characters and underscores for file names;
For type-hinting in PHPDocs and casting, use bool (instead of boolean or Boolean), int (instead of integer), float (instead of double or real);
Don’t forget to look at the more verbose Conventions document for more subjective naming considerations.

Service Naming Conventions¶

A service name contains groups, separated by dots;
The DI alias of the bundle is the first group (e.g. swp_user);
Use lowercase letters for service and parameter names;
A group name uses the underscore notation.

Documentation¶

Add PHPDoc blocks for all classes, methods, and functions;
Group annotations together so that annotations of the same type immediately follow each other, and annotations of a different type are separated by a single blank line;
Omit the @return tag if the method does not return anything;
The @package and @subpackage annotations are not used.

License¶

Superdesk Publisher is released under the license, and the license block has to be present at the top of every PHP file, before the namespace.

Conventions¶

The Coding Standards document describes the coding standards for the Superdesk Publisher projects and the internal and third-party bundles. This document describes coding standards and conventions used in the core project to make it more consistent and predictable.

Method Names¶

When an object has a “main” many relation with related “things” (objects, parameters, …), the method names are normalized:

get()
set()
has()
all()
replace()
remove()
clear()
isEmpty()
add()
register()
count()
keys()

The usage of these methods are only allowed when it is clear that there is a main relation:

a CookieJar has many Cookie objects;
a Service Container has many services and many parameters (as services is the main relation, the naming convention is used for this relation);
a Console Input has many arguments and many options. There is no “main” relation, and so the naming convention does not apply.

For many relations where the convention does not apply, the following methods must be used instead (where XXX is the name of the related thing):

Main Relation	Other Relations
`get()`	`getXXX()`
`set()`	`setXXX()`
n/a	`replaceXXX()`
`has()`	`hasXXX()`
`all()`	`getXXXs()`
`replace()`	`setXXXs()`
`remove()`	`removeXXX()`
`clear()`	`clearXXX()`
`isEmpty()`	`isEmptyXXX()`
`add()`	`addXXX()`
`register()`	`registerXXX()`
`count()`	`countXXX()`
`keys()`	n/a

Note

While “setXXX” and “replaceXXX” are very similar, there is one notable difference: “setXXX” may replace, or add new elements to the relation. “replaceXXX”, on the other hand, cannot add new elements. If an unrecognized key is passed to “replaceXXX” it must throw an exception.

Deprecations¶

From time to time, some classes and/or methods are deprecated in the project; that happens when a feature implementation cannot be changed because of backward compatibility issues, but we still want to propose a “better” alternative. In that case, the old implementation can simply be deprecated.

A feature is marked as deprecated by adding a @deprecated phpdoc to relevant classes, methods, properties, and so on:

/**
 * @deprecated Deprecated since version 1.0, to be removed in 2.0. Use XXX instead.
 */

The deprecation message should indicate the version when the class/method was deprecated, the version when it will be removed, and whenever possible, how the feature was replaced.

A PHP E_USER_DEPRECATED error must also be triggered to help people with the migration starting one or two minor versions before the version where the feature will be removed (depending on the criticality of the removal):

@trigger_error('XXX() is deprecated since version 1.0 and will be removed in 2.0. Use XXX instead.', E_USER_DEPRECATED);

Without the @-silencing operator, users would need to opt-out from deprecation notices. Silencing swaps this behaviour and allows users to opt-in when they are ready to cope with them (by adding a custom error handler like the one used by the Web Debug Toolbar or by the PHPUnit bridge).

Git¶

This document explains some conventions and specificities in the way we manage the Superdesk Publisher code with Git.

Pull Requests¶

Whenever a pull request is merged, all the information contained in the pull request (including comments) is saved in the repository.

You can easily spot pull request merges as the commit message always follows this pattern:

1	merged branch USER_NAME/BRANCH_NAME (PR #11)

The PR reference allows you to have a look at the original pull request on GitHub: https://github.com/superdesk/web-publisher/pull/11. But all the information you can get on GitHub is also available from the repository itself.

The merge commit message contains the original message from the author of the changes. Often, this can help understand what the changes were about and the reasoning behind the changes.

The License¶

Superdesk Publisher is released under the GNU Affero General Public License, version 3.

Contributing Documentation¶

Contributing to the Documentation¶

One of the essential principles of the Superdesk Publisher project is that documentation is as important as code. That’s why a great amount of resources are dedicated to documenting new features and to keeping the rest of the documentation up-to-date.

Before Your First Contribution¶

Before contributing, you should consider the following:

Superdesk Publisher documentation is written using reStructuredText markup language. If you are not familiar with this format, read this article for a quick overview of its basic features.
Superdesk Publisher documentation is hosted on GitHub. You’ll need a GitHub user account to contribute to the documentation.
Superdesk Publisher documentation is published under a Creative Commons BY-SA 3.0 License and all your contributions will implicitly adhere to that license.

Your First Documentation Contribution¶

In this section, you’ll learn how to contribute to the Superdesk Publisher documentation for the first time. The next section will explain the shorter process you’ll follow in the future for every contribution after your first one.

Let’s imagine that you want to improve the installation chapter of the Superdesk Publisher. In order to make your changes, follow these steps:

Step 1. Go to the official Superdesk Publisher repository located at github.com/superdesk/web-publisher and fork the repository to your personal account. This is only needed the first time you contribute to Superdesk Publisher.

Step 2. Clone the forked repository to your local machine (this example uses the projects/web-publisher/ directory to store the Superdesk Publisher; change this value accordingly):

cd projects/
git clone git://github.com/<YOUR GITHUB USERNAME>/web-publisher.git

Step 3. Create a dedicated new branch for your changes. This greatly simplifies the work of reviewing and merging your changes. Use a short and memorable name for the new branch:

1	git checkout -b improve_install_chapter

Step 4. Now make your changes in the documentation. Add, tweak, reword and even remove any content, but make sure that you comply with the Documentation Standards.

Note

Documentation is stored under the docs directory in the root Superdesk Publisher folder.

Step 5. Push the changes to your forked repository:

git commit book/installation.rst
git push origin improve_install_chapter

Step 6. Everything is now ready to initiate a pull request. Go to your forked repository at https//github.com/<YOUR GITHUB USERNAME>/web-publisher and click on the Pull Requests link located in the sidebar.

Then, click on the big New pull request button. As GitHub cannot guess the exact changes that you want to propose, select the appropriate branches where changes should be applied.

The base fork should be superdesk/web-publisher and the base branch should be the master, which is the branch that you selected to base your changes on. The head fork should be your forked copy of web-publisher and the compare branch should be improve_install_chapter, which is the name of the branch you created and where you made your changes.

Step 7. The last step is to prepare the description of the pull request. To ensure that your work is reviewed quickly, please add the following table at the beginning of your pull request description:

| Q             | A
| ------------- | ---
| Doc fix?      | [yes|no]
| New docs?     | [yes|no] (PR # on superdesk/web-publisher if applicable)
| Applies to    | [Superdesk Publisher version numbers this applies to]
| Fixed tickets | [comma separated list of tickets fixed by the PR]

In this example, this table would look as follows:

| Q             | A
| ------------- | ---
| Doc fix?      | yes
| New docs?     | no
| Applies to    | all
| Fixed tickets | #10575

Step 8. Now that you’ve successfully submitted your first contribution to the Superdesk Publisher documentation, go and celebrate! The documentation managers will carefully review your work in short time and they will let you know about any required change.

In case you need to add or modify anything, there is no need to create a new pull request. Just make sure that you are on the correct branch, make your changes and push them:

cd projects/web-publisher/
git checkout improve_install_chapter

# ... do your changes

git push

Step 9. After your pull request is eventually accepted and merged in the Superdesk Publisher, you will be included in the Superdesk Publisher Contributors list.

Your Second Documentation Contribution¶

The first contribution took some time because you had to fork the repository, learn how to write documentation, comply with the pull requests standards, etc. The second contribution will be much easier, except for one detail: given the furious update activity of the Superdesk Publisher documentation repository, odds are that your fork is now out of date with the official repository.

Solving this problem requires you to sync your fork with the original repository. To do this, execute this command first to tell git about the original repository:

cd projects/web-publisher/
git remote add upstream https://github.com/superdesk/web-publisher.git

Now you can sync your fork by executing the following command:

cd projects/web-publisher/
git fetch upstream
git checkout master
git merge upstream/master

This command will update the master branch, which is the one you used to create the new branch for your changes. If you have used another base branch, e.g. testing, replace the master with the appropriate branch name.

Great! Now you can proceed by following the same steps explained in the previous section:

# create a new branch to store your changes based on the master branch
cd projects/web-publisher/
git checkout master
git checkout -b my_changes

# ... do your changes

# submit the changes to your forked repository
git add xxx.rst     # (optional) only if this is a new content
git commit xxx.rst
git push origin my_changes

# go to GitHub and create the Pull Request
#
# Include this table in the description:
# | Q             | A
# | ------------- | ---
# | Doc fix?      | [yes|no]
# | New docs?     | [yes|no]
# | Applies to    | [Superdesk Publisher version numbers this applies to]
# | Fixed tickets | [comma separated list of tickets fixed by the PR]

Your second contribution is now complete, so go and celebrate again! You can also see how your ranking improves in the list of Superdesk Publisher Contributors.

Your Next Documentation Contributions¶

Now that you’ve made two contributions to the Superdesk Publisher documentation, you are probably comfortable with all the Git-magic involved in the process. That’s why your next contributions would be much faster. Here you can find the complete steps to contribute to the Superdesk Publisher documentation, which you can use as a checklist:

# sync your fork with the official Superdesk Publisher repository
cd projects/web-publisher/
git fetch upstream
git checkout master
git merge upstream/master

# create a new branch from the maintained version
git checkout master
git checkout -b my_changes

# ... do your changes

# add and commit your changes
git add xxx.rst     # (optional) only if this is a new content
git commit xxx.rst
git push origin my_changes

# go to GitHub and create the Pull Request
#
# Include this table in the description:
# | Q             | A
# | ------------- | ---
# | Doc fix?      | [yes|no]
# | New docs?     | [yes|no]
# | Applies to    | [Superdesk Publisher version numbers this applies to]
# | Fixed tickets | [comma separated list of tickets fixed by the PR]

# (optional) make the changes requested by reviewers and commit them
git commit xxx.rst
git push

You guessed right: after all this hard work, it’s time to celebrate again!

Review your changes¶

Every GitHub Pull Request when merged, is automatically deployed to http://superdesk-publisher.readthedocs.io/en/latest/

Minor Changes (e.g. Typos)¶

You may find just a typo and want to fix it. Due to GitHub’s functional frontend, it is quite simple to create Pull Requests right in your browser while reading the docs on http://superdesk-publisher.readthedocs.io/en/latest/. To do this, just click the edit this page button on the upper right corner. Beforehand, please switch to the right branch as mentioned before. Now you are able to edit the content and describe your changes within the GitHub frontend. When your work is done, click Propose file change to create a commit and, in case it is your first contribution, also your fork. A new branch is created automatically in order to provide a base for your Pull Request. Then fill out the form to create the Pull Request as described above.

Frequently Asked Questions¶

Why Do my Changes Take so Long to Be Reviewed and/or Merged?¶

Please be patient. It can take up to several days before your pull request can be fully reviewed. After merging the changes, it could take again several hours before your changes appear on the http://superdesk-publisher.readthedocs.io/en/latest/ website.

What If I Want to Submit my Work without Fully Finishing It?¶

You can do it. But please use one of these two prefixes to let reviewers know about the state of your work:

[WIP] (Work in Progress) is used when you are not yet finished with your pull request, but you would like it to be reviewed. The pull request won’t be merged until you say it is ready.
[WCM] (Waiting Code Merge) is used when you’re documenting a new feature or change that hasn’t been accepted yet into the core code. The pull request will not be merged until it is merged in the core code (or closed if the change is rejected).

Would You Accept a Huge Pull Request with Lots of Changes?¶

First, make sure that the changes are somewhat related. Otherwise, please create separate pull requests. Anyway, before submitting a huge change, it’s probably a good idea to open an issue in the Superdesk Web Publisher repository to ask the managers if they agree with your proposed changes. Otherwise, they could refuse your proposal after you put all that hard work into making the changes. We definitely don’t want you to waste your time!

Documentation Format¶

The Superdesk Publisher documentation uses reStructuredText as its markup language and Sphinx for generating the documentation in the formats read by the end users, such as HTML and PDF.

reStructuredText¶

reStructuredText is a plaintext markup syntax similar to Markdown, but much stricter with its syntax. If you are new to reStructuredText, take some time to familiarize with this format by reading the existing Superdesk Publisher documentation source code.

If you want to learn more about this format, check out the reStructuredText Primer tutorial and the reStructuredText Reference.

Caution

If you are familiar with Markdown, be careful as things are sometimes very similar but different:

Lists starts at the beginning of a line (no indentation is allowed);
Inline code blocks use double-ticks (``like this``).

Sphinx¶

Sphinx is a build system that provides tools to create documentation from reStructuredText documents. As such, it adds new directives and interpreted text roles to the standard reST markup. Read more about the Sphinx Markup Constructs.

Syntax Highlighting¶

PHP is the default syntax highlighter applied to all code blocks. You can change it with the code-block directive:

.. code-block:: yaml

    { foo: bar, bar: { foo: bar, bar: baz } }

Note

Besides all of the major programming languages, the syntax highlighter supports all kinds of markup and configuration languages. Check out the list of supported languages on the syntax highlighter website.

Configuration Blocks¶

Whenever you include a configuration sample, use the configuration-block directive to show the configuration in all supported configuration formats (PHP, YAML and XML). Example:

.. configuration-block::

    .. code-block:: yaml

        # Configuration in YAML

    .. code-block:: xml

        <!-- Configuration in XML -->

    .. code-block:: php

        // Configuration in PHP

The previous reST snippet renders as follow:

YAML
1
# Configuration in YAML
XML
1
PHP
1
// Configuration in PHP

The current list of supported formats are the following:

Markup Format	Use It to Display
`html`	HTML
`xml`	XML
`php`	PHP
`yaml`	YAML
`twig`	Pure Twig markup
`html+twig`	Twig markup blended with HTML
`html+php`	PHP code blended with HTML
`ini`	INI
`php-annotations`	PHP Annotations

Adding Links¶

The most common type of links are internal links to other documentation pages, which use the following syntax:

1	:doc:`/absolute/path/to/page`

The page name should not include the file extension (.rst). For example:

:doc:`/book/controller`

:doc:`/components/event_dispatcher/introduction`

:doc:`/cookbook/configuration/environments`

The title of the linked page will be automatically used as the text of the link. If you want to modify that title, use this alternative syntax:

1	:doc:`Spooling Email </cookbook/email/spool>`

Note

Although they are technically correct, avoid the use of relative internal links such as the following, because they break the references in the generated PDF documentation:

:doc:`controller`

:doc:`event_dispatcher/introduction`

:doc:`environments`

Links to the API follow a different syntax, where you must specify the type of the linked resource (namespace, class or method):

:namespace:`SWP\\Component\\MultiTenancy`

:class:`SWP\\Component\\MultiTenancy\\Context\\TenantContext`

:method:`SWP\\Component\\MultiTenancy\\Context\\TenantContext::setTenant`

Links to the PHP documentation follow a pretty similar syntax:

:phpclass:`SimpleXMLElement`

:phpmethod:`DateTime::createFromFormat`

:phpfunction:`iterator_to_array`

New Features or Behaviour Changes¶

If you’re documenting a brand new feature or a change that’s been made in Superdesk Publisher, you should precede your description of the change with a .. versionadded:: 2.X directive and a short description:

.. versionadded:: 2.3
    The ``askHiddenResponse`` method was introduced in Superdesk Publisher 2.3.

You can also ask a question and hide the response. This is particularly [...]

If you’re documenting a behaviour change, it may be helpful to briefly describe how the behaviour has changed:

.. versionadded:: 2.3
    The ``include()`` function is a new MultiTenancy feature that's available in
    Superdesk Publisher 2.3. Prior, the ``{% include %}`` tag was used.

At this point, all the versionadded tags for Superdesk Publisher versions that have reached end-of-maintenance will be removed. For example, if Superdesk Publisher 2.5 were released today, and 2.2 had recently reached its end-of-life, the 2.2 versionadded tags would be removed from the new 2.5 branch.

Testing Documentation¶

When submitting new content to the documentation repository or when changing any existing resource, an automatic process will check if your documentation is free of syntax errors and is ready to be reviewed.

Nevertheless, if you prefer to do this check locally on your own machine before submitting your documentation, follow these steps:

Install Sphinx;
Install the Sphinx extensions using git submodules: git submodule update --init;
Run make html and view the generated HTML in the _build/html directory.

Documentation Standards¶

In order to help the reader as much as possible and to create code examples that look and feel familiar, you should follow these standards.

Sphinx¶

The following characters are chosen for different heading levels: level 1 is = (equal sign), level 2 - (dash), level 3 ~ (tilde), level 4 . (dot) and level 5 " (double quote);
Each line should break approximately after the first word that crosses the 72nd character (so most lines end up being 72-78 characters);
The :: shorthand is preferred over .. code-block:: php to begin a PHP code block (read the Sphinx documentation to see when you should use the shorthand);
Inline hyperlinks are not used. Separate the link and their target definition, which you add on the bottom of the page;
Inline markup should be closed on the same line as the open-string;

Example¶

Example
=======

When you are working on the docs, you should follow the
`Superdesk Publisher Documentation`_ standards.

Level 2
-------

A PHP example would be::

    echo 'Hello World';

Level 3
~~~~~~~

.. code-block:: php

    echo 'You cannot use the :: shortcut here';

.. _`Superdesk Publisher Documentation`: http://superdesk-publisher.readthedocs.io/en/latest

Code Examples¶

The code follows the Superdesk Publisher Coding Standards as well as the Twig Coding Standards;
The code examples should look real for a web application context. Avoid abstract or trivial examples (foo, bar, demo, etc.);
The code should follow the Symfony Best Practices. Unless the example requires a custom bundle, make sure to always use the AppBundle bundle to store your code;
Use Acme when the code requires a vendor name;
Use example.com as the domain of sample URLs and example.org and example.net when additional domains are required. All of these domains are reserved by the IANA.
To avoid horizontal scrolling on code blocks, we prefer to break a line correctly if it crosses the 85th character;
When you fold one or more lines of code, place ... in a comment at the point of the fold. These comments are: // ... (php), # ... (yaml/bash), {# ... #} (twig),  (xml/html), ; ... (ini), ... (text);
When you fold a part of a line, e.g. a variable value, put ... (without comment) at the place of the fold;
Description of the folded code: (optional)
- If you fold several lines: the description of the fold can be placed after the ...;
- If you fold only part of a line: the description can be placed before the line;
If useful to the reader, a PHP code example should start with the namespace declaration;
When referencing classes, be sure to show the use statements at the top of your code block. You don’t need to show all use statements in every example, just show what is actually being used in the code block;
If useful, a codeblock should begin with a comment containing the filename of the file in the code block. Don’t place a blank line after this comment, unless the next line is also a comment;
You should not put a $ in front of every bash line, as it doesn’t work when readers copy and paste into the console.

Formats¶

Configuration examples should show all supported formats using configuration blocks. The supported formats (and their orders) are:

Configuration (including services): YAML, XML, PHP
Routing: Annotations, YAML, XML, PHP
Validation: Annotations, YAML, XML, PHP
Doctrine Mapping: Annotations, YAML, XML, PHP
Translation: XML, YAML, PHP

Example¶

// src/Foo/Bar.php
namespace Foo;

use Acme\Demo\Cat;
// ...

class Bar
{
    // ...

    public function foo($bar)
    {
        // set foo with a value of bar
        $foo = ...;

        $cat = new Cat($foo);

        // ... check if $bar has the correct value

        return $cat->baz($bar, ...);
    }
}

Caution

In YAML you should put a space after { and before } (e.g. { _controller: ... }), but this should not be done in Twig (e.g. {'hello' : 'value'}).

Files and Directories¶

When referencing directories, always add a trailing slash to avoid confusions with regular files (e.g. “execute the console script located at the app/ directory”).
When referencing file extensions explicitly, you should include a leading dot for every extension (e.g. “XML files use the .xml extension”).
When you list a Superdesk Publisher file/directory hierarchy, use your-project/ as the top level directory. E.g.
1 2 3 4 5
your-project/ ├─ app/ ├─ src/ ├─ vendor/ └─ ...

English Language Standards¶

Superdesk Publisher documentation uses the English dialect, sometimes called British English by people who don’t know any better. Collins Dictionary is used as the vocabulary reference.

In addition, documentation follows these rules:

Section titles: use a variant of the title case, where the first word is always capitalized and all other words are capitalized, except for the closed-class words (read Wikipedia article about headings and titles).

E.g.: The Vitamins are in my Fresh California Raisins
Punctuation: avoid the use of Serial (Oxford) Commas;
Pronouns: avoid the use of nosism and always use you instead of we. (i.e. avoid the first person point of view: use the second instead);
Gender-neutral language: when referencing a hypothetical person, such as “a user with a session cookie”, use gender-neutral pronouns (they/their/them). For example, instead of:
- he or she, use they
- him or her, use them
- his or her, use their
- his or hers, use theirs
- himself or herself, use themselves

Contributed Documentation License¶

The Superdesk Publisher documentation is licensed under a Creative Commons Attribution-Share Alike 3.0 Unported License (CC BY-SA 3.0).

You are free:

to Share — to copy, distribute and transmit the work;
to Remix — to adapt the work.

Under the following conditions:

Attribution — You must attribute the work in the manner specified by the author or licensor (but not in any way that suggests that they endorse you or your use of the work);
Share Alike — If you alter, transform, or build upon this work, you may distribute the resulting work only under the same or similar license to this one.

With the understanding that:

Waiver — Any of the above conditions can be waived if you get permission from the copyright holder;
Public Domain — Where the work or any of its elements is in the public domain under applicable law, that status is in no way affected by the license;
Other Rights — In no way are any of the following rights affected by the license:
- Your fair dealing or fair use rights, or other applicable copyright exceptions and limitations;
- The author’s moral rights;
- Rights other persons may have either in the work itself or in how the work is used, such as publicity or privacy rights.
Notice — For any reuse or distribution, you must make clear to others the license terms of this work. The best way to do this is with a link to this web page.

This is a human-readable summary of the Legal Code (the full license).

Translations¶

The Superdesk Publisher documentation has not been translated yet.

If you would like to help translate this documentation, please email us at contact@sourcefabric.org or find us on GitHub (https://github.com/superdesk/web-publisher).

Community¶

Community Reviews¶

Superdesk Publisher is an open-source project driven by its community. If you don’t feel ready to contribute code or patches, reviewing issues and pull requests (PRs) can be a great start to get involved and give back. In fact, people who “triage” issues are the backbone to Superdesk Publisher’s success!

Why Reviewing Is Important¶

Community reviews are essential for the development of the Superdesk Publisher project. On the Superdesk Publisher JIRA bug tracker and GitHub, you can find many items to work on:

Bug Reports: Bug reports need to be checked for completeness. Is any important information missing? Can the bug be easily reproduced?
Pull Requests: Pull requests contain code that fixes a bug or implements new functionality. Reviews of pull requests ensure that they are implemented properly, are covered by test cases, don’t introduce new bugs and maintain backwards compatibility.

Note that anyone who has some basic familiarity with Symfony and PHP can review bug reports and pull requests. You don’t need to be an expert to help.

Be Constructive¶

Before you begin, remember that you are looking at the result of someone else’s hard work. A good review comment thanks the contributor for their work, identifies what was done well, identifies what should be improved and suggests a next step.

Create a GitHub Account¶

Superdesk Publisher uses GitHub to manage pull requests. If you want to do reviews, you need to create a GitHub account and log in.

Create a JIRA Account¶

Superdesk Publisher uses JIRA to manage bug reports. If you want to report a bug, you need to create a JIRA account and log in.

The Pull Request Review Process¶

Reviews of pull requests usually take a little longer since you need to understand the functionality that has been fixed or added and find out whether the implementation is complete.

It is okay to do partial reviews! If you do a partial review, comment how far you got and leave the PR in “needs review” state.

Pick a pull request from the PRs in need of review and follow these steps:

Is the PR Complete? Every pull request must contain a header that gives some basic information about the PR. You can find the template for that header in the Contribution Guidelines.
Is the Base Branch Correct? GitHub displays the branch that a PR is based on below the title of the pull request. Is that branch correct?
Reproduce the Problem Read the issue that the pull request is supposed to fix. Reproduce the problem on a clean Superdesk Web Publisher project and try to understand why it exists. If the linked issue already contains such a project, install it and run it on your system.
Review the Code Read the code of the pull request and check it against some common criteria:
- Does the code address the issue the PR is intended to fix/implement?
- Does the PR stay within scope to address only that issue?
- Does the PR contain automated tests? Do those tests cover all relevant edge cases?
- Does the PR contain sufficient comments to easily understand its code?
- Does the code break backwards compatibility? If yes, does the PR header say so?
- Does the PR contain deprecations? If yes, does the PR header say so? Does the code contain trigger_error() statements for all deprecated features?
- Are all deprecations and backwards compatibility breaks documented in the latest UPGRADE-X.X.md file? Do those explanations contain “Before”/”After” examples with clear upgrade instructions?
Note

Eventually, some of these aspects will be checked automatically.
Test the Code
Update the PR Status

At last, add a comment to the PR. Thank the contributor for working on the PR.

Example

Here is a sample comment for a PR that is not yet ready for merge:

Thank you @takeit for working on this! It seems that your test
cases don't cover the cases when the counter is zero or smaller.
Could you please add some tests for that?

Other Resources¶

In order to follow what is happening in the community you might find helpful these additional resources:

List of open pull requests
List of recent commits
List of open bugs and enhancements

Code
Documentation
Community
- Community Reviews
- Other Resources

Code
Documentation
Community
- Community Reviews
- Other Resources

Components¶

Note

This section is based on Sylius documentation released under a Creative Commons Attribution-Share Alike 3.0 Unported License.

The Components are a set of decoupled PHP libraries, which are the foundation of the Superdesk Web Publisher project, but they can also be used in other PHP projects which don’t rely on the Symfony framework itself.

The goal of these components is to solve the problems related to content publishing and content rendering.

How to Install and Use the Components¶

Note

This section is based on Symfony2 documentation.

If you’re starting a new project (or already have a project) that will use one or more components, the easiest way to integrate everything is with Composer. Composer is smart enough to download the component(s) that you need and take care of autoloading so that you can begin using the libraries immediately.

This article will take you through using The MultiTenancy Component, though this applies to using any component.

Using the MultiTenancy Component¶

1. If you’re creating a new project, create a new empty directory for it.

2. Open a console and use Composer to grab the library.

1	composer require swp/multi-tenancy

The name swp/multi-tenancy is written at the top of the documentation for the component.

Tip

Install composer if you don’t have it already present on your system. Depending on how you install, you may end up with a composer.phar file in your directory. In that case, no worries! Just run php composer.phar require swp/multi-tenancy.

3. Write your code!

Once Composer has downloaded the component(s), all you need to do is include the vendor/autoload.php file that was generated by Composer. This file takes care of autoloading all of the libraries so that you can use them immediately.

<?php
// File example: src/script.php

// update this to the path to the "vendor/"
// directory, relative to this file
require_once __DIR__.'/../vendor/autoload.php';

use SWP\Component\MultiTenancy\Context\TenantContext;
use SWP\Component\MultiTenancy\Model\Tenant;

$tenant = new Tenant();
$tenantContext = new TenantContext();

$tenantContext->setTenant($tenant);

var_dump($tenantContext->getTenant());

// ...

Now what?¶

Now that the component is installed and autoloaded, read the specific component’s documentation to find out more about how to use it.

And have fun!

The MultiTenancy Component¶

This component provides basic services, models and interfaces to build multi-tenant PHP apps.

Installation¶

You can install the component in two different ways:

Install it via Composer (swp/multi-tenancy on Packagist);
Use the official Git repository (https://github.com/SuperdeskWebPublisher/multi-tenancy).

Note

This section is based on Symfony2 documentation.

Then, require the vendor/autoload.php file to enable the autoloading mechanism provided by Composer. Otherwise, your application won’t be able to find the classes of this Symfony component.

Usage¶

TenantContext¶

The TenantContext allows you to manage the currently used tenant.

<?php

// ..
use SWP\Component\MultiTenancy\Context\TenantContext;
use SWP\Component\MultiTenancy\Model\Tenant;

$tenant = new Tenant();
$tenantContext = new TenantContext();

$tenantContext->setTenant($tenant);

var_dump($tenantContext->getTenant());

Note

This service implements TenantContextInterface.

TenantProvider¶

The TenantProvider allows you to get all available tenants.

<?php

 // ..
use SWP\Component\MultiTenancy\Provider\TenantProvider;

$tenantProvider = new TenantProvider(/* SWP\Component\MultiTenancy\Repository\TenantRepositoryInterface repository */);

var_dump($tenantProvider->getAvailableTenants());

The getAvailableTenants method retrieves all tenants which have the enabled property set to true and have been inserted in the given repository.

Note

This service implements the TenantProviderInterface.

TenantResolver¶

The TenantResolver allows you to resolve the current tenant from the request.

<?php

// ..
use SWP\Component\MultiTenancy\Resolver\TenantResolver;

$tenantResolver = new TenantResolver('example.com', /* SWP\Component\MultiTenancy\Repository\TenantRepositoryInterface repository */);

var_dump($tenantResolver->resolve('tenant.example.com')); // will return an instance of TenantInterface.

The resolve method resolves the tenant based on the current subdomain name. For example, when the host is tenant.example.com, it will resolve the subdomain (tenant) and then it will search for the tenant in the given repository, by the resolved subdomain name. If the subdomain tenant is not found, it always returns the default tenant.

Note

This service implements the TenantResolverInterface.

TenantAwarePathBuilder¶

The TenantAwarePathBuilder responsibility is to build PHPCR base paths which are tenant-aware. This can build whatever path is needed by the provided paths’ names and the given context.

<?php

// ..
use SWP\Component\MultiTenancy\PathBuilder\TenantAwarePathBuilder;
use SWP\Component\MultiTenancy\Context\TenantContext;
use SWP\Component\MultiTenancy\Model\Tenant;

$tenant = new Tenant();
$tenant->setSubdomain('example');
$tenant->setName('Example tenant');
$tenantContext = new TenantContext();
$tenantContext->setTenant($tenant);

$pathBuilder = new TenantAwarePathBuilder($tenantContext, '/swp');

var_dump($pathBuilder->build('routes')); // will return: /swp/example/routes.
var_dump($pathBuilder->build(['routes', 'content'])); // will return an array: ['/swp/example/routes', '/swp/example/routes']
var_dump($pathBuilder->build('/')); // will return: /swp/default

The build method method builds the PHPCR path. It accepts as a first argument, a string or an array of routes’ names. The second argument is the context for the given path(s) name(s).

In order to build the base paths, the TenantAwarePathBuilder’s construct requires an object of type TenantResolverInterface to be provided as a first argument, the object of type TenantContextInterface as a second argument, and the root base path as a third argument.

Note

This service implements the TenantResolverInterface.

TenantFactory¶

The TenantFactory allows you to create an objects of type TenantInterface.

<?php

 // ..
use SWP\Component\MultiTenancy\Model\Tenant;
use SWP\Component\MultiTenancy\Factory\TenantFactory;

$tenantFactory = new TenantFactory(Tenant::class);
$tenant = $tenantFactory->create();

var_dump($tenant);

OrganizationFactory¶

The OrganizationFactory allows you to create objects of type OrganizationInterface.

<?php

 // ..
use SWP\Component\MultiTenancy\Model\Organization;
use SWP\Component\MultiTenancy\Factory\OrganizationFactory;
use SWP\Component\Common\Generator\RandomStringGenerator;

$organizationFactory = new OrganizationFactory(Organization::class, new RandomStringGenerator());
$organization = $organizationFactory->create();
$organizationWithCode = $organizationFactory->createWithCode();

var_dump($organization);

Models¶

Tenant¶

Every tenant is represented by a Tenant model which by default has the following properties:

Method	Description
id	Unique identifier
subdomain	Tenant’s subdomain
name	Tenant’s name
code	Tenant’s code
enabled	Indicates whether the tenant is enabled
createdAt	Date of creation
updatedAt	Date of last update
deletedAt	Indicates whether the tenant is deleted

Note

This model implements TenantInterface.

Organization¶

Every organization is represented by a Organization model which by default has the following properties:

Method	Description
id	Unique identifier
name	Tenant’s name
code	Tenant’s code
enabled	Indicates whether the tenant is enabled
createdAt	Date of creation
updatedAt	Date of last update
deletedAt	Indicates whether the tenant is deleted

Note

This model implements OrganizationInterface.

Interfaces¶

Model Interfaces¶

TenantInterface¶

This interface should be implemented by every tenant.

Note

This interface extends TimestampableInterface, EnableableInterface and SoftDeleteableInterface.

OrganizationInterface¶

This interface should be implemented by every organization.

TenantAwareInterface¶

This interface should be implemented by models associated with a specific tenant.

Service Interfaces¶

TenantContextInterface¶

This interface should be implemented by a service responsible for managing the currently used Tenant.

TenantProviderInterface¶

This interface should be implemented by a service responsible for providing all available tenants.

TenantResolverInterface¶

This interface should be implemented by a service responsible for resolving the current Tenant.

Repository Interfaces¶

TenantRepositoryInterface¶

This interface should be implemented by repositories responsible for storing the Tenant objects.

OrganizationRepositoryInterface¶

This interface should be implemented by repositories responsible for storing the OrganizationInterface objects.

Factory Interfaces¶

TenantFactoryInterface¶

This interface should be implemented by tenant factories which are responsible for creating objects of type TenantInterface.

OrganizationFactoryInterface¶

This interface should be implemented by organization factories which are responsible for creating objects of type OrganizationInterface.

The Storage Component¶

This component provides basic models and interfaces to build persistence-agnostic storage.

Installation¶

You can install the component in two different ways:

Install it via Composer (swp/storage on Packagist);
Use the official Git repository (https://github.com/SuperdeskWebPublisher/storage).

Note

This section is based on Symfony2 documentation.

Then, require the vendor/autoload.php file to enable the autoloading mechanism provided by Composer. Otherwise, your application won’t be able to find the classes of this Symfony component.

Usage¶

Creating storage-agnostic objects¶

The Factory allows you to create objects. Let’s say you need to instantiate a new Article object, which you would want to persist in the persistence storage.

You could use the following code to do that:

<?php

// ..
use SWP\Component\Storage\Factory\Factory;
use Acme\DemoBundle\Article;

$factory = new Factory(Article::class);
$article = $factory->create();

var_dump($article); // dumps Article object

By passing the fully qualified class name in Factory’s construct we allow for a flexible way to create storage-agnostic objects. For example, depending on the persistence backend, such as PHPCR or MongoDB, you can instantiate new objects very easily using the same factory class.

Creating storage-agnostic repositories¶

In some cases you would need to implement different repositories for your persistence backend. Let’s say you are using Doctrine PHPCR and you want to have a generic way of adding and removing objects from the storage, even if you decide to change Doctrine PHPCR to Doctrine MongoDB or something else. By implementing RepositoryInterface in your new repository, you will be able to achive that.

Here’s the example PHPCR repository implementation:

<?php

namespace SWP\Bundle\StorageBundle\Doctrine\ODM\PHPCR;

use Doctrine\ODM\PHPCR\DocumentRepository as BaseDocumentRepository;
use SWP\Component\Storage\Model\PersistableInterface;
use SWP\Component\Storage\Repository\RepositoryInterface;

class DocumentRepository extends BaseDocumentRepository implements RepositoryInterface
{
    /**
     * {@inheritdoc}
     */
    public function add(PersistableInterface $object)
    {
        $this->dm->persist($object);
        $this->dm->flush();
    }

    /**
     * {@inheritdoc}
     */
    public function remove(PersistableInterface $object)
    {
        if (null !== $this->find($object->getId())) {
            $this->dm->remove($object);
            $this->dm->flush();
        }
    }
}

In this case, all objects that need to be persisted should implement PersistableInterface which extends Doctrine’s default Doctrine\Common\Persistence\ObjectRepository interface. This component gives you simple interfaces to create storage-agnostic repositories.

The Bridge Component¶

This component provides tools which help to make a connection between Superdesk and Superdesk Web Publisher.

Installation¶

You can install the component in 2 different ways:

Install it via Composer (swp/bridge on Packagist);
Use the official Git repository (https://github.com/SuperdeskWebPublisher/bridge).

Note

This section is based on Symfony2 documentation.

Then, require the vendor/autoload.php file to enable the autoloading mechanism provided by Composer. Otherwise, your application won’t be able to find the classes of this Symfony component.

Usage¶

Validators¶

This component provides one type of validator by default:

ninjs validator

This is a simple implementation of JSON format validation against a concrete schema. There is also a custom implementation of the ninjs validator which validates the specific Superdesk ninjs schema.

Superdesk Ninjs Validator¶

This validator validates against specific Superdesk ninjs schema.

Usage:

<?php
// example.php
// ..

use SWP\Component\Bridge\Validator\NinjsValidator;

$validator = new NinjsValidator();
if ($validator->isValid('{some ninjs format value}')) {
    // valid
}

// not valid

The Superdesk ninjs schema:

{
    "$schema": "http://json-schema.org/draft-03/schema#",
    "id" : "http://www.iptc.org/std/ninjs/ninjs-schema_1.1.json#",
    "type" : "object",
    "title" : "IPTC ninjs - News in JSON - version 1.1 (approved, 2014-03-12) / document revision of 2014-11-15: geometry_* moved under place",
    "description" : "A news item as JSON object -- copyright 2014 IPTC - International Press Telecommunications Council - www.iptc.org - This document is published under the Creative Commons Attribution 3.0 license, see  http://creativecommons.org/licenses/by/3.0/  $$comment: as of 2014-03-13 ",
    "additionalProperties" : false,
    "patternProperties" : {
        "^description_[a-zA-Z0-9_]+" : {
            "description" : "A free-form textual description of the content of the item. (The string appended to description_ in the property name should reflect the format of the text)",
            "type" : "string"
        },
        "^body_[a-zA-Z0-9_]+" : {
            "description" : "The textual content of the news object. (The string appended to body_ in the property name should reflect the format of the text)",
            "type" : "string"
        }
    },
    "properties" : {
        "guid" : {
            "description" : "The identifier for this news object",
            "type" : "string",
            "format" : "guid",
            "required" : true
        },
        "type" : {
            "description" : "The generic news type of this news object",
            "type" : "string",
            "enum" : ["text", "audio", "video", "picture", "graphic", "composite"]
        },
        "slugline" : {
            "description" : "The slugline",
            "type" : "string",
            "required" : true
        },
        "mimetype" : {
            "description" : "A MIME type which applies to this news object",
            "type" : "string"
        },
        "representationtype" : {
            "description" : "Indicates how complete this representation of a news item is",
            "type" : "string",
            "enum" : ["complete", "incomplete"]
        },
        "profile" : {
            "description" : "An identifier for the kind of content of this news object",
            "type" : "string"
        },
        "version" : {
            "description" : "The version of the news object which is identified by the uri property",
            "type" : "string"
        },
        "versioncreated" : {
            "description" : "The date and time when this version of the news object was created",
            "type" : "string",
            "format" : "date-time"
        },
        "embargoed" : {
            "description" : "The date and time before which all versions of the news object are embargoed. If absent, this object is not embargoed.",
            "type" : "string",
            "format" : "date-time"
        },
        "pubstatus" : {
            "description" : "The publishing status of the news object, its value is *usable* by default.",
            "type" : "string",
            "enum" : ["usable", "withheld", "canceled"]
        },
        "urgency" : {
            "description" : "The editorial urgency of the content from 1 to 9. 1 represents the highest urgency, 9 the lowest.",
            "type" : "number"
        },
        "priority" : {
            "description" : "The editorial priority of the content from 1 to 9. 1 represents the highest priority, 9 the lowest.",
            "type" : "number"
        },
        "copyrightholder" : {
            "description" : "The person or organisation claiming the intellectual property for the content.",
            "type" : "string"
        },
        "copyrightnotice" : {
            "description" : "Any necessary copyright notice for claiming the intellectual property for the content.",
            "type" : "string"
        },
        "usageterms" : {
            "description" : "A natural-language statement about the usage terms pertaining to the content.",
            "type" : "string"
        },
        "language" : {
            "description" : "The human language used by the content. The value should follow IETF BCP47",
            "type" : "string"
        },
        "service" : {
            "description" : "A service e.g. World Photos, UK News etc.",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of a service",
                        "type" : "string"
                    },
                    "code" : {
                        "description": "The code for the service in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "person" : {
            "description" : "An individual human being",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of a person",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the person",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the person",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the person in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "organisation" : {
            "description" : "An administrative and functional structure which may act as as a business, as a political party or not-for-profit party",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of the organisation",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the organisation",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the organisation",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the organisation in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    },
                    "symbols" : {
                        "description" : "Symbols used for a finanical instrument linked to the organisation at a specific market place",
                        "type" : "array",
                        "items" : {
                            "type" : "object",
                            "additionalProperties" : false,
                            "properties" : {
                                "ticker" : {
                                    "description" : "Ticker symbol used for the financial instrument",
                                    "type": "string"
                                },
                                "exchange" : {
                                    "description" : "Identifier for the marketplace which uses the ticker symbols of the ticker property",
                                    "type" : "string"
                                }
                            }
                        }
                    }
                }
            }
        },
        "place" : {
            "description" : "A named location",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "patternProperties" : {
                    "^geometry_[a-zA-Z0-9_]+" : {
                        "description" : "An object holding geo data of this place. Could be of any relevant geo data JSON object definition.",
                        "type" : "object"
                    }
                },
                "properties" : {
                    "name" : {
                        "description" : "The name of the place",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the place",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the place",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "qcode" : {
                        "description": "The code for the place in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    },
                    "state" : {
                        "description" : "The state for the place",
                        "type" : "string"
                    },
                    "group" : {
                        "description" : "The place group",
                        "type" : "string"
                    },
                    "name" : {
                        "description" : "The place name",
                        "type" : "string"
                    },
                    "country" : {
                        "description" : "The country name",
                        "type" : "string"
                    },
                    "world_region" : {
                        "description" : "The world region",
                        "type" : "string"
                    }
                }
            }
        },
        "subject" : {
            "description" : "A concept with a relationship to the content",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of the subject",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the subject",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the subject",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the subject in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "event" : {
            "description" : "Something which happens in a planned or unplanned manner",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of the event",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the event",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the event",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the event in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "object" : {
            "description" : "Something material, excluding persons",
            "type" : "array",
            "items" : {
                "type" : "object",
                "additionalProperties" : false,
                "properties" : {
                    "name" : {
                        "description" : "The name of the object",
                        "type" : "string"
                    },
                    "rel" : {
                        "description" : "The relationship of the content of the news object to the object",
                        "type" : "string"
                    },
                    "scheme" : {
                        "description" : "The identifier of a scheme (= controlled vocabulary) which includes a code for the object",
                        "type" : "string",
                        "format" : "uri"
                    },
                    "code" : {
                        "description": "The code for the object in a scheme (= controlled vocabulary) which is identified by the scheme property",
                        "type" : "string"
                    }
                }
            }
        },
        "byline" : {
            "description" : "The name(s) of the creator(s) of the content",
            "type" : "string"
        },
        "headline" : {
            "description" : "A brief and snappy introduction to the content, designed to catch the reader's attention",
            "type" : "string"
        },
        "located" : {
            "description" : "The name of the location from which the content originates.",
            "type" : "string"
        },
        "renditions" : {
            "description" : "Wrapper for different renditions of non-textual content of the news object",
            "type" : "object",
            "additionalProperties" : false,
            "patternProperties" : {
                "^[a-zA-Z0-9]+" : {
                    "description" : "A specific rendition of a non-textual content of the news object.",
                    "type" : "object",
                    "additionalProperties" : false,
                    "properties" : {
                        "href" : {
                            "description" : "The URL for accessing the rendition as a resource",
                            "type" : "string",
                            "format" : "uri"
                        },
                        "mimetype" : {
                            "description" : "A MIME type which applies to the rendition",
                            "type" : "string"
                        },
                        "title" : {
                            "description" : "A title for the link to the rendition resource",
                            "type" : "string"
                        },
                        "height" : {
                            "description" : "For still and moving images: the height of the display area measured in pixels",
                            "type" : "number"
                        },
                        "width" : {
                            "description" : "For still and moving images: the width of the display area measured in pixels",
                            "type" : "number"
                        },
                        "sizeinbytes" : {
                            "description" : "The size of the rendition resource in bytes",
                            "type" : "number"
                        }
                    }
                }
            }
        },
        "associations" : {
            "description" : "Content of news objects which are associated with this news object.",
            "type" : "object",
            "additionalProperties" : false,
            "patternProperties" : {
                "^[a-zA-Z0-9]+" :  { "$ref": "http://www.iptc.org/std/ninjs/ninjs-schema_1.0.json#" }
            }
        }
    }
}

Validator Chain¶

You could also use Validator Chain to validate the json value with many validators at once:

<?php
// example.php
// ..

use SWP\Component\Bridge\ValidatorChain;
use SWP\Component\Bridge\Validator\NinjsValidator;

$validatorChain = ValidatorChain();
$validatorChain->addValidator(new NinjsValidator(), 'ninjs');
$validatorChain->addValidator(new CustomValidator(), 'custom');

if ($validatorChain->isValid('{json value}')) {
    // valid
}

// not valid

Data Transformers¶

Transformers are meant to transform an incoming value to an object representation.

This component supports one transformer by default:

JSON to Package

JSON to Package Data Transformer¶

This transforms a JSON string which is first validated by the Validator Chain. If the validation is a success, it serializes the JSON value to a Package object.

The Package object is a one-to-one representation of Superdesk Package.

Usage:

<?php
// example.php
// ..

use SWP\Component\Bridge\Transformer\JsonToPackageTransformer;

$transformer = new JsonToPackageTransformer();
$package = $transformer->transform('{json value}');

var_dump($package);die; // will dump an instance of ``SWP\Component\Bridge\Model\Package`` object.

This transformer could support reverse transform, but it is not supported at the moment.

The example below will throw an SWP\Component\Bridge\Exception\MethodNotSupportedException exception:

<?php
// example.php
// ..

use SWP\Component\Bridge\Model\Package;
use SWP\Component\Bridge\Transformer\JsonToPackageTransformer;

$package = new Package();
// ..
$transformer = new JsonToPackageTransformer();
$package = $transformer->reverseTransform($package);

Note

If the transformation fails for some reason, an exception SWP\Component\Bridge\Exception\TransformationFailedException will be thrown.

Data Transformer Chain¶

You can use Transformer Chain to transform any value by many transformers at once:

<?php
// example.php
// ..

use SWP\Component\Bridge\Transformer\DataTransformerChain;
use SWP\Component\Bridge\Transformer\JsonToPackageTransformer;

$transformerChain = DataTransformerChain(new JsonToPackageTransformer(), /* new CustomTransformer() */);
$result = $transformer->transform(/* some value or object */);

var_dump($result); // result of transformation

Note

If the transformation fails for some reason an exception SWP\Component\Bridge\Exception\TransformationFailedException will be thrown.

To reverse transform, use the reverseTransform method:

<?php
// example.php
// ..

use SWP\Component\Bridge\Transformer\DataTransformerChain;
use SWP\Component\Bridge\Transformer\JsonToPackageTransformer;

$transformerChain = DataTransformerChain(new JsonToPackageTransformer(), /* new CustomTransformer() */);
$result = $transformer->reverseTransform(/* some value or object */);

var_dump($result); // result of transformation

The Templates System Component¶

This component provides useful features for templates and themes systems based on Twig templating engine.

Installation¶

You can install the component in two different ways:

Install it via Composer (swp/templates-system on Packagist);
Use the official Git repository (https://github.com/SuperdeskWebPublisher/templates-system) and autoload files manually.

Features¶

Context and Meta objects¶

Why Meta objects?¶

Meta objects provides extra layer between your internal documents/entities and this what is available for theme developer (Templator). Thanks to this feature you can make more changes in Your project code and data structures without breaking templates.

How to create Meta?¶

Every Meta object requires Context, Value and Configuration.

Context - it’s special service class used for collecting all meta’s and resolving meta objects inside other meta’s
Value - object or array with data
Configuration - array with configuration definitions for provided object.

Create Meta manually:

<?php

...
use SWP\Component\TemplatesSystem\Gimme\Meta\Meta;

return new Meta($context, $value, $configuration);

Create Meta with MetaFactory:

<?php

...
use SWP\Component\TemplatesSystem\Gimme\Factory\MetaFactory;
use SWP\Component\TemplatesSystem\Gimme\Meta\Meta;

$metaFactory = new MetaFactory($context);

return $metaFactory->create($value, $configuration);

What is Context?¶

Context is a special service class used for collecting all meta’s and resolving meta objects inside other meta’s. It can collect all available configurations for meta’s, and convert provided objects into meta’s when there is configuration for it.

Note

When property of Meta object can be itself a Meta instance (there is configuration for it) Context will automatically process it.

Example yaml configuration file for object (context can read config from .yml files).

name: article
class: "SWP\\Component\\TemplatesSystem\\Tests\\Article"
description: Article Meta is representation of Article in Superdesk Web Publisher.
properties:
    title:
        description: "Article title, max 160 characters, can't have any html tags"
        type: text
    keywords:
        description: "Article keywords"
        type: text
to_string: title

Note

Configurations are used to manually expose properties from provided data, and create documentation for templators.

All objects passed to template should be Meta’s.

Loading Meta from template¶

Meta Loaders¶

Meta Loader provides easy way for fetching data directly from template file. Loader can return single meta or collection of meta’s (with MetaCollection class).

Library provides ChainLoader class witch can simplify work with many loaders.

Meta Loader must implement Loader Interface.

How to load single Meta?¶

gimme¶

The tag gimme has one required parameter and one optional parameter:

(required) Meta object type (and name of variable available inside block), for example: article

(optional) Keword with and parameters for Meta Loader, for example: { param: "value" }

{% gimme article %}
    {# article Meta will be available under "article" variable inside block #}
    {{ article.title }}
{% endgimme %}

Meta Loaders sometimes require special parameters - like the article number, language of the article, user id, etc..

{% gimme article with { articleNumber: 1 } %}
    {# Meta Loader will use provided parameters to load article Meta #}
    {{ article.title }}
{% endgimme %}

How to load Meta collection?¶

gimmelist¶

The gimmelist tag has two required parameters and two optional parameters:

(required) Name of variable available inside block: article

(required) Keyword from and type of requested Metas in collection: from articles with filters passed to Meta Loader as extra parameters (start, limit, order)

(optional) Keyword with and parameters for Meta Loader, for example: with {foo: 'bar', param1: 'value1'}

(optional) Keyword without and parameters for Meta Loader, for example: without {source: 'AAP'}

(optional) Keyword if and expression used for results filtering

(optional) Keyword ignoreContext and optional array of selected meta to be ignored

Example of the required parameters:

{% gimmelist article from articles %}
    {{ article.title }}
{% endgimmelist %}

Example with ignoring selected context parameters:

{% gimmelist article from articles ignoreContext ['route', 'article'] %}
...

Example with ignoring whole context

{% gimmelist article from articles ignoreContext [] %}
...

Or even without empty array

{% gimmelist article from articles ignoreContext %}
...

Example with filtering articles by metadata:

{% gimmelist article from articles with {metadata: {byline: "Karen Ruhiger", located: "Sydney"}} %}
    {{ article.title }}
{% endgimmelist %}

The above example will list all articles by metadata which contain byline equals to Karen Ruhiger AND located equals to Sydney.

To list articles by authors you can also do:

{% gimmelist article from articles with {author: ["Karen Ruhiger", "Doe"]} %}
    {{ article.title }}
    Author(s): {% for author in article.authors %}<img src="{{ url(author.avatar) }}" />{{ author.name }} ({{ author.role }}) {{ author.biography }} - {{ author.jobTitle.name }},{% endfor %}
{% endgimmelist %}

It will then list all articles written by Karen Ruhiger AND Doe.

To list articles from the Forbes source but without an AAP source you can also do:

{% gimmelist article from articles with {source: ["Forbes"]} without {source: ["AAP"]} %}
    {% for source in article.sources %} {{ source.name }} {% endfor %}
{% endgimmelist %}

It will then list all articles with source Forbes and without AAP.

Listing article’s custom fields:

{% gimmelist article from articles %}
    {{ article.title }}
    {{ article.extra['my-custom-field'] }}
{% endgimmelist %}

Example with usage of all parameters:

{% gimmelist article from articles|start(0)|limit(10)|order('id', 'desc')
    with {foo: 'bar', param1: 'value1'}
    contextIgnore ['route', 'article']
    if article.title == "New Article 1"
%}
    {{ article.title }}
{% endgimmelist %}

Custom Twig tags¶

Gimme allows you to fetch the Meta object you need in any place of your template. It supports single Meta objects (with gimme ) and collections of Meta objects (with gimmelist).

gimme¶

The tag gimme has one required parameter and one optional parameter:

(required) Meta object type (and name of variable available inside block), for example: article

(optional) Keword with and parameters for Meta Loader, for example: { param: "value" }

{% gimme article %}
    {# article Meta will be available under "article" variable inside block #}
    {{ article.title }}
{% endgimme %}

Meta Loaders sometimes require special parameters - like the article number, language of the article, user id, etc..

{% gimme article with { articleNumber: 1 } %}
    {# Meta Loader will use provided parameters to load article Meta #}
    {{ article.title }}
{% endgimme %}

gimmelist¶

The gimmelist tag has two required parameters and two optional parameters:

(required) Name of variable available inside block: article

(required) Keyword from and type of requested Metas in collection: from articles with filters passed to Meta Loader as extra parameters (start, limit, order)

(optional) Keyword with and parameters for Meta Loader, for example: with {foo: 'bar', param1: 'value1'}

(optional) Keyword without and parameters for Meta Loader, for example: without {source: 'AAP'}

(optional) Keyword if and expression used for results filtering

(optional) Keyword ignoreContext and optional array of selected meta to be ignored

Example of the required parameters:

{% gimmelist article from articles %}
    {{ article.title }}
{% endgimmelist %}

Example with ignoring selected context parameters:

{% gimmelist article from articles ignoreContext ['route', 'article'] %}
...

Example with ignoring whole context

{% gimmelist article from articles ignoreContext [] %}
...

Or even without empty array

{% gimmelist article from articles ignoreContext %}
...

Example with filtering articles by metadata:

{% gimmelist article from articles with {metadata: {byline: "Karen Ruhiger", located: "Sydney"}} %}
    {{ article.title }}
{% endgimmelist %}

The above example will list all articles by metadata which contain byline equals to Karen Ruhiger AND located equals to Sydney.

To list articles by authors you can also do:

{% gimmelist article from articles with {author: ["Karen Ruhiger", "Doe"]} %}
    {{ article.title }}
    Author(s): {% for author in article.authors %}<img src="{{ url(author.avatar) }}" />{{ author.name }} ({{ author.role }}) {{ author.biography }} - {{ author.jobTitle.name }},{% endfor %}
{% endgimmelist %}

It will then list all articles written by Karen Ruhiger AND Doe.

To list articles from the Forbes source but without an AAP source you can also do:

{% gimmelist article from articles with {source: ["Forbes"]} without {source: ["AAP"]} %}
    {% for source in article.sources %} {{ source.name }} {% endfor %}
{% endgimmelist %}

It will then list all articles with source Forbes and without AAP.

Listing article’s custom fields:

{% gimmelist article from articles %}
    {{ article.title }}
    {{ article.extra['my-custom-field'] }}
{% endgimmelist %}

Example with usage of all parameters:

{% gimmelist article from articles|start(0)|limit(10)|order('id', 'desc')
    with {foo: 'bar', param1: 'value1'}
    contextIgnore ['route', 'article']
    if article.title == "New Article 1"
%}
    {{ article.title }}
{% endgimmelist %}

How to work with `gimmelist` pagination?¶

gimmelist is based on Twig for tag, like in Twig there is loop variable available. In addition to default loop properties there is also totalLength. It’s filled by loader with number of total elements in storage which are matching criteria. Thanks to this addition we can build real pagination.

TemplateEngine Bundle provides simple default pagination template file: pagination.html.twig.

Note

You can override that template with SWPTemplatesSystemBundle/views/pagination.html.twig file in Your theme. Or You can use own file used for pagination rendering.

Here is commented example of pagination:

{# Setup list and pagination parameters #}
{% set itemsPerPage, currentPage = 1, app.request.get('page', 1) %}
{% set start = (currentPage / itemsPerPage) - 1 %}

{# List all articles from route '/news' and limit them to `itemsPerPage` value starting from `start` value #}
{% gimmelist article from articles|start(start)|limit(itemsPerPage) with {'route': '/news'} %}

    <li><a href="{{ url(article) }}">{{ article.title }} </a></li>

    {# Render pagination only at end of list #}
    {% if loop.last  %}
        {#
            Use provided by default pagination template

            Parameters:
            * currentFilters (array) : associative array that contains the current route-arguments
            * currentPage (int) : the current page you are in
            * paginationPath (Meta|string) : the route name (or supported by router Meta object) to use for links
            * lastPage (int) : represents the total number of existing pages
            * showAlwaysFirstAndLast (bool) : Always show first and last link (just disabled)
        #}
        {% include '@SWPTemplatesSystem/pagination.html.twig' with {
            currentFilters: {}|merge(app.request.query.all()),
            currentPage: currentPage,
            paginationPath: gimme.route,
            lastPage: (loop.totalLength/itemsPerPage)|round(1, 'ceil'),
            showAlwaysFirstAndLast: true
        } only %}
    {% endif %}
{% endgimmelist %}

For referrence, see original pagination.html.twig template (if you want to customize it and use instead of default one):

{#
  Source: http://dev.dbl-a.com/symfony-2-0/symfony2-and-twig-pagination/
  Updated by: Simon Schick <simonsimcity@gmail.com>

  Parameters:
    * currentFilters (array) : associative array that contains the current route-arguments
    * currentPage (int) : the current page you are in
    * paginationPath (string) : the route name to use for links
    * showAlwaysFirstAndLast (bool) : Always show first and last link (just disabled)
    * lastPage (int) : represents the total number of existing pages
#}
{% spaceless %}
    {% if lastPage > 1 %}

        {# the number of first and last pages to be displayed #}
        {% set extremePagesLimit = 3 %}

        {# the number of pages that are displayed around the active page #}
        {% set nearbyPagesLimit = 2 %}

        <nav class="pagination">
            <div class="numbers">
                <ul>
                    {% if currentPage > 1 %}
                        <li><a href="{{ path(paginationPath, currentFilters|merge({page: currentPage-1})) }}">Previous</a></li>

                        {% for i in range(1, extremePagesLimit) if ( i < currentPage - nearbyPagesLimit ) %}
                            <li><a href="{{ path(paginationPath, currentFilters|merge({page: i})) }}">{{ i }}</a></li>
                        {% endfor %}

                        {% if extremePagesLimit + 1 < currentPage - nearbyPagesLimit %}
                            <span class="sep-dots">...</span>
                        {% endif %}

                        {% for i in range(currentPage-nearbyPagesLimit, currentPage-1) if ( i > 0 ) %}
                            <li><a href="{{ path(paginationPath, currentFilters|merge({page: i})) }}">{{ i }}</a></li>
                        {% endfor %}
                    {% elseif showAlwaysFirstAndLast %}
                        <span class="disabled">Previous</span>
                    {% endif %}

                    <li class="current"><a href="{{ path(paginationPath, currentFilters|merge({ page: currentPage })) }}">{{ currentPage }}</a></li>

                    {% if currentPage < lastPage %}
                        {% for i in range(currentPage+1, currentPage + nearbyPagesLimit) if ( i <= lastPage ) %}
                            <li><a href="{{ path(paginationPath, currentFilters|merge({page: i})) }}">{{ i }}</a></li>
                        {% endfor %}

                        {% if  (lastPage - extremePagesLimit) > (currentPage + nearbyPagesLimit) %}
                            <li><span class="sep-dots">...</span></li>
                        {% endif %}

                        {% for i in range(lastPage - extremePagesLimit+1, lastPage) if ( i > currentPage + nearbyPagesLimit ) %}
                            <li><a href="{{ path(paginationPath, currentFilters|merge({page: i})) }}">{{ i }}</a></li>
                        {% endfor %}

                        <li><a href="{{ path(paginationPath, currentFilters|merge({page: currentPage+1})) }}">Next</a></li>
                    {% elseif showAlwaysFirstAndLast %}
                        <li><span class="disabled">Next</span></li>
                    {% endif %}
                </ul>
            </div>
        </nav>
    {% endif %}
{% endspaceless %}

How to work with Meta objects¶

On the template level, every variable in Context and fetched by gimme and gimmelist is a representation of Meta objects.

dump

{{ dump(article) }}

print

{{ article }}

If the meta configuration has the to_string property then the value of this property will be printed, otherwise it will be represented as JSON.

access property

{{ article.title }}
{{ article['title']}}

generate url

{{ url(article) }}    // absolute url
{{ path(article) }}   // relative path

Here’s an example using gimmelist:

{% gimmelist article from articles %}
    <li><a href="{{ url(article) }}">{{ article.title }} </a></li>
{% endgimmelist %}

The Rule Component¶

This component provides basic business rules engine tools.

Installation¶

You can install the component in two different ways:

Install it via Composer (swp/rule on Packagist);
Use the official Git repository (https://github.com/SuperdeskWebPublisher/rule).

Note

This section is based on Symfony2 documentation.

Then, require the vendor/autoload.php file to enable the autoloading mechanism provided by Composer. Otherwise, your application won’t be able to find the classes of this Symfony component.

Usage¶

Using Rule Applicator Chain Service¶

The Rule Applicator Chain service is used to execute all rule applicators added to the chain.

Usage:

<?php
 // example.php
 // ..

 use Acme\DemoBundle\Model\Subject;
 use SWP\Component\Rule\Applicator\RuleApplicatorChain;
 use SWP\Component\Rule\Model\Rule;
 // ..

 $applicatorChain = new RuleApplicatorChain();
 $applicatorChain->addApplicator(/* instance of RuleApplicatorInterface */)

 $subject = new Subject(); // an instance of RuleSubjectInterface
 $applicatorChain->isSupported($subject); // return true or false

 $rule = new Rule();
 // ..

 $applicatorChain->apply($rule, $subject);

What is Rule Evaluator?¶

Rule evaluators are used to evaluate rule on an given object and make sure it matches the rule’s criteria. By default, the Symfony Expression Language Component is used which perfectly fits into the business rules engine concept.

Adding new Rule Evaluator¶

There is a possibility to create your custom implementation for Rule Evaluator. All you need to do is to create a new class and implement SWP\Component\Rule\Evaluator\RuleEvaluatorInterface interface.

What is Rule Processor?¶

Rule processor is responsible for processing all rules and apply them respectively to an object, based on the defined rule priority. The greater the priority value, the higher the priority.

Rule Processor implements SWP\Component\Rule\Processor\RuleProcessorInterface interface.

What is Rule Applicator?¶

Rule applicators are used to apply given rule’s configuration to an object. You create your custom rule applicators and register them in Rule Applicator Chain service which triggers apply method on them if the given applicator is supported by the rule subject.

Adding new Rule Applicator¶

There is a possibility to create your custom implementation for Rule Applicator. All you need to do is to create a new class and implement SWP\Component\Rule\Applicator\RuleApplicatorInterface interface.

The Content List Component¶

This component provides basic classes to build dynamic and flexible content lists.

Installation¶

You can install the component in two different ways:

Install it via Composer (swp/content-list on Packagist);
Use the official Git repository (https://github.com/SuperdeskWebPublisher/content-list).

Note

This section is based on Symfony2 documentation.

Then, require the vendor/autoload.php file to enable the autoloading mechanism provided by Composer. Otherwise, your application won’t be able to find the classes of this Symfony component.

Models¶

ContentList¶

Every content list is represented by a ContentList model which by default has the following properties:

Method	Description
id	Unique identifier
description	List’s description
name	List’s name
type	List’s type (`automatic` or `manual`)
cacheLifeTime	List cache life time in seconds
limit	List limit
items	Collection of list items
enabled	Indicates whether the list is enabled
createdAt	Date of creation
updatedAt	Date of last update
deletedAt	Indicates whether the list is deleted

Note

This model implements SWP\Component\ContentList\Model\ContentListInterface.

ContentListItem¶

Every content list item is represented by a ContentListItem model which by default has the following properties:

Method	Description
id	Unique identifier
position	List item position
content	Object of type `ListContentInterface`
enabled	Indicates whether the item is enabled
sticky	Defines whether content is sticky or not
filters	An array of list criteria/filters
createdAt	Date of creation
updatedAt	Date of last update
deletedAt	Indicates whether the item is deleted

Note

Repository Interface¶

This component contains SWP\Component\ContentList\Repository\ContentListRepositoryInterface interface which can be used by your custom entity repository, in order to make it working with ContentListBundle.

The Output Channel Component¶

This component provides the implementation of output channels.

Installation¶

You can install the component in two different ways:

Install it via Composer (swp/output-channel on Packagist);
Use the official Git repository (https://github.com/SuperdeskWebPublisher/OutputChannel).

Note

This section is based on Symfony2 documentation.

Then, require the vendor/autoload.php file to enable the autoloading mechanism provided by Composer. Otherwise, your application won’t be able to find the classes of this Symfony component.

Models¶

OutputChannel¶

Every output channel is represented by a OutputChannel model which by default has the following properties:

Method	Description
id	Unique identifier
type	Output channel’s type (`wordpress` etc.)
config	Output channel configuration per type

Note

This model implements SWP\Component\OutputChannel\Model\OutputChannelInterface.

Output Channel Aware Interface¶

This component contains SWP\Component\OutputChannel\Model\OutputChannelAwareInterface interface to make your custom entity/model output channel aware.

The Paywall Component¶

This component provides the paywall implementation.

Installation¶

You can install the component in two different ways:

Install it via Composer (swp/paywall on Packagist);
Use the official Git repository (https://github.com/SuperdeskWebPublisher/Paywall).

Note

This section is based on Symfony2 documentation.

Then, require the vendor/autoload.php file to enable the autoloading mechanism provided by Composer. Otherwise, your application won’t be able to find the classes of this Symfony component.

Models¶

Subscription¶

Every subscription is represented by a Subscription model which by default has the following properties:

Method	Description
id	Unique identifier
type	Subscription’s type (`recurring` etc.)
details	Subscription details
code	Subscription’s unique code
active	Subscription’s status
updatedAt	Subscription updated at datetime
createdAt	Subscription created at datetime

Note

This model implements SWP\Component\Paywall\Model\SubscriptionInterface.

Interfaces¶

Subscriber Interface¶

This component contains SWP\Component\Paywall\Model\SubscriberInterface interface which should be implemented by your user class.

Paywall Secured Interface¶

This component contains SWP\Component\Paywall\Model\PaywallSecuredInterface interface which should be implemented by classes that must be flagged as “paywall secured”.

Paywall Secured Trait¶

This component contains SWP\Component\Paywall\Model\PaywallSecuredTrait trait which adds a special property along with getter and setter. By using this trait it is possible to flag your objects as paywall secured.

Factories¶

Subscription Factory¶

A factory class to create new instances of SWP\Component\Paywall\Model\Subscription class.

Method	Description
create()	Create a new instance of Subscription

Adapters¶

Adapters are used to interact with the external subscription systems to fetch existing subscriptions.

PaymentsHub Adapter¶

This adapter interacts with the Payments Hub API to fetch the subscriptions data.

Note

This model implements SWP\Component\Paywall\Adapter\PaywallAdapterInterface.

Paywall Adapter Interface¶

This interface should be implemented by your custom adapter.

The SEO Component¶

This component provides the implementation of SEO metadata (Facebook, LinkedIn etc.)

Installation¶

You can install the component in two different ways:

Install it via Composer (swp/seo on Packagist);
Use the official Git repository (https://github.com/SuperdeskWebPublisher/seo).

Note

This section is based on Symfony2 documentation.

Then, require the vendor/autoload.php file to enable the autoloading mechanism provided by Composer. Otherwise, your application won’t be able to find the classes of this Symfony component.

(Un)publishing content¶

Deciding if content is visible for user in Publisher is done dynamically - based on user roles.

Note

By default all users can see only published content, but if user have role ROLE_CAN_VIEW_NON_PUBLISHED then also not published content will be fetched and rendered.

How published content is marked in database?¶

Published content need to match those criteria:

status property set to published

publishedAt property filled with date and time

isPublishable property set to true

How to Publish content?¶

By REST API call

curl -X "PATCH" -d "article[status]=published" -H "Content-type:\ application/x-www-form-urlencoded" /api/v1/content/articles/get-involved

By code

// $this->container - instance of Service Container
// $article - ArticleInterface implementation
$articleService = $this->container->get('swp.service.article');
$articleService->publish($article);

How to check (in code) if content is published?¶

use Symfony\Cmf\Bundle\CoreBundle\PublishWorkflow\PublishWorkflowChecker;
....
$publishWorkflowChecker = $this->serviceContainer->get('swp.publish_workflow.checker');

if ($publishWorkflowChecker->isGranted(PublishWorkflowChecker::VIEW_ATTRIBUTE, $article)) {
    // give access for article
}

Note

Content assigned to routes is automatically checked (if it’s publishable for current user) by event listener.

How to Un-publish Content?¶

If an article is published you can easily un-publish it via API as described above. Another way to un-publish already published article is by killing the article. It can be achieved by setting the value of pubStatus property to canceled in the JSON (Ninjs) content according to IPTC standards. Once that status will be set, and the content will be send to Publisher (/content/push API endpoint), article will be un-published immediately and its status will be set to canceled.

Publisher API¶

API authentication¶

Internal API endpoints require user authentication (user need to have ROLE_INTERNAL_API role assigned).

Authentication data (token) must be attached to every request with Authorization header or auth_token query parameter.

Get authentication token for registered user¶

To get authentication token you need to call /api/v1/auth with your username and password - in response you will get your user information’s and token data.

Example:

curl 'http://publisher.dev/api/v1/auth' -d 'auth%5Busername%5D=username&auth%5Bpassword%5D=password' --compressed

Note

Publisher token will be valid for 48 hours

Get authentication token for superdesk user¶

To get authentication token you need to call /api/v1/auth/superdesk with superdesk legged in user session_id and token - in response you will get your user information’s and token data.

Example:

curl 'http://publisher.dev/api/v1/auth/superdesk' -d 'auth_superdesk%5Bsession_id%5D=5831599634d0c100405d84c7&auth_superdesk%5Btoken%5D=Basic YTRmMWMzMTItODlkNS00MzQzLTkzYjctZWMyMmM5ZGMzYWEwOg==' --compressed

Publisher in background will ask authorized superdesk server for user session (and user data). If Superdesk will confirm session information then Publisher will get internal user (or create one if not exists) and create token for him.

Note

Publisher token will be this same as the one from superdesk (provided in /api/v1/auth/superdesk request).

Generate Authentication URL for Livesite Editor¶

You can create with API special authentication URL for tenant website. To do that you need to call /api/v1/livesite/auth/livesite_editor as authorized user (with token in request header or url).

curl 'http://publisher.dev/api/v1/livesite/auth/livesite_editor' -H 'Authorization: d6O3UorCHZ2Pd8PRs/0aXGg1qnT0bKUPWW43dgKqYm3CI4U4Og==' --compressed

In response you will get JSON with Your token details and special URL which can be used for authentication and Livesite Editor activation.

After following that url you will be redirected to tenant homepage. Meantime special cookie with name activate_livesite_editor will be set. This cookie will have API token set as it’s value. It would best if you will set token value in browser local storage and remove cookie (so it will not be send to server with every request).

Contributing to Web Publisher¶

Code
Documentation
Community
- Community Reviews
- Other Resources

Welcome to Superdesk Publisher documentation¶

Manual¶

Introduction¶

Why choose Publisher?¶

What is Publisher’s focus?¶

How is Publisher structured?¶

Notable features¶

Website architecture¶

Getting started¶

Installing Superdesk Publisher¶

Installing Publisher Prerequisites¶

Install PHP-FPM 7.3 and Extensions¶

Install PostgreSQL¶

Install Memcached¶

Install ElasticSearch¶

Install and Configure Nginx Server¶

Install RabbitMQ Server¶

Install Supervisor¶

Installing Publisher¶

Clone the Publisher Repository¶

Install Publisher Dependencies¶

Create and Populate the Database¶

Generate the SSH keys to properly use the authentication (readers)¶

Create an Organization and Tenant¶

Install the Demo Theme¶

Configure Supervisor¶

Next Steps¶

How to Install and Configure Superdesk Publisher with Superdesk¶

Install the Superdesk Publisher Component in Superdesk¶

Update Your Superdesk Configuration File¶

Update Your Publisher Configuration File¶

Configure Subscriber to Publish Content from Superdesk to Superdesk Publisher¶

Configuring Publisher¶

Managing websites and routes¶

Navigation management¶

Configure image formats¶

Templates System¶

Rendering pages with Twig¶

Creating a page templates¶

Creating custom templates¶

Setting error pages from theme¶

Testing error pages during theme development¶

Properties¶

Templates features¶

Custom Twig tags¶

gimme¶

gimmelist¶

How to work with gimmelist pagination?¶

How to work with Meta objects¶

Stringy twig extensions¶

Redirects¶

redirect¶

notFound¶

Handling Articles¶

Listing Articles¶

Fetching first article url (when it’s changed after slug or route change)¶

Handling Related Articles¶

Listing a collection of Related Articles¶

Handling Article Media¶

Listing Article Media¶

Article Media¶

Image Renditions¶

Feature Media¶

Handling Article Authors¶

Listing a Single Article’s Author¶

Listing a collection of Article’s Authors¶

Handling Article Slideshows¶

Listing a Single Article’s Slideshow¶

Listing a collection of Article’s Slideshows¶

Listing all Article’s Slideshows Items¶

Listing all Article’s Slideshows and its Items¶

Listing a Single Slideshow and its Items by Name¶

Handling Routes¶

Listing a Single Route¶

Listing a Routes Collection¶

Handling Content List Items¶

Listing Content List Items¶

Content List¶

Content List Items¶

Keywords¶

How to work with `gimmelist` pagination?¶

Load assets from the theme’s public directory (`app/themes/<theme-name>/public`)¶

Load assets from the public `web` directory¶

Create new route with extension (example: `feed/sitemap.rss`)¶