Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
Jackalope implementation with the jackrabbit backend
PHP Perl

This branch is 189 commits behind jackalope:master

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.
bin
doc/config
src/Jackalope
tests
.gitignore
.travis.yml
LICENSE
README.md
cli-config.php.dist
composer.json

README.md

Jackalope Build Status

A powerful implementation of the PHPCR API.

Jackalope binding for the jackrabbit backend server.

Discuss on jackalope-dev@googlegroups.com or visit #jackalope on irc.freenode.net

License: This code is licenced under the apache license. Please see the file LICENSE in this folder.

Preconditions

Installation

If you do not yet have composer, install it like this

curl -s http://getcomposer.org/installer | sudo php -- --install-dir=/usr/local/bin

To install jackalope itselves, run the following in the parent directory of where you want jackalope

git clone git://github.com/jackalope/jackalope-jackrabbit.git
cd jackalope-jackrabbit
php /usr/local/bin/composer.phar install --dev

Note that the --dev parameter is only needed if you want to be able to run the test suite. If you already installed jackalope without the test suite, you need to remove composer.lock before running composer again with the --dev parameter.

Jackrabbit storage server

Besides the Jackalope repository, you need the Jackrabbit server component. For instructions, see Jackalope Wiki Make sure you have at least the version specified in the VERSION constant of the protocol implementation

phpunit tests

If you want to run the tests, please see the README file in the tests folder and check if you told composer to install the suggested dependencies (see Installation)

Enable the commands

There are a couple of useful commands to interact with the repository.

To use the console, copy cli-config.php.dist to cli-config.php and configure the connection parameters. Then you can run the commands from the jackalope directory with ./bin/jackalope

NOTE: If you are using PHPCR inside of Symfony, the DoctrinePHPCRBundle provides the commands inside the normal Symfony console and you don't need to prepare anything special.

Jackalope specific commands:

  • jackalope:run:jackrabbit [--jackrabbit_jar[="..."]] [start|stop|status]: Start and stop the Jackrabbit server

Commands available from the phpcr-utils:

  • phpcr:workspace:create <name>: Create a workspace name in the repository
  • phpcr:register-node-types --allow-update [cnd-file]: Register namespaces and node types from a "Compact Node Type Definition" .cnd file
  • phpcr:dump [--sys_nodes[="..."]] [--props[="..."]] [path]: Show the node names under the specified path. If you set sys_nodes=yes you will also see system nodes. If you set props=yes you will additionally see all properties of the dumped nodes.
  • phpcr:purge: Remove all content from the configured repository in the configured workspace
  • phpcr:sql2: Run a query in the JCR SQL2 language against the repository and dump the resulting rows to the console.

Bootstrapping

Jackalope relies on autoloading. Namespaces and folders are compliant with PSR-0. You should use the autoload file generated by composer: vendor/.composer/autoload.php

If you want to integrate jackalope into other PSR-0 compliant code and use your own classloader, find the mapping in vendor/.composer/autoload_namespaces.php

Once you have autoloading, you need to bootstrap the library. A minimalist sample code to get a PHPCR session with the jackrabbit backend:

$jackrabbit_url = 'http://127.0.0.1:8080/server/';
$user           = 'admin';
$pass           = 'admin';
$workspace      = 'default';

$repository = \Jackalope\RepositoryFactoryJackrabbit::getRepository(
    array('jackalope.jackrabbit_uri' => $jackrabbit_url)
);
$credentials = new \PHPCR\SimpleCredentials($user, $pass);
$session = $repository->login($credentials, $workspace);

To use a workspace different than default you need to create it first. The easiest is to run the command bin/jackalope phpcr:workspace:create <myworkspace> but you can of course also use the PHPCR API to create workspaces from your code.

Usage

The entry point is to create the repository factory. The factory specifies the storage backend as well. From this point on, there are no differences in the usage (except for supported features, that is).

// see Bootstrapping for how to get the session.

$rootNode = $session->getNode("/");
$whitewashing = $rootNode->addNode("www-whitewashing-de");
$session->save();

$posts = $whitewashing->addNode("posts");
$session->save();

$post = $posts->addNode("welcome-to-blog");
$post->addMixin("mix:title");
$post->setProperty("jcr:title", "Welcome to my Blog!");
$post->setProperty("jcr:description", "This is the first post on my blog! Do you like it?");

$session->save();

See PHPCR Tutorial for a more detailed tutorial on how to use the PHPCR API.

Query Languages

Jackalope supports the PHPCR standard query language SQL2 as well as the Query Object Model (QOM) to build queries programmatically. We recommend using the QOM or the QueryBuilder mentioned in the PHPCR Tutorial. They are built to use the best possible query language depending on the capabilities of the backend. A later switching to another PHPCR implementation shouldn't cause any issues then.

Jackalope-Jackrabbit also supports the depricated SQL and XPath query languages from JCR 1.0. Those languages will be suported by Jackrabbit for the forseeable future, but almost certainly won't be supported by other PHPCR implementations. So use them with care and only if you know what you are doing.

One reason for using SQL or XPath is that the newer and more capable SQL2 is not as optimized as the older languages on the Jackrabbit side. Queries with large resultsets are much slower with SQL2 than with XPath or SQL.

However, the best is to use the QueryBuilder mentioned above to let the implementation chose the most efficient query language for your implementation.

Performance tweaks

If you know that you will need many child nodes of a node you are about to request, you can use Session::setSessionOption with Session::OPTION_FETCH_DEPTH to something bigger than 1. This will prefetch the children to reduce the round trips to the database.

Implementation notes

See doc/architecture.md for an introduction how Jackalope is built. Have a look at the source files and generate the phpdoc.

TODO

The best overview of what needs to be done are the skipped API tests. Have a look at JackrabbitImplementationLoader to see what is currently not working and start hacking :-)

Contributors

Something went wrong with that request. Please try again.