++ed by:

1 PAUSE user
1 non-PAUSE user.

Author image Dave Rolsky


Silki::Manual::Devel - Hack on Silki


version 0.29


In order to hack on Silki, you'll need to make sure that you can connect to Postgres without a username or password as yourself, and that you can create and drop databases.

In the future, this requirement may change, and patches are certainly welcome.


You'll need to install several additional prereqs, including Catalyst::Devel, Catalyst::Plugin::Static::Simple, and Catalyst::Plugin::StackTrace.

Silki uses Dist::Zilla to build the distribution. Unless you are planning on hacking on the release tool portion of Silki, you won't need this, but installing it may be useful. You'll also need a number of dzil plugins.


The Silki distribution includes a number of tools that are useful for development:

  • bin/silki_server.pl

    This runs a standalone server for Silki. This is very handy for development.

    You can make it auto-restart when relevant files change. Here's the best command to do this:

      bin/silki_server.pl -r \
        --restart_directory lib \
        --restart_directory share/i18n \
        -rr (?:\/|^)(?!\.\#).+(?:\.pm|\.po)$ \
  • dev-bin/extract-po

    This extracts strings which need localization. If you're just using English, you don't need to extract new strings immediately. By default, the English string is used as both the msgid and the translation of that string in English.

  • dev-bin/manage-db

    This rebuilds the Silki schema and optionally populates it with some initial users and wikis. Run it with the --help flag to see options. If you want to drop the existing schema you'll have to pass the --drop flag.

    Typically, you'll run it like this:

      > dev-bin/manage-db --db-name Silki --drop --seed

    Be careful not to run this on a system where you have a production install of Silki! You could easily destroy a real database.

  • dev-bin/make-test-schema

    Some of the tests create a Fey::Schema object without talking to Postgres. This is faster than using Fey::Loader. This is done by using Storable to store a copy of the schema in a file. You can regenerate the serialized schema by running this command.

  • dev-bin/populate-db

    This inserts a whole bunch of pages into the database. This script could be improved to add more variations in the data, more user, revisions, etc.


Dave Rolsky <autarch@urth.org>


This software is Copyright (c) 2011 by Dave Rolsky.

This is free software, licensed under:

  The GNU Affero General Public License, Version 3, November 2007