Quick setup guide¶
This is just a quick guide for installing and starting to use Weblate for testing purposes. Please check Installation instructions for more real world setup instructions.
Installing from sources¶
Install all required dependencies, see Software requirements.
Grab Weblate sources (either using Git or download a tarball) and unpack them, see Installing Weblate.
weblate/settings.pyand adjust it to match your setup. You will at least need to configure database connection (possibly adding user and creating the database). Check Configuration for Weblate specific configuration options.
Create database which will be used by Weblate, see Database setup for Weblate.
./manage.py migrate ./manage.py collectstatic ./scripts/generate-locales # If you are using Git checkout
Configure webserver to serve Weblate, see Running server.
Using prebuilt appliance¶
- Download the appliance and start it. You need to choose format depending on your target environment.
- Everything should be set up immediately after boot, though you will want to adjust some settings to improve security, see SUSE Studio appliance.
Installing on OpenShift¶
You can install Weblate on OpenShift PaaS directly from its git repository using the OpenShift Client Tools:
rhc -aweblate app create -t python-2.7 --from-code https://github.com/WeblateOrg/weblate.git --no-git
After installation everything should be preconfigured and you can immediately start to add a translation project as described below. For more information, including on how to retrieve the generated admin password, see Weblate on OpenShift.
All you need to specify here is project name and its website.
Create component which is the real object for translating - it points to VCS repository and selects which files to translate. See Component configuration for more details.
The important fields here being component name, VCS repository address and mask for finding translatable files. Weblate supports wide range of formats including Gettext PO files, Android resource strings, OS X string properties, Java properties or Qt Linguist files, see Supported formats for more details.
Once above is completed (it can be lengthy process depending on size of your VCS repository and number of messages to translate), you can start translating.