Working with Message Catalogs
The gettext translation system enables you to mark any strings used in your application as subject to localization, by wrapping them in functions such as gettext(str) and ngettext(singular, plural, num). For brevity, the gettext function is often aliased to _(str), so you can write:
instead of just:
to make the string "Hello" localizable.
Message catalogs are collections of translations for such localizable messages used in an application. They are commonly stored in PO (Portable Object) and MO (Machine Object) files, the formats of which are defined by the GNU gettext tools and the GNU translation project.
The general procedure for building message catalogs looks something like this:
- use a tool (such as xgettext) to extract localizable strings from the code base and write them to a POT (PO Template) file.
- make a copy of the POT file for a specific locale (for example, "en_US") and start translating the messages
- use a tool such as msgfmt to compile the locale PO file into an binary MO file
- later, when code changes make it necessary to update the translations, you regenerate the POT file and merge the changes into the various locale-specific PO files, for example using msgmerge
Python provides the gettext module as part of the standard library, which enables applications to work with appropriately generated MO files.
As gettext provides a solid and well supported foundation for translating application messages, Babel does not reinvent the wheel, but rather reuses this infrastructure, and makes it easier to build message catalogs for Python applications.
Babel provides functionality similar to that of the xgettext program, except that only extraction from Python source files is built-in, while support for other file formats can be added using a simple extension mechanism.
Unlike xgettext, which is usually invoked once for every file, the routines for message extraction in Babel operate on directories. While the per-file approach of xgettext works nicely with projects using a Makefile, Python projects rarely use make, and thus a different mechanism is needed for extracting messages from the heterogeneous collection of source files that many Python projects are composed of.
When message extraction is based on directories instead of individual files, there needs to be a way to configure which files should be treated in which manner. For example, while many projects may contain .html files, some of those files may be static HTML files that don't contain localizable message, while others may be Django templates, and still others may contain Genshi markup templates. Some projects may even mix HTML files for different templates languages (for whatever reason). Therefore the way in which messages are extracted from source files can not only depend on the file extension, but needs to be controllable in a precise manner.
Babel accepts a configuration file to specify this mapping of files to extraction methods, which is described below.
The mapping of extraction methods to files in Babel is done via a configuration file. This file maps extended glob patterns to the names of the extraction methods, and can also set various options for each pattern (which options are available depends on the specific extraction method).
For example, the following configuration adds extraction of messages from both Genshi markup templates and text templates:
# Extraction from Python source files [python: foobar/**.py] # Extraction from Genshi HTML and text templates [genshi: foobar/**/templates/**.html] ignore_tags = script,style include_attrs = alt title summary [genshi: foobar/**/templates/**.txt] template_class = genshi.template:TextTemplate encoding = ISO-8819-15
The configuration file syntax is based on the format commonly found in .INI files on Windows systems, and as supported by the ConfigParser module in the Python standard libraries. Section names (the strings enclosed in square brackets) specify both the name of the extraction method, and the extended glob pattern to specify the files that this extraction method should be used for, separated by a colon. The options in the sections are passed to the extraction method. Which options are available is specific to the extraction method used.
The extended glob patterns used in this configuration are similar to the glob patterns provided by most shells. A single asterisk (*) is a wildcard for any number of characters (except for the pathname component separator "/"), while a question mark (?) only matches a single character. In addition, two subsequent asterisk characters (**) can be used to make the wildcard match any directory level, so the pattern **.txt matches any file with the extension .txt in any directory.
Lines that start with a # or ; character are ignored and can be used for comments. Empty lines are also ignored, too.
if you're performing message extraction using the command Babel provides for integration into setup.py scripts (see below), you can also provide this configuration in a different way, namely as a keyword argument to the setup() function.
Babel provides two different front-ends to access its functionality for working with message catalogs:
Which one you choose depends on the nature of your project. For most modern Python projects, the distutils/setuptools integration is probably more convenient.
Adding new methods for extracting localizable methods is easy. First, you'll need to implement a function that complies with the following interface:
def extract_xxx(fileobj, keywords, comment_tags, options): """Extract messages from XXX files. :param fileobj: the file-like object the messages should be extracted from :param keywords: a list of keywords (i.e. function names) that should be recognized as translation functions :param comment_tags: a list of translator tags to search for and include in the results :param options: a dictionary of additional options (optional) :return: an iterator over ``(lineno, funcname, message, comments)`` tuples :rtype: ``iterator`` """
Any strings in the tuples produced by this function must be either unicode objects, or str objects using plain ASCII characters. That means that if sources contain strings using other encodings, it is the job of the extractor implementation to do the decoding to unicode objects.
Next, you should register that function as an entry point. This requires your setup.py script to use setuptools, and your package to be installed with the necessary metadata. If that's taken care of, add something like the following to your setup.py script:
def setup(... entry_points = """ [babel.extractors] xxx = your.package:extract_xxx """,
That is, add your extraction method to the entry point group babel.extractors, where the name of the entry point is the name that people will use to reference the extraction method, and the value being the module and the name of the function (separated by a colon) implementing the actual extraction.