Checks and fixups

Custom automatic fixups

You can also implement own automatic fixup in addition to standard ones and include them in AUTOFIX_LIST.

The automatic fixes are powerful, but can also cause damage, be careful when writing one.

For example following automatic fixup would replace every occurrence of string foo in translation with bar:

# -*- coding: utf-8 -*-
#
# Copyright © 2012 - 2014 Michal Čihař <michal@cihar.com>
#
# This file is part of Weblate <http://weblate.org/>
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
#

from weblate.trans.autofixes.base import AutoFix
from django.utils.translation import ugettext_lazy as _


class ReplaceFooWithBar(AutoFix):
    '''
    Replaces foo with bar.
    '''

    name = _('Foobar')

    def fix_single_target(self, target, source, unit):
        if 'foo' in target:
            return target.replace('foo', 'bar'), True
        return target, False

Customizing checks

Fine tuning existing checks

You can fine tune checks for each source strings or in the Component configuration, here is current list of flags accepted:

rst-text
Treat text as RST document, affects Unchanged translation.
python-format, c-format, php-format, python-brace-format
Treats all string like format strings, affects Format strings, Format strings, Format strings, Format strings, Unchanged translation.
ignore-*
Ignores given check for a component.

These flags are understood both in Component configuration settings and in translation file itself (eg. in GNU Gettext).

Writing own checks

Weblate comes with wide range of quality checks (see Quality checks), though they might not 100% cover all you want to check. The list of performed checks can be adjusted using CHECK_LIST and you can also add custom checks. All you need to do is to subclass trans.checks.Check, set few attributes and implement either check or check_single methods (first one if you want to deal with plurals in your code, the latter one does this for you). You will find below some examples.

Checking translation text does not contain “foo”

This is pretty simple check which just checks whether translation does not contain string “foo”.

# -*- coding: utf-8 -*-
#
# Copyright © 2012 - 2014 Michal Čihař <michal@cihar.com>
#
# This file is part of Weblate <http://weblate.org/>
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
'''
Simple quality check example.
'''

from weblate.trans.checks.base import TargetCheck
from django.utils.translation import ugettext_lazy as _


class FooCheck(TargetCheck):

    # Used as identifier for check, should be unique
    check_id = 'foo'

    # Short name used to display failing check
    name = _('Foo check')

    # Description for failing check
    description = _('Your translation is foo')

    # Real check code
    def check_single(self, source, target, unit):
        return 'foo' in target

Checking Czech translation text plurals differ

Check using language information to verify that two plural forms in Czech language are not same.

# -*- coding: utf-8 -*-
#
# Copyright © 2012 - 2014 Michal Čihař <michal@cihar.com>
#
# This file is part of Weblate <http://weblate.org/>
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.
#
'''
Quality check example for Czech plurals.
'''

from weblate.trans.checks.base import TargetCheck
from django.utils.translation import ugettext_lazy as _


class PluralCzechCheck(TargetCheck):

    # Used as identifier for check, should be unique
    check_id = 'foo'

    # Short name used to display failing check
    name = _('Foo check')

    # Description for failing check
    description = _('Your translation is foo')

    # Real check code
    def check_target_unit(self, sources, targets, unit):
        if self.is_language(unit, ('cs', )):
            return targets[1] == targets[2]
        return False

    def check_single(self, source, target, unit, cache_slot):
        '''
        We don't check target strings here.
        '''
        return False