The existing function pull_from_zanata would only work for the simple
Python project case of pulling translations from Zanata -- with only
locales directory and trivial rules to apply, with no real hope of
working for Horizon or OpenStack Manuals, given the current
functionality of the Zanata CLI utility.
Rewrite it to download every translation from Zanata and perform some
checks by hand if we keep the file or remove it. This also required
refactoring a large portion of cleanup_po_files so both functions could
leverage the same code. OpenStack Manuals treats downloaded
translations sufficiently different that a similar function was required.
This also adds full support for Horizon and OpenStack Manuals projects,
including jobs, as well as new job templates to make it easier to
transition projects to pulling from Zanata. Adding -zanata jobs for the
Python projects and django_openstack_auth is not required, they are
already present.
Change-Id: I5b9f2eca7f099c56057fc5d09b4bbfe5ea5c596d
To help move to Zanata, we should pull down translations from
the server. Add a function that should behave the same way as
the existing Transifex function, as well as changing two existing
scripts to call it based on a command-line parameter. Add two new
jobs with a -zanata suffix so we can switch projects one at a time.
This change deals with all of the Python projects and
django_openstack_auth. Horizon and OpenStack manuals will be dealt with
in a future patch.
Change-Id: I8003ac37aeb89d59ceeba2e75eed2f8100b82367
In preperation for switching projects one at a time to pulling
translations from Zanata, refactor the two calls to tx pull into
a common function.
Change-Id: I21051fbdcf7bd210606333ba46e5abb0c64e7540
Cleanup every use of `` for subshells in the jenkins/scripts
directory, replacing them with $(), and finally making the
scripts consistent.
Change-Id: Id3e45fb873c1ebeae6ab63a0caa4c427a7ccbb62
Remove .tx file at the end to have a clean repository.
Run setup.py with parameter --quiet, the long list of "extracting
messages from file XXX" are not needed.
Change-Id: I735f02a968aed8b362803781498ccc8085fa0e03
Today's nova import had some file changes where only the
POT-Creation-Date was changed which we don't need to import.
With the new compression step, we need to filter out commits *after* the
compression, otherwise counting is wrong.
Change-Id: Ib8fe6ca5f77c805147b73fe94905c72ccf3369d4
We have some strange files in the tree that contain no or little
translation after merging - and these are downloaded from transifex.
So, these files are not known to git and git rm fails.
To not special case every git rm and git reset for both old and new
files, re-add a "git add" at the beginning so that we can operate on
files known to git.
Note: This happens when we have a file like glance-log-critical.po that has no
strings - but we had to create it with initial content in transifex,
since empty files are not allowed there.
Change-Id: Ie9bb3e59560d50829e56928b2ed79e3999dd39cb
Remove untranslated strings and location information from PO files. The
location information is still in the POT files as reference for easy
regeneration using "msgmerge POTFILE POFILE -o COMPLETEPOFILE".
This moves the git add command to the end since compress_po_files
changes the files.
Change-Id: I8d9f80276085ffe4dd614960457637330ca28370
Download translations for all languages that are in the repo.
this avoids the situation that a file gets downloaded with 75 %
translations and then not updated as contents changed and translations
do not catch up.
Change-Id: Ic3bc5fa2642aeaabee5d4c1e0c3764baed811a5d
Also create tools/run-bashate.sh to wrap a find command to search for
files to run bashate on, and use it in tox.
Change-Id: I5ab07425f566a556f158a4b76b00aa35f9fb4385
Translations might become stale and obsolete over time, let's cleanup
slowly:
* Remove empty translations files (those were added by accident in the
past or content gets removed over time)
* Remove files that have less than 20 per cent translated content.
Change-Id: I0c3f6de436d342d032fa38ce2df8a90dd878ea08
This repo was created from filter branching the openstack-infra/
config repo. This process brought a lot of cruft with it in the
form of directories that we no longer need. This patch removes
that cruft so we begin with a tidier repo.
Change-Id: Ibffad1b11c0c5f84eedfb0365369f60c4961a0f3