During the last year the Docs Project has made significant strides in
working with the XML in our CVS repository. Our toolchain has improved
along with our understanding of how to make better use of our XML
content assets. We are making a change that will turn quite a number of
translations "fuzzy" but the changes themselves will be quite minimal.
Some time ago, because of some toolchain deficiencies, we elected to use
the "xml2po -e" option to parse entities when creating POT and PO files.
So if our XML looked like this:
<para>This paragraph is about &FED;.</para>
The POT would use contain this msgid:
This paragraph is about Fedora.
We are moving now to a better XML handling infrastructure where the
POT's msgid will instead look like this:
This paragraph is about &FED;.
These entity definitions, such as &FED; or &FCVER;, are defined in our
docs-common/ module under docs-common/common/entities, and can be
translated using normal POT/PO tools. By translating that content, you
will allow these entities to be "filled in" after translation, and
during the build process.
Previously, if an entity changed, the msgid's using that content would
all change too. Now changes in the entities can be entirely independent
of the content and if we change &FCVER; from "7" to "8," you will
need to update any content. (Entities with actual language content
would need to be kept up-to-date, of course!)
So, while there will be some temporary minor inconvenience, in the long
run this will improve the translation process and lighten your workload.
Thank you to everyone who works so hard on the translations of Fedora
Paul W. Frields, RHCE http://paul.frields.org/
gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233 5906 ACDB C937 BD11 3717
Fedora Project: http://fedoraproject.org/wiki/PaulWFrields
: stickster @ #fedora-docs, #fedora-devel, #fredlug