[open-data-manual] Sphinx/Transifex translation tool chain

[Dinu Gherman]

Hi,

I'm sorry, but after sifting through the archives and quite
a few web pages I still don't understand how the .po(t)
files for transifex.net are generated.

I hoped to find a Sphinx extension that would convert each
paragraph and headline into the .po(t) file, but haven't
found one, so I assume this is done by some dedicated script
that splits the input at empty lines and feeds the content
into some template. Is that how it works? If so, I wonder
why this tool isn't in the bin folder, but anyway.

The real question is about the process behind that. People
write and edit the text, but how and when is it carried
over to transifex? Is it automatic? If it is what prevents
authors/editors from rearranging paragraphs such that the
.po(t) entries get screwed-up?

So far I have only translated inside transifex but not set
up my own project, so I'd like to understand how this is
designed to work, but as I said, I haven't found any infor-
mation whatsoever. If I missed something I'd be glad to re-
ceive a pointer (or more).

Thanks,

Dinu