Mats Bengtsson wrote: > John Mandereau wrote: > > > > It's possible to make lilypond-book translate comments and variable > > names, and still have one lily-* snippet, through a gettext mechanism. > > I estimate it'd take me 2 hours for implementing and testing this > > properly. lilypond-book would try to translate all variable names and > > comments in snippets for verbatim output, and translators would simply > > have to translate the desired strings in Documentation/po/LANG.po. It > > would require no more work for doc editors.
I've been thinking too much about it, so it's sold now. Graham, you can definitely revert the "no comments in snippets" policy. > I don't think it would have to add so many lines of code to > lilypond-book, since > gettext already is used in lilypond-book to translate error messages. Yes, but we need to load lilypond-doc domain, which lives in Documentation/po/out-www. I did this in buildscripts through a command-line argument, but lilypond-book has already a lot of command-line options. I'm thinking about adding loading of lilypond-doc message catalogs in langdefs.py, and moving this module to python/, so lilypond-book can import langdefs and use lilypond-doc domain seamlessly. > Also, I don't > see any advantage in terms of maintainability to modularize it. I'm not an experienced programmer, so if you say it, I believe you :-). I raised the modularization issue because it's in the TODO in the script, and I wonder if 2000 lines is too long for a Python script. Cheers, John _______________________________________________ lilypond-devel mailing list [email protected] http://lists.gnu.org/mailman/listinfo/lilypond-devel
