One issue that comes to my mind is how to make the fonts and TeX header files available to teTeX. So far, we have done it by setting up a separate texmf/ tree with all these files and adding it to the teTeX search tree by setting TEXMF. If we split into separate distributions, we would have to take care of two separate texmf/ file trees (at least with the current naming convention with the release number in the directory name). Even though we can just duplicate the current solution, I can still think of several problems.
Also, one version of the fonts mostly works together with some other version of LilyPond, but I can think of several occasions where an update involved changes to both the program and the fonts so both had to be updated together.
One step in the direction you outline would be for the package maintainers to split into two RPMs or two Debian packages. However, it needs some extra job to make sure to keep the dependencies up to date. For the Debian package, the teTeX integration is already done by linking the files into the ordinary texmf/ tree, so that's not a problem.
/Mats
Ferenc Wagner wrote:
Dear Gurus,
I think better separating the font stuff from the main distribution would have the following advantages:
1. Tracking LilyPond development would be easier for the non-uber-hackers due to a) shorter build times and b) less build dependencies. 2. Creating and distributing development branch packages would be easier for the (non-uber-hacker) packagers for the same a) reason. 3. Downloading development packages would require less bandwidth. 4. Storing packages for various architectures would require much less storage space in the distribution archives and their mirrors (this issue has already been raised by Debian, which hosts the most architectures AFAIK. The lilypond-doc package also goes in this direction.)
So I'd like to hear your word:
1. Is this split feasible at all, ie. are the fonts really architecture independent? 2. Are you willing to move them in a separate repository and issue separate release notes? 3. Have you got a better idea how this should be done instead? 4. Can you think of any related issues?
-- ============================================= Mats Bengtsson Signal Processing Signals, Sensors and Systems Royal Institute of Technology SE-100 44 STOCKHOLM Sweden Phone: (+46) 8 790 8463 Fax: (+46) 8 790 7260 Email: [EMAIL PROTECTED] WWW: http://www.s3.kth.se/~mabe =============================================
_______________________________________________ Lilypond-devel mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/lilypond-devel
