Werner LEMBERG wrote: > Hmm. But the current master branch doesn't call mf2pt1 at all, > AFAIK... Han-Wen hasn't merged yet his changes into the master > branch, has he?
Yes, he has, and he made some other changes too; I played a bit with git bisect, running "make clean && make" each time, and found that at least one of these commits caused this font selection problem: d855b626f5778137af77881672b21efe3f17bd1b Bugfix: uniq-list uses equal? iso. eq?. 6311cce5a6f2ba8ad59f1fa533953cb4c12c6cc4 Pfa vs. pfb fixes. 5c6a64bb795c80febcc6ce1ebcb7a25a862ebc30 Mf2pt1 fixes: run mf2pt1 in a temp dir to fix filesystem pollution. c8af5675d8b865905bad068598778269d71e3294 Change build system for mf2pt1 font generation. > Well, mf2pt1 only creates PFAs, nothing else. Huh!? after compiling with current Git, I see a lot of .pfb files in mf/out, but no .pfa; is it the source of the problem? With the old build system with mftrace, .pfa files were generated too. When looking at mf/GNUmakefile history, it looks like Han-Wen deliberately choosed to generate .pfb iso .pfa. > The wrong font selection is done elsewhere. Yes, certainly. I can't help more on this, because I'm ignorant about font selection, and I have no time to learn and investigate. Cheers, John _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-devel
