Andy> I'm throwing out this patch for discussion. It's kind of hard to discuss something which is probably about half (or more) unnecessary. This is more or less just an #ifdef merge of the version before Matt's patch and current 21.4, right? Eg, I saw dozens of changes from "re_bool" to "boolean" and others like changing "comstyle" back to "mask", which are probably purely cosmetic, and if so, undesirable. Please do work up something more concise. Andy> This patch makes the syntax table support optional through Andy> configure. If we're not going to revert the feature entirely, we should make it a runtime option. Perl programmers (cperl being the only current mode that uses the syntax-table property, although Alan Mackenzie is threatening us with another one) program C and Java too! -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN My nostalgia for Icon makes me forget about any of the bad things. I don't have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py

Yay!! Stefan

Andy> What confuses me is that the code in question was ported Andy> from Emacs 20.7. Did this change significantly in later Andy> versions? Or is it just that it is incompatible with our Andy> extents? Well, for starters Matt didn't understand the basic buffer structures; Ben had a patch to fix the buffer movement. Second, according to the discussion between Ben and Jan, it appears that the SOE which makes our redisplay relatively fast is pessimized for the kind of usage that the syntax cache makes of extents. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN My nostalgia for Icon makes me forget about any of the bad things. I don't have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py