Re: fontlock-speed

Andy> Some datapoints would be great. There are a number of Andy> possible culprits: He sent me some profiles, which point straight at at scan-lists. CC'd xemacs-beta. I took a look at syntax.c (scan_lists). It looks like it's related to Matt's syntax stuff after all. What's happening is that the UPDATE_SYNTAX_CACHE macros are getting called at least once per movement of point. These in turn call get-text-properties when the cache gets invalidated, which is every time point crosses an extent boundary. get-text-properties is a pretty heavy function to be calling in a buffer scan routine, I think, since it does a map_extents_bytind. I suspect this could be tested by defining all those .*SYNTAX_CACHE.* macros in syntax.h to the empty string. Still, one scan shouldn't be _that_ bad. The next question is why does one invocation of `c-electric-semi&comma' call `scan-lists' 998 times? I think there's a pretty good chance there's something broken in cc-mode or font-lock here. Time for bed; may I expect a solution in the morning? :) :) -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Don't ask how you can "do" free software business; ask what your business can "do for" free software.