Re: Slow cursor movements with XEmacs 21.4.9 "Informed Management (Windows [2])"

Andy> *sigh* oh dear. Maybe I will have to revert the change. I think you should. We have a crash report, we have performance problems being reported, we have a known reversion of a patch that Ben said prevents crashes, and we have _many_ reversions that have at best unclear connection to syntax-table text properties, some of which look like they are bug-fixes for things like infloops and not advancing pointers and things like that. Andy> I have exactly the opposite experience with large java Andy> files. This is not surprising; semantic exercises all of the relevent functionality hard, and JDE provides more intrusive^Wsupportive functionality than pretty much any other mode. In the meantime I've reduced your patch to about half its original size. I think I can halve it again, at which point it should be pretty obvious where to start looking for "Matt's molasses". One interesting fact that has been identified already is that Matt substituted parse-partial-sexp for syntactically-sectionalize in the core font-lock function dealing with syntactic fontification. Interestingly enough, syntactically-sectionalize has _no_ update_syntax_cache code in it. Why don't you revert your patch and just reapply the font-lock.el portion, and see what happens to JDE performance? -- 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