Re: [Bug: 21.5-b11] java core.

Samuel> Dear Bug Team! I am trying to edit a java program and Samuel> xemacs cores every time. Don't send core files. They only make sense on your machine with your binary. Get a backtrace with the debugger on your system, and send that to us. Samuel> It is trying to fontify the file and then it crashes. Java files stretch the fontifier to its limits, because they don't satisfy the heuristics used. If you need to get work done, I suggest turning off font-lock (font-lock-maximum-size to 0 should do it) until the bug is squashed. Samuel> XEmacs 21.5-b11 "cabbage" configured for `sparc-sun-solaris2.8'. Not current version. Update to current CVS, if you can do it again you'll get priority attention, but many changes have been made to allocation and regexp routines since then. Since you may have crashed in the regexp code (note the `looking-at' and `scan-sexps'): Samuel> byte-code("..." [limit 0 looking-at " \\s <file://\\s> *\\( Samuel> \\(,\\)\\|;\\|$\\ <file://\\(,\\)\\|;\\|$\\> )" scan-sexps 1 2] 3) it's possible the bug has been fixed by changes to the regexp code (it used to use alloca, which guaranteed a smashed stack on "hard" matches) around 21.5.12. Samuel> Fatal error: assertion failed, file gui-x.c, line 284, Samuel> RECORD_TYPEP (obj, lrecord_type_cons) OTOH, this looks familiar. The only excuse for being in gui-x.c during fontlock that I can think of is the progress bar; turn it off with (setq progress-feedback-use-echo-area t). -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN Ask not how you can "do" free software business; ask what your business can "do for" free software.