16-Jan-2003 17:57 Thomas Dickey wrote: >> Another new patch: > thanks (I'm looking through your next-most-recent one now).
Seems this letter were lost during integration. (ALLOC_IN_POOL macros vs functions may be ignored, but complexity issue still very important). From: "Leonid Pauzner" <[EMAIL PROTECTED]> Date: Sat, 21 Dec 2002 20:15:13 +0300 (MSK) Subject: Re: lynx-dev valgrind-reverse.log.gz 20-Dec-2002 18:33 Thomas Dickey wrote: >> Any ideas? > not offhand (I'll spend some time on lynx this weekend when I have more > time to do this; right now I'm checking on easier-to-fix bugs). Three unrelated patches, against dev.12 (1) define HAVE_ALLOCA for DJGPP (2) remove quadratic complexity from insert_blanks_in_line() usage with large tables (Stbl). It occasionally cleanup split_line() a bit. CPU load anomaly reported by BL. (3) ALLOC_IN_POOL, POOL_NEW, POOL_FREE macros now became functions, suggested by BL. p.s. (1) was sent to the list previously; (2) was sent previously, now updated. --ATTACHMENT-- Binary file <htutils.dif> Description: for WWW/.../HTUtils.h --ATTACHMENT-- Binary file <q_comple.dif> Description: for GridText.c --ATTACHMENT-- Binary file <alloc_in.dif> Description: for GridText.c ; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to [EMAIL PROTECTED]
