1*66382Sbostic# @(#)WHATSNEW 8.3 (Berkeley) 03/18/94 260200Sbostic 3*66382SbosticNew in alpha3.4: The complex bug alluded to below has been fixed (in a 4*66382Sbosticslightly kludgey temporary way that may hurt efficiency a bit; this is 5*66382Sbosticanother "get it out the door for 4.4" release). The tests at the end of 6*66382Sbosticthe tests file have accordingly been uncommented. The primary sign of 7*66382Sbosticthe bug was that something like a?b matching ab matched b rather than ab. 8*66382Sbostic(The bug was essentially specific to this exact situation, else it would 9*66382Sbostichave shown up earlier.) 10*66382Sbostic 1166361SbosticNew in alpha3.3: The definition of word boundaries has been altered 1266361Sbosticslightly, to more closely match the usual programming notion that "_" 1366361Sbosticis an alphabetic. Stuff used for pre-ANSI systems is now in a subdir, 1466361Sbosticand the makefile no longer alludes to it in mysterious ways. The 1566361Sbosticmakefile has generally been cleaned up some. Fixes have been made 1666361Sbostic(again!) so that the regression test will run without -DREDEBUG, at 1766361Sbosticthe cost of weaker checking. A workaround for a bug in some folks' 1866361Sbostic<assert.h> has been added. And some more things have been added to 1966361Sbostictests, including a couple right at the end which are commented out 2066361Sbosticbecause the code currently flunks them (complex bug; fix coming). 2166361SbosticPlus the usual minor cleanup. 2266361Sbostic 2366361SbosticNew in alpha3.2: Assorted bits of cleanup and portability improvement 2466361Sbostic(the development base is now a BSDI system using GCC instead of an ancient 2566361SbosticSun system, and the newer compiler exposed some glitches). Fix for a 2666361Sbosticserious bug that affected REs using many [] (including REG_ICASE REs 2766361Sbosticbecause of the way they are implemented), *sometimes*, depending on 2866361Sbosticmemory-allocation patterns. The header-file prototypes no longer name 2966361Sbosticthe parameters, avoiding possible name conflicts. The possibility that 3066361Sbosticsome clot has defined CHAR_MIN as (say) `-128' instead of `(-128)' is 3166361Sbosticnow handled gracefully. "uchar" is no longer used as an internal type 3266361Sbosticname (too many people have the same idea). Still the same old lousy 3366361Sbosticperformance, alas. 3466361Sbostic 3566361SbosticNew in alpha3.1: Basically nothing, this release is just a bookkeeping 3666361Sbosticconvenience. Stay tuned. 3766361Sbostic 3860200SbosticNew in alpha3.0: Performance is no better, alas, but some fixes have been 3960200Sbosticmade and some functionality has been added. (This is basically the "get 4060200Sbosticit out the door in time for 4.4" release.) One bug fix: regfree() didn't 4160200Sbosticfree the main internal structure (how embarrassing). It is now possible 4260200Sbosticto put NULs in either the RE or the target string, using (resp.) a new 4360200SbosticREG_PEND flag and the old REG_STARTEND flag. The REG_NOSPEC flag to 4460200Sbosticregcomp() makes all characters ordinary, so you can match a literal 4560200Sbosticstring easily (this will become more useful when performance improves!). 4660200SbosticThere are now primitives to match beginnings and ends of words, although 4760200Sbosticthe syntax is disgusting and so is the implementation. The REG_ATOI 4860200Sbosticdebugging interface has changed a bit. And there has been considerable 4960200Sbosticinternal cleanup of various kinds. 5060200Sbostic 5160200SbosticNew in alpha2.3: Split change list out of README, and moved flags notes 5260200Sbosticinto Makefile. Macro-ized the name of regex(7) in regex(3), since it has 5360200Sbosticto change for 4.4BSD. Cleanup work in engine.c, and some new regression 5460200Sbostictests to catch tricky cases thereof. 5560200Sbostic 5660200SbosticNew in alpha2.2: Out-of-date manpages updated. Regerror() acquires two 5760200Sbosticsmall extensions -- REG_ITOA and REG_ATOI -- which avoid debugging kludges 5860200Sbosticin my own test program and might be useful to others for similar purposes. 5960200SbosticThe regression test will now compile (and run) without REDEBUG. The 6060200SbosticBRE \$ bug is fixed. Most uses of "uchar" are gone; it's all chars now. 6160200SbosticChar/uchar parameters are now written int/unsigned, to avoid possible 6260200Sbosticportability problems with unpromoted parameters. Some unsigned casts have 6360200Sbosticbeen introduced to minimize portability problems with shifting into sign 6460200Sbosticbits. 6560200Sbostic 6660200SbosticNew in alpha2.1: Lots of little stuff, cleanup and fixes. The one big 6760200Sbosticthing is that regex.h is now generated, using mkh, rather than being 6860200Sbosticsupplied in the distribution; due to circularities in dependencies, 6960200Sbosticyou have to build regex.h explicitly by "make h". The two known bugs 7060200Sbostichave been fixed (and the regression test now checks for them), as has a 7160200Sbosticproblem with assertions not being suppressed in the absence of REDEBUG. 7260200SbosticNo performance work yet. 7360200Sbostic 7460200SbosticNew in alpha2: Backslash-anything is an ordinary character, not an 7560200Sbosticerror (except, of course, for the handful of backslashed metacharacters 7660200Sbosticin BREs), which should reduce script breakage. The regression test 7760200Sbosticchecks *where* null strings are supposed to match, and has generally 7860200Sbosticbeen tightened up somewhat. Small bug fixes in parameter passing (not 7960200Sbosticharmful, but technically errors) and some other areas. Debugging 8060200Sbosticinvoked by defining REDEBUG rather than not defining NDEBUG. 8160200Sbostic 8260200SbosticNew in alpha+3: full prototyping for internal routines, using a little 8360200Sbostichelper program, mkh, which extracts prototypes given in stylized comments. 8460200SbosticMore minor cleanup. Buglet fix: it's CHAR_BIT, not CHAR_BITS. Simple 8560200Sbosticpre-screening of input when a literal string is known to be part of the 8660200SbosticRE; this does wonders for performance. 8760200Sbostic 8860200SbosticNew in alpha+2: minor bits of cleanup. Notably, the number "32" for the 8960200Sbosticword width isn't hardwired into regexec.c any more, the public header 9060200Sbosticfile prototypes the functions if __STDC__ is defined, and some small typos 9160200Sbosticin the manpages have been fixed. 9260200Sbostic 9360200SbosticNew in alpha+1: improvements to the manual pages, and an important 9460200Sbosticextension, the REG_STARTEND option to regexec(). 95