1New in alpha3.3: The definition of word boundaries has been altered 2slightly, to more closely match the usual programming notion that "_" 3is an alphabetic. Stuff used for pre-ANSI systems is now in a subdir, 4and the makefile no longer alludes to it in mysterious ways. The 5makefile has generally been cleaned up some. Fixes have been made 6(again!) so that the regression test will run without -DREDEBUG, at 7the cost of weaker checking. A workaround for a bug in some folks' 8<assert.h> has been added. And some more things have been added to 9tests, including a couple right at the end which are commented out 10because the code currently flunks them (complex bug; fix coming). 11Plus the usual minor cleanup. 12 13New in alpha3.2: Assorted bits of cleanup and portability improvement 14(the development base is now a BSDI system using GCC instead of an ancient 15Sun system, and the newer compiler exposed some glitches). Fix for a 16serious bug that affected REs using many [] (including REG_ICASE REs 17because of the way they are implemented), *sometimes*, depending on 18memory-allocation patterns. The header-file prototypes no longer name 19the parameters, avoiding possible name conflicts. The possibility that 20some clot has defined CHAR_MIN as (say) `-128' instead of `(-128)' is 21now handled gracefully. "uchar" is no longer used as an internal type 22name (too many people have the same idea). Still the same old lousy 23performance, alas. 24 25New in alpha3.1: Basically nothing, this release is just a bookkeeping 26convenience. Stay tuned. 27 28New in alpha3.0: Performance is no better, alas, but some fixes have been 29made and some functionality has been added. (This is basically the "get 30it out the door in time for 4.4" release.) One bug fix: regfree() didn't 31free the main internal structure (how embarrassing). It is now possible 32to put NULs in either the RE or the target string, using (resp.) a new 33REG_PEND flag and the old REG_STARTEND flag. The REG_NOSPEC flag to 34regcomp() makes all characters ordinary, so you can match a literal 35string easily (this will become more useful when performance improves!). 36There are now primitives to match beginnings and ends of words, although 37the syntax is disgusting and so is the implementation. The REG_ATOI 38debugging interface has changed a bit. And there has been considerable 39internal cleanup of various kinds. 40 41New in alpha2.3: Split change list out of README, and moved flags notes 42into Makefile. Macro-ized the name of regex(7) in regex(3), since it has 43to change for 4.4BSD. Cleanup work in engine.c, and some new regression 44tests to catch tricky cases thereof. 45 46New in alpha2.2: Out-of-date manpages updated. Regerror() acquires two 47small extensions -- REG_ITOA and REG_ATOI -- which avoid debugging kludges 48in my own test program and might be useful to others for similar purposes. 49The regression test will now compile (and run) without REDEBUG. The 50BRE \$ bug is fixed. Most uses of "uchar" are gone; it's all chars now. 51Char/uchar parameters are now written int/unsigned, to avoid possible 52portability problems with unpromoted parameters. Some unsigned casts have 53been introduced to minimize portability problems with shifting into sign 54bits. 55 56New in alpha2.1: Lots of little stuff, cleanup and fixes. The one big 57thing is that regex.h is now generated, using mkh, rather than being 58supplied in the distribution; due to circularities in dependencies, 59you have to build regex.h explicitly by "make h". The two known bugs 60have been fixed (and the regression test now checks for them), as has a 61problem with assertions not being suppressed in the absence of REDEBUG. 62No performance work yet. 63 64New in alpha2: Backslash-anything is an ordinary character, not an 65error (except, of course, for the handful of backslashed metacharacters 66in BREs), which should reduce script breakage. The regression test 67checks *where* null strings are supposed to match, and has generally 68been tightened up somewhat. Small bug fixes in parameter passing (not 69harmful, but technically errors) and some other areas. Debugging 70invoked by defining REDEBUG rather than not defining NDEBUG. 71 72New in alpha+3: full prototyping for internal routines, using a little 73helper program, mkh, which extracts prototypes given in stylized comments. 74More minor cleanup. Buglet fix: it's CHAR_BIT, not CHAR_BITS. Simple 75pre-screening of input when a literal string is known to be part of the 76RE; this does wonders for performance. 77 78New in alpha+2: minor bits of cleanup. Notably, the number "32" for the 79word width isn't hardwired into regexec.c any more, the public header 80file prototypes the functions if __STDC__ is defined, and some small typos 81in the manpages have been fixed. 82 83New in alpha+1: improvements to the manual pages, and an important 84extension, the REG_STARTEND option to regexec(). 85