1 2New in alpha3.4: The complex bug alluded to below has been fixed (in a 3slightly kludgey temporary way that may hurt efficiency a bit; this is 4another "get it out the door for 4.4" release). The tests at the end of 5the tests file have accordingly been uncommented. The primary sign of 6the bug was that something like a?b matching ab matched b rather than ab. 7(The bug was essentially specific to this exact situation, else it would 8have shown up earlier.) 9 10New in alpha3.3: The definition of word boundaries has been altered 11slightly, to more closely match the usual programming notion that "_" 12is an alphabetic. Stuff used for pre-ANSI systems is now in a subdir, 13and the makefile no longer alludes to it in mysterious ways. The 14makefile has generally been cleaned up some. Fixes have been made 15(again!) so that the regression test will run without -DREDEBUG, at 16the cost of weaker checking. A workaround for a bug in some folks' 17<assert.h> has been added. And some more things have been added to 18tests, including a couple right at the end which are commented out 19because the code currently flunks them (complex bug; fix coming). 20Plus the usual minor cleanup. 21 22New in alpha3.2: Assorted bits of cleanup and portability improvement 23(the development base is now a BSDI system using GCC instead of an ancient 24Sun system, and the newer compiler exposed some glitches). Fix for a 25serious bug that affected REs using many [] (including REG_ICASE REs 26because of the way they are implemented), *sometimes*, depending on 27memory-allocation patterns. The header-file prototypes no longer name 28the parameters, avoiding possible name conflicts. The possibility that 29some clot has defined CHAR_MIN as (say) `-128' instead of `(-128)' is 30now handled gracefully. "uchar" is no longer used as an internal type 31name (too many people have the same idea). Still the same old lousy 32performance, alas. 33 34New in alpha3.1: Basically nothing, this release is just a bookkeeping 35convenience. Stay tuned. 36 37New in alpha3.0: Performance is no better, alas, but some fixes have been 38made and some functionality has been added. (This is basically the "get 39it out the door in time for 4.4" release.) One bug fix: regfree() didn't 40free the main internal structure (how embarrassing). It is now possible 41to put NULs in either the RE or the target string, using (resp.) a new 42REG_PEND flag and the old REG_STARTEND flag. The REG_NOSPEC flag to 43regcomp() makes all characters ordinary, so you can match a literal 44string easily (this will become more useful when performance improves!). 45There are now primitives to match beginnings and ends of words, although 46the syntax is disgusting and so is the implementation. The REG_ATOI 47debugging interface has changed a bit. And there has been considerable 48internal cleanup of various kinds. 49 50New in alpha2.3: Split change list out of README, and moved flags notes 51into Makefile. Macro-ized the name of regex(7) in regex(3), since it has 52to change for 4.4BSD. Cleanup work in engine.c, and some new regression 53tests to catch tricky cases thereof. 54 55New in alpha2.2: Out-of-date manpages updated. Regerror() acquires two 56small extensions -- REG_ITOA and REG_ATOI -- which avoid debugging kludges 57in my own test program and might be useful to others for similar purposes. 58The regression test will now compile (and run) without REDEBUG. The 59BRE \$ bug is fixed. Most uses of "uchar" are gone; it's all chars now. 60Char/uchar parameters are now written int/unsigned, to avoid possible 61portability problems with unpromoted parameters. Some unsigned casts have 62been introduced to minimize portability problems with shifting into sign 63bits. 64 65New in alpha2.1: Lots of little stuff, cleanup and fixes. The one big 66thing is that regex.h is now generated, using mkh, rather than being 67supplied in the distribution; due to circularities in dependencies, 68you have to build regex.h explicitly by "make h". The two known bugs 69have been fixed (and the regression test now checks for them), as has a 70problem with assertions not being suppressed in the absence of REDEBUG. 71No performance work yet. 72 73New in alpha2: Backslash-anything is an ordinary character, not an 74error (except, of course, for the handful of backslashed metacharacters 75in BREs), which should reduce script breakage. The regression test 76checks *where* null strings are supposed to match, and has generally 77been tightened up somewhat. Small bug fixes in parameter passing (not 78harmful, but technically errors) and some other areas. Debugging 79invoked by defining REDEBUG rather than not defining NDEBUG. 80 81New in alpha+3: full prototyping for internal routines, using a little 82helper program, mkh, which extracts prototypes given in stylized comments. 83More minor cleanup. Buglet fix: it's CHAR_BIT, not CHAR_BITS. Simple 84pre-screening of input when a literal string is known to be part of the 85RE; this does wonders for performance. 86 87New in alpha+2: minor bits of cleanup. Notably, the number "32" for the 88word width isn't hardwired into regexec.c any more, the public header 89file prototypes the functions if __STDC__ is defined, and some small typos 90in the manpages have been fixed. 91 92New in alpha+1: improvements to the manual pages, and an important 93extension, the REG_STARTEND option to regexec(). 94