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