Annotation of src/lib/libc/regex/WHATSNEW, Revision 1.6
1.6 ! cgd 1: # $NetBSD$
1.5 cgd 2: # @(#)WHATSNEW 8.3 (Berkeley) 3/18/94
3:
1.4 jtc 4: New in alpha3.4: The complex bug alluded to below has been fixed (in a
5: slightly kludgey temporary way that may hurt efficiency a bit; this is
6: another "get it out the door for 4.4" release). The tests at the end of
7: the tests file have accordingly been uncommented. The primary sign of
8: the bug was that something like a?b matching ab matched b rather than ab.
9: (The bug was essentially specific to this exact situation, else it would
10: have shown up earlier.)
11:
1.3 jtc 12: New in alpha3.3: The definition of word boundaries has been altered
13: slightly, to more closely match the usual programming notion that "_"
14: is an alphabetic. Stuff used for pre-ANSI systems is now in a subdir,
15: and the makefile no longer alludes to it in mysterious ways. The
16: makefile has generally been cleaned up some. Fixes have been made
17: (again!) so that the regression test will run without -DREDEBUG, at
18: the cost of weaker checking. A workaround for a bug in some folks'
19: <assert.h> has been added. And some more things have been added to
20: tests, including a couple right at the end which are commented out
21: because the code currently flunks them (complex bug; fix coming).
22: Plus the usual minor cleanup.
23:
1.2 jtc 24: New in alpha3.2: Assorted bits of cleanup and portability improvement
25: (the development base is now a BSDI system using GCC instead of an ancient
26: Sun system, and the newer compiler exposed some glitches). Fix for a
27: serious bug that affected REs using many [] (including REG_ICASE REs
28: because of the way they are implemented), *sometimes*, depending on
29: memory-allocation patterns. The header-file prototypes no longer name
30: the parameters, avoiding possible name conflicts. The possibility that
31: some clot has defined CHAR_MIN as (say) `-128' instead of `(-128)' is
32: now handled gracefully. "uchar" is no longer used as an internal type
33: name (too many people have the same idea). Still the same old lousy
34: performance, alas.
35:
1.1 jtc 36: New in alpha3.1: Basically nothing, this release is just a bookkeeping
37: convenience. Stay tuned.
38:
39: New in alpha3.0: Performance is no better, alas, but some fixes have been
40: made and some functionality has been added. (This is basically the "get
41: it out the door in time for 4.4" release.) One bug fix: regfree() didn't
42: free the main internal structure (how embarrassing). It is now possible
43: to put NULs in either the RE or the target string, using (resp.) a new
44: REG_PEND flag and the old REG_STARTEND flag. The REG_NOSPEC flag to
45: regcomp() makes all characters ordinary, so you can match a literal
46: string easily (this will become more useful when performance improves!).
47: There are now primitives to match beginnings and ends of words, although
48: the syntax is disgusting and so is the implementation. The REG_ATOI
49: debugging interface has changed a bit. And there has been considerable
50: internal cleanup of various kinds.
51:
52: New in alpha2.3: Split change list out of README, and moved flags notes
53: into Makefile. Macro-ized the name of regex(7) in regex(3), since it has
54: to change for 4.4BSD. Cleanup work in engine.c, and some new regression
55: tests to catch tricky cases thereof.
56:
57: New in alpha2.2: Out-of-date manpages updated. Regerror() acquires two
58: small extensions -- REG_ITOA and REG_ATOI -- which avoid debugging kludges
59: in my own test program and might be useful to others for similar purposes.
60: The regression test will now compile (and run) without REDEBUG. The
61: BRE \$ bug is fixed. Most uses of "uchar" are gone; it's all chars now.
62: Char/uchar parameters are now written int/unsigned, to avoid possible
63: portability problems with unpromoted parameters. Some unsigned casts have
64: been introduced to minimize portability problems with shifting into sign
65: bits.
66:
67: New in alpha2.1: Lots of little stuff, cleanup and fixes. The one big
68: thing is that regex.h is now generated, using mkh, rather than being
69: supplied in the distribution; due to circularities in dependencies,
70: you have to build regex.h explicitly by "make h". The two known bugs
71: have been fixed (and the regression test now checks for them), as has a
72: problem with assertions not being suppressed in the absence of REDEBUG.
73: No performance work yet.
74:
75: New in alpha2: Backslash-anything is an ordinary character, not an
76: error (except, of course, for the handful of backslashed metacharacters
77: in BREs), which should reduce script breakage. The regression test
78: checks *where* null strings are supposed to match, and has generally
79: been tightened up somewhat. Small bug fixes in parameter passing (not
80: harmful, but technically errors) and some other areas. Debugging
81: invoked by defining REDEBUG rather than not defining NDEBUG.
82:
83: New in alpha+3: full prototyping for internal routines, using a little
84: helper program, mkh, which extracts prototypes given in stylized comments.
85: More minor cleanup. Buglet fix: it's CHAR_BIT, not CHAR_BITS. Simple
86: pre-screening of input when a literal string is known to be part of the
87: RE; this does wonders for performance.
88:
89: New in alpha+2: minor bits of cleanup. Notably, the number "32" for the
90: word width isn't hardwired into regexec.c any more, the public header
91: file prototypes the functions if __STDC__ is defined, and some small typos
92: in the manpages have been fixed.
93:
94: New in alpha+1: improvements to the manual pages, and an important
95: extension, the REG_STARTEND option to regexec().
CVSweb <webmaster@jp.NetBSD.org>