Annotation of src/usr.bin/awk/FIXES, Revision 1.26
1.26 ! millert 1: /* $OpenBSD: FIXES,v 1.25 2020/06/10 21:02:33 millert Exp $ */
1.1 tholo 2: /****************************************************************
1.6 kstailey 3: Copyright (C) Lucent Technologies 1997
1.1 tholo 4: All Rights Reserved
5:
6: Permission to use, copy, modify, and distribute this software and
7: its documentation for any purpose and without fee is hereby
8: granted, provided that the above copyright notice appear in all
9: copies and that both that the copyright notice and this
10: permission notice and warranty disclaimer appear in supporting
1.6 kstailey 11: documentation, and that the name Lucent Technologies or any of
12: its entities not be used in advertising or publicity pertaining
13: to distribution of the software without specific, written prior
14: permission.
15:
16: LUCENT DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE,
17: INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS.
18: IN NO EVENT SHALL LUCENT OR ANY OF ITS ENTITIES BE LIABLE FOR ANY
19: SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
20: WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER
21: IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION,
22: ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF
23: THIS SOFTWARE.
1.1 tholo 24: ****************************************************************/
25:
26: This file lists all bug fixes, changes, etc., made since the AWK book
27: was sent to the printers in August, 1987.
1.26 ! millert 28:
! 29: October 6, 2019:
! 30: Import code from NetBSD awk that implements RS as a regular
! 31: expression.
1.23 millert 32:
1.25 millert 33: September 10, 2019:
34: Fixes for various array / memory overruns found via gcc's
35: -fsanitize=unknown. Thanks to Alexander Richardson (Github
36: user arichardson). Merges PRs 47 and 48.
37:
38: July 28, 2019:
39: Import grammar optimization from NetBSD: Two string constants
40: concatenated together get turned into a single string.
41:
42: July 26, 2019:
43: Support POSIX-specified C-style escape sequences "\a" (alarm)
44: and "\v" (vertical tab) in command line arguments and regular
45: expressions, further to the support for them in strings added on
46: Apr 9, 1989. These now no longer match as literal "a" and "v"
47: characters (as they don't on other awk implementations).
48: Thanks to Martijn Dekker.
49:
50: July 17, 2019:
51: Pull in a number of code cleanups and minor fixes from
52: Warner Losh's bsd-ota branch. The only user visible change
53: is the use of random(3) as the random number generator.
54: Thanks to Warner Losh for collecting all these fixes in
55: one easy place to get them from.
56:
57: July 16, 2019:
58: Fix field splitting to use FS value as of the time a record
59: was read or assigned to. Thanks to GitHub user Cody Mello (melloc)
60: for the fix. (Merged from his branch, via PR #42.) Updated
61: testdir/T.split per said PR as well.
62:
63: June 24, 2019:
64: Extract awktest.tar into testdir directory. Add some very
65: simple mechanics to the makefile for running the tests and
66: for cleaning up. No changes to awk itself.
67:
1.24 millert 68: June 17, 2019:
69: Disallow deleting SYMTAB and its elements, which creates
70: use-after-free bugs. Thanks to GitHub user Cody Mello (melloc)
71: for the fix. (Merged from PR #43.)
72:
73: June 5, 2019:
74: Allow unmatched right parenthesis in a regular expression to
75: be treated literally. Fixes Issue #40. Thanks to GitHub user
76: Warner Losh (bsdimp) for the report. Thanks to Arnold Robbins
77: for the fix.
78:
79: May 29,2019:
80: Fix check for command line arguments to no longer require that
81: first character after '=' not be another '='. Reverts change of
82: August 11, 1989. Thanks to GitHub user Jamie Landeg Jones for
83: pointing out the issue; from Issue #38.
84:
85: Apr 7, 2019:
86: Update awktest.tar(p.50) to use modern options to sort. Needed
87: for Android development. Thanks to GitHub user mohd-akram (Mohamed
88: Akram). From Issue #33.
89:
90: Mar 12, 2019:
91: Added very simplistic support for cross-compiling in the
92: makefile. We are NOT going to go in the direction of the
93: autotools, though. Thanks to GitHub user nee-san for
94: the basic change. (Merged from PR #34.)
95:
1.23 millert 96: Mar 5, 2019:
97: Added support for POSIX-standard interval expressions (a.k.a.
98: bounds, a.k.a. repetition expressions) in regular expressions,
99: backported (via NetBSD) from Apple awk-24 (20070501).
100: Thanks to Martijn Dekker <martijn@inlv.org> for the port.
101: (Merged from PR #30.)
102:
103: Mar 3, 2019:
104: Merge PRs as follows:
105: #12: Avoid undefined behaviour when using ctype(3) functions in
106: relex(). Thanks to GitHub user iamleot.
107: #31: Make getline handle numeric strings, and update FIXES. Thanks
1.24 millert 108: to GitHub user arnoldrobbins.
1.23 millert 109: #32: maketab: support build systems with read-only source. Thanks
110: to GitHub user enh.
1.22 millert 111:
112: Jan 25, 2019:
113: Make getline handle numeric strings properly in all cases.
114: (Thanks, Arnold.)
115:
116: Jan 21, 2019:
117: Merged a number of small fixes from GitHub pull requests.
118: Thanks to GitHub users Arnold Robbins (arnoldrobbins),
119: Cody Mello (melloc) and Christoph Junghans (junghans).
120: PR numbers: 13-21, 23, 24, 27.
1.21 millert 121:
122: Oct 25, 2018:
123: Added test in maketab.c to prevent generating a proctab entry
124: for YYSTYPE_IS_DEFINED. It was harmless but some gcc settings
125: generated a warning message. Thanks to Nan Xiao for report.
1.20 millert 126:
127: Aug 27, 2018:
128: Disallow '$' in printf formats; arguments evaluated in order
129: and printed in order.
130:
131: Added some casts to silence warnings on debugging printfs.
132: (Thanks, Arnold.)
1.18 millert 133:
1.19 millert 134: Aug 23, 2018:
135: A long list of fixes courtesy of Arnold Robbins,
136: to whom profound thanks.
137:
138: 1. ofs-rebuild: OFS value used to rebuild the record was incorrect.
139: Fixed August 19, 2014. Revised fix August 2018.
140:
141: 2. system-status: Instead of a floating-point division by 256, use
142: the wait(2) macros to create a reasonable exit status.
143: Fixed March 12, 2016.
144:
145: 3. space: Use provided xisblank() function instead of ispace() for
146: matching [[:blank:]].
147:
148: 4. a-format: Add POSIX standard %a and %A to supported formats. Check
149: at runtime that this format is available.
150:
151: 5. decr-NF: Decrementing NF did not change $0. This is a decades-old
152: bug. There are interactions with the old and new value of OFS as well.
153: Most of the fix came from the NetBSD awk.
154:
155: 6. string-conv: String conversions of scalars were sticky. Once a
156: conversion to string happened, even with OFMT, that value was used until
157: a new numeric value was assigned, even if OFMT differed from CONVFMT,
158: and also if CONVFMT changed.
159:
160: 7. unary-plus: Unary plus on a string constant returned the string.
161: Instead, it should convert the value to numeric and give that value.
162:
163: Also added Arnold's tests for these to awktest.tar as T.arnold.
164:
165: Aug 15, 2018:
166: fixed mangled awktest.tar (thanks, Arnold), posted all
167: current (very minor) fixes to github / onetrueawk
168:
1.18 millert 169: Jun 7, 2018:
170: (yes, a long layoff)
171: Updated some broken tests (beebe.tar, T.lilly)
172: [thanks to Arnold Robbins]
173:
174: Mar 26, 2015:
175: buffer overflow in error reporting; thanks to tobias ulmer
176: and john-mark gurney for spotting it and the fix.
177:
178: Feb 4, 2013:
179: cleaned up a handful of tests that didn't seem to actually
180: test for correct behavior: T.latin1, T.gawk.
181:
182: Jan 5, 2013:
183: added ,NULL initializer to static Cells in run.c; not really
184: needed but cleaner. Thanks to Michael Bombardieri.
1.17 millert 185:
186: Dec 20, 2012:
187: fiddled makefile to get correct yacc and bison flags. pick yacc
188: (linux) or bison (mac) as necessary.
189:
190: added __attribute__((__noreturn__)) to a couple of lines in
191: proto.h, to silence someone's enthusiastic checker.
192:
193: fixed obscure call by value bug in split(a[1],a) reported on
194: 9fans. the management of temporary values is just a mess; i
195: took a shortcut by making an extra string copy. thanks
196: to paul patience and arnold robbins for passing it on and for
197: proposed patches.
198:
199: tiny fiddle in setfval to eliminate -0 results in T.expr, which
200: has irritated me for 20+ years.
1.16 millert 201:
202: Aug 10, 2011:
203: another fix to avoid core dump with delete(ARGV); again, many thanks
204: to ruslan ermilov.
205:
206: Aug 7, 2011:
207: split(s, a, //) now behaves the same as split(s, a, "")
208:
209: Jun 12, 2011:
210: /pat/, \n /pat/ {...} is now legal, though bad style to use.
211:
212: added checks to new -v code that permits -vnospace; thanks to
1.25 millert 213: ruslan ermilov for spotting this and providing the patch.
1.16 millert 214:
215: removed fixed limit on number of open files; thanks to aleksey
1.25 millert 216: cheusov and christos zoulos.
1.16 millert 217:
218: fixed day 1 bug that resurrected deleted elements of ARGV when
219: used as filenames (in lib.c).
220:
221: minor type fiddles to make gcc -Wall -pedantic happier (but not
222: totally so); turned on -fno-strict-aliasing in makefile.
223:
224: May 6, 2011:
225: added #ifdef for isblank.
226: now allows -ffoo as well as -f foo arguments.
227: (thanks, ruslan)
228:
229: May 1, 2011:
230: after advice from todd miller, kevin lo, ruslan ermilov,
231: and arnold robbins, changed srand() to return the previous
232: seed (which is 1 on the first call of srand). the seed is
233: an Awkfloat internally though converted to unsigned int to
1.25 millert 234: pass to the library srand(). thanks, everyone.
1.16 millert 235:
236: fixed a subtle (and i hope low-probability) overflow error
1.25 millert 237: in fldbld, by adding space for one extra \0. thanks to
1.16 millert 238: robert bassett for spotting this one and providing a fix.
239:
240: removed the files related to compilation on windows. i no
241: longer have anything like a current windows environment, so
242: i can't test any of it.
1.15 millert 243:
244: May 23, 2010:
245: fixed long-standing overflow bug in run.c; many thanks to
246: nelson beebe for spotting it and providing the fix.
247:
248: fixed bug that didn't parse -vd=1 properly; thanks to santiago
249: vila for spotting it.
250:
251: Feb 8, 2010:
252: i give up. replaced isblank with isspace in b.c; there are
253: no consistent header files.
254:
255: Nov 26, 2009:
256: fixed a long-standing issue with when FS takes effect. a
257: change to FS is now noticed immediately for subsequent splits.
258:
259: changed the name getline() to awkgetline() to avoid yet another
260: name conflict somewhere.
261:
262: Feb 11, 2009:
263: temporarily for now defined HAS_ISBLANK, since that seems to
264: be the best way through the thicket. isblank arrived in C99,
265: but seems to be arriving at different systems at different
266: times.
267:
268: Oct 8, 2008:
269: fixed typo in b.c that set tmpvec wrongly. no one had ever
270: run into the problem, apparently. thanks to alistair crooks.
1.14 millert 271:
272: Oct 23, 2007:
273: minor fix in lib.c: increase inputFS to 100, change malloc
1.25 millert 274: for fields to n+1.
1.14 millert 275:
276: fixed memory fault caused by out of order test in setsval.
277:
278: thanks to david o'brien, freebsd, for both fixes.
279:
280: May 1, 2007:
281: fiddle in makefile to fix for BSD make; thanks to igor sobrado.
282:
283: Mar 31, 2007:
284: fixed some null pointer refs calling adjbuf.
285:
286: Feb 21, 2007:
287: fixed a bug in matching the null RE in sub and gsub. thanks to al aho
288: who actually did the fix (in b.c), and to wolfgang seeberg for finding
289: it and providing a very compact test case.
290:
291: fixed quotation in b.c; thanks to Hal Pratt and the Princeton Dante
292: Project.
293:
294: removed some no-effect asserts in run.c.
295:
296: fiddled maketab.c to not complain about bison-generated values.
297:
298: removed the obsolete -V argument; fixed --version to print the
299: version and exit.
300:
301: fixed wording and an outright error in the usage message; thanks to igor
302: sobrado and jason mcintyre.
303:
304: fixed a bug in -d that caused core dump if no program followed.
305:
306: Jan 1, 2007:
307: dropped mac.code from makefile; there are few non-MacOSX
308: mac's these days.
309:
310: Jan 17, 2006:
311: system() not flagged as unsafe in the unadvertised -safe option.
312: found it while enhancing tests before shipping the ;login: article.
313: practice what you preach.
314:
315: removed the 9-years-obsolete -mr and -mf flags.
316:
317: added -version and --version options.
318:
319: core dump on linux with BEGIN {nextfile}, now fixed.
320:
1.25 millert 321: removed some #ifdef's in run.c and lex.c that appear to no
1.14 millert 322: longer be necessary.
323:
324: Apr 24, 2005:
325: modified lib.c so that values of $0 et al are preserved in the END
326: block, apparently as required by posix. thanks to havard eidnes
327: for the report and code.
328:
329: Jan 14, 2005:
330: fixed infinite loop in parsing, originally found by brian tsang.
331: thanks to arnold robbins for a suggestion that started me
332: rethinking it.
333:
334: Dec 31, 2004:
1.25 millert 335: prevent overflow of -f array in main, head off potential error in
336: call of SYNTAX(), test malloc return in lib.c, all with thanks to
1.14 millert 337: todd miller.
1.13 millert 338:
339: Dec 22, 2004:
340: cranked up size of NCHARS; coverity thinks it can be overrun with
341: smaller size, and i think that's right. added some assertions to b.c
342: to catch places where it might overrun. the RE code is still fragile.
343:
344: Dec 5, 2004:
345: fixed a couple of overflow problems with ridiculous field numbers:
346: e.g., print $(2^32-1). thanks to ruslan ermilov, giorgos keramidas
347: and david o'brien at freebsd.org for patches. this really should
348: be re-done from scratch.
349:
350: Nov 21, 2004:
351: fixed another 25-year-old RE bug, in split. it's another failure
352: to (re-)initialize. thanks to steve fisher for spotting this and
353: providing a good test case.
354:
355: Nov 22, 2003:
356: fixed a bug in regular expressions that dates (so help me) from 1977;
357: it's been there from the beginning. an anchored longest match that
358: was longer than the number of states triggered a failure to initialize
359: the machine properly. many thanks to moinak ghosh for not only finding
360: this one but for providing a fix, in some of the most mysterious
361: code known to man.
362:
363: fixed a storage leak in call() that appears to have been there since
1.25 millert 364: 1983 or so -- a function without an explicit return that assigns a
365: string to a parameter leaked a Cell. thanks to moinak ghosh for
1.13 millert 366: spotting this very subtle one.
367:
368: Jul 31, 2003:
369: fixed, thanks to andrey chernov and ruslan ermilov, a bug in lex.c
370: that mis-handled the character 255 in input. (it was being compared
371: to EOF with a signed comparison.)
372:
373: Jul 29, 2003:
374: fixed (i think) the long-standing botch that included the beginning of
375: line state ^ for RE's in the set of valid characters; this led to a
376: variety of odd problems, including failure to properly match certain
377: regular expressions in non-US locales. thanks to ruslan for keeping
378: at this one.
379:
380: Jul 28, 2003:
381: n-th try at getting internationalization right, with thanks to volker
382: kiefel, arnold robbins and ruslan ermilov for advice, though they
383: should not be blamed for the outcome. according to posix, "." is the
384: radix character in programs and command line arguments regardless of
385: the locale; otherwise, the locale should prevail for input and output
386: of numbers. so it's intended to work that way.
1.25 millert 387:
1.13 millert 388: i have rescinded the attempt to use strcoll in expanding shorthands in
389: regular expressions (cclenter). its properties are much too
390: surprising; for example [a-c] matches aAbBc in locale en_US but abBcC
391: in locale fr_CA. i can see how this might arise by implementation
392: but i cannot explain it to a human user. (this behavior can be seen
393: in gawk as well; we're leaning on the same library.)
394:
395: the issue appears to be that strcoll is meant for sorting, where
396: merging upper and lower case may make sense (though note that unix
397: sort does not do this by default either). it is not appropriate
398: for regular expressions, where the goal is to match specific
399: patterns of characters. in any case, the notations [:lower:], etc.,
400: are available in awk, and they are more likely to work correctly in
401: most locales.
402:
403: a moratorium is hereby declared on internationalization changes.
404: i apologize to friends and colleagues in other parts of the world.
405: i would truly like to get this "right", but i don't know what
406: that is, and i do not want to keep making changes until it's clear.
407:
408: Jul 4, 2003:
409: fixed bug that permitted non-terminated RE, as in "awk /x".
410:
411: Jun 1, 2003:
412: subtle change to split: if source is empty, number of elems
413: is always 0 and the array is not set.
414:
415: Mar 21, 2003:
416: added some parens to isblank, in another attempt to make things
417: internationally portable.
418:
419: Mar 14, 2003:
420: the internationalization changes, somewhat modified, are now
421: reinstated. in theory awk will now do character comparisons
422: and case conversions in national language, but "." will always
423: be the decimal point separator on input and output regardless
424: of national language. isblank(){} has an #ifndef.
425:
426: this no longer compiles on windows: LC_MESSAGES isn't defined
427: in vc6++.
428:
429: fixed subtle behavior in field and record splitting: if FS is
430: a single character and RS is not empty, \n is NOT a separator.
431: this tortuous reading is found in the awk book; behavior now
432: matches gawk and mawk.
1.11 millert 433:
434: Dec 13, 2002:
435: for the moment, the internationalization changes of nov 29 are
436: rolled back -- programs like x = 1.2 don't work in some locales,
437: because the parser is expecting x = 1,2. until i understand this
438: better, this will have to wait.
439:
440: Nov 29, 2002:
441: modified b.c (with tiny changes in main and run) to support
442: locales, using strcoll and iswhatever tests for posix character
443: classes. thanks to ruslan ermilov (ru@freebsd.org) for code.
444: the function isblank doesn't seem to have propagated to any
445: header file near me, so it's there explicitly. not properly
446: tested on non-ascii character sets by me.
447:
448: Jun 28, 2002:
449: modified run/format() and tran/getsval() to do a slightly better
450: job on using OFMT for output from print and CONVFMT for other
1.25 millert 451: number->string conversions, as promised by posix and done by
1.11 millert 452: gawk and mawk. there are still places where it doesn't work
453: right if CONVFMT is changed; by then the STR attribute of the
454: variable has been irrevocably set. thanks to arnold robbins for
455: code and examples.
456:
457: fixed subtle bug in format that could get core dump. thanks to
458: Jaromir Dolecek <jdolecek@NetBSD.org> for finding and fixing.
459: minor cleanup in run.c / format() at the same time.
460:
461: added some tests for null pointers to debugging printf's, which
462: were never intended for external consumption. thanks to dave
463: kerns (dkerns@lucent.com) for pointing this out.
464:
465: GNU compatibility: an empty regexp matches anything (thanks to
466: dag-erling smorgrav, des@ofug.org). subject to reversion if
467: this does more harm than good.
468:
469: pervasive small changes to make things more const-correct, as
470: reported by gcc's -Wwrite-strings. as it says in the gcc manual,
471: this may be more nuisance than useful. provoked by a suggestion
472: and code from arnaud desitter, arnaud@nimbus.geog.ox.ac.uk
473:
474: minor documentation changes to note that this now compiles out
475: of the box on Mac OS X.
476:
477: Feb 10, 2002:
478: changed types in posix chars structure to quiet solaris cc.
479:
480: Jan 1, 2002:
481: fflush() or fflush("") flushes all files and pipes.
482:
1.25 millert 483: length(arrayname) returns number of elements; thanks to
1.11 millert 484: arnold robbins for suggestion.
485:
486: added a makefile.win to make it easier to build on windows.
487: based on dan allen's buildwin.bat.
488:
489: Nov 16, 2001:
490: added support for posix character class names like [:digit:],
491: which are not exactly shorter than [0-9] and perhaps no more
492: portable. thanks to dag-erling smorgrav for code.
493:
494: Feb 16, 2001:
495: removed -m option; no longer needed, and it was actually
496: broken (noted thanks to volker kiefel).
497:
498: Feb 10, 2001:
499: fixed an appalling bug in gettok: any sequence of digits, +,-, E, e,
500: and period was accepted as a valid number if it started with a period.
501: this would never have happened with the lex version.
502:
503: other 1-character botches, now fixed, include a bare $ and a
504: bare " at the end of the input.
505:
506: Feb 7, 2001:
507: more (const char *) casts in b.c and tran.c to silence warnings.
1.8 millert 508:
1.10 millert 509: Nov 15, 2000:
510: fixed a bug introduced in august 1997 that caused expressions
511: like $f[1] to be syntax errors. thanks to arnold robbins for
512: noticing this and providing a fix.
513:
514: Oct 30, 2000:
515: fixed some nextfile bugs: not handling all cases. thanks to
516: arnold robbins for pointing this out. new regressions added.
517:
518: close() is now a function. it returns whatever the library
519: fclose returns, and -1 for closing a file or pipe that wasn't
520: opened.
521:
522: Sep 24, 2000:
523: permit \n explicitly in character classes; won't work right
524: if comes in as "[\n]" but ok as /[\n]/, because of multiple
525: processing of \'s. thanks to arnold robbins.
526:
527: July 5, 2000:
528: minor fiddles in tran.c to keep compilers happy about uschar.
529: thanks to norman wilson.
530:
531: May 25, 2000:
532: yet another attempt at making 8-bit input work, with another
1.25 millert 533: band-aid in b.c (member()), and some (uschar) casts to head
1.10 millert 534: off potential errors in subscripts (like isdigit). also
535: changed HAT to NCHARS-2. thanks again to santiago vila.
536:
537: changed maketab.c to ignore apparently out of range definitions
538: instead of halting; new freeBSD generates one. thanks to
539: jon snader <jsnader@ix.netcom.com> for pointing out the problem.
540:
541: May 2, 2000:
542: fixed an 8-bit problem in b.c by making several char*'s into
543: unsigned char*'s. not clear i have them all yet. thanks to
544: Santiago Vila <sanvila@unex.es> for the bug report.
545:
546: Apr 21, 2000:
547: finally found and fixed a memory leak in function call; it's
548: been there since functions were added ~1983. thanks to
549: jon bentley for the test case that found it.
550:
551: added test in envinit to catch environment "variables" with
1.12 otto 552: names beginning with '='; thanks to Berend Hasselman.
1.10 millert 553:
554: Jul 28, 1999:
555: added test in defn() to catch function foo(foo), which
556: otherwise recurses until core dump. thanks to arnold
557: robbins for noticing this.
558:
1.9 millert 559: Jun 20, 1999:
560: added *bp in gettok in lex.c; appears possible to exit function
561: without terminating the string. thanks to russ cox.
562:
563: Jun 2, 1999:
564: added function stdinit() to run to initialize files[] array,
565: in case stdin, etc., are not constants; some compilers care.
566:
567: May 10, 1999:
568: replaced the ERROR ... FATAL, etc., macros with functions
569: based on vprintf, to avoid problems caused by overrunning
570: fixed-size errbuf array. thanks to ralph corderoy for the
571: impetus, and for pointing out a string termination bug in
572: qstring as well.
573:
574: Apr 21, 1999:
575: fixed bug that caused occasional core dumps with commandline
576: variable with value ending in \. (thanks to nelson beebe for
577: the test case.)
578:
1.8 millert 579: Apr 16, 1999:
1.25 millert 580: with code kindly provided by Bruce Lilly, awk now parses
1.8 millert 581: /=/ and similar constructs more sensibly in more places.
1.10 millert 582: Bruce also provided some helpful test cases.
1.8 millert 583:
584: Apr 5, 1999:
585: changed true/false to True/False in run.c to make it
586: easier to compile with C++. Added some casts on malloc
587: and realloc to be honest about casts; ditto. changed
588: ltype int to long in struct rrow to reduce some 64-bit
589: complaints; other changes scattered throughout for the
590: same purpose. thanks to Nelson Beebe for these portability
591: improvements.
592:
593: removed some horrible pointer-int casting in b.c and elsewhere
594: by adding ptoi and itonp to localize the casts, which are
595: all benign. fixed one incipient bug that showed up on sgi
596: in 64-bit mode.
597:
598: reset lineno for new source file; include filename in error
599: message. also fixed line number error in continuation lines.
600: (thanks to Nelson Beebe for both of these.)
601:
602: Mar 24, 1999:
603: Nelson Beebe notes that irix 5.3 yacc dies with a bogus
604: error; use a newer version or switch to bison, since sgi
605: is unlikely to fix it.
1.1 tholo 606:
1.7 millert 607: Mar 5, 1999:
1.9 millert 608: changed isnumber to is_number to avoid the problem caused by
609: versions of ctype.h that include the name isnumber.
1.7 millert 610:
611: distribution now includes a script for building on a Mac,
612: thanks to Dan Allen.
613:
614: Feb 20, 1999:
615: fixed memory leaks in run.c (call) and tran.c (setfval).
616: thanks to Stephen Nutt for finding these and providing the fixes.
617:
618: Jan 13, 1999:
619: replaced srand argument by (unsigned int) in run.c;
620: avoids problem on Mac and potentially on Unix & Windows.
621: thanks to Dan Allen.
622:
623: added a few (int) casts to silence useless compiler warnings.
624: e.g., errorflag= in run.c jump().
625:
626: added proctab.c to the bundle outout; one less thing
627: to have to compile out of the box.
628:
629: added calls to _popen and _pclose to the win95 stub for
630: pipes (thanks to Steve Adams for this helpful suggestion).
631: seems to work, though properties are not well understood
632: by me, and it appears that under some circumstances the
633: pipe output is truncated. Be careful.
634:
635: Oct 19, 1998:
636: fixed a couple of bugs in getrec: could fail to update $0
1.25 millert 637: after a getline var; because inputFS wasn't initialized,
1.7 millert 638: could split $0 on every character, a misleading diversion.
639:
640: fixed caching bug in makedfa: LRU was actually removing
641: least often used.
642:
643: thanks to ross ridge for finding these, and for providing
644: great bug reports.
645:
646: May 12, 1998:
647: fixed potential bug in readrec: might fail to update record
648: pointer after growing. thanks to dan levy for spotting this
649: and suggesting the fix.
650:
651: Mar 12, 1998:
652: added -V to print version number and die.
1.19 millert 653:
654: [notify dave kerns, dkerns@dacsoup.ih.lucent.com]
1.7 millert 655:
656: Feb 11, 1998:
657: subtle silent bug in lex.c: if the program ended with a number
658: longer than 1 digit, part of the input would be pushed back and
659: parsed again because token buffer wasn't terminated right.
660: example: awk 'length($0) > 10'. blush. at least i found it
661: myself.
662:
663: Aug 31, 1997:
664: s/adelete/awkdelete/: SGI uses this in malloc.h.
665: thanks to nelson beebe for pointing this one out.
666:
667: Aug 21, 1997:
668: fixed some bugs in sub and gsub when replacement includes \\.
669: this is a dark, horrible corner, but at least now i believe that
670: the behavior is the same as gawk and the intended posix standard.
671: thanks to arnold robbins for advice here.
672:
673: Aug 9, 1997:
674: somewhat regretfully, replaced the ancient lex-based lexical
675: analyzer with one written in C. it's longer, generates less code,
676: and more portable; the old one depended too much on mysterious
677: properties of lex that were not preserved in other environments.
678: in theory these recognize the same language.
679:
680: now using strtod to test whether a string is a number, instead of
681: the convoluted original function. should be more portable and
682: reliable if strtod is implemented right.
683:
684: removed now-pointless optimization in makefile that tries to avoid
685: recompilation when awkgram.y is changed but symbols are not.
686:
687: removed most fixed-size arrays, though a handful remain, some
688: of which are unchecked. you have been warned.
689:
690: Aug 4, 1997:
691: with some trepidation, replaced the ancient code that managed
692: fields and $0 in fixed-size arrays with arrays that grow on
693: demand. there is still some tension between trying to make this
694: run fast and making it clean; not sure it's right yet.
695:
696: the ill-conceived -mr and -mf arguments are now useful only
697: for debugging. previous dynamic string code removed.
698:
699: numerous other minor cleanups along the way.
1.1 tholo 700:
1.7 millert 701: Jul 30, 1997:
702: using code provided by dan levy (to whom profuse thanks), replaced
703: fixed-size arrays and awkward kludges by a fairly uniform mechanism
704: to grow arrays as needed for printf, sub, gsub, etc.
1.1 tholo 705:
1.7 millert 706: Jul 23, 1997:
707: falling off the end of a function returns "" and 0, not 0.
708: thanks to arnold robbins.
1.1 tholo 709:
1.7 millert 710: Jun 17, 1997:
711: replaced several fixed-size arrays by dynamically-created ones
712: in run.c; added overflow tests to some previously unchecked cases.
713: getline, toupper, tolower.
1.1 tholo 714:
1.7 millert 715: getline code is still broken in that recursive calls may wind
716: up using the same space. [fixed later]
1.1 tholo 717:
1.7 millert 718: increased RECSIZE to 8192 to push problems further over the horizon.
1.1 tholo 719:
1.7 millert 720: added \r to \n as input line separator for programs, not data.
721: damn CRLFs.
1.1 tholo 722:
1.7 millert 723: modified format() to permit explicit printf("%c", 0) to include
724: a null byte in output. thanks to ken stailey for the fix.
1.1 tholo 725:
1.7 millert 726: added a "-safe" argument that disables file output (print >,
727: print >>), process creation (cmd|getline, print |, system), and
728: access to the environment (ENVIRON). this is a first approximation
729: to a "safe" version of awk, but don't rely on it too much. thanks
730: to joan feigenbaum and matt blaze for the inspiration long ago.
1.1 tholo 731:
1.7 millert 732: Jul 8, 1996:
733: fixed long-standing bug in sub, gsub(/a/, "\\\\&"); thanks to
734: ralph corderoy.
1.1 tholo 735:
1.7 millert 736: Jun 29, 1996:
737: fixed awful bug in new field splitting; didn't get all the places
738: where input was done.
1.1 tholo 739:
1.7 millert 740: Jun 28, 1996:
741: changed field-splitting to conform to posix definition: fields are
742: split using the value of FS at the time of input; it used to be
743: the value when the field or NF was first referred to, a much less
744: predictable definition. thanks to arnold robbins for encouragement
745: to do the right thing.
1.1 tholo 746:
1.7 millert 747: May 28, 1996:
748: fixed appalling but apparently unimportant bug in parsing octal
749: numbers in reg exprs.
1.1 tholo 750:
1.7 millert 751: explicit hex in reg exprs now limited to 2 chars: \xa, \xaa.
1.1 tholo 752:
1.7 millert 753: May 27, 1996:
754: cleaned up some declarations so gcc -Wall is now almost silent.
1.1 tholo 755:
1.7 millert 756: makefile now includes backup copies of ytab.c and lexyy.c in case
757: one makes before looking; it also avoids recreating lexyy.c unless
758: really needed.
1.1 tholo 759:
1.7 millert 760: s/aprintf/awkprint, s/asprintf/awksprintf/ to avoid some name clashes
761: with unwisely-written header files.
1.1 tholo 762:
1.7 millert 763: thanks to jeffrey friedl for several of these.
1.1 tholo 764:
1.7 millert 765: May 26, 1996:
766: an attempt to rationalize the (unsigned) char issue. almost all
767: instances of unsigned char have been removed; the handful of places
768: in b.c where chars are used as table indices have been hand-crafted.
769: added some latin-1 tests to the regression, but i'm not confident;
770: none of my compilers seem to care much. thanks to nelson beebe for
771: pointing out some others that do care.
1.1 tholo 772:
1.7 millert 773: May 2, 1996:
774: removed all register declarations.
1.1 tholo 775:
1.7 millert 776: enhanced split(), as in gawk, etc: split(s, a, "") splits s into
777: a[1]...a[length(s)] with each character a single element.
1.1 tholo 778:
1.7 millert 779: made the same changes for field-splitting if FS is "".
1.1 tholo 780:
1.7 millert 781: added nextfile, as in gawk: causes immediate advance to next
782: input file. (thanks to arnold robbins for inspiration and code).
1.1 tholo 783:
1.7 millert 784: small fixes to regexpr code: can now handle []], [[], and
1.25 millert 785: variants; [] is now a syntax error, rather than matching
1.7 millert 786: everything; [z-a] is now empty, not z. far from complete
787: or correct, however. (thanks to jeffrey friedl for pointing out
788: some awful behaviors.)
1.1 tholo 789:
1.7 millert 790: Apr 29, 1996:
1.12 otto 791: replaced uchar by uschar everywhere; apparently some compilers
1.7 millert 792: usurp this name and this causes conflicts.
1.1 tholo 793:
1.7 millert 794: fixed call to time in run.c (bltin); arg is time_t *.
1.1 tholo 795:
1.7 millert 796: replaced horrible pointer/long punning in b.c by a legitimate
797: union. should be safer on 64-bit machines and cleaner everywhere.
798: (thanks to nelson beebe for pointing out some of these problems.)
1.1 tholo 799:
1.7 millert 800: replaced nested comments by #if 0...#endif in run.c, lib.c.
1.1 tholo 801:
1.7 millert 802: removed getsval, setsval, execute macros from run.c and lib.c.
803: machines are 100x faster than they were when these macros were
804: first used.
1.1 tholo 805:
1.7 millert 806: revised filenames: awk.g.y => awkgram.y, awk.lx.l => awklex.l,
807: y.tab.[ch] => ytab.[ch], lex.yy.c => lexyy.c, all in the aid of
808: portability to nameless systems.
1.1 tholo 809:
1.7 millert 810: "make bundle" now includes yacc and lex output files for recipients
811: who don't have yacc or lex.
1.1 tholo 812:
1.7 millert 813: Aug 15, 1995:
814: initialized Cells in setsymtab more carefully; some fields
815: were not set. (thanks to purify, all of whose complaints i
816: think i now understand.)
1.1 tholo 817:
1.7 millert 818: fixed at least one error in gsub that looked at -1-th element
819: of an array when substituting for a null match (e.g., $).
1.1 tholo 820:
1.7 millert 821: delete arrayname is now legal; it clears the elements but leaves
822: the array, which may not be the right behavior.
1.1 tholo 823:
1.7 millert 824: modified makefile: my current make can't cope with the test used
825: to avoid unnecessary yacc invocations.
1.1 tholo 826:
1.7 millert 827: Jul 17, 1995:
828: added dynamically growing strings to awk.lx.l and b.c
829: to permit regular expressions to be much bigger.
830: the state arrays can still overflow.
1.1 tholo 831:
1.7 millert 832: Aug 24, 1994:
833: detect duplicate arguments in function definitions (mdm).
1.1 tholo 834:
1.7 millert 835: May 11, 1994:
836: trivial fix to printf to limit string size in sub().
1.1 tholo 837:
1.7 millert 838: Apr 22, 1994:
839: fixed yet another subtle self-assignment problem:
840: $1 = $2; $1 = $1 clobbered $1.
1.1 tholo 841:
1.7 millert 842: Regression tests now use private echo, to avoid quoting problems.
1.1 tholo 843:
1.7 millert 844: Feb 2, 1994:
845: changed error() to print line number as %d, not %g.
1.1 tholo 846:
1.7 millert 847: Jul 23, 1993:
848: cosmetic changes: increased sizes of some arrays,
849: reworded some error messages.
1.1 tholo 850:
1.7 millert 851: added CONVFMT as in posix (just replaced OFMT in getsval)
1.1 tholo 852:
1.7 millert 853: FILENAME is now "" until the first thing that causes a file
854: to be opened.
1.1 tholo 855:
1.7 millert 856: Nov 28, 1992:
857: deleted yyunput and yyoutput from proto.h;
858: different versions of lex give these different declarations.
1.1 tholo 859:
1.7 millert 860: May 31, 1992:
861: added -mr N and -mf N options: more record and fields.
862: these really ought to adjust automatically.
1.1 tholo 863:
1.7 millert 864: cleaned up some error messages; "out of space" now means
865: malloc returned NULL in all cases.
1.1 tholo 866:
1.7 millert 867: changed rehash so that if it runs out, it just returns;
868: things will continue to run slow, but maybe a bit longer.
1.1 tholo 869:
1.7 millert 870: Apr 24, 1992:
871: remove redundant close of stdin when using -f -.
1.1 tholo 872:
1.7 millert 873: got rid of core dump with -d; awk -d just prints date.
1.1 tholo 874:
1.7 millert 875: Apr 12, 1992:
876: added explicit check for /dev/std(in,out,err) in redirection.
877: unlike gawk, no /dev/fd/n yet.
1.1 tholo 878:
1.7 millert 879: added (file/pipe) builtin. hard to test satisfactorily.
880: not posix.
1.1 tholo 881:
1.7 millert 882: Feb 20, 1992:
883: recompile after abortive changes; should be unchanged.
1.1 tholo 884:
1.7 millert 885: Dec 2, 1991:
886: die-casting time: converted to ansi C, installed that.
1.1 tholo 887:
1.7 millert 888: Nov 30, 1991:
889: fixed storage leak in freefa, failing to recover [N]CCL.
890: thanks to Bill Jones (jones@cs.usask.ca)
1.1 tholo 891:
1.7 millert 892: Nov 19, 1991:
893: use RAND_MAX instead of literal in builtin().
1.1 tholo 894:
1.7 millert 895: Nov 12, 1991:
896: cranked up some fixed-size arrays in b.c, and added a test for
897: overflow in penter. thanks to mark larsen.
1.1 tholo 898:
1.7 millert 899: Sep 24, 1991:
900: increased buffer in gsub. a very crude fix to a general problem.
901: and again on Sep 26.
1.1 tholo 902:
1.7 millert 903: Aug 18, 1991:
904: enforce variable name syntax for commandline variables: has to
905: start with letter or _.
1.1 tholo 906:
1.7 millert 907: Jul 27, 1991:
908: allow newline after ; in for statements.
1.1 tholo 909:
1.7 millert 910: Jul 21, 1991:
911: fixed so that in self-assignment like $1=$1, side effects
912: like recomputing $0 take place. (this is getting subtle.)
1.1 tholo 913:
1.7 millert 914: Jun 30, 1991:
915: better test for detecting too-long output record.
1.1 tholo 916:
1.7 millert 917: Jun 2, 1991:
918: better defense against very long printf strings.
919: made break and continue illegal outside of loops.
1.1 tholo 920:
1.7 millert 921: May 13, 1991:
922: removed extra arg on gettemp, tempfree. minor error message rewording.
1.1 tholo 923:
924: May 6, 1991:
925: fixed silly bug in hex parsing in hexstr().
926: removed an apparently unnecessary test in isnumber().
927: warn about weird printf conversions.
928: fixed unchecked array overwrite in relex().
929:
930: changed for (i in array) to access elements in sorted order.
931: then unchanged it -- it really does run slower in too many cases.
932: left the code in place, commented out.
933:
1.7 millert 934: Feb 10, 1991:
935: check error status on all writes, to avoid banging on full disks.
1.1 tholo 936:
1.7 millert 937: Jan 28, 1991:
938: awk -f - reads the program from stdin.
1.1 tholo 939:
1.7 millert 940: Jan 11, 1991:
941: failed to set numeric state on $0 in cmd|getline context in run.c.
1.1 tholo 942:
1.7 millert 943: Nov 2, 1990:
944: fixed sleazy test for integrality in getsval; use modf.
1.1 tholo 945:
1.7 millert 946: Oct 29, 1990:
947: fixed sleazy buggy code in lib.c that looked (incorrectly) for
948: too long input lines.
1.1 tholo 949:
1.7 millert 950: Oct 14, 1990:
951: fixed the bug on p. 198 in which it couldn't deduce that an
952: argument was an array in some contexts. replaced the error
953: message in intest() by code that damn well makes it an array.
1.1 tholo 954:
1.7 millert 955: Oct 8, 1990:
956: fixed horrible bug: types and values were not preserved in
957: some kinds of self-assignment. (in assign().)
1.1 tholo 958:
1.7 millert 959: Aug 24, 1990:
960: changed NCHARS to 256 to handle 8-bit characters in strings
961: presented to match(), etc.
1.1 tholo 962:
1.7 millert 963: Jun 26, 1990:
964: changed struct rrow (awk.h) to use long instead of int for lval,
965: since cfoll() stores a pointer in it. now works better when int's
966: are smaller than pointers!
1.1 tholo 967:
1.7 millert 968: May 6, 1990:
969: AVA fixed the grammar so that ! is uniformly of the same precedence as
970: unary + and -. This renders illegal some constructs like !x=y, which
971: now has to be parenthesized as !(x=y), and makes others work properly:
972: !x+y is (!x)+y, and x!y is x !y, not two pattern-action statements.
973: (These problems were pointed out by Bob Lenk of Posix.)
1.1 tholo 974:
1.7 millert 975: Added \x to regular expressions (already in strings).
976: Limited octal to octal digits; \8 and \9 are not octal.
977: Centralized the code for parsing escapes in regular expressions.
978: Added a bunch of tests to T.re and T.sub to verify some of this.
1.1 tholo 979:
1.7 millert 980: Feb 9, 1990:
981: fixed null pointer dereference bug in main.c: -F[nothing]. sigh.
1.1 tholo 982:
1.7 millert 983: restored srand behavior: it returns the current seed.
1.1 tholo 984:
1.7 millert 985: Jan 18, 1990:
986: srand now returns previous seed value (0 to start).
1.1 tholo 987:
1.7 millert 988: Jan 5, 1990:
989: fix potential problem in tran.c -- something was freed,
990: then used in freesymtab.
1.1 tholo 991:
1.7 millert 992: Oct 18, 1989:
993: another try to get the max number of open files set with
994: relatively machine-independent code.
1.1 tholo 995:
1.7 millert 996: small fix to input() in case of multiple reads after EOF.
1.1 tholo 997:
1.7 millert 998: Oct 11, 1989:
999: FILENAME is now defined in the BEGIN block -- too many old
1000: programs broke.
1.1 tholo 1001:
1.7 millert 1002: "-" means stdin in getline as well as on the commandline.
1.1 tholo 1003:
1.7 millert 1004: added a bunch of casts to the code to tell the truth about
1005: char * vs. unsigned char *, a right royal pain. added a
1006: setlocale call to the front of main, though probably no one
1007: has it usefully implemented yet.
1.1 tholo 1008:
1.7 millert 1009: Aug 24, 1989:
1010: removed redundant relational tests against nullnode if parse
1011: tree already had a relational at that point.
1.1 tholo 1012:
1.7 millert 1013: Aug 11, 1989:
1014: fixed bug: commandline variable assignment has to look like
1015: var=something. (consider the man page for =, in file =.1)
1.1 tholo 1016:
1.7 millert 1017: changed number of arguments to functions to static arrays
1018: to avoid repeated malloc calls.
1.1 tholo 1019:
1.7 millert 1020: Aug 2, 1989:
1021: restored -F (space) separator
1.1 tholo 1022:
1.7 millert 1023: Jul 30, 1989:
1024: added -v x=1 y=2 ... for immediate commandline variable assignment;
1025: done before the BEGIN block for sure. they have to precede the
1026: program if the program is on the commandline.
1027: Modified Aug 2 to require a separate -v for each assignment.
1.1 tholo 1028:
1.7 millert 1029: Jul 10, 1989:
1030: fixed ref-thru-zero bug in environment code in tran.c
1.1 tholo 1031:
1.7 millert 1032: Jun 23, 1989:
1033: add newline to usage message.
1.1 tholo 1034:
1.7 millert 1035: Jun 14, 1989:
1036: added some missing ansi printf conversion letters: %i %X %E %G.
1037: no sensible meaning for h or L, so they may not do what one expects.
1.1 tholo 1038:
1.7 millert 1039: made %* conversions work.
1.1 tholo 1040:
1.7 millert 1041: changed x^y so that if n is a positive integer, it's done
1042: by explicit multiplication, thus achieving maximum accuracy.
1043: (this should be done by pow() but it seems not to be locally.)
1044: done to x ^= y as well.
1.1 tholo 1045:
1.7 millert 1046: Jun 4, 1989:
1047: ENVIRON array contains environment: if shell variable V=thing,
1048: ENVIRON["V"] is "thing"
1.1 tholo 1049:
1.7 millert 1050: multiple -f arguments permitted. error reporting is naive.
1051: (they were permitted before, but only the last was used.)
1.1 tholo 1052:
1.7 millert 1053: fixed a really stupid botch in the debugging macro dprintf
1.1 tholo 1054:
1.7 millert 1055: fixed order of evaluation of commandline assignments to match
1056: what the book claims: an argument of the form x=e is evaluated
1057: at the time it would have been opened if it were a filename (p 63).
1058: this invalidates the suggested answer to ex 4-1 (p 195).
1.1 tholo 1059:
1.7 millert 1060: removed some code that permitted -F (space) fieldseparator,
1061: since it didn't quite work right anyway. (restored aug 2)
1.1 tholo 1062:
1.7 millert 1063: Apr 27, 1989:
1064: Line number now accumulated correctly for comment lines.
1.1 tholo 1065:
1.7 millert 1066: Apr 26, 1989:
1067: Debugging output now includes a version date,
1068: if one compiles it into the source each time.
1.1 tholo 1069:
1.7 millert 1070: Apr 9, 1989:
1071: Changed grammar to prohibit constants as 3rd arg of sub and gsub;
1072: prevents class of overwriting-a-constant errors. (Last one?)
1073: This invalidates the "banana" example on page 43 of the book.
1.1 tholo 1074:
1.7 millert 1075: Added \a ("alert"), \v (vertical tab), \xhhh (hexadecimal),
1076: as in ANSI, for strings. Rescinded the sloppiness that permitted
1077: non-octal digits in \ooo. Warning: not all compilers and libraries
1078: will be able to deal with \x correctly.
1.1 tholo 1079:
1.7 millert 1080: Jan 9, 1989:
1081: Fixed bug that caused tempcell list to contain a duplicate.
1082: The fix is kludgy.
1.1 tholo 1083:
1.7 millert 1084: Dec 17, 1988:
1085: Catches some more commandline errors in main.
1086: Removed redundant decl of modf in run.c (confuses some compilers).
1087: Warning: there's no single declaration of malloc, etc., in awk.h
1088: that seems to satisfy all compilers.
1.1 tholo 1089:
1.7 millert 1090: Dec 7, 1988:
1091: Added a bit of code to error printing to avoid printing nulls.
1092: (Not clear that it actually would.)
1.1 tholo 1093:
1.7 millert 1094: Nov 27, 1988:
1095: With fear and trembling, modified the grammar to permit
1096: multiple pattern-action statements on one line without
1097: an explicit separator. By definition, this capitulation
1098: to the ghost of ancient implementations remains undefined
1099: and thus subject to change without notice or apology.
1100: DO NOT COUNT ON IT.
1.1 tholo 1101:
1.7 millert 1102: Oct 30, 1988:
1103: Fixed bug in call() that failed to recover storage.
1.1 tholo 1104:
1.7 millert 1105: A warning is now generated if there are more arguments
1106: in the call than in the definition (in lieu of fixing
1107: another storage leak).
1.2 millert 1108:
1.7 millert 1109: Oct 20, 1988:
1110: Fixed %c: if expr is numeric, use numeric value;
1111: otherwise print 1st char of string value. still
1112: doesn't work if the value is 0 -- won't print \0.
1.2 millert 1113:
1.7 millert 1114: Added a few more checks for running out of malloc.
1.2 millert 1115:
1.7 millert 1116: Oct 12, 1988:
1117: Fixed bug in call() that freed local arrays twice.
1.5 kstailey 1118:
1.7 millert 1119: Fixed to handle deletion of non-existent array right;
1120: complains about attempt to delete non-array element.
1.6 kstailey 1121:
1.7 millert 1122: Sep 30, 1988:
1123: Now guarantees to evaluate all arguments of built-in
1124: functions, as in C; the appearance is that arguments
1125: are evaluated before the function is called. Places
1126: affected are sub (gsub was ok), substr, printf, and
1127: all the built-in arithmetic functions in bltin().
1128: A warning is generated if a bltin() is called with
1129: the wrong number of arguments.
1.6 kstailey 1130:
1.7 millert 1131: This requires changing makeprof on p167 of the book.
1.6 kstailey 1132:
1.7 millert 1133: Aug 23, 1988:
1134: setting FILENAME in BEGIN caused core dump, apparently
1135: because it was freeing space not allocated by malloc.
1.6 kstailey 1136:
1.7 millert 1137: July 24, 1988:
1138: fixed egregious error in toupper/tolower functions.
1139: still subject to rescinding, however.
1.6 kstailey 1140:
1.7 millert 1141: July 2, 1988:
1142: flush stdout before opening file or pipe
1.6 kstailey 1143:
1.7 millert 1144: July 2, 1988:
1145: performance bug in b.c/cgoto(): not freeing some sets of states.
1146: partial fix only right now, and the number of states increased
1147: to make it less obvious.
1.6 kstailey 1148:
1.7 millert 1149: June 1, 1988:
1150: check error status on close
1.6 kstailey 1151:
1.7 millert 1152: May 28, 1988:
1153: srand returns seed value it's using.
1154: see 1/18/90
1.6 kstailey 1155:
1.7 millert 1156: May 22, 1988:
1157: Removed limit on depth of function calls.
1.6 kstailey 1158:
1.7 millert 1159: May 10, 1988:
1160: Fixed lib.c to permit _ in commandline variable names.
1.6 kstailey 1161:
1.7 millert 1162: Mar 25, 1988:
1163: main.c fixed to recognize -- as terminator of command-
1164: line options. Illegal options flagged.
1165: Error reporting slightly cleaned up.
1.6 kstailey 1166:
1.7 millert 1167: Dec 2, 1987:
1168: Newer C compilers apply a strict scope rule to extern
1169: declarations within functions. Two extern declarations in
1170: lib.c and tran.c have been moved to obviate this problem.
1.6 kstailey 1171:
1.7 millert 1172: Oct xx, 1987:
1173: Reluctantly added toupper and tolower functions.
1174: Subject to rescinding without notice.
1.6 kstailey 1175:
1.7 millert 1176: Sep 17, 1987:
1177: Error-message printer had printf(s) instead of
1178: printf("%s",s); got core dumps when the message
1179: included a %.
1.6 kstailey 1180:
1.7 millert 1181: Sep 12, 1987:
1182: Very long printf strings caused core dump;
1183: fixed aprintf, asprintf, format to catch them.
1184: Can still get a core dump in printf itself.
1.6 kstailey 1185:
1186: