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