Annotation of src/usr.bin/awk/FIXES, Revision 1.32
1.32 ! millert 1: /* $OpenBSD: FIXES,v 1.31 2020/06/10 21:05:02 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.
1.30 millert 28:
1.32 ! millert 29: February 28, 2020:
! 30: More cleanups from Christos Zoulas: notably backslash continuation
! 31: inside strings removes the newline and a fix for RS = "^a".
! 32: Fix for address sanitizer-found problem. Thanks to GitHub user
! 33: enh-google.
! 34:
! 35: February 19, 2020:
! 36: More small cleanups from Christos Zoulas.
! 37:
! 38: February 18, 2020:
! 39: Additional cleanups from Christos Zoulas. It's no longer necessary
! 40: to use the -y flag to bison.
! 41:
! 42: February 6, 2020:
! 43: Additional small cleanups from Christos Zoulas. awk is now
! 44: a little more robust about reporting I/O errors upon exit.
! 45:
1.31 millert 46: January 31, 2020:
47: Merge PR #70, which avoids use of variable length arrays. Thanks
48: to GitHub user michaelforney. Fix issue #60 ({0} in interval
1.32 ! millert 49: expressions doesn't work). Also get all tests working again.
! 50: Thanks to Arnold Robbins.
1.31 millert 51:
52: January 24, 2020:
53: A number of small cleanups from Christos Zoulas. Add the close
54: on exec flag to files/pipes opened for redirection; courtesy of
55: Arnold Robbins.
56:
57: January 19, 2020:
58: If POSIXLY_CORRECT is set in the environment, then sub and gsub
59: use POSIX rules for multiple backslashes. This fixes Issue #66,
60: while maintaining backwards compatibility.
61:
62: January 9, 2020:
63: Input/output errors on closing files are now fatal instead of
64: mere warnings. Thanks to Martijn Dekker <martijn@inlv.org>.
65:
1.30 millert 66: January 5, 2020:
67: Fix a bug in the concatentation of two string constants into
68: one done in the grammar. Fixes GitHub issue #61. Thanks
69: to GitHub user awkfan77 for pointing out the direction for
70: the fix. New test T.concat added to the test suite.
71: Fix a few memory leaks reported by valgrind, as well.
72:
73: December 27, 2019:
74: Fix a bug whereby a{0,3} could match four a's. Thanks to
75: "Anonymous AWK fan" for the report.
76:
77: December 11, 2019:
78: Further printf-related fixes for 32 bit systems.
79: Thanks again to Christos Zoulas.
80:
81: December 8, 2019:
82: Fix the return value of sprintf("%d") on 32 bit systems.
83: Thanks to Jim Lowe for the report and to Christos Zoulas
84: for the fix.
1.29 millert 85:
86: November 10, 2019:
87: Convert a number of Boolean integer variables into
88: actual bools. Convert compile_time variable into an
89: enum and simplify some of the related code. Thanks
90: to Arnold Robbins.
91:
92: November 8, 2019:
93: Fix from Ori Bernstein to get UTF-8 characters instead of
94: bytes when FS = "". This is currently the only bit of
95: the One True Awk that understands multibyte characters.
96: From Arnold Robbins, apply some cleanups in the test suite.
1.28 millert 97:
98: October 25, 2019:
99: More fixes and cleanups from NetBSD, courtesy of Christos
100: Zoulas. Merges PRs 54 and 55.
101:
102: October 24, 2019:
103: Import second round of code cleanups from NetBSD. Much thanks
1.31 millert 104: to Christos Zoulas (GitHub user zoulasc). Merges PR 53.
1.28 millert 105: Add an optimization for string concatenation, also from
106: Christos.
1.27 millert 107:
108: October 17, 2019:
109: Import code cleanups from NetBSD. Much thanks to Christos
1.31 millert 110: Zoulas (GitHub user zoulasc). Merges PR 51.
1.26 millert 111:
112: October 6, 2019:
113: Import code from NetBSD awk that implements RS as a regular
114: expression.
1.23 millert 115:
1.25 millert 116: September 10, 2019:
117: Fixes for various array / memory overruns found via gcc's
1.31 millert 118: -fsanitize=unknown. Thanks to Alexander Richardson (GitHub
1.25 millert 119: user arichardson). Merges PRs 47 and 48.
120:
121: July 28, 2019:
122: Import grammar optimization from NetBSD: Two string constants
123: concatenated together get turned into a single string.
124:
125: July 26, 2019:
126: Support POSIX-specified C-style escape sequences "\a" (alarm)
127: and "\v" (vertical tab) in command line arguments and regular
128: expressions, further to the support for them in strings added on
129: Apr 9, 1989. These now no longer match as literal "a" and "v"
130: characters (as they don't on other awk implementations).
131: Thanks to Martijn Dekker.
132:
133: July 17, 2019:
134: Pull in a number of code cleanups and minor fixes from
135: Warner Losh's bsd-ota branch. The only user visible change
136: is the use of random(3) as the random number generator.
137: Thanks to Warner Losh for collecting all these fixes in
138: one easy place to get them from.
139:
140: July 16, 2019:
141: Fix field splitting to use FS value as of the time a record
142: was read or assigned to. Thanks to GitHub user Cody Mello (melloc)
143: for the fix. (Merged from his branch, via PR #42.) Updated
144: testdir/T.split per said PR as well.
145:
146: June 24, 2019:
147: Extract awktest.tar into testdir directory. Add some very
148: simple mechanics to the makefile for running the tests and
149: for cleaning up. No changes to awk itself.
150:
1.24 millert 151: June 17, 2019:
152: Disallow deleting SYMTAB and its elements, which creates
153: use-after-free bugs. Thanks to GitHub user Cody Mello (melloc)
154: for the fix. (Merged from PR #43.)
155:
156: June 5, 2019:
157: Allow unmatched right parenthesis in a regular expression to
158: be treated literally. Fixes Issue #40. Thanks to GitHub user
159: Warner Losh (bsdimp) for the report. Thanks to Arnold Robbins
160: for the fix.
161:
162: May 29,2019:
163: Fix check for command line arguments to no longer require that
164: first character after '=' not be another '='. Reverts change of
165: August 11, 1989. Thanks to GitHub user Jamie Landeg Jones for
166: pointing out the issue; from Issue #38.
167:
168: Apr 7, 2019:
169: Update awktest.tar(p.50) to use modern options to sort. Needed
170: for Android development. Thanks to GitHub user mohd-akram (Mohamed
171: Akram). From Issue #33.
172:
173: Mar 12, 2019:
174: Added very simplistic support for cross-compiling in the
175: makefile. We are NOT going to go in the direction of the
176: autotools, though. Thanks to GitHub user nee-san for
177: the basic change. (Merged from PR #34.)
178:
1.23 millert 179: Mar 5, 2019:
180: Added support for POSIX-standard interval expressions (a.k.a.
181: bounds, a.k.a. repetition expressions) in regular expressions,
182: backported (via NetBSD) from Apple awk-24 (20070501).
183: Thanks to Martijn Dekker <martijn@inlv.org> for the port.
184: (Merged from PR #30.)
185:
186: Mar 3, 2019:
187: Merge PRs as follows:
188: #12: Avoid undefined behaviour when using ctype(3) functions in
189: relex(). Thanks to GitHub user iamleot.
190: #31: Make getline handle numeric strings, and update FIXES. Thanks
1.24 millert 191: to GitHub user arnoldrobbins.
1.23 millert 192: #32: maketab: support build systems with read-only source. Thanks
193: to GitHub user enh.
1.22 millert 194:
195: Jan 25, 2019:
196: Make getline handle numeric strings properly in all cases.
197: (Thanks, Arnold.)
198:
199: Jan 21, 2019:
200: Merged a number of small fixes from GitHub pull requests.
201: Thanks to GitHub users Arnold Robbins (arnoldrobbins),
202: Cody Mello (melloc) and Christoph Junghans (junghans).
203: PR numbers: 13-21, 23, 24, 27.
1.21 millert 204:
205: Oct 25, 2018:
206: Added test in maketab.c to prevent generating a proctab entry
207: for YYSTYPE_IS_DEFINED. It was harmless but some gcc settings
208: generated a warning message. Thanks to Nan Xiao for report.
1.20 millert 209:
210: Aug 27, 2018:
211: Disallow '$' in printf formats; arguments evaluated in order
212: and printed in order.
213:
214: Added some casts to silence warnings on debugging printfs.
215: (Thanks, Arnold.)
1.18 millert 216:
1.19 millert 217: Aug 23, 2018:
218: A long list of fixes courtesy of Arnold Robbins,
219: to whom profound thanks.
220:
221: 1. ofs-rebuild: OFS value used to rebuild the record was incorrect.
222: Fixed August 19, 2014. Revised fix August 2018.
223:
224: 2. system-status: Instead of a floating-point division by 256, use
225: the wait(2) macros to create a reasonable exit status.
226: Fixed March 12, 2016.
227:
228: 3. space: Use provided xisblank() function instead of ispace() for
229: matching [[:blank:]].
230:
231: 4. a-format: Add POSIX standard %a and %A to supported formats. Check
232: at runtime that this format is available.
233:
234: 5. decr-NF: Decrementing NF did not change $0. This is a decades-old
235: bug. There are interactions with the old and new value of OFS as well.
236: Most of the fix came from the NetBSD awk.
237:
238: 6. string-conv: String conversions of scalars were sticky. Once a
239: conversion to string happened, even with OFMT, that value was used until
240: a new numeric value was assigned, even if OFMT differed from CONVFMT,
241: and also if CONVFMT changed.
242:
243: 7. unary-plus: Unary plus on a string constant returned the string.
244: Instead, it should convert the value to numeric and give that value.
245:
246: Also added Arnold's tests for these to awktest.tar as T.arnold.
247:
248: Aug 15, 2018:
249: fixed mangled awktest.tar (thanks, Arnold), posted all
250: current (very minor) fixes to github / onetrueawk
251:
1.18 millert 252: Jun 7, 2018:
253: (yes, a long layoff)
254: Updated some broken tests (beebe.tar, T.lilly)
255: [thanks to Arnold Robbins]
256:
257: Mar 26, 2015:
258: buffer overflow in error reporting; thanks to tobias ulmer
259: and john-mark gurney for spotting it and the fix.
260:
261: Feb 4, 2013:
262: cleaned up a handful of tests that didn't seem to actually
263: test for correct behavior: T.latin1, T.gawk.
264:
265: Jan 5, 2013:
266: added ,NULL initializer to static Cells in run.c; not really
267: needed but cleaner. Thanks to Michael Bombardieri.
1.17 millert 268:
269: Dec 20, 2012:
270: fiddled makefile to get correct yacc and bison flags. pick yacc
271: (linux) or bison (mac) as necessary.
272:
273: added __attribute__((__noreturn__)) to a couple of lines in
274: proto.h, to silence someone's enthusiastic checker.
275:
276: fixed obscure call by value bug in split(a[1],a) reported on
277: 9fans. the management of temporary values is just a mess; i
278: took a shortcut by making an extra string copy. thanks
279: to paul patience and arnold robbins for passing it on and for
280: proposed patches.
281:
282: tiny fiddle in setfval to eliminate -0 results in T.expr, which
283: has irritated me for 20+ years.
1.16 millert 284:
285: Aug 10, 2011:
286: another fix to avoid core dump with delete(ARGV); again, many thanks
287: to ruslan ermilov.
288:
289: Aug 7, 2011:
290: split(s, a, //) now behaves the same as split(s, a, "")
291:
292: Jun 12, 2011:
293: /pat/, \n /pat/ {...} is now legal, though bad style to use.
294:
295: added checks to new -v code that permits -vnospace; thanks to
1.25 millert 296: ruslan ermilov for spotting this and providing the patch.
1.16 millert 297:
298: removed fixed limit on number of open files; thanks to aleksey
1.25 millert 299: cheusov and christos zoulos.
1.16 millert 300:
301: fixed day 1 bug that resurrected deleted elements of ARGV when
302: used as filenames (in lib.c).
303:
304: minor type fiddles to make gcc -Wall -pedantic happier (but not
305: totally so); turned on -fno-strict-aliasing in makefile.
306:
307: May 6, 2011:
308: added #ifdef for isblank.
309: now allows -ffoo as well as -f foo arguments.
310: (thanks, ruslan)
311:
312: May 1, 2011:
313: after advice from todd miller, kevin lo, ruslan ermilov,
314: and arnold robbins, changed srand() to return the previous
315: seed (which is 1 on the first call of srand). the seed is
316: an Awkfloat internally though converted to unsigned int to
1.25 millert 317: pass to the library srand(). thanks, everyone.
1.16 millert 318:
319: fixed a subtle (and i hope low-probability) overflow error
1.25 millert 320: in fldbld, by adding space for one extra \0. thanks to
1.16 millert 321: robert bassett for spotting this one and providing a fix.
322:
323: removed the files related to compilation on windows. i no
324: longer have anything like a current windows environment, so
325: i can't test any of it.
1.15 millert 326:
327: May 23, 2010:
328: fixed long-standing overflow bug in run.c; many thanks to
329: nelson beebe for spotting it and providing the fix.
330:
331: fixed bug that didn't parse -vd=1 properly; thanks to santiago
332: vila for spotting it.
333:
334: Feb 8, 2010:
335: i give up. replaced isblank with isspace in b.c; there are
336: no consistent header files.
337:
338: Nov 26, 2009:
339: fixed a long-standing issue with when FS takes effect. a
340: change to FS is now noticed immediately for subsequent splits.
341:
342: changed the name getline() to awkgetline() to avoid yet another
343: name conflict somewhere.
344:
345: Feb 11, 2009:
346: temporarily for now defined HAS_ISBLANK, since that seems to
347: be the best way through the thicket. isblank arrived in C99,
348: but seems to be arriving at different systems at different
349: times.
350:
351: Oct 8, 2008:
352: fixed typo in b.c that set tmpvec wrongly. no one had ever
353: run into the problem, apparently. thanks to alistair crooks.
1.14 millert 354:
355: Oct 23, 2007:
356: minor fix in lib.c: increase inputFS to 100, change malloc
1.25 millert 357: for fields to n+1.
1.14 millert 358:
359: fixed memory fault caused by out of order test in setsval.
360:
361: thanks to david o'brien, freebsd, for both fixes.
362:
363: May 1, 2007:
364: fiddle in makefile to fix for BSD make; thanks to igor sobrado.
365:
366: Mar 31, 2007:
367: fixed some null pointer refs calling adjbuf.
368:
369: Feb 21, 2007:
370: fixed a bug in matching the null RE in sub and gsub. thanks to al aho
371: who actually did the fix (in b.c), and to wolfgang seeberg for finding
372: it and providing a very compact test case.
373:
374: fixed quotation in b.c; thanks to Hal Pratt and the Princeton Dante
375: Project.
376:
377: removed some no-effect asserts in run.c.
378:
379: fiddled maketab.c to not complain about bison-generated values.
380:
381: removed the obsolete -V argument; fixed --version to print the
382: version and exit.
383:
384: fixed wording and an outright error in the usage message; thanks to igor
385: sobrado and jason mcintyre.
386:
387: fixed a bug in -d that caused core dump if no program followed.
388:
389: Jan 1, 2007:
390: dropped mac.code from makefile; there are few non-MacOSX
391: mac's these days.
392:
393: Jan 17, 2006:
394: system() not flagged as unsafe in the unadvertised -safe option.
395: found it while enhancing tests before shipping the ;login: article.
396: practice what you preach.
397:
398: removed the 9-years-obsolete -mr and -mf flags.
399:
400: added -version and --version options.
401:
402: core dump on linux with BEGIN {nextfile}, now fixed.
403:
1.25 millert 404: removed some #ifdef's in run.c and lex.c that appear to no
1.14 millert 405: longer be necessary.
406:
407: Apr 24, 2005:
408: modified lib.c so that values of $0 et al are preserved in the END
409: block, apparently as required by posix. thanks to havard eidnes
410: for the report and code.
411:
412: Jan 14, 2005:
413: fixed infinite loop in parsing, originally found by brian tsang.
414: thanks to arnold robbins for a suggestion that started me
415: rethinking it.
416:
417: Dec 31, 2004:
1.25 millert 418: prevent overflow of -f array in main, head off potential error in
419: call of SYNTAX(), test malloc return in lib.c, all with thanks to
1.14 millert 420: todd miller.
1.13 millert 421:
422: Dec 22, 2004:
423: cranked up size of NCHARS; coverity thinks it can be overrun with
424: smaller size, and i think that's right. added some assertions to b.c
425: to catch places where it might overrun. the RE code is still fragile.
426:
427: Dec 5, 2004:
428: fixed a couple of overflow problems with ridiculous field numbers:
429: e.g., print $(2^32-1). thanks to ruslan ermilov, giorgos keramidas
430: and david o'brien at freebsd.org for patches. this really should
431: be re-done from scratch.
432:
433: Nov 21, 2004:
434: fixed another 25-year-old RE bug, in split. it's another failure
435: to (re-)initialize. thanks to steve fisher for spotting this and
436: providing a good test case.
437:
438: Nov 22, 2003:
439: fixed a bug in regular expressions that dates (so help me) from 1977;
440: it's been there from the beginning. an anchored longest match that
441: was longer than the number of states triggered a failure to initialize
442: the machine properly. many thanks to moinak ghosh for not only finding
443: this one but for providing a fix, in some of the most mysterious
444: code known to man.
445:
446: fixed a storage leak in call() that appears to have been there since
1.25 millert 447: 1983 or so -- a function without an explicit return that assigns a
448: string to a parameter leaked a Cell. thanks to moinak ghosh for
1.13 millert 449: spotting this very subtle one.
450:
451: Jul 31, 2003:
452: fixed, thanks to andrey chernov and ruslan ermilov, a bug in lex.c
453: that mis-handled the character 255 in input. (it was being compared
454: to EOF with a signed comparison.)
455:
456: Jul 29, 2003:
457: fixed (i think) the long-standing botch that included the beginning of
458: line state ^ for RE's in the set of valid characters; this led to a
459: variety of odd problems, including failure to properly match certain
460: regular expressions in non-US locales. thanks to ruslan for keeping
461: at this one.
462:
463: Jul 28, 2003:
464: n-th try at getting internationalization right, with thanks to volker
465: kiefel, arnold robbins and ruslan ermilov for advice, though they
466: should not be blamed for the outcome. according to posix, "." is the
467: radix character in programs and command line arguments regardless of
468: the locale; otherwise, the locale should prevail for input and output
469: of numbers. so it's intended to work that way.
1.25 millert 470:
1.13 millert 471: i have rescinded the attempt to use strcoll in expanding shorthands in
472: regular expressions (cclenter). its properties are much too
473: surprising; for example [a-c] matches aAbBc in locale en_US but abBcC
474: in locale fr_CA. i can see how this might arise by implementation
475: but i cannot explain it to a human user. (this behavior can be seen
476: in gawk as well; we're leaning on the same library.)
477:
478: the issue appears to be that strcoll is meant for sorting, where
479: merging upper and lower case may make sense (though note that unix
480: sort does not do this by default either). it is not appropriate
481: for regular expressions, where the goal is to match specific
482: patterns of characters. in any case, the notations [:lower:], etc.,
483: are available in awk, and they are more likely to work correctly in
484: most locales.
485:
486: a moratorium is hereby declared on internationalization changes.
487: i apologize to friends and colleagues in other parts of the world.
488: i would truly like to get this "right", but i don't know what
489: that is, and i do not want to keep making changes until it's clear.
490:
491: Jul 4, 2003:
492: fixed bug that permitted non-terminated RE, as in "awk /x".
493:
494: Jun 1, 2003:
495: subtle change to split: if source is empty, number of elems
496: is always 0 and the array is not set.
497:
498: Mar 21, 2003:
499: added some parens to isblank, in another attempt to make things
500: internationally portable.
501:
502: Mar 14, 2003:
503: the internationalization changes, somewhat modified, are now
504: reinstated. in theory awk will now do character comparisons
505: and case conversions in national language, but "." will always
506: be the decimal point separator on input and output regardless
507: of national language. isblank(){} has an #ifndef.
508:
509: this no longer compiles on windows: LC_MESSAGES isn't defined
510: in vc6++.
511:
512: fixed subtle behavior in field and record splitting: if FS is
513: a single character and RS is not empty, \n is NOT a separator.
514: this tortuous reading is found in the awk book; behavior now
515: matches gawk and mawk.
1.11 millert 516:
517: Dec 13, 2002:
518: for the moment, the internationalization changes of nov 29 are
519: rolled back -- programs like x = 1.2 don't work in some locales,
520: because the parser is expecting x = 1,2. until i understand this
521: better, this will have to wait.
522:
523: Nov 29, 2002:
524: modified b.c (with tiny changes in main and run) to support
525: locales, using strcoll and iswhatever tests for posix character
526: classes. thanks to ruslan ermilov (ru@freebsd.org) for code.
527: the function isblank doesn't seem to have propagated to any
528: header file near me, so it's there explicitly. not properly
529: tested on non-ascii character sets by me.
530:
531: Jun 28, 2002:
532: modified run/format() and tran/getsval() to do a slightly better
533: job on using OFMT for output from print and CONVFMT for other
1.25 millert 534: number->string conversions, as promised by posix and done by
1.11 millert 535: gawk and mawk. there are still places where it doesn't work
536: right if CONVFMT is changed; by then the STR attribute of the
537: variable has been irrevocably set. thanks to arnold robbins for
538: code and examples.
539:
540: fixed subtle bug in format that could get core dump. thanks to
541: Jaromir Dolecek <jdolecek@NetBSD.org> for finding and fixing.
542: minor cleanup in run.c / format() at the same time.
543:
544: added some tests for null pointers to debugging printf's, which
545: were never intended for external consumption. thanks to dave
546: kerns (dkerns@lucent.com) for pointing this out.
547:
548: GNU compatibility: an empty regexp matches anything (thanks to
549: dag-erling smorgrav, des@ofug.org). subject to reversion if
550: this does more harm than good.
551:
552: pervasive small changes to make things more const-correct, as
553: reported by gcc's -Wwrite-strings. as it says in the gcc manual,
554: this may be more nuisance than useful. provoked by a suggestion
555: and code from arnaud desitter, arnaud@nimbus.geog.ox.ac.uk
556:
557: minor documentation changes to note that this now compiles out
558: of the box on Mac OS X.
559:
560: Feb 10, 2002:
561: changed types in posix chars structure to quiet solaris cc.
562:
563: Jan 1, 2002:
564: fflush() or fflush("") flushes all files and pipes.
565:
1.25 millert 566: length(arrayname) returns number of elements; thanks to
1.11 millert 567: arnold robbins for suggestion.
568:
569: added a makefile.win to make it easier to build on windows.
570: based on dan allen's buildwin.bat.
571:
572: Nov 16, 2001:
573: added support for posix character class names like [:digit:],
574: which are not exactly shorter than [0-9] and perhaps no more
575: portable. thanks to dag-erling smorgrav for code.
576:
577: Feb 16, 2001:
578: removed -m option; no longer needed, and it was actually
579: broken (noted thanks to volker kiefel).
580:
581: Feb 10, 2001:
582: fixed an appalling bug in gettok: any sequence of digits, +,-, E, e,
583: and period was accepted as a valid number if it started with a period.
584: this would never have happened with the lex version.
585:
586: other 1-character botches, now fixed, include a bare $ and a
587: bare " at the end of the input.
588:
589: Feb 7, 2001:
590: more (const char *) casts in b.c and tran.c to silence warnings.
1.8 millert 591:
1.10 millert 592: Nov 15, 2000:
593: fixed a bug introduced in august 1997 that caused expressions
594: like $f[1] to be syntax errors. thanks to arnold robbins for
595: noticing this and providing a fix.
596:
597: Oct 30, 2000:
598: fixed some nextfile bugs: not handling all cases. thanks to
599: arnold robbins for pointing this out. new regressions added.
600:
601: close() is now a function. it returns whatever the library
602: fclose returns, and -1 for closing a file or pipe that wasn't
603: opened.
604:
605: Sep 24, 2000:
606: permit \n explicitly in character classes; won't work right
607: if comes in as "[\n]" but ok as /[\n]/, because of multiple
608: processing of \'s. thanks to arnold robbins.
609:
610: July 5, 2000:
611: minor fiddles in tran.c to keep compilers happy about uschar.
612: thanks to norman wilson.
613:
614: May 25, 2000:
615: yet another attempt at making 8-bit input work, with another
1.25 millert 616: band-aid in b.c (member()), and some (uschar) casts to head
1.10 millert 617: off potential errors in subscripts (like isdigit). also
618: changed HAT to NCHARS-2. thanks again to santiago vila.
619:
620: changed maketab.c to ignore apparently out of range definitions
621: instead of halting; new freeBSD generates one. thanks to
622: jon snader <jsnader@ix.netcom.com> for pointing out the problem.
623:
624: May 2, 2000:
625: fixed an 8-bit problem in b.c by making several char*'s into
626: unsigned char*'s. not clear i have them all yet. thanks to
627: Santiago Vila <sanvila@unex.es> for the bug report.
628:
629: Apr 21, 2000:
630: finally found and fixed a memory leak in function call; it's
631: been there since functions were added ~1983. thanks to
632: jon bentley for the test case that found it.
633:
634: added test in envinit to catch environment "variables" with
1.12 otto 635: names beginning with '='; thanks to Berend Hasselman.
1.10 millert 636:
637: Jul 28, 1999:
638: added test in defn() to catch function foo(foo), which
639: otherwise recurses until core dump. thanks to arnold
640: robbins for noticing this.
641:
1.9 millert 642: Jun 20, 1999:
643: added *bp in gettok in lex.c; appears possible to exit function
644: without terminating the string. thanks to russ cox.
645:
646: Jun 2, 1999:
647: added function stdinit() to run to initialize files[] array,
648: in case stdin, etc., are not constants; some compilers care.
649:
650: May 10, 1999:
651: replaced the ERROR ... FATAL, etc., macros with functions
652: based on vprintf, to avoid problems caused by overrunning
653: fixed-size errbuf array. thanks to ralph corderoy for the
654: impetus, and for pointing out a string termination bug in
655: qstring as well.
656:
657: Apr 21, 1999:
658: fixed bug that caused occasional core dumps with commandline
659: variable with value ending in \. (thanks to nelson beebe for
660: the test case.)
661:
1.8 millert 662: Apr 16, 1999:
1.25 millert 663: with code kindly provided by Bruce Lilly, awk now parses
1.8 millert 664: /=/ and similar constructs more sensibly in more places.
1.10 millert 665: Bruce also provided some helpful test cases.
1.8 millert 666:
667: Apr 5, 1999:
668: changed true/false to True/False in run.c to make it
669: easier to compile with C++. Added some casts on malloc
670: and realloc to be honest about casts; ditto. changed
671: ltype int to long in struct rrow to reduce some 64-bit
672: complaints; other changes scattered throughout for the
673: same purpose. thanks to Nelson Beebe for these portability
674: improvements.
675:
676: removed some horrible pointer-int casting in b.c and elsewhere
677: by adding ptoi and itonp to localize the casts, which are
678: all benign. fixed one incipient bug that showed up on sgi
679: in 64-bit mode.
680:
681: reset lineno for new source file; include filename in error
682: message. also fixed line number error in continuation lines.
683: (thanks to Nelson Beebe for both of these.)
684:
685: Mar 24, 1999:
686: Nelson Beebe notes that irix 5.3 yacc dies with a bogus
687: error; use a newer version or switch to bison, since sgi
688: is unlikely to fix it.
1.1 tholo 689:
1.7 millert 690: Mar 5, 1999:
1.9 millert 691: changed isnumber to is_number to avoid the problem caused by
692: versions of ctype.h that include the name isnumber.
1.7 millert 693:
694: distribution now includes a script for building on a Mac,
695: thanks to Dan Allen.
696:
697: Feb 20, 1999:
698: fixed memory leaks in run.c (call) and tran.c (setfval).
699: thanks to Stephen Nutt for finding these and providing the fixes.
700:
701: Jan 13, 1999:
702: replaced srand argument by (unsigned int) in run.c;
703: avoids problem on Mac and potentially on Unix & Windows.
704: thanks to Dan Allen.
705:
706: added a few (int) casts to silence useless compiler warnings.
707: e.g., errorflag= in run.c jump().
708:
709: added proctab.c to the bundle outout; one less thing
710: to have to compile out of the box.
711:
712: added calls to _popen and _pclose to the win95 stub for
713: pipes (thanks to Steve Adams for this helpful suggestion).
714: seems to work, though properties are not well understood
715: by me, and it appears that under some circumstances the
716: pipe output is truncated. Be careful.
717:
718: Oct 19, 1998:
719: fixed a couple of bugs in getrec: could fail to update $0
1.25 millert 720: after a getline var; because inputFS wasn't initialized,
1.7 millert 721: could split $0 on every character, a misleading diversion.
722:
723: fixed caching bug in makedfa: LRU was actually removing
724: least often used.
725:
726: thanks to ross ridge for finding these, and for providing
727: great bug reports.
728:
729: May 12, 1998:
730: fixed potential bug in readrec: might fail to update record
731: pointer after growing. thanks to dan levy for spotting this
732: and suggesting the fix.
733:
734: Mar 12, 1998:
735: added -V to print version number and die.
1.19 millert 736:
737: [notify dave kerns, dkerns@dacsoup.ih.lucent.com]
1.7 millert 738:
739: Feb 11, 1998:
740: subtle silent bug in lex.c: if the program ended with a number
741: longer than 1 digit, part of the input would be pushed back and
742: parsed again because token buffer wasn't terminated right.
743: example: awk 'length($0) > 10'. blush. at least i found it
744: myself.
745:
746: Aug 31, 1997:
747: s/adelete/awkdelete/: SGI uses this in malloc.h.
748: thanks to nelson beebe for pointing this one out.
749:
750: Aug 21, 1997:
751: fixed some bugs in sub and gsub when replacement includes \\.
752: this is a dark, horrible corner, but at least now i believe that
753: the behavior is the same as gawk and the intended posix standard.
754: thanks to arnold robbins for advice here.
755:
756: Aug 9, 1997:
757: somewhat regretfully, replaced the ancient lex-based lexical
758: analyzer with one written in C. it's longer, generates less code,
759: and more portable; the old one depended too much on mysterious
760: properties of lex that were not preserved in other environments.
761: in theory these recognize the same language.
762:
763: now using strtod to test whether a string is a number, instead of
764: the convoluted original function. should be more portable and
765: reliable if strtod is implemented right.
766:
767: removed now-pointless optimization in makefile that tries to avoid
768: recompilation when awkgram.y is changed but symbols are not.
769:
770: removed most fixed-size arrays, though a handful remain, some
771: of which are unchecked. you have been warned.
772:
773: Aug 4, 1997:
774: with some trepidation, replaced the ancient code that managed
775: fields and $0 in fixed-size arrays with arrays that grow on
776: demand. there is still some tension between trying to make this
777: run fast and making it clean; not sure it's right yet.
778:
779: the ill-conceived -mr and -mf arguments are now useful only
780: for debugging. previous dynamic string code removed.
781:
782: numerous other minor cleanups along the way.
1.1 tholo 783:
1.7 millert 784: Jul 30, 1997:
785: using code provided by dan levy (to whom profuse thanks), replaced
786: fixed-size arrays and awkward kludges by a fairly uniform mechanism
787: to grow arrays as needed for printf, sub, gsub, etc.
1.1 tholo 788:
1.7 millert 789: Jul 23, 1997:
790: falling off the end of a function returns "" and 0, not 0.
791: thanks to arnold robbins.
1.1 tholo 792:
1.7 millert 793: Jun 17, 1997:
794: replaced several fixed-size arrays by dynamically-created ones
795: in run.c; added overflow tests to some previously unchecked cases.
796: getline, toupper, tolower.
1.1 tholo 797:
1.7 millert 798: getline code is still broken in that recursive calls may wind
799: up using the same space. [fixed later]
1.1 tholo 800:
1.7 millert 801: increased RECSIZE to 8192 to push problems further over the horizon.
1.1 tholo 802:
1.7 millert 803: added \r to \n as input line separator for programs, not data.
804: damn CRLFs.
1.1 tholo 805:
1.7 millert 806: modified format() to permit explicit printf("%c", 0) to include
807: a null byte in output. thanks to ken stailey for the fix.
1.1 tholo 808:
1.7 millert 809: added a "-safe" argument that disables file output (print >,
810: print >>), process creation (cmd|getline, print |, system), and
811: access to the environment (ENVIRON). this is a first approximation
812: to a "safe" version of awk, but don't rely on it too much. thanks
813: to joan feigenbaum and matt blaze for the inspiration long ago.
1.1 tholo 814:
1.7 millert 815: Jul 8, 1996:
816: fixed long-standing bug in sub, gsub(/a/, "\\\\&"); thanks to
817: ralph corderoy.
1.1 tholo 818:
1.7 millert 819: Jun 29, 1996:
820: fixed awful bug in new field splitting; didn't get all the places
821: where input was done.
1.1 tholo 822:
1.7 millert 823: Jun 28, 1996:
824: changed field-splitting to conform to posix definition: fields are
825: split using the value of FS at the time of input; it used to be
826: the value when the field or NF was first referred to, a much less
827: predictable definition. thanks to arnold robbins for encouragement
828: to do the right thing.
1.1 tholo 829:
1.7 millert 830: May 28, 1996:
831: fixed appalling but apparently unimportant bug in parsing octal
832: numbers in reg exprs.
1.1 tholo 833:
1.7 millert 834: explicit hex in reg exprs now limited to 2 chars: \xa, \xaa.
1.1 tholo 835:
1.7 millert 836: May 27, 1996:
837: cleaned up some declarations so gcc -Wall is now almost silent.
1.1 tholo 838:
1.7 millert 839: makefile now includes backup copies of ytab.c and lexyy.c in case
840: one makes before looking; it also avoids recreating lexyy.c unless
841: really needed.
1.1 tholo 842:
1.7 millert 843: s/aprintf/awkprint, s/asprintf/awksprintf/ to avoid some name clashes
844: with unwisely-written header files.
1.1 tholo 845:
1.7 millert 846: thanks to jeffrey friedl for several of these.
1.1 tholo 847:
1.7 millert 848: May 26, 1996:
849: an attempt to rationalize the (unsigned) char issue. almost all
850: instances of unsigned char have been removed; the handful of places
851: in b.c where chars are used as table indices have been hand-crafted.
852: added some latin-1 tests to the regression, but i'm not confident;
853: none of my compilers seem to care much. thanks to nelson beebe for
854: pointing out some others that do care.
1.1 tholo 855:
1.7 millert 856: May 2, 1996:
857: removed all register declarations.
1.1 tholo 858:
1.7 millert 859: enhanced split(), as in gawk, etc: split(s, a, "") splits s into
860: a[1]...a[length(s)] with each character a single element.
1.1 tholo 861:
1.7 millert 862: made the same changes for field-splitting if FS is "".
1.1 tholo 863:
1.7 millert 864: added nextfile, as in gawk: causes immediate advance to next
865: input file. (thanks to arnold robbins for inspiration and code).
1.1 tholo 866:
1.7 millert 867: small fixes to regexpr code: can now handle []], [[], and
1.25 millert 868: variants; [] is now a syntax error, rather than matching
1.7 millert 869: everything; [z-a] is now empty, not z. far from complete
870: or correct, however. (thanks to jeffrey friedl for pointing out
871: some awful behaviors.)
1.1 tholo 872:
1.7 millert 873: Apr 29, 1996:
1.12 otto 874: replaced uchar by uschar everywhere; apparently some compilers
1.7 millert 875: usurp this name and this causes conflicts.
1.1 tholo 876:
1.7 millert 877: fixed call to time in run.c (bltin); arg is time_t *.
1.1 tholo 878:
1.7 millert 879: replaced horrible pointer/long punning in b.c by a legitimate
880: union. should be safer on 64-bit machines and cleaner everywhere.
881: (thanks to nelson beebe for pointing out some of these problems.)
1.1 tholo 882:
1.7 millert 883: replaced nested comments by #if 0...#endif in run.c, lib.c.
1.1 tholo 884:
1.7 millert 885: removed getsval, setsval, execute macros from run.c and lib.c.
886: machines are 100x faster than they were when these macros were
887: first used.
1.1 tholo 888:
1.7 millert 889: revised filenames: awk.g.y => awkgram.y, awk.lx.l => awklex.l,
890: y.tab.[ch] => ytab.[ch], lex.yy.c => lexyy.c, all in the aid of
891: portability to nameless systems.
1.1 tholo 892:
1.7 millert 893: "make bundle" now includes yacc and lex output files for recipients
894: who don't have yacc or lex.
1.1 tholo 895:
1.7 millert 896: Aug 15, 1995:
897: initialized Cells in setsymtab more carefully; some fields
898: were not set. (thanks to purify, all of whose complaints i
899: think i now understand.)
1.1 tholo 900:
1.7 millert 901: fixed at least one error in gsub that looked at -1-th element
902: of an array when substituting for a null match (e.g., $).
1.1 tholo 903:
1.7 millert 904: delete arrayname is now legal; it clears the elements but leaves
905: the array, which may not be the right behavior.
1.1 tholo 906:
1.7 millert 907: modified makefile: my current make can't cope with the test used
908: to avoid unnecessary yacc invocations.
1.1 tholo 909:
1.7 millert 910: Jul 17, 1995:
911: added dynamically growing strings to awk.lx.l and b.c
912: to permit regular expressions to be much bigger.
913: the state arrays can still overflow.
1.1 tholo 914:
1.7 millert 915: Aug 24, 1994:
916: detect duplicate arguments in function definitions (mdm).
1.1 tholo 917:
1.7 millert 918: May 11, 1994:
919: trivial fix to printf to limit string size in sub().
1.1 tholo 920:
1.7 millert 921: Apr 22, 1994:
922: fixed yet another subtle self-assignment problem:
923: $1 = $2; $1 = $1 clobbered $1.
1.1 tholo 924:
1.7 millert 925: Regression tests now use private echo, to avoid quoting problems.
1.1 tholo 926:
1.7 millert 927: Feb 2, 1994:
928: changed error() to print line number as %d, not %g.
1.1 tholo 929:
1.7 millert 930: Jul 23, 1993:
931: cosmetic changes: increased sizes of some arrays,
932: reworded some error messages.
1.1 tholo 933:
1.7 millert 934: added CONVFMT as in posix (just replaced OFMT in getsval)
1.1 tholo 935:
1.7 millert 936: FILENAME is now "" until the first thing that causes a file
937: to be opened.
1.1 tholo 938:
1.7 millert 939: Nov 28, 1992:
940: deleted yyunput and yyoutput from proto.h;
941: different versions of lex give these different declarations.
1.1 tholo 942:
1.7 millert 943: May 31, 1992:
944: added -mr N and -mf N options: more record and fields.
945: these really ought to adjust automatically.
1.1 tholo 946:
1.7 millert 947: cleaned up some error messages; "out of space" now means
948: malloc returned NULL in all cases.
1.1 tholo 949:
1.7 millert 950: changed rehash so that if it runs out, it just returns;
951: things will continue to run slow, but maybe a bit longer.
1.1 tholo 952:
1.7 millert 953: Apr 24, 1992:
954: remove redundant close of stdin when using -f -.
1.1 tholo 955:
1.7 millert 956: got rid of core dump with -d; awk -d just prints date.
1.1 tholo 957:
1.7 millert 958: Apr 12, 1992:
959: added explicit check for /dev/std(in,out,err) in redirection.
960: unlike gawk, no /dev/fd/n yet.
1.1 tholo 961:
1.7 millert 962: added (file/pipe) builtin. hard to test satisfactorily.
963: not posix.
1.1 tholo 964:
1.7 millert 965: Feb 20, 1992:
966: recompile after abortive changes; should be unchanged.
1.1 tholo 967:
1.7 millert 968: Dec 2, 1991:
969: die-casting time: converted to ansi C, installed that.
1.1 tholo 970:
1.7 millert 971: Nov 30, 1991:
972: fixed storage leak in freefa, failing to recover [N]CCL.
973: thanks to Bill Jones (jones@cs.usask.ca)
1.1 tholo 974:
1.7 millert 975: Nov 19, 1991:
976: use RAND_MAX instead of literal in builtin().
1.1 tholo 977:
1.7 millert 978: Nov 12, 1991:
979: cranked up some fixed-size arrays in b.c, and added a test for
980: overflow in penter. thanks to mark larsen.
1.1 tholo 981:
1.7 millert 982: Sep 24, 1991:
983: increased buffer in gsub. a very crude fix to a general problem.
984: and again on Sep 26.
1.1 tholo 985:
1.7 millert 986: Aug 18, 1991:
987: enforce variable name syntax for commandline variables: has to
988: start with letter or _.
1.1 tholo 989:
1.7 millert 990: Jul 27, 1991:
991: allow newline after ; in for statements.
1.1 tholo 992:
1.7 millert 993: Jul 21, 1991:
994: fixed so that in self-assignment like $1=$1, side effects
995: like recomputing $0 take place. (this is getting subtle.)
1.1 tholo 996:
1.7 millert 997: Jun 30, 1991:
998: better test for detecting too-long output record.
1.1 tholo 999:
1.7 millert 1000: Jun 2, 1991:
1001: better defense against very long printf strings.
1002: made break and continue illegal outside of loops.
1.1 tholo 1003:
1.7 millert 1004: May 13, 1991:
1005: removed extra arg on gettemp, tempfree. minor error message rewording.
1.1 tholo 1006:
1007: May 6, 1991:
1008: fixed silly bug in hex parsing in hexstr().
1009: removed an apparently unnecessary test in isnumber().
1010: warn about weird printf conversions.
1011: fixed unchecked array overwrite in relex().
1012:
1013: changed for (i in array) to access elements in sorted order.
1014: then unchanged it -- it really does run slower in too many cases.
1015: left the code in place, commented out.
1016:
1.7 millert 1017: Feb 10, 1991:
1018: check error status on all writes, to avoid banging on full disks.
1.1 tholo 1019:
1.7 millert 1020: Jan 28, 1991:
1021: awk -f - reads the program from stdin.
1.1 tholo 1022:
1.7 millert 1023: Jan 11, 1991:
1024: failed to set numeric state on $0 in cmd|getline context in run.c.
1.1 tholo 1025:
1.7 millert 1026: Nov 2, 1990:
1027: fixed sleazy test for integrality in getsval; use modf.
1.1 tholo 1028:
1.7 millert 1029: Oct 29, 1990:
1030: fixed sleazy buggy code in lib.c that looked (incorrectly) for
1031: too long input lines.
1.1 tholo 1032:
1.7 millert 1033: Oct 14, 1990:
1034: fixed the bug on p. 198 in which it couldn't deduce that an
1035: argument was an array in some contexts. replaced the error
1036: message in intest() by code that damn well makes it an array.
1.1 tholo 1037:
1.7 millert 1038: Oct 8, 1990:
1039: fixed horrible bug: types and values were not preserved in
1040: some kinds of self-assignment. (in assign().)
1.1 tholo 1041:
1.7 millert 1042: Aug 24, 1990:
1043: changed NCHARS to 256 to handle 8-bit characters in strings
1044: presented to match(), etc.
1.1 tholo 1045:
1.7 millert 1046: Jun 26, 1990:
1047: changed struct rrow (awk.h) to use long instead of int for lval,
1048: since cfoll() stores a pointer in it. now works better when int's
1049: are smaller than pointers!
1.1 tholo 1050:
1.7 millert 1051: May 6, 1990:
1052: AVA fixed the grammar so that ! is uniformly of the same precedence as
1053: unary + and -. This renders illegal some constructs like !x=y, which
1054: now has to be parenthesized as !(x=y), and makes others work properly:
1055: !x+y is (!x)+y, and x!y is x !y, not two pattern-action statements.
1056: (These problems were pointed out by Bob Lenk of Posix.)
1.1 tholo 1057:
1.7 millert 1058: Added \x to regular expressions (already in strings).
1059: Limited octal to octal digits; \8 and \9 are not octal.
1060: Centralized the code for parsing escapes in regular expressions.
1061: Added a bunch of tests to T.re and T.sub to verify some of this.
1.1 tholo 1062:
1.7 millert 1063: Feb 9, 1990:
1064: fixed null pointer dereference bug in main.c: -F[nothing]. sigh.
1.1 tholo 1065:
1.7 millert 1066: restored srand behavior: it returns the current seed.
1.1 tholo 1067:
1.7 millert 1068: Jan 18, 1990:
1069: srand now returns previous seed value (0 to start).
1.1 tholo 1070:
1.7 millert 1071: Jan 5, 1990:
1072: fix potential problem in tran.c -- something was freed,
1073: then used in freesymtab.
1.1 tholo 1074:
1.7 millert 1075: Oct 18, 1989:
1076: another try to get the max number of open files set with
1077: relatively machine-independent code.
1.1 tholo 1078:
1.7 millert 1079: small fix to input() in case of multiple reads after EOF.
1.1 tholo 1080:
1.7 millert 1081: Oct 11, 1989:
1082: FILENAME is now defined in the BEGIN block -- too many old
1083: programs broke.
1.1 tholo 1084:
1.7 millert 1085: "-" means stdin in getline as well as on the commandline.
1.1 tholo 1086:
1.7 millert 1087: added a bunch of casts to the code to tell the truth about
1088: char * vs. unsigned char *, a right royal pain. added a
1089: setlocale call to the front of main, though probably no one
1090: has it usefully implemented yet.
1.1 tholo 1091:
1.7 millert 1092: Aug 24, 1989:
1093: removed redundant relational tests against nullnode if parse
1094: tree already had a relational at that point.
1.1 tholo 1095:
1.7 millert 1096: Aug 11, 1989:
1097: fixed bug: commandline variable assignment has to look like
1098: var=something. (consider the man page for =, in file =.1)
1.1 tholo 1099:
1.7 millert 1100: changed number of arguments to functions to static arrays
1101: to avoid repeated malloc calls.
1.1 tholo 1102:
1.7 millert 1103: Aug 2, 1989:
1104: restored -F (space) separator
1.1 tholo 1105:
1.7 millert 1106: Jul 30, 1989:
1107: added -v x=1 y=2 ... for immediate commandline variable assignment;
1108: done before the BEGIN block for sure. they have to precede the
1109: program if the program is on the commandline.
1110: Modified Aug 2 to require a separate -v for each assignment.
1.1 tholo 1111:
1.7 millert 1112: Jul 10, 1989:
1113: fixed ref-thru-zero bug in environment code in tran.c
1.1 tholo 1114:
1.7 millert 1115: Jun 23, 1989:
1116: add newline to usage message.
1.1 tholo 1117:
1.7 millert 1118: Jun 14, 1989:
1119: added some missing ansi printf conversion letters: %i %X %E %G.
1120: no sensible meaning for h or L, so they may not do what one expects.
1.1 tholo 1121:
1.7 millert 1122: made %* conversions work.
1.1 tholo 1123:
1.7 millert 1124: changed x^y so that if n is a positive integer, it's done
1125: by explicit multiplication, thus achieving maximum accuracy.
1126: (this should be done by pow() but it seems not to be locally.)
1127: done to x ^= y as well.
1.1 tholo 1128:
1.7 millert 1129: Jun 4, 1989:
1130: ENVIRON array contains environment: if shell variable V=thing,
1131: ENVIRON["V"] is "thing"
1.1 tholo 1132:
1.7 millert 1133: multiple -f arguments permitted. error reporting is naive.
1134: (they were permitted before, but only the last was used.)
1.1 tholo 1135:
1.7 millert 1136: fixed a really stupid botch in the debugging macro dprintf
1.1 tholo 1137:
1.7 millert 1138: fixed order of evaluation of commandline assignments to match
1139: what the book claims: an argument of the form x=e is evaluated
1140: at the time it would have been opened if it were a filename (p 63).
1141: this invalidates the suggested answer to ex 4-1 (p 195).
1.1 tholo 1142:
1.7 millert 1143: removed some code that permitted -F (space) fieldseparator,
1144: since it didn't quite work right anyway. (restored aug 2)
1.1 tholo 1145:
1.7 millert 1146: Apr 27, 1989:
1147: Line number now accumulated correctly for comment lines.
1.1 tholo 1148:
1.7 millert 1149: Apr 26, 1989:
1150: Debugging output now includes a version date,
1151: if one compiles it into the source each time.
1.1 tholo 1152:
1.7 millert 1153: Apr 9, 1989:
1154: Changed grammar to prohibit constants as 3rd arg of sub and gsub;
1155: prevents class of overwriting-a-constant errors. (Last one?)
1156: This invalidates the "banana" example on page 43 of the book.
1.1 tholo 1157:
1.7 millert 1158: Added \a ("alert"), \v (vertical tab), \xhhh (hexadecimal),
1159: as in ANSI, for strings. Rescinded the sloppiness that permitted
1160: non-octal digits in \ooo. Warning: not all compilers and libraries
1161: will be able to deal with \x correctly.
1.1 tholo 1162:
1.7 millert 1163: Jan 9, 1989:
1164: Fixed bug that caused tempcell list to contain a duplicate.
1165: The fix is kludgy.
1.1 tholo 1166:
1.7 millert 1167: Dec 17, 1988:
1168: Catches some more commandline errors in main.
1169: Removed redundant decl of modf in run.c (confuses some compilers).
1170: Warning: there's no single declaration of malloc, etc., in awk.h
1171: that seems to satisfy all compilers.
1.1 tholo 1172:
1.7 millert 1173: Dec 7, 1988:
1174: Added a bit of code to error printing to avoid printing nulls.
1175: (Not clear that it actually would.)
1.1 tholo 1176:
1.7 millert 1177: Nov 27, 1988:
1178: With fear and trembling, modified the grammar to permit
1179: multiple pattern-action statements on one line without
1180: an explicit separator. By definition, this capitulation
1181: to the ghost of ancient implementations remains undefined
1182: and thus subject to change without notice or apology.
1183: DO NOT COUNT ON IT.
1.1 tholo 1184:
1.7 millert 1185: Oct 30, 1988:
1186: Fixed bug in call() that failed to recover storage.
1.1 tholo 1187:
1.7 millert 1188: A warning is now generated if there are more arguments
1189: in the call than in the definition (in lieu of fixing
1190: another storage leak).
1.2 millert 1191:
1.7 millert 1192: Oct 20, 1988:
1193: Fixed %c: if expr is numeric, use numeric value;
1194: otherwise print 1st char of string value. still
1195: doesn't work if the value is 0 -- won't print \0.
1.2 millert 1196:
1.7 millert 1197: Added a few more checks for running out of malloc.
1.2 millert 1198:
1.7 millert 1199: Oct 12, 1988:
1200: Fixed bug in call() that freed local arrays twice.
1.5 kstailey 1201:
1.7 millert 1202: Fixed to handle deletion of non-existent array right;
1203: complains about attempt to delete non-array element.
1.6 kstailey 1204:
1.7 millert 1205: Sep 30, 1988:
1206: Now guarantees to evaluate all arguments of built-in
1207: functions, as in C; the appearance is that arguments
1208: are evaluated before the function is called. Places
1209: affected are sub (gsub was ok), substr, printf, and
1210: all the built-in arithmetic functions in bltin().
1211: A warning is generated if a bltin() is called with
1212: the wrong number of arguments.
1.6 kstailey 1213:
1.7 millert 1214: This requires changing makeprof on p167 of the book.
1.6 kstailey 1215:
1.7 millert 1216: Aug 23, 1988:
1217: setting FILENAME in BEGIN caused core dump, apparently
1218: because it was freeing space not allocated by malloc.
1.6 kstailey 1219:
1.7 millert 1220: July 24, 1988:
1221: fixed egregious error in toupper/tolower functions.
1222: still subject to rescinding, however.
1.6 kstailey 1223:
1.7 millert 1224: July 2, 1988:
1225: flush stdout before opening file or pipe
1.6 kstailey 1226:
1.7 millert 1227: July 2, 1988:
1228: performance bug in b.c/cgoto(): not freeing some sets of states.
1229: partial fix only right now, and the number of states increased
1230: to make it less obvious.
1.6 kstailey 1231:
1.7 millert 1232: June 1, 1988:
1233: check error status on close
1.6 kstailey 1234:
1.7 millert 1235: May 28, 1988:
1236: srand returns seed value it's using.
1237: see 1/18/90
1.6 kstailey 1238:
1.7 millert 1239: May 22, 1988:
1240: Removed limit on depth of function calls.
1.6 kstailey 1241:
1.7 millert 1242: May 10, 1988:
1243: Fixed lib.c to permit _ in commandline variable names.
1.6 kstailey 1244:
1.7 millert 1245: Mar 25, 1988:
1246: main.c fixed to recognize -- as terminator of command-
1247: line options. Illegal options flagged.
1248: Error reporting slightly cleaned up.
1.6 kstailey 1249:
1.7 millert 1250: Dec 2, 1987:
1251: Newer C compilers apply a strict scope rule to extern
1252: declarations within functions. Two extern declarations in
1253: lib.c and tran.c have been moved to obviate this problem.
1.6 kstailey 1254:
1.7 millert 1255: Oct xx, 1987:
1256: Reluctantly added toupper and tolower functions.
1257: Subject to rescinding without notice.
1.6 kstailey 1258:
1.7 millert 1259: Sep 17, 1987:
1260: Error-message printer had printf(s) instead of
1261: printf("%s",s); got core dumps when the message
1262: included a %.
1.6 kstailey 1263:
1.7 millert 1264: Sep 12, 1987:
1265: Very long printf strings caused core dump;
1266: fixed aprintf, asprintf, format to catch them.
1267: Can still get a core dump in printf itself.
1.6 kstailey 1268:
1269: