Annotation of src/usr.bin/awk/FIXES, Revision 1.33
1.33 ! millert 1: /* $OpenBSD: FIXES,v 1.32 2020/06/10 21:05:50 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.
1.33 ! millert 28:
! 29: June 5, 2020:
! 30: In fldbld(), make sure that inputFS is set before trying to
! 31: use it. Thanks to Steffen Nurpmeso <steffen@sdaoden.eu>
! 32: for the report.
! 33:
! 34: May 5, 2020:
! 35: Fix checks for compilers that can handle noreturn. Thanks to
! 36: GitHub user enh-google for pointing it out. Closes Issue #79.
! 37:
! 38: April 16, 2020:
! 39: Handle old compilers that don't support C11 (for noreturn).
! 40: Thanks to Arnold Robbins.
! 41:
! 42: April 5, 2020:
! 43: Use <stdnoreturn.h> and noreturn instead of GCC attributes.
! 44: Thanks to GitHub user awkfan77. Closes PR #77.
1.30 millert 45:
1.32 millert 46: February 28, 2020:
47: More cleanups from Christos Zoulas: notably backslash continuation
48: inside strings removes the newline and a fix for RS = "^a".
49: Fix for address sanitizer-found problem. Thanks to GitHub user
50: enh-google.
51:
52: February 19, 2020:
53: More small cleanups from Christos Zoulas.
54:
55: February 18, 2020:
56: Additional cleanups from Christos Zoulas. It's no longer necessary
57: to use the -y flag to bison.
58:
59: February 6, 2020:
60: Additional small cleanups from Christos Zoulas. awk is now
61: a little more robust about reporting I/O errors upon exit.
62:
1.31 millert 63: January 31, 2020:
64: Merge PR #70, which avoids use of variable length arrays. Thanks
65: to GitHub user michaelforney. Fix issue #60 ({0} in interval
1.32 millert 66: expressions doesn't work). Also get all tests working again.
67: Thanks to Arnold Robbins.
1.31 millert 68:
69: January 24, 2020:
70: A number of small cleanups from Christos Zoulas. Add the close
71: on exec flag to files/pipes opened for redirection; courtesy of
72: Arnold Robbins.
73:
74: January 19, 2020:
75: If POSIXLY_CORRECT is set in the environment, then sub and gsub
76: use POSIX rules for multiple backslashes. This fixes Issue #66,
77: while maintaining backwards compatibility.
78:
79: January 9, 2020:
80: Input/output errors on closing files are now fatal instead of
81: mere warnings. Thanks to Martijn Dekker <martijn@inlv.org>.
82:
1.30 millert 83: January 5, 2020:
84: Fix a bug in the concatentation of two string constants into
85: one done in the grammar. Fixes GitHub issue #61. Thanks
86: to GitHub user awkfan77 for pointing out the direction for
87: the fix. New test T.concat added to the test suite.
88: Fix a few memory leaks reported by valgrind, as well.
89:
90: December 27, 2019:
91: Fix a bug whereby a{0,3} could match four a's. Thanks to
92: "Anonymous AWK fan" for the report.
93:
94: December 11, 2019:
95: Further printf-related fixes for 32 bit systems.
96: Thanks again to Christos Zoulas.
97:
98: December 8, 2019:
99: Fix the return value of sprintf("%d") on 32 bit systems.
100: Thanks to Jim Lowe for the report and to Christos Zoulas
101: for the fix.
1.29 millert 102:
103: November 10, 2019:
104: Convert a number of Boolean integer variables into
105: actual bools. Convert compile_time variable into an
106: enum and simplify some of the related code. Thanks
107: to Arnold Robbins.
108:
109: November 8, 2019:
110: Fix from Ori Bernstein to get UTF-8 characters instead of
111: bytes when FS = "". This is currently the only bit of
112: the One True Awk that understands multibyte characters.
113: From Arnold Robbins, apply some cleanups in the test suite.
1.28 millert 114:
115: October 25, 2019:
116: More fixes and cleanups from NetBSD, courtesy of Christos
117: Zoulas. Merges PRs 54 and 55.
118:
119: October 24, 2019:
120: Import second round of code cleanups from NetBSD. Much thanks
1.31 millert 121: to Christos Zoulas (GitHub user zoulasc). Merges PR 53.
1.28 millert 122: Add an optimization for string concatenation, also from
123: Christos.
1.27 millert 124:
125: October 17, 2019:
126: Import code cleanups from NetBSD. Much thanks to Christos
1.31 millert 127: Zoulas (GitHub user zoulasc). Merges PR 51.
1.26 millert 128:
129: October 6, 2019:
130: Import code from NetBSD awk that implements RS as a regular
131: expression.
1.23 millert 132:
1.25 millert 133: September 10, 2019:
134: Fixes for various array / memory overruns found via gcc's
1.31 millert 135: -fsanitize=unknown. Thanks to Alexander Richardson (GitHub
1.25 millert 136: user arichardson). Merges PRs 47 and 48.
137:
138: July 28, 2019:
139: Import grammar optimization from NetBSD: Two string constants
140: concatenated together get turned into a single string.
141:
142: July 26, 2019:
143: Support POSIX-specified C-style escape sequences "\a" (alarm)
144: and "\v" (vertical tab) in command line arguments and regular
145: expressions, further to the support for them in strings added on
146: Apr 9, 1989. These now no longer match as literal "a" and "v"
147: characters (as they don't on other awk implementations).
148: Thanks to Martijn Dekker.
149:
150: July 17, 2019:
151: Pull in a number of code cleanups and minor fixes from
152: Warner Losh's bsd-ota branch. The only user visible change
153: is the use of random(3) as the random number generator.
154: Thanks to Warner Losh for collecting all these fixes in
155: one easy place to get them from.
156:
157: July 16, 2019:
158: Fix field splitting to use FS value as of the time a record
159: was read or assigned to. Thanks to GitHub user Cody Mello (melloc)
160: for the fix. (Merged from his branch, via PR #42.) Updated
161: testdir/T.split per said PR as well.
162:
163: June 24, 2019:
164: Extract awktest.tar into testdir directory. Add some very
165: simple mechanics to the makefile for running the tests and
166: for cleaning up. No changes to awk itself.
167:
1.24 millert 168: June 17, 2019:
169: Disallow deleting SYMTAB and its elements, which creates
170: use-after-free bugs. Thanks to GitHub user Cody Mello (melloc)
171: for the fix. (Merged from PR #43.)
172:
173: June 5, 2019:
174: Allow unmatched right parenthesis in a regular expression to
175: be treated literally. Fixes Issue #40. Thanks to GitHub user
176: Warner Losh (bsdimp) for the report. Thanks to Arnold Robbins
177: for the fix.
178:
179: May 29,2019:
180: Fix check for command line arguments to no longer require that
181: first character after '=' not be another '='. Reverts change of
182: August 11, 1989. Thanks to GitHub user Jamie Landeg Jones for
183: pointing out the issue; from Issue #38.
184:
185: Apr 7, 2019:
186: Update awktest.tar(p.50) to use modern options to sort. Needed
187: for Android development. Thanks to GitHub user mohd-akram (Mohamed
188: Akram). From Issue #33.
189:
190: Mar 12, 2019:
191: Added very simplistic support for cross-compiling in the
192: makefile. We are NOT going to go in the direction of the
193: autotools, though. Thanks to GitHub user nee-san for
194: the basic change. (Merged from PR #34.)
195:
1.23 millert 196: Mar 5, 2019:
197: Added support for POSIX-standard interval expressions (a.k.a.
198: bounds, a.k.a. repetition expressions) in regular expressions,
199: backported (via NetBSD) from Apple awk-24 (20070501).
200: Thanks to Martijn Dekker <martijn@inlv.org> for the port.
201: (Merged from PR #30.)
202:
203: Mar 3, 2019:
204: Merge PRs as follows:
205: #12: Avoid undefined behaviour when using ctype(3) functions in
206: relex(). Thanks to GitHub user iamleot.
207: #31: Make getline handle numeric strings, and update FIXES. Thanks
1.24 millert 208: to GitHub user arnoldrobbins.
1.23 millert 209: #32: maketab: support build systems with read-only source. Thanks
210: to GitHub user enh.
1.22 millert 211:
212: Jan 25, 2019:
213: Make getline handle numeric strings properly in all cases.
214: (Thanks, Arnold.)
215:
216: Jan 21, 2019:
217: Merged a number of small fixes from GitHub pull requests.
218: Thanks to GitHub users Arnold Robbins (arnoldrobbins),
219: Cody Mello (melloc) and Christoph Junghans (junghans).
220: PR numbers: 13-21, 23, 24, 27.
1.21 millert 221:
222: Oct 25, 2018:
223: Added test in maketab.c to prevent generating a proctab entry
224: for YYSTYPE_IS_DEFINED. It was harmless but some gcc settings
225: generated a warning message. Thanks to Nan Xiao for report.
1.20 millert 226:
227: Aug 27, 2018:
228: Disallow '$' in printf formats; arguments evaluated in order
229: and printed in order.
230:
231: Added some casts to silence warnings on debugging printfs.
232: (Thanks, Arnold.)
1.18 millert 233:
1.19 millert 234: Aug 23, 2018:
235: A long list of fixes courtesy of Arnold Robbins,
236: to whom profound thanks.
237:
238: 1. ofs-rebuild: OFS value used to rebuild the record was incorrect.
239: Fixed August 19, 2014. Revised fix August 2018.
240:
241: 2. system-status: Instead of a floating-point division by 256, use
242: the wait(2) macros to create a reasonable exit status.
243: Fixed March 12, 2016.
244:
245: 3. space: Use provided xisblank() function instead of ispace() for
246: matching [[:blank:]].
247:
248: 4. a-format: Add POSIX standard %a and %A to supported formats. Check
249: at runtime that this format is available.
250:
251: 5. decr-NF: Decrementing NF did not change $0. This is a decades-old
252: bug. There are interactions with the old and new value of OFS as well.
253: Most of the fix came from the NetBSD awk.
254:
255: 6. string-conv: String conversions of scalars were sticky. Once a
256: conversion to string happened, even with OFMT, that value was used until
257: a new numeric value was assigned, even if OFMT differed from CONVFMT,
258: and also if CONVFMT changed.
259:
260: 7. unary-plus: Unary plus on a string constant returned the string.
261: Instead, it should convert the value to numeric and give that value.
262:
263: Also added Arnold's tests for these to awktest.tar as T.arnold.
264:
265: Aug 15, 2018:
266: fixed mangled awktest.tar (thanks, Arnold), posted all
267: current (very minor) fixes to github / onetrueawk
268:
1.18 millert 269: Jun 7, 2018:
270: (yes, a long layoff)
271: Updated some broken tests (beebe.tar, T.lilly)
272: [thanks to Arnold Robbins]
273:
274: Mar 26, 2015:
275: buffer overflow in error reporting; thanks to tobias ulmer
276: and john-mark gurney for spotting it and the fix.
277:
278: Feb 4, 2013:
279: cleaned up a handful of tests that didn't seem to actually
280: test for correct behavior: T.latin1, T.gawk.
281:
282: Jan 5, 2013:
283: added ,NULL initializer to static Cells in run.c; not really
284: needed but cleaner. Thanks to Michael Bombardieri.
1.17 millert 285:
286: Dec 20, 2012:
287: fiddled makefile to get correct yacc and bison flags. pick yacc
288: (linux) or bison (mac) as necessary.
289:
290: added __attribute__((__noreturn__)) to a couple of lines in
291: proto.h, to silence someone's enthusiastic checker.
292:
293: fixed obscure call by value bug in split(a[1],a) reported on
294: 9fans. the management of temporary values is just a mess; i
295: took a shortcut by making an extra string copy. thanks
296: to paul patience and arnold robbins for passing it on and for
297: proposed patches.
298:
299: tiny fiddle in setfval to eliminate -0 results in T.expr, which
300: has irritated me for 20+ years.
1.16 millert 301:
302: Aug 10, 2011:
303: another fix to avoid core dump with delete(ARGV); again, many thanks
304: to ruslan ermilov.
305:
306: Aug 7, 2011:
307: split(s, a, //) now behaves the same as split(s, a, "")
308:
309: Jun 12, 2011:
310: /pat/, \n /pat/ {...} is now legal, though bad style to use.
311:
312: added checks to new -v code that permits -vnospace; thanks to
1.25 millert 313: ruslan ermilov for spotting this and providing the patch.
1.16 millert 314:
315: removed fixed limit on number of open files; thanks to aleksey
1.25 millert 316: cheusov and christos zoulos.
1.16 millert 317:
318: fixed day 1 bug that resurrected deleted elements of ARGV when
319: used as filenames (in lib.c).
320:
321: minor type fiddles to make gcc -Wall -pedantic happier (but not
322: totally so); turned on -fno-strict-aliasing in makefile.
323:
324: May 6, 2011:
325: added #ifdef for isblank.
326: now allows -ffoo as well as -f foo arguments.
327: (thanks, ruslan)
328:
329: May 1, 2011:
330: after advice from todd miller, kevin lo, ruslan ermilov,
331: and arnold robbins, changed srand() to return the previous
332: seed (which is 1 on the first call of srand). the seed is
333: an Awkfloat internally though converted to unsigned int to
1.25 millert 334: pass to the library srand(). thanks, everyone.
1.16 millert 335:
336: fixed a subtle (and i hope low-probability) overflow error
1.25 millert 337: in fldbld, by adding space for one extra \0. thanks to
1.16 millert 338: robert bassett for spotting this one and providing a fix.
339:
340: removed the files related to compilation on windows. i no
341: longer have anything like a current windows environment, so
342: i can't test any of it.
1.15 millert 343:
344: May 23, 2010:
345: fixed long-standing overflow bug in run.c; many thanks to
346: nelson beebe for spotting it and providing the fix.
347:
348: fixed bug that didn't parse -vd=1 properly; thanks to santiago
349: vila for spotting it.
350:
351: Feb 8, 2010:
352: i give up. replaced isblank with isspace in b.c; there are
353: no consistent header files.
354:
355: Nov 26, 2009:
356: fixed a long-standing issue with when FS takes effect. a
357: change to FS is now noticed immediately for subsequent splits.
358:
359: changed the name getline() to awkgetline() to avoid yet another
360: name conflict somewhere.
361:
362: Feb 11, 2009:
363: temporarily for now defined HAS_ISBLANK, since that seems to
364: be the best way through the thicket. isblank arrived in C99,
365: but seems to be arriving at different systems at different
366: times.
367:
368: Oct 8, 2008:
369: fixed typo in b.c that set tmpvec wrongly. no one had ever
370: run into the problem, apparently. thanks to alistair crooks.
1.14 millert 371:
372: Oct 23, 2007:
373: minor fix in lib.c: increase inputFS to 100, change malloc
1.25 millert 374: for fields to n+1.
1.14 millert 375:
376: fixed memory fault caused by out of order test in setsval.
377:
378: thanks to david o'brien, freebsd, for both fixes.
379:
380: May 1, 2007:
381: fiddle in makefile to fix for BSD make; thanks to igor sobrado.
382:
383: Mar 31, 2007:
384: fixed some null pointer refs calling adjbuf.
385:
386: Feb 21, 2007:
387: fixed a bug in matching the null RE in sub and gsub. thanks to al aho
388: who actually did the fix (in b.c), and to wolfgang seeberg for finding
389: it and providing a very compact test case.
390:
391: fixed quotation in b.c; thanks to Hal Pratt and the Princeton Dante
392: Project.
393:
394: removed some no-effect asserts in run.c.
395:
396: fiddled maketab.c to not complain about bison-generated values.
397:
398: removed the obsolete -V argument; fixed --version to print the
399: version and exit.
400:
401: fixed wording and an outright error in the usage message; thanks to igor
402: sobrado and jason mcintyre.
403:
404: fixed a bug in -d that caused core dump if no program followed.
405:
406: Jan 1, 2007:
407: dropped mac.code from makefile; there are few non-MacOSX
408: mac's these days.
409:
410: Jan 17, 2006:
411: system() not flagged as unsafe in the unadvertised -safe option.
412: found it while enhancing tests before shipping the ;login: article.
413: practice what you preach.
414:
415: removed the 9-years-obsolete -mr and -mf flags.
416:
417: added -version and --version options.
418:
419: core dump on linux with BEGIN {nextfile}, now fixed.
420:
1.25 millert 421: removed some #ifdef's in run.c and lex.c that appear to no
1.14 millert 422: longer be necessary.
423:
424: Apr 24, 2005:
425: modified lib.c so that values of $0 et al are preserved in the END
426: block, apparently as required by posix. thanks to havard eidnes
427: for the report and code.
428:
429: Jan 14, 2005:
430: fixed infinite loop in parsing, originally found by brian tsang.
431: thanks to arnold robbins for a suggestion that started me
432: rethinking it.
433:
434: Dec 31, 2004:
1.25 millert 435: prevent overflow of -f array in main, head off potential error in
436: call of SYNTAX(), test malloc return in lib.c, all with thanks to
1.14 millert 437: todd miller.
1.13 millert 438:
439: Dec 22, 2004:
440: cranked up size of NCHARS; coverity thinks it can be overrun with
441: smaller size, and i think that's right. added some assertions to b.c
442: to catch places where it might overrun. the RE code is still fragile.
443:
444: Dec 5, 2004:
445: fixed a couple of overflow problems with ridiculous field numbers:
446: e.g., print $(2^32-1). thanks to ruslan ermilov, giorgos keramidas
447: and david o'brien at freebsd.org for patches. this really should
448: be re-done from scratch.
449:
450: Nov 21, 2004:
451: fixed another 25-year-old RE bug, in split. it's another failure
452: to (re-)initialize. thanks to steve fisher for spotting this and
453: providing a good test case.
454:
455: Nov 22, 2003:
456: fixed a bug in regular expressions that dates (so help me) from 1977;
457: it's been there from the beginning. an anchored longest match that
458: was longer than the number of states triggered a failure to initialize
459: the machine properly. many thanks to moinak ghosh for not only finding
460: this one but for providing a fix, in some of the most mysterious
461: code known to man.
462:
463: fixed a storage leak in call() that appears to have been there since
1.25 millert 464: 1983 or so -- a function without an explicit return that assigns a
465: string to a parameter leaked a Cell. thanks to moinak ghosh for
1.13 millert 466: spotting this very subtle one.
467:
468: Jul 31, 2003:
469: fixed, thanks to andrey chernov and ruslan ermilov, a bug in lex.c
470: that mis-handled the character 255 in input. (it was being compared
471: to EOF with a signed comparison.)
472:
473: Jul 29, 2003:
474: fixed (i think) the long-standing botch that included the beginning of
475: line state ^ for RE's in the set of valid characters; this led to a
476: variety of odd problems, including failure to properly match certain
477: regular expressions in non-US locales. thanks to ruslan for keeping
478: at this one.
479:
480: Jul 28, 2003:
481: n-th try at getting internationalization right, with thanks to volker
482: kiefel, arnold robbins and ruslan ermilov for advice, though they
483: should not be blamed for the outcome. according to posix, "." is the
484: radix character in programs and command line arguments regardless of
485: the locale; otherwise, the locale should prevail for input and output
486: of numbers. so it's intended to work that way.
1.25 millert 487:
1.13 millert 488: i have rescinded the attempt to use strcoll in expanding shorthands in
489: regular expressions (cclenter). its properties are much too
490: surprising; for example [a-c] matches aAbBc in locale en_US but abBcC
491: in locale fr_CA. i can see how this might arise by implementation
492: but i cannot explain it to a human user. (this behavior can be seen
493: in gawk as well; we're leaning on the same library.)
494:
495: the issue appears to be that strcoll is meant for sorting, where
496: merging upper and lower case may make sense (though note that unix
497: sort does not do this by default either). it is not appropriate
498: for regular expressions, where the goal is to match specific
499: patterns of characters. in any case, the notations [:lower:], etc.,
500: are available in awk, and they are more likely to work correctly in
501: most locales.
502:
503: a moratorium is hereby declared on internationalization changes.
504: i apologize to friends and colleagues in other parts of the world.
505: i would truly like to get this "right", but i don't know what
506: that is, and i do not want to keep making changes until it's clear.
507:
508: Jul 4, 2003:
509: fixed bug that permitted non-terminated RE, as in "awk /x".
510:
511: Jun 1, 2003:
512: subtle change to split: if source is empty, number of elems
513: is always 0 and the array is not set.
514:
515: Mar 21, 2003:
516: added some parens to isblank, in another attempt to make things
517: internationally portable.
518:
519: Mar 14, 2003:
520: the internationalization changes, somewhat modified, are now
521: reinstated. in theory awk will now do character comparisons
522: and case conversions in national language, but "." will always
523: be the decimal point separator on input and output regardless
524: of national language. isblank(){} has an #ifndef.
525:
526: this no longer compiles on windows: LC_MESSAGES isn't defined
527: in vc6++.
528:
529: fixed subtle behavior in field and record splitting: if FS is
530: a single character and RS is not empty, \n is NOT a separator.
531: this tortuous reading is found in the awk book; behavior now
532: matches gawk and mawk.
1.11 millert 533:
534: Dec 13, 2002:
535: for the moment, the internationalization changes of nov 29 are
536: rolled back -- programs like x = 1.2 don't work in some locales,
537: because the parser is expecting x = 1,2. until i understand this
538: better, this will have to wait.
539:
540: Nov 29, 2002:
541: modified b.c (with tiny changes in main and run) to support
542: locales, using strcoll and iswhatever tests for posix character
543: classes. thanks to ruslan ermilov (ru@freebsd.org) for code.
544: the function isblank doesn't seem to have propagated to any
545: header file near me, so it's there explicitly. not properly
546: tested on non-ascii character sets by me.
547:
548: Jun 28, 2002:
549: modified run/format() and tran/getsval() to do a slightly better
550: job on using OFMT for output from print and CONVFMT for other
1.25 millert 551: number->string conversions, as promised by posix and done by
1.11 millert 552: gawk and mawk. there are still places where it doesn't work
553: right if CONVFMT is changed; by then the STR attribute of the
554: variable has been irrevocably set. thanks to arnold robbins for
555: code and examples.
556:
557: fixed subtle bug in format that could get core dump. thanks to
558: Jaromir Dolecek <jdolecek@NetBSD.org> for finding and fixing.
559: minor cleanup in run.c / format() at the same time.
560:
561: added some tests for null pointers to debugging printf's, which
562: were never intended for external consumption. thanks to dave
563: kerns (dkerns@lucent.com) for pointing this out.
564:
565: GNU compatibility: an empty regexp matches anything (thanks to
566: dag-erling smorgrav, des@ofug.org). subject to reversion if
567: this does more harm than good.
568:
569: pervasive small changes to make things more const-correct, as
570: reported by gcc's -Wwrite-strings. as it says in the gcc manual,
571: this may be more nuisance than useful. provoked by a suggestion
572: and code from arnaud desitter, arnaud@nimbus.geog.ox.ac.uk
573:
574: minor documentation changes to note that this now compiles out
575: of the box on Mac OS X.
576:
577: Feb 10, 2002:
578: changed types in posix chars structure to quiet solaris cc.
579:
580: Jan 1, 2002:
581: fflush() or fflush("") flushes all files and pipes.
582:
1.25 millert 583: length(arrayname) returns number of elements; thanks to
1.11 millert 584: arnold robbins for suggestion.
585:
586: added a makefile.win to make it easier to build on windows.
587: based on dan allen's buildwin.bat.
588:
589: Nov 16, 2001:
590: added support for posix character class names like [:digit:],
591: which are not exactly shorter than [0-9] and perhaps no more
592: portable. thanks to dag-erling smorgrav for code.
593:
594: Feb 16, 2001:
595: removed -m option; no longer needed, and it was actually
596: broken (noted thanks to volker kiefel).
597:
598: Feb 10, 2001:
599: fixed an appalling bug in gettok: any sequence of digits, +,-, E, e,
600: and period was accepted as a valid number if it started with a period.
601: this would never have happened with the lex version.
602:
603: other 1-character botches, now fixed, include a bare $ and a
604: bare " at the end of the input.
605:
606: Feb 7, 2001:
607: more (const char *) casts in b.c and tran.c to silence warnings.
1.8 millert 608:
1.10 millert 609: Nov 15, 2000:
610: fixed a bug introduced in august 1997 that caused expressions
611: like $f[1] to be syntax errors. thanks to arnold robbins for
612: noticing this and providing a fix.
613:
614: Oct 30, 2000:
615: fixed some nextfile bugs: not handling all cases. thanks to
616: arnold robbins for pointing this out. new regressions added.
617:
618: close() is now a function. it returns whatever the library
619: fclose returns, and -1 for closing a file or pipe that wasn't
620: opened.
621:
622: Sep 24, 2000:
623: permit \n explicitly in character classes; won't work right
624: if comes in as "[\n]" but ok as /[\n]/, because of multiple
625: processing of \'s. thanks to arnold robbins.
626:
627: July 5, 2000:
628: minor fiddles in tran.c to keep compilers happy about uschar.
629: thanks to norman wilson.
630:
631: May 25, 2000:
632: yet another attempt at making 8-bit input work, with another
1.25 millert 633: band-aid in b.c (member()), and some (uschar) casts to head
1.10 millert 634: off potential errors in subscripts (like isdigit). also
635: changed HAT to NCHARS-2. thanks again to santiago vila.
636:
637: changed maketab.c to ignore apparently out of range definitions
638: instead of halting; new freeBSD generates one. thanks to
639: jon snader <jsnader@ix.netcom.com> for pointing out the problem.
640:
641: May 2, 2000:
642: fixed an 8-bit problem in b.c by making several char*'s into
643: unsigned char*'s. not clear i have them all yet. thanks to
644: Santiago Vila <sanvila@unex.es> for the bug report.
645:
646: Apr 21, 2000:
647: finally found and fixed a memory leak in function call; it's
648: been there since functions were added ~1983. thanks to
649: jon bentley for the test case that found it.
650:
651: added test in envinit to catch environment "variables" with
1.12 otto 652: names beginning with '='; thanks to Berend Hasselman.
1.10 millert 653:
654: Jul 28, 1999:
655: added test in defn() to catch function foo(foo), which
656: otherwise recurses until core dump. thanks to arnold
657: robbins for noticing this.
658:
1.9 millert 659: Jun 20, 1999:
660: added *bp in gettok in lex.c; appears possible to exit function
661: without terminating the string. thanks to russ cox.
662:
663: Jun 2, 1999:
664: added function stdinit() to run to initialize files[] array,
665: in case stdin, etc., are not constants; some compilers care.
666:
667: May 10, 1999:
668: replaced the ERROR ... FATAL, etc., macros with functions
669: based on vprintf, to avoid problems caused by overrunning
670: fixed-size errbuf array. thanks to ralph corderoy for the
671: impetus, and for pointing out a string termination bug in
672: qstring as well.
673:
674: Apr 21, 1999:
675: fixed bug that caused occasional core dumps with commandline
676: variable with value ending in \. (thanks to nelson beebe for
677: the test case.)
678:
1.8 millert 679: Apr 16, 1999:
1.25 millert 680: with code kindly provided by Bruce Lilly, awk now parses
1.8 millert 681: /=/ and similar constructs more sensibly in more places.
1.10 millert 682: Bruce also provided some helpful test cases.
1.8 millert 683:
684: Apr 5, 1999:
685: changed true/false to True/False in run.c to make it
686: easier to compile with C++. Added some casts on malloc
687: and realloc to be honest about casts; ditto. changed
688: ltype int to long in struct rrow to reduce some 64-bit
689: complaints; other changes scattered throughout for the
690: same purpose. thanks to Nelson Beebe for these portability
691: improvements.
692:
693: removed some horrible pointer-int casting in b.c and elsewhere
694: by adding ptoi and itonp to localize the casts, which are
695: all benign. fixed one incipient bug that showed up on sgi
696: in 64-bit mode.
697:
698: reset lineno for new source file; include filename in error
699: message. also fixed line number error in continuation lines.
700: (thanks to Nelson Beebe for both of these.)
701:
702: Mar 24, 1999:
703: Nelson Beebe notes that irix 5.3 yacc dies with a bogus
704: error; use a newer version or switch to bison, since sgi
705: is unlikely to fix it.
1.1 tholo 706:
1.7 millert 707: Mar 5, 1999:
1.9 millert 708: changed isnumber to is_number to avoid the problem caused by
709: versions of ctype.h that include the name isnumber.
1.7 millert 710:
711: distribution now includes a script for building on a Mac,
712: thanks to Dan Allen.
713:
714: Feb 20, 1999:
715: fixed memory leaks in run.c (call) and tran.c (setfval).
716: thanks to Stephen Nutt for finding these and providing the fixes.
717:
718: Jan 13, 1999:
719: replaced srand argument by (unsigned int) in run.c;
720: avoids problem on Mac and potentially on Unix & Windows.
721: thanks to Dan Allen.
722:
723: added a few (int) casts to silence useless compiler warnings.
724: e.g., errorflag= in run.c jump().
725:
726: added proctab.c to the bundle outout; one less thing
727: to have to compile out of the box.
728:
729: added calls to _popen and _pclose to the win95 stub for
730: pipes (thanks to Steve Adams for this helpful suggestion).
731: seems to work, though properties are not well understood
732: by me, and it appears that under some circumstances the
733: pipe output is truncated. Be careful.
734:
735: Oct 19, 1998:
736: fixed a couple of bugs in getrec: could fail to update $0
1.25 millert 737: after a getline var; because inputFS wasn't initialized,
1.7 millert 738: could split $0 on every character, a misleading diversion.
739:
740: fixed caching bug in makedfa: LRU was actually removing
741: least often used.
742:
743: thanks to ross ridge for finding these, and for providing
744: great bug reports.
745:
746: May 12, 1998:
747: fixed potential bug in readrec: might fail to update record
748: pointer after growing. thanks to dan levy for spotting this
749: and suggesting the fix.
750:
751: Mar 12, 1998:
752: added -V to print version number and die.
1.19 millert 753:
754: [notify dave kerns, dkerns@dacsoup.ih.lucent.com]
1.7 millert 755:
756: Feb 11, 1998:
757: subtle silent bug in lex.c: if the program ended with a number
758: longer than 1 digit, part of the input would be pushed back and
759: parsed again because token buffer wasn't terminated right.
760: example: awk 'length($0) > 10'. blush. at least i found it
761: myself.
762:
763: Aug 31, 1997:
764: s/adelete/awkdelete/: SGI uses this in malloc.h.
765: thanks to nelson beebe for pointing this one out.
766:
767: Aug 21, 1997:
768: fixed some bugs in sub and gsub when replacement includes \\.
769: this is a dark, horrible corner, but at least now i believe that
770: the behavior is the same as gawk and the intended posix standard.
771: thanks to arnold robbins for advice here.
772:
773: Aug 9, 1997:
774: somewhat regretfully, replaced the ancient lex-based lexical
775: analyzer with one written in C. it's longer, generates less code,
776: and more portable; the old one depended too much on mysterious
777: properties of lex that were not preserved in other environments.
778: in theory these recognize the same language.
779:
780: now using strtod to test whether a string is a number, instead of
781: the convoluted original function. should be more portable and
782: reliable if strtod is implemented right.
783:
784: removed now-pointless optimization in makefile that tries to avoid
785: recompilation when awkgram.y is changed but symbols are not.
786:
787: removed most fixed-size arrays, though a handful remain, some
788: of which are unchecked. you have been warned.
789:
790: Aug 4, 1997:
791: with some trepidation, replaced the ancient code that managed
792: fields and $0 in fixed-size arrays with arrays that grow on
793: demand. there is still some tension between trying to make this
794: run fast and making it clean; not sure it's right yet.
795:
796: the ill-conceived -mr and -mf arguments are now useful only
797: for debugging. previous dynamic string code removed.
798:
799: numerous other minor cleanups along the way.
1.1 tholo 800:
1.7 millert 801: Jul 30, 1997:
802: using code provided by dan levy (to whom profuse thanks), replaced
803: fixed-size arrays and awkward kludges by a fairly uniform mechanism
804: to grow arrays as needed for printf, sub, gsub, etc.
1.1 tholo 805:
1.7 millert 806: Jul 23, 1997:
807: falling off the end of a function returns "" and 0, not 0.
808: thanks to arnold robbins.
1.1 tholo 809:
1.7 millert 810: Jun 17, 1997:
811: replaced several fixed-size arrays by dynamically-created ones
812: in run.c; added overflow tests to some previously unchecked cases.
813: getline, toupper, tolower.
1.1 tholo 814:
1.7 millert 815: getline code is still broken in that recursive calls may wind
816: up using the same space. [fixed later]
1.1 tholo 817:
1.7 millert 818: increased RECSIZE to 8192 to push problems further over the horizon.
1.1 tholo 819:
1.7 millert 820: added \r to \n as input line separator for programs, not data.
821: damn CRLFs.
1.1 tholo 822:
1.7 millert 823: modified format() to permit explicit printf("%c", 0) to include
824: a null byte in output. thanks to ken stailey for the fix.
1.1 tholo 825:
1.7 millert 826: added a "-safe" argument that disables file output (print >,
827: print >>), process creation (cmd|getline, print |, system), and
828: access to the environment (ENVIRON). this is a first approximation
829: to a "safe" version of awk, but don't rely on it too much. thanks
830: to joan feigenbaum and matt blaze for the inspiration long ago.
1.1 tholo 831:
1.7 millert 832: Jul 8, 1996:
833: fixed long-standing bug in sub, gsub(/a/, "\\\\&"); thanks to
834: ralph corderoy.
1.1 tholo 835:
1.7 millert 836: Jun 29, 1996:
837: fixed awful bug in new field splitting; didn't get all the places
838: where input was done.
1.1 tholo 839:
1.7 millert 840: Jun 28, 1996:
841: changed field-splitting to conform to posix definition: fields are
842: split using the value of FS at the time of input; it used to be
843: the value when the field or NF was first referred to, a much less
844: predictable definition. thanks to arnold robbins for encouragement
845: to do the right thing.
1.1 tholo 846:
1.7 millert 847: May 28, 1996:
848: fixed appalling but apparently unimportant bug in parsing octal
849: numbers in reg exprs.
1.1 tholo 850:
1.7 millert 851: explicit hex in reg exprs now limited to 2 chars: \xa, \xaa.
1.1 tholo 852:
1.7 millert 853: May 27, 1996:
854: cleaned up some declarations so gcc -Wall is now almost silent.
1.1 tholo 855:
1.7 millert 856: makefile now includes backup copies of ytab.c and lexyy.c in case
857: one makes before looking; it also avoids recreating lexyy.c unless
858: really needed.
1.1 tholo 859:
1.7 millert 860: s/aprintf/awkprint, s/asprintf/awksprintf/ to avoid some name clashes
861: with unwisely-written header files.
1.1 tholo 862:
1.7 millert 863: thanks to jeffrey friedl for several of these.
1.1 tholo 864:
1.7 millert 865: May 26, 1996:
866: an attempt to rationalize the (unsigned) char issue. almost all
867: instances of unsigned char have been removed; the handful of places
868: in b.c where chars are used as table indices have been hand-crafted.
869: added some latin-1 tests to the regression, but i'm not confident;
870: none of my compilers seem to care much. thanks to nelson beebe for
871: pointing out some others that do care.
1.1 tholo 872:
1.7 millert 873: May 2, 1996:
874: removed all register declarations.
1.1 tholo 875:
1.7 millert 876: enhanced split(), as in gawk, etc: split(s, a, "") splits s into
877: a[1]...a[length(s)] with each character a single element.
1.1 tholo 878:
1.7 millert 879: made the same changes for field-splitting if FS is "".
1.1 tholo 880:
1.7 millert 881: added nextfile, as in gawk: causes immediate advance to next
882: input file. (thanks to arnold robbins for inspiration and code).
1.1 tholo 883:
1.7 millert 884: small fixes to regexpr code: can now handle []], [[], and
1.25 millert 885: variants; [] is now a syntax error, rather than matching
1.7 millert 886: everything; [z-a] is now empty, not z. far from complete
887: or correct, however. (thanks to jeffrey friedl for pointing out
888: some awful behaviors.)
1.1 tholo 889:
1.7 millert 890: Apr 29, 1996:
1.12 otto 891: replaced uchar by uschar everywhere; apparently some compilers
1.7 millert 892: usurp this name and this causes conflicts.
1.1 tholo 893:
1.7 millert 894: fixed call to time in run.c (bltin); arg is time_t *.
1.1 tholo 895:
1.7 millert 896: replaced horrible pointer/long punning in b.c by a legitimate
897: union. should be safer on 64-bit machines and cleaner everywhere.
898: (thanks to nelson beebe for pointing out some of these problems.)
1.1 tholo 899:
1.7 millert 900: replaced nested comments by #if 0...#endif in run.c, lib.c.
1.1 tholo 901:
1.7 millert 902: removed getsval, setsval, execute macros from run.c and lib.c.
903: machines are 100x faster than they were when these macros were
904: first used.
1.1 tholo 905:
1.7 millert 906: revised filenames: awk.g.y => awkgram.y, awk.lx.l => awklex.l,
907: y.tab.[ch] => ytab.[ch], lex.yy.c => lexyy.c, all in the aid of
908: portability to nameless systems.
1.1 tholo 909:
1.7 millert 910: "make bundle" now includes yacc and lex output files for recipients
911: who don't have yacc or lex.
1.1 tholo 912:
1.7 millert 913: Aug 15, 1995:
914: initialized Cells in setsymtab more carefully; some fields
915: were not set. (thanks to purify, all of whose complaints i
916: think i now understand.)
1.1 tholo 917:
1.7 millert 918: fixed at least one error in gsub that looked at -1-th element
919: of an array when substituting for a null match (e.g., $).
1.1 tholo 920:
1.7 millert 921: delete arrayname is now legal; it clears the elements but leaves
922: the array, which may not be the right behavior.
1.1 tholo 923:
1.7 millert 924: modified makefile: my current make can't cope with the test used
925: to avoid unnecessary yacc invocations.
1.1 tholo 926:
1.7 millert 927: Jul 17, 1995:
928: added dynamically growing strings to awk.lx.l and b.c
929: to permit regular expressions to be much bigger.
930: the state arrays can still overflow.
1.1 tholo 931:
1.7 millert 932: Aug 24, 1994:
933: detect duplicate arguments in function definitions (mdm).
1.1 tholo 934:
1.7 millert 935: May 11, 1994:
936: trivial fix to printf to limit string size in sub().
1.1 tholo 937:
1.7 millert 938: Apr 22, 1994:
939: fixed yet another subtle self-assignment problem:
940: $1 = $2; $1 = $1 clobbered $1.
1.1 tholo 941:
1.7 millert 942: Regression tests now use private echo, to avoid quoting problems.
1.1 tholo 943:
1.7 millert 944: Feb 2, 1994:
945: changed error() to print line number as %d, not %g.
1.1 tholo 946:
1.7 millert 947: Jul 23, 1993:
948: cosmetic changes: increased sizes of some arrays,
949: reworded some error messages.
1.1 tholo 950:
1.7 millert 951: added CONVFMT as in posix (just replaced OFMT in getsval)
1.1 tholo 952:
1.7 millert 953: FILENAME is now "" until the first thing that causes a file
954: to be opened.
1.1 tholo 955:
1.7 millert 956: Nov 28, 1992:
957: deleted yyunput and yyoutput from proto.h;
958: different versions of lex give these different declarations.
1.1 tholo 959:
1.7 millert 960: May 31, 1992:
961: added -mr N and -mf N options: more record and fields.
962: these really ought to adjust automatically.
1.1 tholo 963:
1.7 millert 964: cleaned up some error messages; "out of space" now means
965: malloc returned NULL in all cases.
1.1 tholo 966:
1.7 millert 967: changed rehash so that if it runs out, it just returns;
968: things will continue to run slow, but maybe a bit longer.
1.1 tholo 969:
1.7 millert 970: Apr 24, 1992:
971: remove redundant close of stdin when using -f -.
1.1 tholo 972:
1.7 millert 973: got rid of core dump with -d; awk -d just prints date.
1.1 tholo 974:
1.7 millert 975: Apr 12, 1992:
976: added explicit check for /dev/std(in,out,err) in redirection.
977: unlike gawk, no /dev/fd/n yet.
1.1 tholo 978:
1.7 millert 979: added (file/pipe) builtin. hard to test satisfactorily.
980: not posix.
1.1 tholo 981:
1.7 millert 982: Feb 20, 1992:
983: recompile after abortive changes; should be unchanged.
1.1 tholo 984:
1.7 millert 985: Dec 2, 1991:
986: die-casting time: converted to ansi C, installed that.
1.1 tholo 987:
1.7 millert 988: Nov 30, 1991:
989: fixed storage leak in freefa, failing to recover [N]CCL.
990: thanks to Bill Jones (jones@cs.usask.ca)
1.1 tholo 991:
1.7 millert 992: Nov 19, 1991:
993: use RAND_MAX instead of literal in builtin().
1.1 tholo 994:
1.7 millert 995: Nov 12, 1991:
996: cranked up some fixed-size arrays in b.c, and added a test for
997: overflow in penter. thanks to mark larsen.
1.1 tholo 998:
1.7 millert 999: Sep 24, 1991:
1000: increased buffer in gsub. a very crude fix to a general problem.
1001: and again on Sep 26.
1.1 tholo 1002:
1.7 millert 1003: Aug 18, 1991:
1004: enforce variable name syntax for commandline variables: has to
1005: start with letter or _.
1.1 tholo 1006:
1.7 millert 1007: Jul 27, 1991:
1008: allow newline after ; in for statements.
1.1 tholo 1009:
1.7 millert 1010: Jul 21, 1991:
1011: fixed so that in self-assignment like $1=$1, side effects
1012: like recomputing $0 take place. (this is getting subtle.)
1.1 tholo 1013:
1.7 millert 1014: Jun 30, 1991:
1015: better test for detecting too-long output record.
1.1 tholo 1016:
1.7 millert 1017: Jun 2, 1991:
1018: better defense against very long printf strings.
1019: made break and continue illegal outside of loops.
1.1 tholo 1020:
1.7 millert 1021: May 13, 1991:
1022: removed extra arg on gettemp, tempfree. minor error message rewording.
1.1 tholo 1023:
1024: May 6, 1991:
1025: fixed silly bug in hex parsing in hexstr().
1026: removed an apparently unnecessary test in isnumber().
1027: warn about weird printf conversions.
1028: fixed unchecked array overwrite in relex().
1029:
1030: changed for (i in array) to access elements in sorted order.
1031: then unchanged it -- it really does run slower in too many cases.
1032: left the code in place, commented out.
1033:
1.7 millert 1034: Feb 10, 1991:
1035: check error status on all writes, to avoid banging on full disks.
1.1 tholo 1036:
1.7 millert 1037: Jan 28, 1991:
1038: awk -f - reads the program from stdin.
1.1 tholo 1039:
1.7 millert 1040: Jan 11, 1991:
1041: failed to set numeric state on $0 in cmd|getline context in run.c.
1.1 tholo 1042:
1.7 millert 1043: Nov 2, 1990:
1044: fixed sleazy test for integrality in getsval; use modf.
1.1 tholo 1045:
1.7 millert 1046: Oct 29, 1990:
1047: fixed sleazy buggy code in lib.c that looked (incorrectly) for
1048: too long input lines.
1.1 tholo 1049:
1.7 millert 1050: Oct 14, 1990:
1051: fixed the bug on p. 198 in which it couldn't deduce that an
1052: argument was an array in some contexts. replaced the error
1053: message in intest() by code that damn well makes it an array.
1.1 tholo 1054:
1.7 millert 1055: Oct 8, 1990:
1056: fixed horrible bug: types and values were not preserved in
1057: some kinds of self-assignment. (in assign().)
1.1 tholo 1058:
1.7 millert 1059: Aug 24, 1990:
1060: changed NCHARS to 256 to handle 8-bit characters in strings
1061: presented to match(), etc.
1.1 tholo 1062:
1.7 millert 1063: Jun 26, 1990:
1064: changed struct rrow (awk.h) to use long instead of int for lval,
1065: since cfoll() stores a pointer in it. now works better when int's
1066: are smaller than pointers!
1.1 tholo 1067:
1.7 millert 1068: May 6, 1990:
1069: AVA fixed the grammar so that ! is uniformly of the same precedence as
1070: unary + and -. This renders illegal some constructs like !x=y, which
1071: now has to be parenthesized as !(x=y), and makes others work properly:
1072: !x+y is (!x)+y, and x!y is x !y, not two pattern-action statements.
1073: (These problems were pointed out by Bob Lenk of Posix.)
1.1 tholo 1074:
1.7 millert 1075: Added \x to regular expressions (already in strings).
1076: Limited octal to octal digits; \8 and \9 are not octal.
1077: Centralized the code for parsing escapes in regular expressions.
1078: Added a bunch of tests to T.re and T.sub to verify some of this.
1.1 tholo 1079:
1.7 millert 1080: Feb 9, 1990:
1081: fixed null pointer dereference bug in main.c: -F[nothing]. sigh.
1.1 tholo 1082:
1.7 millert 1083: restored srand behavior: it returns the current seed.
1.1 tholo 1084:
1.7 millert 1085: Jan 18, 1990:
1086: srand now returns previous seed value (0 to start).
1.1 tholo 1087:
1.7 millert 1088: Jan 5, 1990:
1089: fix potential problem in tran.c -- something was freed,
1090: then used in freesymtab.
1.1 tholo 1091:
1.7 millert 1092: Oct 18, 1989:
1093: another try to get the max number of open files set with
1094: relatively machine-independent code.
1.1 tholo 1095:
1.7 millert 1096: small fix to input() in case of multiple reads after EOF.
1.1 tholo 1097:
1.7 millert 1098: Oct 11, 1989:
1099: FILENAME is now defined in the BEGIN block -- too many old
1100: programs broke.
1.1 tholo 1101:
1.7 millert 1102: "-" means stdin in getline as well as on the commandline.
1.1 tholo 1103:
1.7 millert 1104: added a bunch of casts to the code to tell the truth about
1105: char * vs. unsigned char *, a right royal pain. added a
1106: setlocale call to the front of main, though probably no one
1107: has it usefully implemented yet.
1.1 tholo 1108:
1.7 millert 1109: Aug 24, 1989:
1110: removed redundant relational tests against nullnode if parse
1111: tree already had a relational at that point.
1.1 tholo 1112:
1.7 millert 1113: Aug 11, 1989:
1114: fixed bug: commandline variable assignment has to look like
1115: var=something. (consider the man page for =, in file =.1)
1.1 tholo 1116:
1.7 millert 1117: changed number of arguments to functions to static arrays
1118: to avoid repeated malloc calls.
1.1 tholo 1119:
1.7 millert 1120: Aug 2, 1989:
1121: restored -F (space) separator
1.1 tholo 1122:
1.7 millert 1123: Jul 30, 1989:
1124: added -v x=1 y=2 ... for immediate commandline variable assignment;
1125: done before the BEGIN block for sure. they have to precede the
1126: program if the program is on the commandline.
1127: Modified Aug 2 to require a separate -v for each assignment.
1.1 tholo 1128:
1.7 millert 1129: Jul 10, 1989:
1130: fixed ref-thru-zero bug in environment code in tran.c
1.1 tholo 1131:
1.7 millert 1132: Jun 23, 1989:
1133: add newline to usage message.
1.1 tholo 1134:
1.7 millert 1135: Jun 14, 1989:
1136: added some missing ansi printf conversion letters: %i %X %E %G.
1137: no sensible meaning for h or L, so they may not do what one expects.
1.1 tholo 1138:
1.7 millert 1139: made %* conversions work.
1.1 tholo 1140:
1.7 millert 1141: changed x^y so that if n is a positive integer, it's done
1142: by explicit multiplication, thus achieving maximum accuracy.
1143: (this should be done by pow() but it seems not to be locally.)
1144: done to x ^= y as well.
1.1 tholo 1145:
1.7 millert 1146: Jun 4, 1989:
1147: ENVIRON array contains environment: if shell variable V=thing,
1148: ENVIRON["V"] is "thing"
1.1 tholo 1149:
1.7 millert 1150: multiple -f arguments permitted. error reporting is naive.
1151: (they were permitted before, but only the last was used.)
1.1 tholo 1152:
1.7 millert 1153: fixed a really stupid botch in the debugging macro dprintf
1.1 tholo 1154:
1.7 millert 1155: fixed order of evaluation of commandline assignments to match
1156: what the book claims: an argument of the form x=e is evaluated
1157: at the time it would have been opened if it were a filename (p 63).
1158: this invalidates the suggested answer to ex 4-1 (p 195).
1.1 tholo 1159:
1.7 millert 1160: removed some code that permitted -F (space) fieldseparator,
1161: since it didn't quite work right anyway. (restored aug 2)
1.1 tholo 1162:
1.7 millert 1163: Apr 27, 1989:
1164: Line number now accumulated correctly for comment lines.
1.1 tholo 1165:
1.7 millert 1166: Apr 26, 1989:
1167: Debugging output now includes a version date,
1168: if one compiles it into the source each time.
1.1 tholo 1169:
1.7 millert 1170: Apr 9, 1989:
1171: Changed grammar to prohibit constants as 3rd arg of sub and gsub;
1172: prevents class of overwriting-a-constant errors. (Last one?)
1173: This invalidates the "banana" example on page 43 of the book.
1.1 tholo 1174:
1.7 millert 1175: Added \a ("alert"), \v (vertical tab), \xhhh (hexadecimal),
1176: as in ANSI, for strings. Rescinded the sloppiness that permitted
1177: non-octal digits in \ooo. Warning: not all compilers and libraries
1178: will be able to deal with \x correctly.
1.1 tholo 1179:
1.7 millert 1180: Jan 9, 1989:
1181: Fixed bug that caused tempcell list to contain a duplicate.
1182: The fix is kludgy.
1.1 tholo 1183:
1.7 millert 1184: Dec 17, 1988:
1185: Catches some more commandline errors in main.
1186: Removed redundant decl of modf in run.c (confuses some compilers).
1187: Warning: there's no single declaration of malloc, etc., in awk.h
1188: that seems to satisfy all compilers.
1.1 tholo 1189:
1.7 millert 1190: Dec 7, 1988:
1191: Added a bit of code to error printing to avoid printing nulls.
1192: (Not clear that it actually would.)
1.1 tholo 1193:
1.7 millert 1194: Nov 27, 1988:
1195: With fear and trembling, modified the grammar to permit
1196: multiple pattern-action statements on one line without
1197: an explicit separator. By definition, this capitulation
1198: to the ghost of ancient implementations remains undefined
1199: and thus subject to change without notice or apology.
1200: DO NOT COUNT ON IT.
1.1 tholo 1201:
1.7 millert 1202: Oct 30, 1988:
1203: Fixed bug in call() that failed to recover storage.
1.1 tholo 1204:
1.7 millert 1205: A warning is now generated if there are more arguments
1206: in the call than in the definition (in lieu of fixing
1207: another storage leak).
1.2 millert 1208:
1.7 millert 1209: Oct 20, 1988:
1210: Fixed %c: if expr is numeric, use numeric value;
1211: otherwise print 1st char of string value. still
1212: doesn't work if the value is 0 -- won't print \0.
1.2 millert 1213:
1.7 millert 1214: Added a few more checks for running out of malloc.
1.2 millert 1215:
1.7 millert 1216: Oct 12, 1988:
1217: Fixed bug in call() that freed local arrays twice.
1.5 kstailey 1218:
1.7 millert 1219: Fixed to handle deletion of non-existent array right;
1220: complains about attempt to delete non-array element.
1.6 kstailey 1221:
1.7 millert 1222: Sep 30, 1988:
1223: Now guarantees to evaluate all arguments of built-in
1224: functions, as in C; the appearance is that arguments
1225: are evaluated before the function is called. Places
1226: affected are sub (gsub was ok), substr, printf, and
1227: all the built-in arithmetic functions in bltin().
1228: A warning is generated if a bltin() is called with
1229: the wrong number of arguments.
1.6 kstailey 1230:
1.7 millert 1231: This requires changing makeprof on p167 of the book.
1.6 kstailey 1232:
1.7 millert 1233: Aug 23, 1988:
1234: setting FILENAME in BEGIN caused core dump, apparently
1235: because it was freeing space not allocated by malloc.
1.6 kstailey 1236:
1.7 millert 1237: July 24, 1988:
1238: fixed egregious error in toupper/tolower functions.
1239: still subject to rescinding, however.
1.6 kstailey 1240:
1.7 millert 1241: July 2, 1988:
1242: flush stdout before opening file or pipe
1.6 kstailey 1243:
1.7 millert 1244: July 2, 1988:
1245: performance bug in b.c/cgoto(): not freeing some sets of states.
1246: partial fix only right now, and the number of states increased
1247: to make it less obvious.
1.6 kstailey 1248:
1.7 millert 1249: June 1, 1988:
1250: check error status on close
1.6 kstailey 1251:
1.7 millert 1252: May 28, 1988:
1253: srand returns seed value it's using.
1254: see 1/18/90
1.6 kstailey 1255:
1.7 millert 1256: May 22, 1988:
1257: Removed limit on depth of function calls.
1.6 kstailey 1258:
1.7 millert 1259: May 10, 1988:
1260: Fixed lib.c to permit _ in commandline variable names.
1.6 kstailey 1261:
1.7 millert 1262: Mar 25, 1988:
1263: main.c fixed to recognize -- as terminator of command-
1264: line options. Illegal options flagged.
1265: Error reporting slightly cleaned up.
1.6 kstailey 1266:
1.7 millert 1267: Dec 2, 1987:
1268: Newer C compilers apply a strict scope rule to extern
1269: declarations within functions. Two extern declarations in
1270: lib.c and tran.c have been moved to obviate this problem.
1.6 kstailey 1271:
1.7 millert 1272: Oct xx, 1987:
1273: Reluctantly added toupper and tolower functions.
1274: Subject to rescinding without notice.
1.6 kstailey 1275:
1.7 millert 1276: Sep 17, 1987:
1277: Error-message printer had printf(s) instead of
1278: printf("%s",s); got core dumps when the message
1279: included a %.
1.6 kstailey 1280:
1.7 millert 1281: Sep 12, 1987:
1282: Very long printf strings caused core dump;
1283: fixed aprintf, asprintf, format to catch them.
1284: Can still get a core dump in printf itself.
1.6 kstailey 1285:
1286: