File: diffstat.c

package info (click to toggle)
diffstat 1.39-1
  • links: PTS
  • area: main
  • in suites: sarge
  • size: 532 kB
  • ctags: 120
  • sloc: sh: 2,769; ansic: 1,102; makefile: 130
file content (1195 lines) | stat: -rw-r--r-- 30,626 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
/******************************************************************************
 * Copyright 1994-2004,2005 by Thomas E. Dickey                               *
 * All Rights Reserved.                                                       *
 *                                                                            *
 * Permission to use, copy, modify, and distribute this software and its      *
 * documentation for any purpose and without fee is hereby granted, provided  *
 * that the above copyright notice appear in all copies and that both that    *
 * copyright notice and this permission notice appear in supporting           *
 * documentation, and that the name of the above listed copyright holder(s)   *
 * not be used in advertising or publicity pertaining to distribution of the  *
 * software without specific, written prior permission.                       *
 *                                                                            *
 * THE ABOVE LISTED COPYRIGHT HOLDER(S) DISCLAIM ALL WARRANTIES WITH REGARD   *
 * TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND  *
 * FITNESS, IN NO EVENT SHALL THE ABOVE LISTED COPYRIGHT HOLDER(S) BE LIABLE  *
 * FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES          *
 * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN      *
 * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR *
 * IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.                *
 ******************************************************************************/

#ifndef	NO_IDENT
static const char *Id = "$Id: diffstat.c,v 1.39 2005/04/10 14:55:15 tom Exp $";
#endif

/*
 * Title:	diffstat.c
 * Author:	T.E.Dickey
 * Created:	02 Feb 1992
 * Modified:
 *		10 Apr 2005, change order of merging and prefix-stripping so
 *			     stripping all prefixes, e.g., with -p9, will be
 *			     sorted as expected (Patch by Jean Delvare
 *			     <khali@linux-fr.org>).
 *		10 Jan 2005, add support for '--help' and '--version' (Patch
 *			     by Eric Blake <ebb9@byu.net>.)
 *		16 Dec 2004, fix a different case for data beginning with "--"
 *			     which was treated as a header line.
 *		14 Dec 2004, Fix allocation problems.  Open files in binary
 *			     mode for reading.  Getopt returns -1, not
 *			     necessarily EOF.  Add const where useful.  Use
 *			     NO_IDENT where necessary.  malloc() comes from
 *			     <stdlib.h> in standard systems (Patch by Eric
 *			     Blake <ebb9@byu.net>.)
 *		08 Nov 2004, minor fix for resync of unified diffs checks for
 *			     range (line beginning with '@' without header
 *			     lines (successive lines beginning with "---" and
 *			     "+++").  Fix a few problems reported by valgrind.
 *		09 Nov 2003, modify check for lines beginning with '-' or '+'
 *			     to treat only "---" in old-style diffs as a
 *			     special case.
 *		14 Feb 2003, modify check for filenames to allow for some cases
 *			     of incomplete dates (the reported example omitted
 *			     the day of the month).  Correct a typo in usage().
 *			     Add -e, -h, -o options.
 *		04 Jan 2003, improve tracking of chunks in unified diff, in
 *			     case the original files contained a '+' or '-' in
 *			     the first column (Debian #155000).  Add -v option
 *			     (Debian #170947).  Modify to allocate buffers big
 *			     enough for long input lines.  Do additional
 *			     merging to handle unusual Index/diff constructs in
 *			     recent makepatch script.
 *		20 Aug 2002, add -u option to tell diffstat to preserve the
 *			     order of filenames as given rather than sort them
 *			     (request by H Peter Anvin <hpa@zytor.com>).  Add
 *			     -k option for completeness.
 *		09 Aug 2002, allow either '/' or '-' as delimiters in dates,
 *			     to accommodate diffutils 2.8 (report by Rik van
 *			     Riel <riel@conectiva.com.br>).
 *		10 Oct 2001, add bzip2 (.bz2) suffix as suggested by
 *			     Gregory T Norris <haphazard@socket.net> in Debian
 *			     bug report #82969).
 *			     add check for diff from RCS archive where the
 *			     "diff" lines do not reference a filename.
 *		29 Mar 2000, add -c option.  Check for compressed input, read
 *			     via pipe.  Change to ANSI C.  Adapted change from
 *			     Troy Engel to add option that displays a number
 *			     only, rather than a histogram.
 *		17 May 1998, handle Debian diff files, which do not contain
 *			     dates on the header lines.
 *		16 Jan 1998, accommodate patches w/o tabs in header lines (e.g.,
 *			     from cut/paste).  Strip suffixes such as ".orig".
 *		24 Mar 1996, corrected -p0 logic, more fixes in do_merging.
 *		16 Mar 1996, corrected state-change for "Binary".  Added -p
 *			     option.
 *		17 Dec 1995, corrected matching algorithm in 'do_merging()'
 *		11 Dec 1995, mods to accommodate diffs against /dev/null or
 *			     /tmp/XXX (tempfiles).
 *		06 May 1995, limit scaling -- only shrink-to-fit.
 *		29 Apr 1995, recognize 'rcsdiff -u' format.
 *		26 Dec 1994, strip common pathname-prefix.
 *		13 Nov 1994, added '-n' option.  Corrected logic of 'match'.
 *		17 Jun 1994, ifdef-<string.h>
 *		12 Jun 1994, recognize unified diff, and output of makepatch.
 *		04 Oct 1993, merge multiple diff-files, busy message when the
 *			     output is piped to a file.
 *
 * Function:	this program reads the output of 'diff' and displays a histogram
 *		of the insertions/deletions/modifications per-file.
 */

#if defined(HAVE_CONFIG_H)
#include <config.h>
#endif

#ifdef WIN32
#define HAVE_STDLIB_H
#define HAVE_STRING_H
#define HAVE_MALLOC_H
#define HAVE_GETOPT_H
#endif

#include <stdio.h>
#include <ctype.h>

#ifdef HAVE_STRING_H
#include <string.h>
#else
#include <strings.h>
#define strchr index
#define strrchr rindex
#endif

#ifdef HAVE_STDLIB_H
#include <stdlib.h>
#else
extern int atoi();
#endif

#ifdef HAVE_UNISTD_H
#include <unistd.h>
#else
extern int isatty();
#endif

#ifdef HAVE_MALLOC_H
#include <malloc.h>
#endif

#ifdef HAVE_GETOPT_H
#include <getopt.h>
#else
extern int getopt();
extern char *optarg;
extern int optind;
#endif

#if !defined(EXIT_SUCCESS)
#define EXIT_SUCCESS 0
#define EXIT_FAILURE 1
#endif

/******************************************************************************/

#ifdef WIN32
#define PATHSEP '\\'
#else
#define PATHSEP '/'
#endif

#define SQUOTE  '\''
#define EOS     '\0'
#define BLANK   ' '

#define UC(c)   ((unsigned char)(c))

#ifdef DEBUG
#define TRACE(p) printf p
#else
#define TRACE(p)		/*nothing */
#endif

#define contain_any(s,reject) (strcspn(s,reject) != strlen(s))

#define HAVE_NOTHING 0
#define HAVE_GENERIC 1		/* e.g., "Index: foo" w/o pathname */
#define HAVE_PATH    2		/* reference-file from "diff dirname/foo" */
#define HAVE_PATH2   4		/* comparison-file from "diff dirname/foo" */

typedef enum comment {
    Normal, Only, Binary
} Comment;

typedef struct _data {
    struct _data *link;
    char *name;			/* the filename */
    int base;			/* beginning of name if -p option used */
    Comment cmt;
    long ins;			/* "+" count inserted lines */
    long del;			/* "-" count deleted lines */
    long mod;			/* "!" count modified lines */
} DATA;

static DATA *all_data;
static char *comment_opt = "";
static int format_opt = 1;
static int max_width;		/* the specified width-limit */
static int merge_names = 1;	/* true if we merge similar filenames */
static int name_wide;		/* the amount reserved for filenames */
static int show_progress;	/* if not writing to tty, show progress */
static int plot_width;		/* the amount left over for histogram */
static int prefix_opt = -1;	/* if positive, controls stripping of PATHSEP */
static int sort_names = 1;	/* true if we sort filenames */
static int verbose = 0;		/* -q/-v options */
static long plot_scale;		/* the effective scale (1:maximum) */

/******************************************************************************/

static void
failed(const char *s)
{
    perror(s);
    exit(EXIT_FAILURE);
}

/* malloc wrapper that never returns NULL */
static void *
xmalloc(size_t s)
{
    void *p;
    if ((p = malloc(s)) == NULL)
	failed("malloc");
    return p;
}

static void
blip(int c)
{
    if (show_progress) {
	(void) fputc(c, stderr);
	(void) fflush(stderr);
    }
}

static char *
new_string(const char *s)
{
    return strcpy((char *) xmalloc((size_t) (strlen(s) + 1)), s);
}

static DATA *
new_data(const char *name)
{
    DATA *p, *q, *r;
    int base = 0;

    TRACE(("new_data(%s)\n", name));

    /* Compute the base offset if the prefix option is used */
    if (prefix_opt >= 0) {
	int n;

	for (n = prefix_opt; n > 0; n--) {
	    char *s = strchr(name + base, PATHSEP);
	    if (s == 0 || *++s == EOS)
		break;
	    base = s - name;
	}
	TRACE(("base set to %d\n", base));
    }

    /* Insert into sorted list (usually sorted).  If we are not sorting or
     * merging names, we fall off the end and link the new entry to the end of
     * the list.  If the prefix option is used, the prefix is ignored by the
     * merge and sort operations.
     */
    for (p = all_data, q = 0; p != 0; q = p, p = p->link) {
	int cmp = strcmp(p->name + p->base, name + base);
	if (merge_names && (cmp == 0))
	    return p;
	if (sort_names && (cmp > 0))
	    break;
    }
    r = (DATA *) xmalloc(sizeof(DATA));
    if (q != 0)
	q->link = r;
    else
	all_data = r;

    r->link = p;
    r->name = new_string(name);
    r->base = base;
    r->cmt = Normal;
    r->ins = 0;
    r->del = 0;
    r->mod = 0;

    return r;
}

/*
 * Remove a unneeded data item from the linked list.  Free the name as well.
 */
static int
delink(DATA * data)
{
    DATA *p, *q;

    TRACE(("delink '%s'\n", data->name));

    for (p = all_data, q = 0; p != 0; q = p, p = p->link) {
	if (p == data) {
	    if (q != 0)
		q->link = p->link;
	    else
		all_data = p->link;
	    free(p->name);
	    free(p);
	    return 1;
	}
    }
    return 0;
}

/* like strncmp, but without the 3rd argument */
static int
match(const char *s, const char *p)
{
    int ok = 0;

    while (*s != EOS) {
	if (*p == EOS) {
	    ok = 1;
	    break;
	}
	if (*s++ != *p++)
	    break;
	if (*s == EOS && *p == EOS) {
	    ok = 1;
	    break;
	}
    }
    return ok;
}

static int
version_num(const char *s)
{
    int main_ver, sub_ver;
    char temp[2];
    return (sscanf(s, "%d.%d%c", &main_ver, &sub_ver, temp) == 2);
}

/*
 * Check for a range of line-numbers, used in editing scripts.
 */
static int
edit_range(const char *s)
{
    int first, last;
    char temp[2];
    return (sscanf(s, "%d,%d%c", &first, &last, temp) == 2)
	|| (sscanf(s, "%d%c", &first, temp) == 1);
}

/*
 * Decode a range for unified diff.  Oddly, the comments in diffutils code
 * claim that both numbers are line-numbers.  However, inspection of the output
 * shows that the numbers are a line-number followed by a count.
 */
static int
decode_range(const char *s, int *first, int *second)
{
    char check;
    if (sscanf(s, "%d,%d%c", first, second, &check) == 2) {
	TRACE(("decode_range #1 first=%d, second=%d\n", *first, *second));
	return 1;
    } else if (sscanf(s, "%d%c", first, &check) == 1) {
	*second = *first;	/* diffutils 2.7 does this */
	TRACE(("decode_range #2 first=%d, second=%d\n", *first, *second));
	return 1;
    }
    return 0;
}

static int
HadDiffs(const DATA * data)
{
    return data->ins != 0
	|| data->del != 0
	|| data->mod != 0
	|| data->cmt != Normal;
}

/*
 * If the given path is not one of the "ignore" paths, then return true.
 */
static int
can_be_merged(const char *path)
{
    if (strcmp(path, "")
	&& strcmp(path, "/dev/null")
	&& strncmp(path, "/tmp/", 5))
	return 1;
    return 0;
}

static int
is_leaf(const char *leaf, const char *path)
{
    char *s;

    if (strchr(leaf, PATHSEP) == 0
	&& (s = strrchr(path, PATHSEP)) != 0
	&& !strcmp(++s, leaf))
	return 1;
    return 0;
}

static char *
do_merging(DATA * data, char *path, int *freed)
{
    TRACE(("do_merging(%s,%s) diffs:%d\n", data->name, path, HadDiffs(data)));

    *freed = 0;
    if (!HadDiffs(data)) {	/* the data was the first of 2 markers */
	if (is_leaf(data->name, path)) {
	    TRACE(("is_leaf: %s vs %s\n", data->name, path));
	    *freed = delink(data);
	} else if (can_be_merged(data->name)
		   && can_be_merged(path)) {
	    size_t len1 = strlen(data->name);
	    size_t len2 = strlen(path);
	    unsigned n;
	    int matched = 0;
	    int diff = 0;

	    /* strip suffixes such as ".orig", ".bak" */
	    if (len1 > len2) {
		if (!strncmp(data->name, path, len2)) {
		    TRACE(("trimming data '%s' to '%.*s'\n",
			   data->name, (int) len2, data->name));
		    data->name[len1 = len2] = EOS;
		}
	    } else if (len1 < len2) {
		if (!strncmp(data->name, path, len1)) {
		    TRACE(("trimming path '%s' to '%.*s'\n",
			   path, (int) len1, path));
		    path[len2 = len1] = EOS;
		}
	    }

	    for (n = 1; n <= len1 && n <= len2; n++) {
		if (data->name[len1 - n] != path[len2 - n]) {
		    diff = n;
		    break;
		}
		if (path[len2 - n] == PATHSEP)
		    matched = n;
	    }

	    if (prefix_opt < 0
		&& matched != 0
		&& diff)
		path += len2 - matched + 1;

	    *freed = delink(data);
	    TRACE(("merge @%d, prefix_opt=%d matched=%d diff=%d\n",
		   __LINE__, prefix_opt, matched, diff));
	} else if (!can_be_merged(path)) {
	    TRACE(("do not merge, retain @%d\n", __LINE__));
	    /* must not merge, retain existing name */
	    path = data->name;
	} else {
	    TRACE(("merge @%d\n", __LINE__));
	    *freed = delink(data);
	}
    } else if (!can_be_merged(path)) {
	path = data->name;
    }
    TRACE(("...do_merging %s\n", path));
    return path;
}

static int
begin_data(const DATA * p)
{
    if (!can_be_merged(p->name)
	&& strchr(p->name, PATHSEP) != 0) {
	TRACE(("begin_data:HAVE_PATH\n"));
	return HAVE_PATH;
    }
    TRACE(("begin_data:HAVE_GENERIC\n"));
    return HAVE_GENERIC;
}

static char *
skip_blanks(char *s)
{
    while (isspace(UC(*s)))
	++s;
    return s;
}

static char *
skip_options(char *params)
{
    while (*params != '\0') {
	params = skip_blanks(params);
	if (*params == '-') {
	    while (isgraph(UC(*params)))
		params++;
	} else {
	    break;
	}
    }
    return params;
}

/*
 * Strip single-quotes from a name (needed for recent makepatch versions).
 */
static void
dequote(char *s)
{
    int len = strlen(s);
    int n;

    if (*s == SQUOTE && len > 2 && s[len - 1] == SQUOTE) {
	for (n = 0; (s[n] = s[n + 1]) != EOS; ++n) {
	    ;
	}
	s[len - 2] = EOS;
    }
}

/*
 * Allocate a fixed-buffer
 */
static void
fixed_buffer(char **buffer, size_t want)
{
    *buffer = xmalloc(want);
}

/*
 * Reallocate a fixed-buffer
 */
static void
adjust_buffer(char **buffer, size_t want)
{
    if ((*buffer = realloc(*buffer, want)) == 0)
	failed("realloc");
}

/*
 * Read until newline or end-of-file, allocating the line-buffer so it is long
 * enough for the input.
 */
static int
get_line(char **buffer, size_t *have, FILE *fp)
{
    int ch;
    size_t used = 0;

    while ((ch = fgetc(fp)) != EOF) {
	if (used + 2 > *have) {
	    adjust_buffer(buffer, *have *= 2);
	}
	(*buffer)[used++] = (char) ch;
	if (ch == '\n')
	    break;
    }
    (*buffer)[used] = '\0';
    return (used != 0);
}

#define date_delims(a,b) (((a)=='/' && (b)=='/') || ((a) == '-' && (b) == '-'))

static void
do_file(FILE *fp)
{
    DATA dummy;
    DATA *that = &dummy;
    DATA *prev = 0;
    char *buffer = 0;
    char *b_fname = 0;
    char *b_temp1 = 0;
    char *b_temp2 = 0;
    char *b_temp3 = 0;
    size_t length = 0;
    size_t fixed = 0;
    int ok = HAVE_NOTHING;
    int marker = -1;
    int unified = 0;
    int freed = 0;
    int old_unify = 0;
    int new_unify = 0;
    int context = 1;
    char *s;
#ifdef DEBUG
    int line_no = 0;
#endif

    memset(&dummy, 0, sizeof(dummy));
    dummy.name = "";

    fixed_buffer(&buffer, fixed = length = BUFSIZ);
    fixed_buffer(&b_fname, length);
    fixed_buffer(&b_temp1, length);
    fixed_buffer(&b_temp2, length);
    fixed_buffer(&b_temp3, length);

    while (get_line(&buffer, &length, fp)) {
	/*
	 * Adjust size of fixed-buffers so that a sscanf cannot overflow.
	 */
	if (length > fixed) {
	    fixed = length;
	    adjust_buffer(&b_fname, length);
	    adjust_buffer(&b_temp1, length);
	    adjust_buffer(&b_temp2, length);
	    adjust_buffer(&b_temp3, length);
	}

	/*
	 * Trim trailing blanks (e.g., newline)
	 */
	for (s = buffer + strlen(buffer); s > buffer; s--) {
	    if (isspace(UC(s[-1])))
		s[-1] = EOS;
	    else
		break;
	}
	TRACE(("[%05d] %s\n", ++line_no, buffer));

	/*
	 * The lines identifying files in a context diff depend on how it was
	 * invoked.  But after the header, each chunk begins with a line
	 * containing 15 *'s.  Each chunk may contain a line-range with '***'
	 * for the "before", and a line-range with '---' for the "after".  The
	 * part of the chunk depicting the deletion may be absent, though the
	 * edit line is present.
	 *
	 * The markers for unified diff are a little different from the normal
	 * context-diff.  Also, the edit-lines in a unified diff won't have a
	 * space in column 2.  Because of the missing space, we have to count
	 * lines to ensure we do not confuse the marker lines.
	 */
	marker = -1;
	if (that != &dummy && !strcmp(buffer, "***************")) {
	    TRACE(("begin context chunk\n"));
	    context = 2;
	} else if (context == 2 && match(buffer, "*** ")) {
	    context = 1;
	} else if (context == 1 && match(buffer, "--- ")) {
	    marker = 1;
	    context = 0;
	} else if (match(buffer, "*** ")) {
	    marker = 0;
	} else if ((old_unify + new_unify) == 0 && match(buffer, "--- ")) {
	    marker = unified = 1;
	} else if ((old_unify + new_unify) == 0 && match(buffer, "+++ ")) {
	    marker = unified = 2;
	} else if (unified == 2
		   || ((old_unify + new_unify) == 0 && (*buffer == '@'))) {
	    unified = 0;
	    if (*buffer == '@') {
		int old_base, new_base, old_size, new_size;
		char test_at;

		old_unify = new_unify = 0;
		if (sscanf(buffer, "@@ -%[0-9,] +%[0-9,] @%c",
			   b_temp1,
			   b_temp2,
			   &test_at) == 3
		    && test_at == '@'
		    && decode_range(b_temp1, &old_base, &old_size)
		    && decode_range(b_temp2, &new_base, &new_size)) {
		    old_unify = old_size;
		    new_unify = new_size;
		    unified = -1;
		}
	    }
	} else if (unified == 1 && !context) {
	    /*
	     * If unified==1, we guessed we would find a "+++" line, but since
	     * we are here, we did not find that.  The context check ensures
	     * we do not mistake the "---" for a unified diff with that for
	     * a context diff's "after" line-range.
	     *
	     * If we guessed wrong, then we probably found a data line with
	     * "--" in the first two columns of the diff'd file.
	     */
	    unified = 0;
	    TRACE(("Expected \"+++\" @%d:%s\n", __LINE__, buffer));
	    if (prev != 0
		&& prev != that
		&& that->ins == 0
		&& that->del == 0
		&& strcmp(prev->name, that->name)) {
		TRACE(("giveup on %ld/%ld %s\n", that->ins, that->del, that->name));
		TRACE(("revert to %ld/%ld %s\n", prev->ins, prev->del, prev->name));
		(void) delink(that);
		that = prev;
		that->del += 1;
	    }
	} else if (old_unify + new_unify) {
	    switch (*buffer) {
	    case '-':
		if (old_unify)
		    --old_unify;
		break;
	    case '+':
		if (new_unify)
		    --new_unify;
		break;
	    case ' ':
		if (old_unify)
		    --old_unify;
		if (new_unify)
		    --new_unify;
		break;
	    default:
		old_unify = new_unify = 0;
		break;
	    }
	} else {
	    unified = 0;
	}

	/*
	 * Override the beginning of the line to simplify the case statement
	 * below.
	 */
	if (marker > 0) {
	    TRACE(("@%d, marker=%d, override %s\n", __LINE__, marker, buffer));
	    (void) strncpy(buffer, "***", 3);
	}

	/*
	 * Use the first character of the input line to determine its
	 * type:
	 */
	switch (*buffer) {
	case 'O':		/* Only */
	    if (match(buffer, "Only in ")) {
		char *path = buffer + 8;
		int found = 0;
		for (s = path; *s != EOS; s++) {
		    if (match(s, ": ")) {
			found = 1;
			*s++ = PATHSEP;
			while ((s[0] = s[1]) != EOS)
			    s++;
			break;
		    }
		}
		if (found) {
		    blip('.');
		    that = new_data(path);
		    that->cmt = Only;
		    ok = HAVE_NOTHING;
		}
	    }
	    break;

	    /*
	     * Several different scripts produce "Index:" lines
	     * (e.g., "makepatch").  Not all bother to put the
	     * pathname of the files; some put only the leaf names.
	     */
	case 'I':
	    if (match(buffer, "Index: ")) {
		s = strrchr(buffer, BLANK);	/* last token is name */
		s = skip_blanks(s);
		dequote(s);
		blip('.');
		s = do_merging(that, s, &freed);
		that = new_data(s);
		ok = begin_data(that);
	    }
	    break;

	case 'd':		/* diff command trace */
	    if (match(buffer, "diff ")
		&& *(s = skip_options(buffer + 5)) != '\0') {
		s = strrchr(buffer, BLANK);
		s = skip_blanks(s);
		dequote(s);
		blip('.');
		s = do_merging(that, s, &freed);
		that = new_data(s);
		ok = begin_data(that);
	    }
	    break;

	case '*':
	    TRACE(("@%d, ok=%d:%s\n", __LINE__, ok, buffer));
	    if (!(ok & HAVE_PATH)) {
		int ddd, hour, minute, second;
		int day, month, year;
		char yrmon, monday;

		/* check for tab-delimited first, so we can
		 * accept filenames containing spaces.
		 */
		if (sscanf(buffer,
			   "*** %[^\t]\t%[^ ] %[^ ] %d %d:%d:%d %d",
			   b_fname,
			   b_temp2, b_temp3, &ddd,
			   &hour, &minute, &second, &year) == 8
		    || (sscanf(buffer,
			       "*** %[^\t]\t%d%c%d%c%d %d:%d:%d",
			       b_fname,
			       &year, &yrmon, &month, &monday, &day,
			       &hour, &minute, &second) == 9
			&& date_delims(yrmon, monday)
			&& !version_num(b_fname))
		    || sscanf(buffer,
			      "*** %[^\t ]%[\t ]%[^ ] %[^ ] %d %d:%d:%d %d",
			      b_fname,
			      b_temp1,
			      b_temp2, b_temp3, &ddd,
			      &hour, &minute, &second, &year) == 9
		    || (sscanf(buffer,
			       "*** %[^\t ]%[\t ]%d%c%d%c%d %d:%d:%d",
			       b_fname,
			       b_temp1,
			       &year, &yrmon, &month, &monday, &day,
			       &hour, &minute, &second) == 10
			&& date_delims(yrmon, monday)
			&& !version_num(b_fname))
		    || (sscanf(buffer,
			       "*** %[^\t ]%[\t ]",
			       b_fname,
			       b_temp1) >= 1
			&& !version_num(b_fname)
			&& !contain_any(b_fname, "*")
			&& !edit_range(b_fname))
		    ) {
		    prev = that;
		    s = do_merging(that, b_fname, &freed);
		    if (freed)
			prev = 0;
		    that = new_data(s);
		    ok = begin_data(that);
		    TRACE(("after merge:%d:%s\n", ok, s));
		}
	    }
	    break;

	case '+':
	    /* FALL-THRU */
	case '>':
	    if (ok)
		that->ins += 1;
	    break;

	case '-':
	    if (!ok)
		break;
	    if (!unified && !strcmp(buffer, "---"))
		break;
	    /* fall-thru */
	case '<':
	    if (ok)
		that->del += 1;
	    break;

	case '!':
	    if (ok)
		that->mod += 1;
	    break;

	    /* Expecting "Binary files XXX and YYY differ" */
	case 'B':		/* Binary */
	    /* FALL-THRU */
	case 'b':		/* binary */
	    if (match(buffer + 1, "inary files ")) {
		s = strrchr(buffer, BLANK);
		if (!strcmp(s, " differ")) {
		    *s = EOS;
		    s = strrchr(buffer, BLANK);
		    blip('.');
		    that = new_data(skip_blanks(s));
		    that->cmt = Binary;
		    ok = HAVE_NOTHING;
		}
	    }
	    break;
	}
    }
    blip('\n');

    if (buffer != 0) {
	free(buffer);
	free(b_fname);
	free(b_temp1);
	free(b_temp2);
	free(b_temp3);
    }
}

/*
 * Each call to 'plot_num()' prints a scaled bar of 'c' characters.  The
 * 'extra' parameter is used to keep the accumulated error in the bar's total
 * length from getting large.
 */
static long
plot_num(long num_value, int c, long extra)
{
    switch (format_opt) {
    case 0:
	printf("\t%ld %c", num_value, c);
	return 0;
    default:
	/* the value to plot */
	/* character to display in the bar */
	/* accumulated error in the bar */
	{
	    long product = (plot_width * num_value) + extra;
	    long count = (product / plot_scale);
	    extra = product - (count * plot_scale);
	    while (--count >= 0)
		(void) putchar(c);
	    return extra;
	}
    }
}

static void
summarize(void)
{
    DATA *p;
    long total_ins = 0, total_del = 0, total_mod = 0, temp;
    int num_files = 0, shortest_name = -1, longest_name = -1, prefix_len = -1;

    plot_scale = 0;
    for (p = all_data; p; p = p->link) {
	int len = strlen(p->name);

	/*
	 * "-p0" gives the whole pathname unmodified.  "-p1" strips
	 * through the first path-separator, etc.
	 */
	if (prefix_opt >= 0) {
	    /* p->base has been computed at node creation */
	    if (name_wide < (len - p->base))
		name_wide = (len - p->base);
	} else {
	    if (len < prefix_len || prefix_len < 0)
		prefix_len = len;
	    while (prefix_len > 0) {
		if (p->name[prefix_len - 1] != PATHSEP)
		    prefix_len--;
		else if (strncmp(all_data->name, p->name, (size_t) prefix_len))
		    prefix_len--;
		else
		    break;
	    }

	    if (len > longest_name)
		longest_name = len;
	    if (len < shortest_name || shortest_name < 0)
		shortest_name = len;
	}

	num_files++;
	total_ins += p->ins;
	total_del += p->del;
	total_mod += p->mod;
	temp = p->ins + p->del + p->mod;
	if (temp > plot_scale)
	    plot_scale = temp;
    }

    if (prefix_opt < 0) {
	if (prefix_len < 0)
	    prefix_len = 0;
	if ((longest_name - prefix_len) > name_wide)
	    name_wide = (longest_name - prefix_len);
    }

    name_wide++;		/* make sure it's nonzero */
    plot_width = (max_width - name_wide - 8);
    if (plot_width < 10)
	plot_width = 10;

    if (plot_scale < plot_width)
	plot_scale = plot_width;	/* 1:1 */

    for (p = all_data; p; p = p->link) {
	printf("%s %-*.*s|",
	       comment_opt,
	       name_wide, name_wide,
	       p->name + (prefix_opt >= 0 ? p->base : prefix_len));
	switch (p->cmt) {
	default:
	case Normal:
	    temp = 0;
	    printf("%5ld ", p->ins + p->del + p->mod);
	    temp = plot_num(p->ins, '+', temp);
	    (void) plot_num(p->del, '-', temp);
	    (void) plot_num(p->mod, '!', temp);
	    break;
	case Binary:
	    printf("binary");
	    break;
	case Only:
	    printf("only");
	    break;
	}
	printf("\n");
    }

    printf("%s %d files changed", comment_opt, num_files);
#define PLURAL(n) n, n != 1 ? "s" : ""
    if (total_ins)
	printf(", %ld insertion%s(+)", PLURAL(total_ins));
    if (total_del)
	printf(", %ld deletion%s(-)", PLURAL(total_del));
    if (total_mod)
	printf(", %ld modification%s(!)", PLURAL(total_mod));
    (void) putchar('\n');
}

#ifdef HAVE_POPEN
static char *
is_compressed(const char *name)
{
    char *verb = 0;
    char *result = 0;
    size_t len = strlen(name);

    if (len > 2 && !strcmp(name + len - 2, ".Z")) {
	verb = "compress -dc %s";
    } else if (len > 3 && !strcmp(name + len - 3, ".gz")) {
	verb = "gzip -dc %s";
    } else if (len > 4 && !strcmp(name + len - 4, ".bz2")) {
	verb = "bzip2 -dc %s";
    }
    if (verb != 0) {
	result = (char *) xmalloc(strlen(verb) + len);
	sprintf(result, verb, name);
    }
    return result;

}
#endif

static void
usage(FILE *fp)
{
    static const char *msg[] =
    {
	"Usage: diffstat [options] [files]",
	"",
	"Reads from one or more input files which contain output from 'diff',",
	"producing a histogram of total lines changed for each file referenced.",
	"If no filename is given on the command line, reads from stdin.",
	"",
	"Options:",
	"  -c      prefix each line with comment (#)",
	"  -e FILE redirect standard error to FILE",
	"  -f NUM  format (0=concise, 1=normal)",
	"  -h      print this message",
	"  -k      do not merge filenames",
	"  -n NUM  specify minimum width for the filenames (default: auto)",
	"  -o FILE redirect standard output to FILE",
	"  -p NUM  specify number of pathname-separators to strip (default: common)",
	"  -u      do not sort the input list",
	"  -v      makes output more verbose",
	"  -V      prints the version number",
	"  -w NUM  specify maximum width of the output (default: 80)",
    };
    unsigned j;
    for (j = 0; j < sizeof(msg) / sizeof(msg[0]); j++)
	fprintf(fp, "%s\n", msg[j]);
}

/* Wrapper around getopt that also parses "--help" and "--version".  
 * argc, argv, opts, return value, and globals optarg, optind,
 * opterr, and optopt are as in getopt().  help and version designate
 * what should be returned if --help or --version are encountered. */
static int
getopt_helper(int argc, char *const argv[], const char *opts,
	      int help, int version)
{
    if (optind < argc && argv[optind] != NULL) {
	if (strcmp(argv[optind], "--help") == 0) {
	    optind++;
	    return help;
	} else if (strcmp(argv[optind], "--version") == 0) {
	    optind++;
	    return version;
	}
    }
    return getopt(argc, argv, opts);
}

int
main(int argc, char *argv[])
{
    int j;
    char version[80];

    max_width = 80;

    while ((j = getopt_helper(argc, argv, "ce:f:hkn:o:p:uvVw:", 'h', 'V'))
	   != -1) {
	switch (j) {
	case 'c':
	    comment_opt = "#";
	    break;
	case 'e':
	    if (freopen(optarg, "w", stderr) == 0)
		failed(optarg);
	    break;
	case 'f':
	    format_opt = atoi(optarg);
	    break;
	case 'h':
	    usage(stdout);
	    return (EXIT_SUCCESS);
	case 'k':
	    merge_names = 0;
	    break;
	case 'n':
	    name_wide = atoi(optarg);
	    break;
	case 'o':
	    if (freopen(optarg, "w", stdout) == 0)
		failed(optarg);
	    break;
	case 'p':
	    prefix_opt = atoi(optarg);
	    break;
	case 'u':
	    sort_names = 0;
	    break;
	case 'v':
	    verbose = 1;
	    break;
	case 'V':
#ifndef	NO_IDENT
	    if (!sscanf(Id, "%*s %*s %s", version))
#endif
		(void) strcpy(version, "?");
	    printf("diffstat version %s\n", version);
	    return (EXIT_SUCCESS);
	case 'w':
	    max_width = atoi(optarg);
	    break;
	default:
	    usage(stderr);
	    return (EXIT_FAILURE);
	}
    }
    show_progress = verbose && (!isatty(fileno(stdout))
				&& isatty(fileno(stderr)));

    if (optind < argc) {
	while (optind < argc) {
	    FILE *fp;
	    char *name = argv[optind++];
#ifdef HAVE_POPEN
	    char *command = is_compressed(name);
	    if (command != 0) {
		if ((fp = popen(command, "r")) != 0) {
		    if (show_progress) {
			(void) fprintf(stderr, "%s\n", name);
			(void) fflush(stderr);
		    }
		    do_file(fp);
		    (void) pclose(fp);
		}
		free(command);
	    } else
#endif
	    if ((fp = fopen(name, "rb")) != 0) {
		if (show_progress) {
		    (void) fprintf(stderr, "%s\n", name);
		    (void) fflush(stderr);
		}
		do_file(fp);
		(void) fclose(fp);
	    } else {
		failed(name);
	    }
	}
    } else {
	do_file(stdin);
    }
    summarize();
#if defined(DEBUG) || defined(NO_LEAKS)
    while (all_data != 0) {
	DATA *p = all_data;
	all_data = all_data->link;
	free(p->name);
	free(p);
    }
#endif
    return (EXIT_SUCCESS);
}