File Coverage

script/cdif

Criterion	Covered	Total	%
statement	442	620	71.2
branch	116	260	44.6
condition	29	75	38.6
subroutine	47	76	61.8
pod			n/a
total	634	1031	61.4

line	stmt	bran	cond	sub	time	code
1						#!/usr/bin/env perl
2
3						##
4						## cdif: word context diff
5						##
6						## Copyright (c) 1992- Kazumasa Utashiro
7						##
8						## Original version on Mar 11 1992
9						##
10
11	8			8	44385	use v5.14;
	8				30
12	8			8	63	use warnings;
	8				17
	8				547
13
14	8			8	4835	use utf8;
	8				2670
	8				56
15	8			8	359	use Carp;
	8				30
	8				1012
16	8			8	5084	use Encode;
	8				171829
	8				971
17	8			8	4441	use Encode::Guess;
	8				37649
	8				63
18
19	8			8	5497	use Pod::Usage;
	8				531763
	8				1497
20	8			8	120	use List::Util qw(first sum pairmap);
	8				35
	8				1425
21	8			8	5001	use Text::ParseWords qw(shellwords);
	8				15322
	8				622
22	8			8	4316	use IO::Divert;
	8				7297
	8				434
23	8			8	15956	use Text::VisualWidth::PP qw(vwidth);
	8				40546
	8				667
24	8			8	5199	use Data::Dumper; {
	8				74752
	8				724
25	8			8	1811839	no warnings 'redefine';
	8				64
	8				15
	8				1050
26	8			0	113	*Data::Dumper::qquote = sub { qq["${\(shift)}"] };
	0				0
	0				0
27	8				29	$Data::Dumper::Terse = 1;
28	8				56	$Data::Dumper::Sortkeys = 1;
29	8				24	$Data::Dumper::Useperl = 1;
30						}
31
32	8			8	4736	use App::sdif;
	8				25
	8				771
33	8				30	our $VERSION = $App::sdif::VERSION;
34
35	8			8	4180	use App::sdif::Util;
	8				59
	8				1289
36	8			8	4273	use App::cdif::Command::Mecab;
	8				34
	8				450
37
38						##
39						## options
40						##
41
42	8			8	4694	use charnames ':full';
	8				97594
	8				60
43	8				517	my %visible = (
44						nul => [ 1, "\000" => "\N{SYMBOL FOR NULL}" ],
45						soh => [ 1, "\001" => "\N{SYMBOL FOR START OF HEADING}" ],
46						bel => [ 1, "\007" => "\N{SYMBOL FOR BELL}" ],
47						bs => [ 1, "\010" => "\N{SYMBOL FOR BACKSPACE}" ],
48						ht => [ 0, "\011" => "\N{SYMBOL FOR HORIZONTAL TABULATION}" ],
49						nl => [ 0, "\012" => "\N{SYMBOL FOR NEWLINE}" . "\n" ],
50						vt => [ 1, "\013" => "\N{SYMBOL FOR VERTICAL TABULATION}" ],
51						np => [ 1, "\014" => "\N{SYMBOL FOR FORM FEED}" ],
52						cr => [ 1, "\015" => "\N{SYMBOL FOR CARRIAGE RETURN}" ],
53						esc => [ 1, "\033" => "\N{SYMBOL FOR ESCAPE}" ],
54						sp => [ 0, "\040" => "\N{SYMBOL FOR SPACE}" ],
55						del => [ 1, "\177" => "\N{SYMBOL FOR DELETE}" ],
56						nbsp => [ 1, "\x{00A0}" => "\N{OPEN BOX}" ],
57						nnbsp => [ 1, "\x{202F}" => "\N{OPEN BOX}" ],
58						ensp => [ 1, "\x{2002}" => "\N{OPEN BOX}" ],
59						emsp => [ 1, "\x{2003}" => "\N{OPEN BOX}" ],
60						thinsp => [ 1, "\x{2009}" => "\N{OPEN BOX}" ],
61						hairsp => [ 1, "\x{200A}" => "\N{OPEN BOX}" ],
62						zwsp => [ 1, "\x{200B}" => "\N{OPEN BOX}" ],
63						idesp => [ 1, "\x{3000}" => "\N{OPEN BOX}" ],
64						);
65	8				205	my %opt_visible = do {
66	8				116	map { $_ => splice @{$visible{$_}}, 0, 1 }
	160				179
	160				369
67						keys %visible;
68						};
69
70	8				50	my @rcsopt;
71						my @diffopts;
72	8				0	my @sub_diffopts;
73
74	8			8	329	binmode STDOUT, ":encoding(utf8)";
	8				7295
	8				144
	8				55
75	8				9787	binmode STDERR, ":encoding(utf8)";
76
77	8	50			264	if (my $env = $ENV{'CDIFOPTS'}) {
78	0				0	unshift(@ARGV, shellwords($env));
79						}
80
81	8				18	my $unit_default = 'word';
82	8				45	my $prefix_default = q/(?:\\\| )(?: )/;
83
84	8				19	my $app;
85
86	8			8	299292	use Getopt::EX::Hashed; {
	8				88923
	8				60
87
88	8				18	Getopt::EX::Hashed->configure( DEFAULT => [ is => 'rw' ] ) ;
	8				138
89
90	8				394	has help => ' ' ;
91	8				607	has version => ' ' ;
92	8				420	has man => ' ' ;
93	8				367	has debug => ' d =s@ ' , default => [] ;
94	8				465	has diff => ' =s ' ;
95	8				397	has subdiff => ' =s ' , default => 'diff' ;
96	8				415	has color => ' ! ' , default => 1 ;
97	8				404	has 256 => ' ! ' , default => 1 ;
98	8				442	has commandcolor => ' cc ! ' , default => 1 ;
99	8				430	has markcolor => ' mc ! ' , default => 1 ;
100	8				531	has textcolor => ' tc ! ' , default => 1 ;
101	8				488	has unknowncolor => ' uc ! ' , default => 1 ;
102	8				427	has colormap => ' cm =s@ ' , default => [] ;
103	8				518	has colordump => ' ' ;
104	8				526	has sdif => ' ' ;
105	8				454	has stat => ' ! ' ;
106	8				459	has unit => ' by :s ' , default => $unit_default ,
107						, any => [ qw(char word letter mecab 0), '' ];
108	8				517	has show_old => ' ! ' , default => 1 , alias => 'old' ;
109	8				523	has show_new => ' ! ' , default => 1 , alias => 'new' ;
110	8				515	has show_mrg => ' ! ' , default => 1 , alias => 'mrg' ;
111	8				498	has command => ' ! ' , default => 1 ;
112	8				541	has unknown => ' ! ' , default => 1 ;
113	8				508	has mark => ' ! ' , default => 1 ;
114	8				581	has prefix => ' ! ' , default => 1 ;
115	8				567	has prefix_pattern => ' =s ' , default => $prefix_default ;
116	8				584	has visible => ' =s@ ' , default => [] ;
117	8				553	has lenience => ' ! ' , default => 1 ;
118	8				565	has linebyline => ' :2 ' , alias => 'lxl';
119	8				692	has style => ' =s ' , default => '' ;
120
121	8				669	has ignore_case => 'i' ;
122	8				629	has ignore_space_change => 'b' ;
123	8				606	has ignore_all_space => 'w' ;
124	8				656	has expand_tabs => 't' ;
125
126	8				627	has rcs => '' ;
127
128						has '+sdif' => sub {
129	4			4	57262	$app->commandcolor = 0;
130	4				70	$app->markcolor = 0;
131	4				35	$app->textcolor = 0;
132	4				32	$app->unknowncolor = 0;
133	8				697	} ;
134						has mecab => '!', action => sub {
135	0	0		0	0	$app->unit = $_[1] ? 'mecab' : $unit_default;
136	8				490	} ;
137	8			0	680	has visible_cr => ' vcr ! ' , action => sub { $opt_visible{cr} = $_[1] } ;
	0				0
138	8			0	618	has visible_esc => ' vesc ! ' , action => sub { $opt_visible{esc} = $_[1] } ;
	0				0
139	8			0	645	has c => ' ' , action => sub { push @diffopts, "-c" } ;
	0				0
140	8			0	842	has u => ' ' , action => sub { push @diffopts, "-u" } ;
	0				0
141	8			0	728	has context => ' C =i ' , action => sub { push @diffopts, "-C" . $_[1] } ;
	0				0
142	8			0	716	has unified => ' U =i ' , action => sub { push @diffopts, "-U" . $_[1] } ;
	0				0
143	8			0	668	has r => ' =s ' , action => sub { push @rcsopt, "-r$_[1]" } ;
	0				0
144	8			0	741	has q => ' ' , action => sub { push @rcsopt, "-q" } ;
	0				0
145
146						has '+help' => sub {
147	0			0	0	pod2usage
148						-verbose => 99,
149						-sections => [ qw(SYNOPSIS VERSION) ];
150	8				760	};
151						has '+man' => sub {
152	0			0	0	pod2usage -verbose => 2;
153	8				359	};
154						has '+version' => sub {
155	0			0	0	print "$VERSION\n";
156	0				0	exit;
157	8				475	};
158
159	8			8	10381	} no Getopt::EX::Hashed;
	8				21
	8				41
160
161	8	50			339	$app = Getopt::EX::Hashed->new() or die;
162
163	8			8	6044	use Getopt::EX::Long qw(:DEFAULT Configure ExConfigure);
	8				931570
	8				6319
164	8				14485	ExConfigure BASECLASS => [ "App::cdif", "Getopt::EX" ];
165	8				377	Configure("bundling");
166	8	50			384	$app->getopt \|\| usage({status => 1});
167
168	8				51501	my %debug = map { $_ => 1 } map { split // } @{$app->debug};
	0				0
	0				0
	8				50
169
170	8				89	$App::sdif::Util::NO_WARNINGS = $app->lenience;
171	8	50			72	$App::cdif::Command::Mecab::debug = 1 if $debug{m};
172
173	8	50			29	$app->rcs++ if @rcsopt;
174	8		33		29	my $diff = $app->diff \|\| ($app->rcs ? 'rcsdiff' : 'diff');
175
176	8				169	push @diffopts, map { $_->[1] } grep { $_->[0] } (
	0				0
	32				221
177						[ $app->ignore_case, "-i" ],
178						[ $app->ignore_space_change, "-b" ],
179						[ $app->ignore_all_space, "-w" ],
180						[ $app->expand_tabs, "-t" ],
181						);
182
183	8				40	push @sub_diffopts, map { $_->[1] } grep { $_->[0] } (
	0				0
	16				108
184						[ $app->ignore_case, "-i" ],
185						[ $app->ignore_all_space, "-w" ],
186						);
187
188	8				17	my %colormap = do {
189	8	50			34	my $col = $app->{256} ? 0 : 1;
190	136	100		136	432	pairmap { $a => (ref $b eq 'ARRAY') ? $b->[$col] : $b } (
191	8				194	UNKNOWN => "" ,
192						UMARK => [ "/444" , "/w" ] ,
193						CMARK => "GS" ,
194						OMARK => "CS" ,
195						NMARK => "MS" ,
196						MMARK => "YS" ,
197						UTEXT => "" ,
198						CTEXT => "G" ,
199						OTEXT => "C" ,
200						NTEXT => "M" ,
201						MTEXT => "Y" ,
202						COMMAND => [ "555/222E" , "w/kE" ] ,
203						OCHANGE => [ "K/445" , "B/w" ] ,
204						NCHANGE => [ "K/445" , "B/w" ] ,
205						DELETE => [ "K/544" , "R/w" ] ,
206						APPEND => [ "K/544" , "R/w" ] ,
207						VISIBLE => "",
208						);
209						};
210
211	8			8	97	use Getopt::EX::Colormap qw(ansi_code);
	8				18
	8				10637
212						my $color_handler = Getopt::EX::Colormap
213						->new(HASH => \%colormap)
214	8				145	->load_params(@{$app->colormap});
	8				693
215
216	8				273	for (
217						[ $app->unknowncolor => q/UNKNOWN/ ],
218						[ $app->commandcolor => q/COMMAND/ ],
219						[ $app->markcolor => q/MARK/ ],
220						[ $app->textcolor => q/TEXT/ ],
221						) {
222	32				219	my($color, $label) = @$_;
223	32	100			100	$color and next;
224	16				526	for (grep /$label/, keys %colormap) {
225	48				136	$colormap{$_} = '';
226						}
227						}
228
229	8	50			44	warn 'colormap = ', Dumper \%colormap if $debug{c};
230
231	8	50			55	if ($app->colordump) {
232	0				0	print $color_handler->colormap(
233						name => '--changeme', option => '--colormap');
234	0				0	exit;
235						}
236
237						sub color {
238	594			594	12467	$color_handler->color(@_);
239						}
240
241	8				72	my $prefix_re = do {
242	8	50			33	if ($app->prefix) {
243	8				277	qr/$app->{prefix_pattern}/;
244						} else {
245	0				0	"";
246						}
247						};
248
249	8				43	my $DIFF;
250						my $OLD;
251	8				0	my $NEW;
252
253	8	50			28	if ($app->rcs) {
		50
		50
254	0		0		0	my $rcsfile = shift \|\| usage("No RCS filename\n\n");
255	0				0	$DIFF = "$diff @diffopts @rcsopt $rcsfile\|";
256						} elsif (@ARGV == 2) {
257	0				0	($OLD, $NEW) = splice(@ARGV, 0, 2);
258	0				0	$DIFF = "$diff @diffopts $OLD $NEW \|";
259						} elsif (@ARGV < 2) {
260	8		100		117	$DIFF = shift \|\| '-';
261						} else {
262	0	0			0	usage("Arguments error.\n\n") if @ARGV;
263						}
264	8	50			41	warn "DIFF = \"$DIFF\"\n" if $debug{f};
265
266	8				17	my %func = do {
267	8	50			33	my $col = $app->color ? 0 : 1;
268	40			40	122	pairmap { $a => $b->[$col] } (
269	0			0	0	DELETE => [ sub { color("DELETE", @_) }, \&bd ],
270	0			0	0	APPEND => [ sub { color("APPEND", @_) }, \&bd ],
271	48			48	445	OLD => [ sub { color("OCHANGE", @_) }, \&ul ],
272	48			48	148	NEW => [ sub { color("NCHANGE", @_) }, \&ul ],
273	8			72	209	UNKNOWN => [ sub { color("UNKNOWN", @_) }, undef ],
	72				225
274						);
275						};
276
277	8				41	my $w_pattern = do {
278	8	50			34	if ($app->unit =~ /^char/) {
279	0				0	qr/\X/s;
280						} else {
281	8	50			74	my $w = $app->unit eq 'letter' ? '' : '_';
282	8				2249	qr{
283						\p{Han} \| \p{InHiragana}+ \| \p{InKatakana}+ \|
284						[$w\p{Latin}]+ \|
285						[$w\p{Hang}]+ \|
286						[$w\p{Cyrillic}]+ \|
287						[$w\p{Arabic}]+ \|
288						[$w\p{Thai}]+ \|
289						\d+ \|
290						# (\p{Punct})\g{-1}* \|
291						[\h\r\f]\n \| \s+ \| (\X)\g{-1}
292						}x;
293						}
294						};
295
296						##
297						## Converter/Effector function for visible characters
298						##
299	8				31	my($converter, $effector);
300	8			8	76	use Getopt::EX::LabeledParam;
	8				16
	8				6627
301						Getopt::EX::LabeledParam
302						->new(HASH => \%opt_visible)
303	8				59	->load_params (@{$app->visible});
	8				292
304	8	50			375	if (my @names = grep $opt_visible{$_}, keys %opt_visible) {
305	8				176	my @chars = map $visible{$_}->[0], @names;
306	8				31	my %hash = map { @$_ } values %visible;
	160				403
307	8				49	my $re = do { local $" = ''; qr/[\Q@chars\E]/ };
	8				27
	8				354
308	8			0	47	my $sub0 = sub { s/($re)/$hash{$1}/g };
	0				0
309	8			0	30	my $sub1 = sub { $_[0] =~ s/($re)/$hash{$1}/gr };
	0				0
310						my $sub2 = sub {
311	78			78	106	my $mark_re = shift;
312	78				137	for (@_) {
313	202		100		413	$_ // next;
314	198				2326	s{^$mark_re\K(?=.$re)(.\R?)}{
315	0				0	$sub1->($1);
316						}mge;
317						}
318	8				36	};
319	8				19	$converter = $sub2;
320
321	8	50			79	if (my $color = $colormap{'VISIBLE'}) {
322	0				0	my $s = ansi_code($color);
323	0				0	my $e = ansi_code($color =~ s/(?=.)/~/gr); # cancel the effect
324						my $symbols = join('',
325	0				0	map { $visible{$_}->[-1] =~ s/\s+//gr }
	0				0
326						@names);
327	0			0	0	$effector = sub { s/([\Q$symbols\E]+)/${s}${1}${e}/g };
	0				0
328						}
329						}
330
331						##
332						## Temporary files
333						##
334	8			8	73	use Command::Run::Tmpfile;
	8				14
	8				54471
335	8				155	my $T1 = Command::Run::Tmpfile->new;
336	8				7853	my $T2 = Command::Run::Tmpfile->new;
337
338						##
339						## Total statistic info
340						##
341	8				2880	my %stat;
342	8				122	@stat{'a', 'd', 'c', 'anl', 'dnl', 'cnl'} = (0, 0, 0, 0, 0, 0);
343	8				40	@stat{'anlb', 'dnlb', 'cnlb'} = (0, 0, 0);
344
345	8	50			221	open(DIFF, $DIFF) \|\| die "$DIFF: $!\n";
346	8				83	binmode DIFF, ":encoding(utf8)";
347
348	8				379	my $stdout = IO::Divert->new;
349	58			58	123	sub sprintln { map { s/(?<=[^\n])\z/\n/r } @_ }
	142				1695
350	58			58	493	sub println { print sprintln @_ }
351						sub print_lxl {
352	0			0	0	my($l1, $l2) = (sprintln($_[0]), sprintln($_[1]));
353	0	0			0	if ($app->style eq 'diff') {
354	0				0	printf "1c1\n". "< $l1". "---\n". "> $l2",
355						} else {
356	0				0	print $l1, $l2;
357						}
358						}
359
360	8				3435	while () {
361						#
362						# --linebyline, --lxl option
363						#
364	94	50			3400	if ($app->linebyline) {
		100
		100
		100
		50
365	0				0	my $old = $_;
366	0		0		0	my $new = // do {
367	0				0	print $old;
368	0				0	next;
369						};
370	0	0			0	compare(\$old, \$new) if $app->unit;
371	0				0	print_lxl(color("OTEXT", $old),
372						color("NTEXT", $new));
373						}
374						#
375						# normal diff
376						#
377						elsif (/^([\d,]+)([adc])([\d,]+)\r?$/) {
378	10				272	my($left, $cmd, $right) = ($1, $2, $3);
379	10				20	my $command_line = $_;
380	10				21	my($old, $del, $new);
381						eval {
382	10	100			62	if ($cmd =~ /[cd]/) {
383	8				51	$old = read_diff(*DIFF, scalar(range($left)));
384	8	50			77	$old =~ /^(?!<)/m and die;
385						}
386	10	100			46	if ($cmd =~ /[c]/) {
387	6				20	$del = read_diff(*DIFF, 1);
388	6	50			29	$del =~ /^(?!---)/m and die;
389						}
390	10	100			72	if ($cmd =~ /[ca]/) {
391	8				32	$new = read_diff(*DIFF, scalar(range($right)));
392	8	50			58	$new =~ /^(?!>)/m and die;
393						}
394	10				35	1;
395						}
396	10	50			22	or do {
397	0		0		0	defined and print for ($command_line, $old, $del, $new);
398	0				0	next;
399						};
400
401	10				50	print_command($command_line);
402
403	10	50			31	if ($converter) {
404	10				151	$converter->(qr/[<>][ \t]/, $old, $new);
405						}
406
407	10	100			54	if ($cmd eq 'c') {
408	6	50			52	compare(\$old, \$new, qr/<[ \t]/, qr/>[ \t]/) if $app->unit;
409						}
410
411	10	50			235	if ($app->color) {
412	10	100			194	$old =~ s{^(<[ \t])(.*)}{
413	12				494	color("OMARK", $1) . color("OTEXT", $2)
414						}mge if $old;
415	10	100			775	$new =~ s{^(>[ \t])(.*)}{
416	12				360	color("NMARK", $1) . color("NTEXT", $2)
417						}mge if $new;
418						}
419
420	10	100	66		815	println $old if $old and $app->show_old;
421	10	100			40	println $del if $del;
422	10	100	66		182	println $new if $new and $app->show_new;
423						}
424						#
425						# diff -c
426						#
427						elsif (/^\\\* ([\d,]+) \\\\\r?$/) {
428	2				39	my $left = $1;
429	2				13	print_command($_);
430	2				4	my(@old, @new);
431	2				13	my $oline = range($left);
432	2				12	@old = read_diffc(*DIFF, $oline);
433	2				69	my $new;
434	2	50	33		23	if (@old and $old[0] =~ /^--- /) {
435	0				0	$new = shift @old;
436	0				0	@old = ("");
437						} else {
438	2				8	$new = ;
439						}
440	2				5	my $dline = map { /^-/mg } @old;
	14				67
441	2	50			25	if ($new =~ /^--- ([\d,]+) ----$/) {
442	2				8	my $right = $1;
443	2				11	my $nline = range($right);
444	2	50	33		79	if (@old == 1 and $old[0] ne "" and $oline - $dline == $nline) {
			33
445	0				0	@new = ("");
446						} else {
447	2				12	@new = read_diffc(*DIFF, $nline);
448						}
449	2	50			9	if ($converter) {
450	2				14	$converter->(qr/[\-\+\!\ ][ \t]/, @old, @new);
451						}
452	2				13	my $mark_re = qr/![ \t]/;
453	2				10	for my $i (keys @old) {
454	14				472	my $cmark = "! ";
455	14	100			54	if ($i % 2) {
456	6	50			66	compare(\$old[$i], \$new[$i], $mark_re) if $app->unit;
457						}
458	14	50			139	if ($app->color) {
459	14				227	$old[$i] =~ s{^([\-\!][ \t])(.*)}{
460	12				515	color("OMARK", $1) . color("OTEXT", $2)
461						}mge;
462	14				526	$new[$i] =~ s{^([\+\!][ \t])(.*)}{
463	12				177	color("NMARK", $1) . color("NTEXT", $2)
464						}mge;
465						}
466						}
467						}
468	2	50			129	println @old if $app->show_old;
469	2				21	println $new;
470	2	50			18	println @new if $app->show_new;
471						}
472						#
473						# diff --combined (generic)
474						#
475						elsif (m{^
476						(? $prefix_re)
477						(?
478						(? \@{2,} ) [ ]
479						(? (?: [-+]\d+(?:,\d+)? [ ] ){2,} )
480						\g{mark}
481						(?s:.*)
482						)
483						}x) {
484	10				404	my($prefix, $command, $lines) = @+{qw(prefix command lines)};
485	10				62	my $column = length $+{mark};
486						my @lines = map {
487	10	50			82	$_ eq ' ' ? 1 : int $_
	20				100
488						} $lines =~ /\d+(?\|,(\d+)\|( ))/g;
489
490	10	50			43	if (@lines != $column) {
491	0				0	print;
492	0				0	next;
493						}
494
495	10				20	my($divert, %read_opt);
496	10	100			29	if ($prefix) {
497	8				21	$read_opt{prefix} = $prefix;
498	8			8	71	$divert = IO::Divert->new(FINAL => sub { s/^/$prefix/mg });
	8				672
499						}
500
501	10				742	print_command($command);
502
503	10				79	my @buf = read_unified \%read_opt, *DIFF, @lines;
504
505	10				23	state @mark_re;
506	10		66		65	my $mark_re = $mark_re[$column] //= do {
507	4				15	my $mark = '.' x ($column - 1);
508	4				125	qr/$mark/;
509						};
510
511	10	50			34	if ($converter) {
512	66				115	map { $converter->($mark_re, @$_) }
513	10				23	map { $_->lists }
	30				69
514						@buf;
515						}
516
517	10				25	for my $buf (@buf) {
518	30				114	my @result = compare_unified($column, $buf, $mark_re);
519	30	50			171	if (@result == 3) {
520	30	50			111	$app->show_new or splice @result, 2, 1;
521	30	50			312	$app->show_old or splice @result, 1, 1;
522						}
523	30				220	println @result;
524						}
525						}
526						#
527						# conflict marker
528						#
529						elsif (/^<<<<<<<\s+(.*)/) {
530						CONFLICT:
531						{
532	0				0	my $c1 = $_;
	0				0
533
534	0			0	0	my @old = read_until { /^=======$/ } *DIFF;
	0				0
535	0		0		0	my $c2 = pop @old // do {
536	0				0	print $c1, @old;
537	0				0	last;
538						};
539
540	0			0	0	my @new = read_until { /^>>>>>>>\s+(.)/ } DIFF;
	0				0
541	0		0		0	my $c3 = pop @new // do {
542	0				0	print $c1, @old, $c2, @new;
543	0				0	last;
544						};
545
546	0			0	0	my $i = first { $old[$_] =~ /^\Q\|\|\|\|\|\|\|\E/ } keys @old;
	0				0
547	0	0			0	my @mrg = defined $i ? splice @old, $i : ();
548
549	0				0	my $old = join '', @old;
550	0				0	my $new = join '', @new;
551	0	0			0	$converter->('', $old, $new) if $converter;
552	0	0			0	compare(\$old, \$new) if $app->unit;
553
554	0	0			0	print $c1 if $app->mark;
555	0	0			0	print color("OTEXT", $old) if $app->show_old;
556	0	0			0	if (@mrg) {
557	0				0	my $c4 = shift @mrg;
558	0	0			0	print $c4 if $app->mark;
559	0				0	my $mrg = join '', @mrg;
560	0	0			0	$converter->('', $mrg) if $converter;
561	0	0			0	print color("MTEXT", $mrg) if $app->show_mrg;
562						}
563	0	0			0	println $c2 if $app->mark;
564	0	0			0	println color("NTEXT", $new) if $app->show_new;
565	0	0			0	println $c3 if $app->mark;
566						}
567						}
568						else {
569	72	50			1888	if ($app->unknown) {
570	72	50			455	if (my $f = $func{UNKNOWN}) {
571	72				134	print $f->($_);
572						} else {
573	0				0	print;
574						}
575						}
576						}
577						}
578						continue {
579	94				2341	$stdout->fh->flush;
580	94				1314	local *_ = $stdout->buffer;
581	94	50			636	if (length) {
582	94				477	$_ = decode 'utf8', $_;
583	94	50			2626	$effector->() if $effector;
584	94				578	STDOUT->printflush($_);
585	94				11039	$stdout->clear;
586						}
587						}
588	8				657	close DIFF;
589	8	50			254	my $exit = $DIFF =~ /\\|$/ ? $? >> 8 : 0;
590
591	8				51	select STDOUT;
592
593	8	50			149	if ($app->stat) {
594	0				0	select STDERR;
595
596	0				0	print("TOTAL: ");
597						printf("append=%d delete=%d change=%d\n",
598	0				0	$stat{'a'}, $stat{'d'}, $stat{'c'});
599	0				0	print("IGNORE WHITESPACE: ");
600						printf("append=%d delete=%d change=%d\n",
601						$stat{'anl'},
602						$stat{'dnl'},
603	0				0	$stat{'cnl'});
604	0				0	print("IGNORE WHITESPACE (bytes): ");
605						printf("append=%d delete=%d change=%d\n",
606						$stat{'anlb'},
607						$stat{'dnlb'},
608	0				0	$stat{'cnlb'});
609						}
610
611	8				240	exit($exit > 1);
612
613						######################################################################
614
615						sub compare_unified {
616	30			30	121	my $column = shift;
617	30	50			78	goto &compare_unified_3 if $column == 3;
618	30				159	goto &compare_unified_generic;
619						}
620
621						sub compare_unified_generic {
622	30			30	69	my($buf, $mark_re) = @_;
623	30				387	my $c = $buf->collect(qr/^[\t ]+$/);
624	30				132	my $o = $buf->collect(qr/[-]/);
625	30				110	my $n = $buf->collect(qr/[+]/);
626	30	50			143	compare(\$o, \$n, $mark_re) if $app->unit;
627	30	50			278	if ($app->color) {
628						map {
629	30				315	my($buf, $m, $t) = @$_;
	90				1226
630	90				618	$$buf =~ s{^($mark_re)(.*)}{
631	154	50			4306	($app->mark ? color($m, $1) : '') . color($t, $2)
632						}mge;
633						} ( [ \$c, 'UMARK', 'UTEXT' ],
634						[ \$o, 'OMARK', 'OTEXT' ],
635						[ \$n, 'NMARK', 'NTEXT' ] );
636						} else {
637	0	0			0	map { s/^$mark_re//mg } $c, $o, $n if not $app->mark;
	0				0
638						}
639	30				1346	($c, $o, $n);
640						}
641
642						sub compare_unified_3 {
643	0			0	0	my($buf, $mark_re) = @_;
644
645	0				0	my @mark = ( " ", "--", "- ", " -", "+ ", " +", "++" );
646	0				0	my %buf = map { $_ => scalar($buf->collect($_)) } @mark;
	0				0
647
648	0	0			0	goto SKIP unless $app->unit;
649
650	0				0	my @r;
651	0				0	$r[0] = compare(\@buf{q/--/, q/++/}, $mark_re);
652	0				0	$r[1] = compare(\@buf{q/- /, q/ -/}, $mark_re);
653	0	0			0	if ($r[1] == 0) {
654	0				0	$r[2] = compare(\@buf{q/- /, q/+ /}, $mark_re);
655	0				0	$r[3] = compare(\@buf{q/ -/, q/ +/}, $mark_re);
656						}
657	0	0			0	if (sum(@r) == 0) {
658	0				0	$r[4] = compare(\@buf{q/- /, q/++/}, $mark_re);
659	0	0			0	$r[5] = compare(\@buf{q/ -/, q/++/}, $mark_re) unless $r[4];
660	0				0	$r[6] = compare(\@buf{q/--/, q/+ /}, $mark_re);
661	0	0			0	$r[7] = compare(\@buf{q/--/, q/ +/}, $mark_re) unless $r[6];
662						}
663
664						SKIP:
665
666	0	0			0	if ($app->color) {
667	0	0			0	map { s/^$mark_re//mg } $buf{q/ /} if not $app->mark;
	0				0
668						map {
669	0				0	my($s, $m, $t) = @$_;
670	0	0			0	$$s =~ s{^($mark_re)(.*)}{
671	0	0			0	($app->mark ? color($m, $1) : '') . color($t, $2)
672						}mge if $$s;
673						} ( [ \$buf{q/--/}, 'CMARK', 'CTEXT' ],
674						[ \$buf{q/- /}, 'OMARK', 'OTEXT' ],
675						[ \$buf{q/ -/}, 'NMARK', 'NTEXT' ],
676						[ \$buf{q/+ /}, 'MMARK', 'MTEXT' ],
677						[ \$buf{q/ +/}, 'MMARK', 'MTEXT' ],
678	0				0	[ \$buf{q/++/}, 'MMARK', 'MTEXT' ] );
679						} else {
680	0	0			0	map { s/^$mark_re//mg } @buf{@mark} if not $app->mark;
	0				0
681						}
682
683	0				0	@buf{$buf->labels};
684						}
685
686						sub print_command {
687	22	50		22	150	return unless $app->command;
688	22				186	my $line = shift;
689	22	50			69	if ($app->color) {
690	22				218	$line = color($colormap{COMMAND}, $line);
691						}
692	22				6109	print $line;
693						}
694
695						sub compare {
696	42			42	539	my($old, $new) = splice @_, 0, 2;
697	42	100	33		489	return 0 unless $old && $new && $$old && $$new;
			66
			100
698
699	24		50		83	my $omark_re = shift \|\| undef;
700	24		66		95	my $nmark_re = shift \|\| $omark_re;
701
702	24				44	my(@omark, @nmark);
703	24	50			68	if ($omark_re) {
704	24				402	$$old =~ s/^($omark_re)/push(@omark, $1), ""/mge;
	32				230
705	24				267	$$new =~ s/^($nmark_re)/push(@nmark, $1), ""/mge;
	32				178
706						}
707
708	24				109	($$old, $$new) = context($$old, $$new);
709
710	24	50			232	$$old =~ s/^/shift @omark/mge if @omark;
	32				195
711	24	50			149	$$new =~ s/^/shift @nmark/mge if @nmark;
	32				121
712
713	24				144	1;
714						}
715
716	8			8	80	use Getopt::EX::Colormap qw(colorize);
	8				20
	8				61010
717
718						sub debug_list {
719	0			0	0	my $i = 0;
720	0				0	my @cmap = qw( K/444 K/333 );
721	0				0	join "", map { colorize $cmap[$i++ % @cmap], $_ } @_;
	0				0
722						}
723
724						sub context {
725	24			24	66	my($old, $new) = @_;
726	24				62	local $_;
727
728	24	50			88	if ($debug{s}) {
729	0				0	print STDERR "****************************** Comparing ...\n";
730	0				0	print STDERR $old;
731	0				0	print STDERR "****************************** and\n";
732	0				0	print STDERR $new;
733	0				0	print STDERR "****************************** .\n";
734						}
735
736	24				110	my @owlist = wordlist($old);
737	24				64	my @nwlist = wordlist($new);
738
739	24	50			86	if ($debug{w}) {
740	0				0	print STDERR "****************************** Comparing ...\n";
741	0				0	print STDERR debug_list @owlist;
742	0				0	print STDERR "****************************** and\n";
743	0				0	print STDERR debug_list @nwlist;
744	0				0	print STDERR "****************************** .\n";
745						}
746
747	24				140	maketmp($T1, @owlist);
748	24				5315	maketmp($T2, @nwlist);
749
750	24				2144	my $diff = sprintf "$app->{subdiff} @sub_diffopts %s %s", $T1->path, $T2->path;
751	24	50			158794	open my $cdif, "$diff \|" or die "diff: $!\n";
752	24				1971	binmode $cdif, ":encoding(utf8)";
753	24				3397	my @command;
754	24				483	my %c = (a => 0, d => 0, c => 0);
755	24				35872	while (<$cdif>) {
756	320	50			1220	warn $_ if $debug{d};
757
758						## Quick hack to read `git diff` unified output
759	320	50			748	if (/^\@\@ \s+ \-(\d+)(?:,(\d+))? \s+ \+(\d+)(?:,(\d+))? \s+ \@\@/x) {
760	0		0		0	my($o, $ol, $n, $nl) = ($1, $2//1, $3, $4//1);
			0
761	0	0			0	my $cmd = ($ol == 0) ? "a" : ($nl == 0) ? "d" : "c";
		0
762	0	0			0	my $old = ($ol <= 1) ? $o : sprintf "%d,%d", $o, $o + $ol - 1;
763	0	0			0	my $new = ($nl <= 1) ? $n : sprintf "%d,%d", $n, $n + $nl - 1;
764	0				0	$_ = sprintf "%s%s%s\n", $old, $cmd, $new;
765						}
766
767	320	100			1776	if (/^[\d,]+([adc])[\d,]+$/) {
768	48				205	push @command, $_;
769	48				591	$c{$1}++;
770						}
771						}
772	24				1655	close $cdif;
773
774	24	50			235	if ($debug{d}) {
775	0				0	printf(STDERR "old=%d new=%d command=%d\n",
776						@owlist+0, @nwlist+0, @command+0);
777	0				0	printf(STDERR "append=$c{a} delete=$c{d} change=$c{c}\n");
778						}
779
780	24				322	my($obuf, $nbuf) = makebuf(\@owlist, \@nwlist, \@command);
781	24				270	my $status = $?>>8;
782	24	50			98	die "Unexpected status of subprocess ($status)\n" if $status > 1;
783
784	24				888	($obuf, $nbuf);
785						}
786
787						sub wordlist {
788	48			48	90	my $text = shift;
789	48	50	33		161	if ($app->unit eq 'mecab' and $text =~ /\P{ASCII}/) {
790	0				0	mecab_words($text);
791						} else {
792	48				496	normal_words($text);
793						}
794						}
795
796						sub normal_words {
797	48			48	77	my $text = shift;
798	48				2143	my @words;
799	48				2406	while ($text =~ /\G($w_pattern)/g) {
800	768				3979	push @words, $1;
801						}
802	48				518	@words;
803						}
804
805						sub mecab_words {
806	0			0	0	my $text = shift;
807	0				0	state $mecab = App::cdif::Command::Mecab->new;
808	0				0	$mecab->wordlist($text);
809						}
810
811						sub maketmp {
812	48			48	318	my($tmpfile, @list) = @_;
813	48				401	$tmpfile->reset;
814	48				9688	for (@list) {
815	768	50	33		11463	s/[ \t]+// if $app->ignore_space_change \|\| $app->ignore_all_space;
816	768				8760	$tmpfile->write($_);
817	768	100			13616	$tmpfile->write("\n") unless /\n\z/;
818						}
819	48				189	$tmpfile->flush->rewind;
820						}
821
822						##
823						## @owlist: old word list
824						## @nwlist: new word list
825						## @command: how different these lists (`diff' result command lines)
826						##
827						sub makebuf {
828	24			24	139	my($ol, $nl, $c) = @_;
829	24				566	my @owlist = @$ol;
830	24				371	my @nwlist = @$nl;
831	24				105	my @command = @$c;
832
833	24				91	my($o, $n) = (0, 0); # pointers
834	24				150	my(@obuf, @nbuf);
835
836	24				94	for (@command) {
837	48	50			617	my($old, $cmd, $new) = /([\d,]+)([adc])([\d,]+)/ or do {
838	0				0	die "Panic! Unexpected diff output";
839						};
840	48				498	my($o1, $o2) = range($old);
841	48				155	my($n1, $n2) = range($new);
842	48				236	map { $_-- } $o1, $o2, $n1, $n2; # make them zero origined
	192				401
843
844	48	100			317	push(@obuf, @owlist[$o .. $o1 - 1]), $o = $o1 if $o < $o1;
845	48	100			282	push(@nbuf, @nwlist[$n .. $n1 - 1]), $n = $n1 if $n < $n1;
846
847	48				220	$stat{$cmd}++;
848
849	48	50			228	if ($cmd eq 'd') {
		50
		0
850	0				0	my $os = join('', @owlist[$o1 .. $o2]);
851	0	0			0	if ($owlist[$o2] =~ /\S/) {
852	0				0	$stat{'dnl'}++;
853	0				0	$stat{'dnlb'} += length($os);
854						}
855	0				0	push(@obuf, $func{DELETE}->($os));
856	0				0	$o = $o2 + 1;
857						}
858						elsif ($cmd eq 'c') {
859	48				256	my $os = join('', @owlist[$o1 .. $o2]);
860	48				287	my $ns = join('', @nwlist[$n1 .. $n2]);
861	48	50	33		476	if (($owlist[$o2] =~ /\S/) \|\| ($nwlist[$n2] =~ /\S/)) {
862	48				121	$stat{'cnl'}++;
863	48				206	$stat{'cnlb'} += length($os);
864	48				123	$stat{'cnlb'} += length($ns);
865						}
866	48				331	push(@obuf, $func{OLD}->($os));
867	48				14572	push(@nbuf, $func{NEW}->($ns));
868	48				3706	$o = $o2 + 1;
869	48				167	$n = $n2 + 1;
870						}
871						elsif ($cmd eq 'a') {
872	0				0	my $ns = join('', @nwlist[$n1 .. $n2]);
873	0	0			0	if ($nwlist[$n2] =~ /\S/) {
874	0				0	$stat{'anl'}++;
875	0				0	$stat{'anlb'} += length($ns);
876						}
877	0				0	push(@nbuf, $func{APPEND}->($ns));
878	0				0	$n = $n2 + 1;
879						}
880						}
881	24				203	push(@obuf, @owlist[$o .. $#owlist]);
882	24				188	push(@nbuf, @nwlist[$n .. $#nwlist]);
883
884	24				405	(join('', @obuf), join('', @nbuf));
885						}
886
887						sub read_diff {
888	22			22	114	my($FH, $c) = @_;
889	22				36	my @buf = ();
890	22				62	while ($c-- > 0) {
891	30				169	push @buf, scalar <$FH>;
892						}
893	22	50			116	wantarray ? @buf : join '', @buf;
894						}
895
896						sub read_diffc {
897	4			4	18	my($FH, $n) = @_;
898	4				6	my @buf;
899	4				26	local $_;
900	4				7	my $i = 0;
901	4		66		36	while ($n-- && ($_ = <$FH>)) {
902	56	100			160	$i++ if ($i % 2) != /^!/;
903	56				109	$buf[$i] .= $_;
904	56	50			277	last if /^--- /;
905						}
906	4		50		12	map { $_ // "" } @buf;
	28				79
907						}
908
909						sub ul {
910	0			0		local $_ = join '', @_;
911	0					s/(.)/["", "_\010", "__\010\010"]->[vwidth($1)].$1/ge;
	0
912	0					$_;
913						}
914						sub bd {
915	0			0		local $_ = join '', @_;
916	0					s/(\S)/$1.["", "\010", "\010\010"]->[vwidth($1)].$1/ge;
	0
917	0					$_;
918						}
919
920						sub wc_l {
921	0			0		my $file = shift;
922	0					my $line;
923	0					$file->rewind;
924	0					$line++ while $file->fh->getline;
925	0					$file->rewind;
926	0					$line;
927						}
928
929						sub eval {
930	0	0	0	0		print STDERR &unctrl($_[0]), "\n" x ($_[0] !~ /\n$/) if $_[1] \|\| $debug{e};
931	0					CORE::eval shift;
932	0	0				die sprintf("eval failed in file %s on line %s\n$@", (caller)[1,2]) if $@;
933						}
934
935						######################################################################
936
937						=head1 NAME
938
939						cdif - word context diff
940
941						=head1 VERSION
942
943						Version 4.44
944
945						=head1 SYNOPSIS
946
947						cdif [option] file1 file2
948
949						cdif [option] [diff-data]
950
951						Options:
952
953						-c, -Cn context diff
954						-u, -Un unified diff
955						-i ignore case
956						-b ignore space change
957						-w ignore whitespace
958						-t expand tabs
959
960						--diff=command specify diff command
961						--subdiff=command specify backend diff command
962						--stat show statistical information
963						--colormap=s specify color map
964						--sdif sdif friendly option
965						--[no]color color or not (default true)
966						--[no]256 ANSI 256 color mode (default true)
967						--[no]cc color command line (default true)
968						--[no]mc color diff mark (default true)
969						--[no]tc color normal text (default true)
970						--[no]uc color unknown text (default true)
971						--[no]old print old text (default true)
972						--[no]new print new text (default true)
973						--[no]mrg print merged text (default true)
974						--[no]command print diff command line (default true)
975						--[no]unknown print unknown line (default true)
976						--[no]mark print mark or not (default true)
977						--[no]prefix read git --graph output (default true)
978						--unit=s word/letter/char/mecab (default word)
979						--[no]mecab use mecab tokenizer (default false)
980						--prefix-pattern prefix pattern
981						--visible char=? set visible attributes
982						--[no]lenience suppress unexpected input warning (default true)
983						--lxl compare input data line-by-line
984						--style=diff print --lxl output in diff style
985						--version show version
986
987
988						=head1 DESCRIPTION
989
990						B is a post-processor of the Unix diff command. It highlights
991						deleted, changed and added words based on word context (B<--unit=word>
992						by default). If you want to compare text character-by-character, use
993						option B<--unit=char>. Option B<--unit=mecab> tells to use external
994						B command as a tokenizer for Japanese text.
995
996						If single or no file is specified, B reads that file or STDIN as
997						an output from diff command. In addition to normal diff, context
998						diff, and unified (combined) diff, the L-compatible conflict
999						marker format is supported as input format.
1000
1001						Lines that don't look like diff output are simply ignored and
1002						printed.
1003
1004						=head2 STARTUP and MODULE
1005
1006						B utilizes Perl L module, and reads I<~/.cdifrc>
1007						file if available when starting up. You can define original and
1008						default option there. The next line enables B<--mecab> option and adds
1009						crossed-out effect for deleted words.
1010
1011						option default --mecab --cm DELETE=+X
1012
1013						Modules under B can be loaded by B<-M> option without
1014						prefix. The next command loads B module.
1015
1016						$ cdif -Mcolors
1017
1018						You can also define options in module file. Read `perldoc
1019						Getopt::EX::Module` for detail.
1020
1021						=head2 COLOR
1022
1023						Each line is displayed in different colors. Each text segment has
1024						own labels, and color for them can be specified by B<--colormap>
1025						option. Read `perldoc Getopt::EX::Colormap` for detail.
1026
1027						Standard module B<-Mcolors> is loaded by default, and defines several
1028						color maps for light and dark screen. If you want to use CMY colors in
1029						dark screen, place next line in your F<~/.cdifrc>.
1030
1031						option default --dark-cmy
1032
1033						Option B<--autocolor> is defined in B module to call
1034						L module. It sets B<--light> or B<--dark>
1035						option according to the brightness of the terminal screen. You can
1036						set preferred color in your F<~/.cdifrc> like:
1037
1038						option --light --cmy
1039						option --dark --dark-cmy
1040
1041						Automatic setting is done by L module and it
1042						works with macOS Terminal.app and iTerm.app, and other XTerm
1043						compatible terminals. This module accepts environment variable
1044						L as a terminal background color. For example, use
1045						C<000> or C<#000000> for black and C<555> or C<#FFFFFF> for white.
1046
1047						Option B<--autocolor> is set by default, so override it to do nothing
1048						to disable.
1049
1050						option --autocolor --nop
1051
1052						=head2 EXIT STATUS
1053
1054						B always exits with status zero unless an error occurs.
1055
1056						=head1 OPTIONS
1057
1058						=over 7
1059
1060						=item B<->[B]
1061
1062						Almost same as B command.
1063
1064						=begin deprecated
1065
1066						=item B<--rcs>, B<-r>I, B<-q>
1067
1068						Use rcsdiff instead of normal diff. Option B<--rcs> is not required
1069						when B<-r>I is supplied.
1070
1071						=end deprecated
1072
1073						=item B<-->B=[C,C,C,C,C<0>,C<>]
1074
1075						=item B<-->B=[C,C,C,C,C<0>,C<>]
1076
1077						Specify the comparing unit. Default is I and compare each line
1078						word-by-word. Specify C if you want to compare them
1079						character-by-character. Unit C is almost same as C but
1080						does not include underscore.
1081
1082						When C is given as an unit, B command is called as a
1083						tokenizer for non-ASCII text. ASCII text is compared word-by-word.
1084						External B command has to be installed.
1085
1086						If you give empty string like C<--unit=>, or C<0>, B does not
1087						compare text in any way. You'll still get colorization effect.
1088
1089						=item B<--mecab>
1090
1091						Shortcut for B<--unit=mecab>.
1092
1093						=item B<--diff>=I
1094
1095						Specify the diff command to use.
1096
1097						=item B<--subdiff>=I
1098
1099						Specify the backend diff command to get word differences. Accept
1100						normal and unified diff format.
1101
1102						If you want to use B command, don't forget to set I<-U0>
1103						option.
1104
1105						--subdiff="git diff -U0 --no-index --histogram"
1106
1107						=item B<-->[B]B
1108
1109						Use ANSI color escape sequence for output.
1110
1111						=item B<--colormap>=I, B<--cm>=I
1112
1113						Basic I format is :
1114
1115						FIELD=COLOR
1116
1117						where the FIELD is one from these :
1118
1119						COMMAND Command line
1120						OMARK Old mark
1121						NMARK New mark
1122						UTEXT Same text
1123						OTEXT Old text
1124						NTEXT New text
1125						OCHANGE Old change part
1126						NCHANGE New change part
1127						APPEND Appended part
1128						DELETE Deleted part
1129
1130						=for comment
1131						VISIBLE Visualized invisible chars
1132
1133						and additional I and I FIELDs for git-diff combined
1134						format.
1135
1136						CMARK Common mark
1137						CTEXT Common text
1138						MMARK Merged mark
1139						MTEXT Merged text
1140
1141						You can make multiple fields same color joining them by = :
1142
1143						FIELD1=FIELD2=...=COLOR
1144
1145						Also wildcard can be used for field name :
1146
1147						*CHANGE=BDw
1148
1149						Multiple fields can be specified by repeating options
1150
1151						--cm FIELD1=COLOR1 --cm FIELD2=COLOR2 ...
1152
1153						or combined with comma (,) :
1154
1155						--cm FIELD1=COLOR1,FIELD2=COLOR2, ...
1156
1157						Color specification is a combination of single uppercase character
1158						representing 8 colors :
1159
1160						R Red
1161						G Green
1162						B Blue
1163						C Cyan
1164						M Magenta
1165						Y Yellow
1166						K Black
1167						W White
1168
1169						and alternative (usually brighter) colors in lowercase :
1170
1171						r, g, b, c, m, y, k, w
1172
1173						or RGB values and 24 grey levels if using ANSI 256 or full color
1174						terminal :
1175
1176						(255,255,255) : 24bit decimal RGB colors
1177						#000000 .. #FFFFFF : 24bit hex RGB colors
1178						#000 .. #FFF : 12bit hex RGB 4096 colors
1179						000 .. 555 : 6x6x6 RGB 216 colors
1180						L00 .. L25 : Black (L00), 24 grey levels, White (L25)
1181
1182						or color names enclosed by angle bracket :
1183
1184
1185
1186
1187
1188						with other special effects :
1189
1190						D Double-struck (boldface)
1191						I Italic
1192						U Underline
1193						S Stand-out (reverse video)
1194
1195						Above color spec is simplified summary so if you want complete
1196						information, read L.
1197
1198						Defaults are :
1199
1200						COMMAND => "555/222E"
1201						OMARK => "CS"
1202						NMARK => "MS"
1203						UTEXT => ""
1204						OTEXT => "C"
1205						NTEXT => "M"
1206						OCHANGE => "K/445"
1207						NCHANGE => "K/445"
1208						DELETE => "K/544"
1209						APPEND => "K/544"
1210
1211						CMARK => "GS"
1212						MMARK => "YS"
1213						CTEXT => "G"
1214						MTEXT => "Y"
1215
1216						This is equivalent to :
1217
1218						cdif --cm 'COMMAND=555/222E,OMARK=CS,NMARK=MS' \
1219						--cm 'UTEXT=,OTEXT=C,NTEXT=M,*CHANGE=BD/445,DELETE=APPEND=RD/544' \
1220						--cm 'CMARK=GS,MMARK=YS,CTEXT=G,MTEXT=Y'
1221
1222						=item B<--colormap>=C<&func>
1223
1224						=item B<--colormap>=C
1225
1226						You can also set the name of perl subroutine name or definition to be
1227						called handling matched words. Target word is passed as variable
1228						C<$_>, and the return value of the subroutine will be displayed.
1229
1230						Next option produces L-like formatted output.
1231
1232						--cm '*'= \
1233						--cm DELETE=OCHANGE='sub{"[-$_-]"}' \
1234						--cm APPEND=NCHANGE='sub{"{+$_+}"}'
1235
1236						See L for detail.
1237
1238						=item B<-->[B]B, B<-->[B]B
1239
1240						=item B<-->[B]B, B<-->[B]B
1241
1242						=item B<-->[B]B, B<-->[B]B
1243
1244						=item B<-->[B]B, B<-->[B]B
1245
1246						Enable/Disable using color for the corresponding field.
1247
1248						=item B<--sdif>
1249
1250						Disable options appropriate to use for B's input:
1251						B<--commandcolor>, B<--markcolor>, B<--textcolor> and
1252						B<--unknowncolor>.
1253
1254						=item B<-->[B]B, B<-->[B]B, B<-->[B]B
1255
1256						Print or not old/new/mrg text in diff output.
1257
1258						=item B<-->[B]B
1259
1260						Print or not command lines preceding diff output.
1261
1262						=item B<-->[B]B
1263
1264						Print or not lines that do not look like diff output.
1265
1266						=item B<-->[B]B
1267
1268						Print or not marks at the top of diff output lines. At this point,
1269						this option is effective only for unified diff.
1270
1271						The next example produces output exactly the same as I except for
1272						visual effects.
1273
1274						cdif -U100 --no-mark --no-old --no-command --no-unknown old new
1275
1276						These options are prepared for watchdiff(1) command.
1277
1278						=item B<-->[B]B
1279
1280						Understand prefix for diff output including B B<--graph> option.
1281						True by default.
1282
1283						=item B<--prefix-pattern>=I
1284
1285						Specify prefix pattern in regex. Default pattern is:
1286
1287						(?:\\| )(?: )
1288
1289						This pattern matches B graph style and whitespace indented diff
1290						output.
1291
1292						=item B<--visible> I=[0,1]
1293
1294						Set visible attribute for specified characters. Visible character is
1295						converted to corresponding Unicode symbol character. Default visible:
1296						nul, bel, bs, vt, np, cr, esc, del, and all non-breaking/special
1297						spaces. Default invisible: ht, nl, sp.
1298
1299						NAME CODE Unicode NAME DEFAULT
1300						------ ------ -------------------------------- -------
1301						nul \000 SYMBOL FOR NULL YES
1302						soh \001 SYMBOL FOR SOH YES
1303						bel \007 SYMBOL FOR BELL YES
1304						bs \010 SYMBOL FOR BACKSPACE YES
1305						ht \011 SYMBOL FOR HORIZONTAL TABULATION NO
1306						nl \012 SYMBOL FOR NEWLINE NO
1307						vt \013 SYMBOL FOR VERTICAL TABULATION YES
1308						np \014 SYMBOL FOR FORM FEED YES
1309						cr \015 SYMBOL FOR CARRIAGE RETURN YES
1310						esc \033 SYMBOL FOR ESCAPE YES
1311						sp \040 SYMBOL FOR SPACE NO
1312						del \177 SYMBOL FOR DELETE YES
1313						nbsp \240 OPEN BOX (No-Break Space) YES
1314						nnbsp U+202F OPEN BOX (Narrow No-Break Space) YES
1315						ensp U+2002 OPEN BOX (En Space) YES
1316						emsp U+2003 OPEN BOX (Em Space) YES
1317						thinsp U+2009 OPEN BOX (Thin Space) YES
1318						hairsp U+200A OPEN BOX (Hair Space) YES
1319						zwsp U+200B OPEN BOX (Zero Width Space) YES
1320						idesp U+3000 OPEN BOX (Ideographic Space) YES
1321
1322						Since there are no dedicated Unicode symbols for nbsp and other
1323						special space characters, OPEN BOX is used instead.
1324
1325						Multiple characters can be specified at once, by assembling them by
1326						comma (C<,>) like C<--visible ht=1,sp=1>; or connecting them by equal
1327						sign (C<=>) like C<--visible ht=sp=1>. Character names accept
1328						wildcard; C<--visible '*=1'>.
1329
1330						=begin comment
1331
1332						Colormap label C is applied to those characters. Default
1333						setting is C~~, and visible characters are displayed in reverse~~
1334						video. Unlike other colormaps, only special effects can be set to
1335						this label. Effect C (double-struck) is exception (See
1336						L).
1337
1338						=end comment
1339
1340						B command also supports B<--visible> option for horizontal tab
1341						with better visibility.
1342
1343						=begin deprecated
1344
1345						=item B<-->[B]B
1346
1347						=item B<-->[B]B
1348
1349						Set CARRIAGE-RETURN and ESCAPE visible attributes. These options will
1350						be deprecated soon. Use B<--visible> option instead.
1351
1352						=end deprecated
1353
1354						=item B<--stat>
1355
1356						Print statistical information at the end of output. It shows number
1357						of total appended/deleted/changed words in the context of cdif. It's
1358						common to have many insertions and deletions of newlines because of
1359						text filling process. So normal information is followed by modified
1360						number which ignores insert/delete newlines.
1361
1362						=item B<-->[B]B
1363
1364						Suppress warning message for unexpected input from diff command. True
1365						by default.
1366
1367						=item B<--linebyline>, B<--lxl>
1368
1369						=item B<--style>=I