File Coverage

script/cdif

Criterion	Covered	Total	%
statement	447	623	71.7
branch	125	264	47.3
condition	31	78	39.7
subroutine	47	76	61.8
pod			n/a
total	650	1041	62.4

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	9			9	36820	use v5.14;
	9				24
12	9			9	34	use warnings;
	9				10
	9				437
13
14	9			9	3612	use utf8;
	9				2025
	9				38
15	9			9	259	use Carp;
	9				13
	9				716
16	9			9	4371	use Encode;
	9				129826
	9				706
17	9			9	3836	use Encode::Guess;
	9				29552
	9				28
18
19	9			9	4734	use Pod::Usage;
	9				372757
	9				1257
20	9			9	110	use List::Util qw(first sum pairmap);
	9				21
	9				1109
21	9			9	4326	use Text::ParseWords qw(shellwords);
	9				11594
	9				477
22	9			9	3375	use IO::Divert;
	9				5709
	9				305
23	9			9	16614	use Text::VisualWidth::PP qw(vwidth);
	9				27847
	9				578
24	9			9	4515	use Data::Dumper; {
	9				55653
	9				564
25	9			9	1403098	no warnings 'redefine';
	9				84
	9				12
	9				858
26	9			0	146	*Data::Dumper::qquote = sub { qq["${\(shift)}"] };
	0				0
	0				0
27	9				27	$Data::Dumper::Terse = 1;
28	9				23	$Data::Dumper::Sortkeys = 1;
29	9				24	$Data::Dumper::Useperl = 1;
30						}
31
32	9			9	3815	use App::sdif;
	9				19
	9				599
33	9				29	our $VERSION = $App::sdif::VERSION;
34
35	9			9	3366	use App::sdif::Util;
	9				24
	9				1041
36	9			9	3202	use App::cdif::Command::Mecab;
	9				27
	9				343
37
38						##
39						## options
40						##
41
42	9			9	3679	use charnames ':full';
	9				61938
	9				44
43	9				516	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	9				144	my %opt_visible = do {
66	9				85	map { $_ => splice @{$visible{$_}}, 0, 1 }
	180				161
	180				320
67						keys %visible;
68						};
69
70	9				44	my @rcsopt;
71						my @diffopts;
72	9				0	my @sub_diffopts;
73
74	9			9	378	binmode STDOUT, ":encoding(utf8)";
	9				7167
	9				167
	9				62
75	9				9059	binmode STDERR, ":encoding(utf8)";
76
77	9	50			239	if (my $env = $ENV{'CDIFOPTS'}) {
78	0				0	unshift(@ARGV, shellwords($env));
79						}
80
81	9				18	my $unit_default = 'word';
82	9				68	my $prefix_default = q/(?:\\\| )(?: )/;
83
84	9				16	my $app;
85
86	9			9	202960	use Getopt::EX::Hashed; {
	9				57082
	9				51
87
88	9				20	Getopt::EX::Hashed->configure( DEFAULT => [ is => 'rw' ] ) ;
	9				114
89
90	9				382	has help => ' ' ;
91	9				590	has version => ' ' ;
92	9				425	has man => ' ' ;
93	9				304	has debug => ' d =s@ ' , default => [] ;
94	9				327	has diff => ' =s ' ;
95	9				303	has subdiff => ' =s ' , default => 'diff' ;
96	9				317	has color => ' ! ' , default => 1 ;
97	9				324	has 256 => ' ! ' , default => 1 ;
98	9				337	has commandcolor => ' cc ! ' , default => 1 ;
99	9				331	has markcolor => ' mc ! ' , default => 1 ;
100	9				361	has textcolor => ' tc ! ' , default => 1 ;
101	9				377	has unknowncolor => ' uc ! ' , default => 1 ;
102	9				432	has colormap => ' cm =s@ ' , default => [] ;
103	9				403	has colordump => ' ' ;
104	9				369	has sdif => ' ' ;
105	9				374	has stat => ' ! ' ;
106	9				433	has unit => ' by :s ' , default => $unit_default ,
107						, any => [ qw(char word letter mecab 0), '' ];
108	9				460	has show_old => ' ! ' , default => 1 , alias => 'old' ;
109	9				434	has show_new => ' ! ' , default => 1 , alias => 'new' ;
110	9				451	has show_mrg => ' ! ' , default => 1 , alias => 'mrg' ;
111	9				467	has command => ' ! ' , default => 1 ;
112	9				426	has unknown => ' ! ' , default => 1 ;
113	9				424	has mark => ' ! ' , default => 1 ;
114	9				450	has prefix => ' ! ' , default => 1 ;
115	9				478	has prefix_pattern => ' =s ' , default => $prefix_default ;
116	9				504	has visible => ' =s@ ' , default => [] ;
117	9				520	has lenience => ' ! ' , default => 1 ;
118	9				512	has limit => ' =s% ' , default => { length => 1000 } ;
119	9				488	has linebyline => ' :2 ' , alias => 'lxl';
120	9				521	has style => ' =s ' , default => '' ;
121
122	9				486	has ignore_case => 'i' ;
123	9				481	has ignore_space_change => 'b' ;
124	9				484	has ignore_all_space => 'w' ;
125	9				519	has expand_tabs => 't' ;
126
127	9				582	has rcs => '' ;
128
129						has '+sdif' => sub {
130	4			4	38174	$app->commandcolor = 0;
131	4				30	$app->markcolor = 0;
132	4				20	$app->textcolor = 0;
133	4				42	$app->unknowncolor = 0;
134	9				598	} ;
135						has mecab => '!', action => sub {
136	0	0		0	0	$app->unit = $_[1] ? 'mecab' : $unit_default;
137	9				478	} ;
138	9			0	573	has visible_cr => ' vcr ! ' , action => sub { $opt_visible{cr} = $_[1] } ;
	0				0
139	9			0	559	has visible_esc => ' vesc ! ' , action => sub { $opt_visible{esc} = $_[1] } ;
	0				0
140	9			0	690	has c => ' ' , action => sub { push @diffopts, "-c" } ;
	0				0
141	9			0	784	has u => ' ' , action => sub { push @diffopts, "-u" } ;
	0				0
142	9			0	631	has context => ' C =i ' , action => sub { push @diffopts, "-C" . $_[1] } ;
	0				0
143	9			0	596	has unified => ' U =i ' , action => sub { push @diffopts, "-U" . $_[1] } ;
	0				0
144	9			0	608	has r => ' =s ' , action => sub { push @rcsopt, "-r$_[1]" } ;
	0				0
145	9			0	652	has q => ' ' , action => sub { push @rcsopt, "-q" } ;
	0				0
146
147						has '+help' => sub {
148	0			0	0	pod2usage
149						-verbose => 99,
150						-sections => [ qw(SYNOPSIS VERSION) ];
151	9				647	};
152						has '+man' => sub {
153	0			0	0	pod2usage -verbose => 2;
154	9				307	};
155						has '+version' => sub {
156	0			0	0	print "$VERSION\n";
157	0				0	exit;
158	9				416	};
159
160	9			9	6495	} no Getopt::EX::Hashed;
	9				14
	9				34
161
162	9	50			284	$app = Getopt::EX::Hashed->new() or die;
163
164	9			9	4365	use Getopt::EX::Long qw(:DEFAULT Configure ExConfigure);
	9				689399
	9				4629
165	9				11874	ExConfigure BASECLASS => [ "App::cdif", "Getopt::EX" ];
166	9				342	Configure("bundling");
167	9	50			446	$app->getopt \|\| usage({status => 1});
168
169	9				50543	my %debug = map { $_ => 1 } map { split // } @{$app->debug};
	0				0
	0				0
	9				37
170
171	9				73	$App::sdif::Util::NO_WARNINGS = $app->lenience;
172	9	50			54	$App::cdif::Command::Mecab::debug = 1 if $debug{m};
173
174	9	50			56	$app->rcs++ if @rcsopt;
175	9		33		25	my $diff = $app->diff \|\| ($app->rcs ? 'rcsdiff' : 'diff');
176
177	9				126	push @diffopts, map { $_->[1] } grep { $_->[0] } (
	0				0
	36				160
178						[ $app->ignore_case, "-i" ],
179						[ $app->ignore_space_change, "-b" ],
180						[ $app->ignore_all_space, "-w" ],
181						[ $app->expand_tabs, "-t" ],
182						);
183
184	9				65	push @sub_diffopts, map { $_->[1] } grep { $_->[0] } (
	0				0
	18				91
185						[ $app->ignore_case, "-i" ],
186						[ $app->ignore_all_space, "-w" ],
187						);
188
189	9				29	my %colormap = do {
190	9	50			67	my $col = $app->{256} ? 0 : 1;
191	153	100		153	359	pairmap { $a => (ref $b eq 'ARRAY') ? $b->[$col] : $b } (
192	9				137	UNKNOWN => "" ,
193						UMARK => [ "/444" , "/w" ] ,
194						CMARK => "GS" ,
195						OMARK => "CS" ,
196						NMARK => "MS" ,
197						MMARK => "YS" ,
198						UTEXT => "" ,
199						CTEXT => "G" ,
200						OTEXT => "C" ,
201						NTEXT => "M" ,
202						MTEXT => "Y" ,
203						COMMAND => [ "555/222E" , "w/kE" ] ,
204						OCHANGE => [ "K/445" , "B/w" ] ,
205						NCHANGE => [ "K/445" , "B/w" ] ,
206						DELETE => [ "K/544" , "R/w" ] ,
207						APPEND => [ "K/544" , "R/w" ] ,
208						VISIBLE => "",
209						);
210						};
211
212	9			9	76	use Getopt::EX::Colormap qw(ansi_code);
	9				12
	9				7370
213						my $color_handler = Getopt::EX::Colormap
214						->new(HASH => \%colormap)
215	9				171	->load_params(@{$app->colormap});
	9				603
216
217	9				245	for (
218						[ $app->unknowncolor => q/UNKNOWN/ ],
219						[ $app->commandcolor => q/COMMAND/ ],
220						[ $app->markcolor => q/MARK/ ],
221						[ $app->textcolor => q/TEXT/ ],
222						) {
223	36				150	my($color, $label) = @$_;
224	36	100			60	$color and next;
225	16				258	for (grep /$label/, keys %colormap) {
226	48				78	$colormap{$_} = '';
227						}
228						}
229
230	9	50			41	warn 'colormap = ', Dumper \%colormap if $debug{c};
231
232	9	50			24	if ($app->colordump) {
233	0				0	print $color_handler->colormap(
234						name => '--changeme', option => '--colormap');
235	0				0	exit;
236						}
237
238						sub color {
239	594			594	8659	$color_handler->color(@_);
240						}
241
242	9				42	my $prefix_re = do {
243	9	50			26	if ($app->prefix) {
244	9				193	qr/$app->{prefix_pattern}/;
245						} else {
246	0				0	"";
247						}
248						};
249
250	9				32	my $DIFF;
251						my $OLD;
252	9				0	my $NEW;
253
254	9	50			23	if ($app->rcs) {
		50
		50
255	0		0		0	my $rcsfile = shift \|\| usage("No RCS filename\n\n");
256	0				0	$DIFF = "$diff @diffopts @rcsopt $rcsfile\|";
257						} elsif (@ARGV == 2) {
258	0				0	($OLD, $NEW) = splice(@ARGV, 0, 2);
259	0				0	$DIFF = "$diff @diffopts $OLD $NEW \|";
260						} elsif (@ARGV < 2) {
261	9		100		121	$DIFF = shift \|\| '-';
262						} else {
263	0	0			0	usage("Arguments error.\n\n") if @ARGV;
264						}
265	9	50			33	warn "DIFF = \"$DIFF\"\n" if $debug{f};
266
267	9				12	my %func = do {
268	9	100			26	my $col = $app->color ? 0 : 1;
269	45			45	96	pairmap { $a => $b->[$col] } (
270	0			0	0	DELETE => [ sub { color("DELETE", @_) }, \&bd ],
271	0			0	0	APPEND => [ sub { color("APPEND", @_) }, \&bd ],
272	48			48	287	OLD => [ sub { color("OCHANGE", @_) }, \&ul ],
273	48			48	92	NEW => [ sub { color("NCHANGE", @_) }, \&ul ],
274	9			72	161	UNKNOWN => [ sub { color("UNKNOWN", @_) }, undef ],
	72				154
275						);
276						};
277
278	9				36	my $w_pattern = do {
279	9	50			25	if ($app->unit =~ /^char/) {
280	0				0	qr/\X/s;
281						} else {
282	9	50			59	my $w = $app->unit eq 'letter' ? '' : '_';
283	9				1598	qr{
284						\p{Han} \| \p{InHiragana}+ \| \p{InKatakana}+ \|
285						[$w\p{Latin}]+ \|
286						[$w\p{Hang}]+ \|
287						[$w\p{Cyrillic}]+ \|
288						[$w\p{Arabic}]+ \|
289						[$w\p{Thai}]+ \|
290						\d+ \|
291						# (\p{Punct})\g{-1}* \|
292						[\h\r\f]\n \| \s+ \| (\X)\g{-1}
293						}x;
294						}
295						};
296
297						##
298						## Converter/Effector function for visible characters
299						##
300	9				26	my($converter, $effector);
301	9			9	64	use Getopt::EX::LabeledParam;
	9				16
	9				5164
302						Getopt::EX::LabeledParam
303						->new(HASH => \%opt_visible)
304	9				43	->load_params (@{$app->visible});
	9				211
305	9	50			272	if (my @names = grep $opt_visible{$_}, keys %opt_visible) {
306	9				149	my @chars = map $visible{$_}->[0], @names;
307	9				26	my %hash = map { @$_ } values %visible;
	180				324
308	9				21	my $re = do { local $" = ''; qr/[\Q@chars\E]/ };
	9				20
	9				245
309	9			0	35	my $sub0 = sub { s/($re)/$hash{$1}/g };
	0				0
310	9			0	22	my $sub1 = sub { $_[0] =~ s/($re)/$hash{$1}/gr };
	0				0
311						my $sub2 = sub {
312	81			81	100	my $mark_re = shift;
313	81				126	for (@_) {
314	204		100		294	$_ // next;
315	200				1699	s{^$mark_re\K(?=.$re)(.\R?)}{
316	0				0	$sub1->($1);
317						}mge;
318						}
319	9				29	};
320	9				27	$converter = $sub2;
321
322	9	50			62	if (my $color = $colormap{'VISIBLE'}) {
323	0				0	my $s = ansi_code($color);
324	0				0	my $e = ansi_code($color =~ s/(?=.)/~/gr); # cancel the effect
325						my $symbols = join('',
326	0				0	map { $visible{$_}->[-1] =~ s/\s+//gr }
	0				0
327						@names);
328	0			0	0	$effector = sub { s/([\Q$symbols\E]+)/${s}${1}${e}/g };
	0				0
329						}
330						}
331
332						##
333						## Temporary files
334						##
335	9			9	78	use Command::Run::Tmpfile;
	9				15
	9				40366
336	9				111	my $T1 = Command::Run::Tmpfile->new;
337	9				5104	my $T2 = Command::Run::Tmpfile->new;
338
339						##
340						## Total statistic info
341						##
342	9				2339	my %stat;
343	9				173	@stat{'a', 'd', 'c', 'anl', 'dnl', 'cnl'} = (0, 0, 0, 0, 0, 0);
344	9				35	@stat{'anlb', 'dnlb', 'cnlb'} = (0, 0, 0);
345
346	9	50			193	open(DIFF, $DIFF) \|\| die "$DIFF: $!\n";
347	9				206	binmode DIFF, ":encoding(utf8)";
348
349	9				302	my $stdout = IO::Divert->new;
350	59			59	94	sub sprintln { map { s/(?<=[^\n])\z/\n/r } @_ }
	145				1420
351	59			59	355	sub println { print sprintln @_ }
352						sub print_lxl {
353	0			0	0	my($l1, $l2) = (sprintln($_[0]), sprintln($_[1]));
354	0	0			0	if ($app->style eq 'diff') {
355	0				0	printf "1c1\n". "< $l1". "---\n". "> $l2",
356						} else {
357	0				0	print $l1, $l2;
358						}
359						}
360
361	9				2607	while () {
362						#
363						# --linebyline, --lxl option
364						#
365	97	50			3113	if ($app->linebyline) {
		100
		100
		100
		50
366	0				0	my $old = $_;
367	0		0		0	my $new = // do {
368	0				0	print $old;
369	0				0	next;
370						};
371	0	0			0	compare(\$old, \$new) if $app->unit;
372	0				0	print_lxl(color("OTEXT", $old),
373						color("NTEXT", $new));
374						}
375						#
376						# normal diff
377						#
378						elsif (/^([\d,]+)([adc])([\d,]+)\r?$/) {
379	10				225	my($left, $cmd, $right) = ($1, $2, $3);
380	10				15	my $command_line = $_;
381	10				14	my($old, $del, $new);
382						eval {
383	10	100			44	if ($cmd =~ /[cd]/) {
384	8				34	$old = read_diff(*DIFF, scalar(range($left)));
385	8	50			69	$old =~ /^(?!<)/m and die;
386						}
387	10	100			40	if ($cmd =~ /[c]/) {
388	6				14	$del = read_diff(*DIFF, 1);
389	6	50			29	$del =~ /^(?!---)/m and die;
390						}
391	10	100			36	if ($cmd =~ /[ca]/) {
392	8				25	$new = read_diff(*DIFF, scalar(range($right)));
393	8	50			29	$new =~ /^(?!>)/m and die;
394						}
395	10				26	1;
396						}
397	10	50			26	or do {
398	0		0		0	defined and print for ($command_line, $old, $del, $new);
399	0				0	next;
400						};
401
402	10				51	print_command($command_line);
403
404	10	50			23	if ($converter) {
405	10				99	$converter->(qr/[<>][ \t]/, $old, $new);
406						}
407
408	10	100			42	if ($cmd eq 'c') {
409	6	50			53	compare(\$old, \$new, qr/<[ \t]/, qr/>[ \t]/) if $app->unit;
410						}
411
412	10	50			143	if ($app->color) {
413	10	100			126	$old =~ s{^(<[ \t])(.*)}{
414	12				264	color("OMARK", $1) . color("OTEXT", $2)
415						}mge if $old;
416	10	100			395	$new =~ s{^(>[ \t])(.*)}{
417	12				246	color("NMARK", $1) . color("NTEXT", $2)
418						}mge if $new;
419						}
420
421	10	100	66		462	println $old if $old and $app->show_old;
422	10	100			33	println $del if $del;
423	10	100	66		48	println $new if $new and $app->show_new;
424						}
425						#
426						# diff -c
427						#
428						elsif (/^\\\* ([\d,]+) \\\\\r?$/) {
429	2				33	my $left = $1;
430	2				12	print_command($_);
431	2				3	my(@old, @new);
432	2				11	my $oline = range($left);
433	2				9	@old = read_diffc(*DIFF, $oline);
434	2				3	my $new;
435	2	50	33		13	if (@old and $old[0] =~ /^--- /) {
436	0				0	$new = shift @old;
437	0				0	@old = ("");
438						} else {
439	2				5	$new = ;
440						}
441	2				3	my $dline = map { /^-/mg } @old;
	14				22
442	2	50			20	if ($new =~ /^--- ([\d,]+) ----$/) {
443	2				37	my $right = $1;
444	2				6	my $nline = range($right);
445	2	50	33		9	if (@old == 1 and $old[0] ne "" and $oline - $dline == $nline) {
			33
446	0				0	@new = ("");
447						} else {
448	2				5	@new = read_diffc(*DIFF, $nline);
449						}
450	2	50			5	if ($converter) {
451	2				9	$converter->(qr/[\-\+\!\ ][ \t]/, @old, @new);
452						}
453	2				6	my $mark_re = qr/![ \t]/;
454	2				8	for my $i (keys @old) {
455	14				331	my $cmark = "! ";
456	14	100			30	if ($i % 2) {
457	6	50			26	compare(\$old[$i], \$new[$i], $mark_re) if $app->unit;
458						}
459	14	50			89	if ($app->color) {
460	14				144	$old[$i] =~ s{^([\-\!][ \t])(.*)}{
461	12				301	color("OMARK", $1) . color("OTEXT", $2)
462						}mge;
463	14				301	$new[$i] =~ s{^([\+\!][ \t])(.*)}{
464	12				126	color("NMARK", $1) . color("NTEXT", $2)
465						}mge;
466						}
467						}
468						}
469	2	50			179	println @old if $app->show_old;
470	2				12	println $new;
471	2	50			14	println @new if $app->show_new;
472						}
473						#
474						# diff --combined (generic)
475						#
476						elsif (m{^
477						(? $prefix_re)
478						(?
479						(? \@{2,} ) [ ]
480						(? (?: [-+]\d+(?:,\d+)? [ ] ){2,} )
481						\g{mark}
482						(?s:.*)
483						)
484						}x) {
485	11				431	my($prefix, $command, $lines) = @+{qw(prefix command lines)};
486	11				66	my $column = length $+{mark};
487						my @lines = map {
488	11	100			122	$_ eq ' ' ? 1 : int $_
	22				178
489						} $lines =~ /\d+(?\|,(\d+)\|( ))/g;
490
491	11	50			45	if (@lines != $column) {
492	0				0	print;
493	0				0	next;
494						}
495
496	11				19	my($divert, %read_opt);
497	11	100			27	if ($prefix) {
498	8				22	$read_opt{prefix} = $prefix;
499	8			8	121	$divert = IO::Divert->new(FINAL => sub { s/^/$prefix/mg });
	8				537
500						}
501
502	11				762	print_command($command);
503
504	11				73	my @buf = read_unified \%read_opt, *DIFF, @lines;
505
506	11				18	state @mark_re;
507	11		66		49	my $mark_re = $mark_re[$column] //= do {
508	5				16	my $mark = '.' x ($column - 1);
509	5				110	qr/$mark/;
510						};
511
512	11	50			30	if ($converter) {
513	69				118	map { $converter->($mark_re, @$_) }
514	11				24	map { $_->lists }
	31				85
515						@buf;
516						}
517
518	11				23	for my $buf (@buf) {
519	31				105	my @result = compare_unified($column, $buf, $mark_re);
520	31	50			101	if (@result == 3) {
521	31	50			133	$app->show_new or splice @result, 2, 1;
522	31	50			211	$app->show_old or splice @result, 1, 1;
523						}
524	31				172	println @result;
525						}
526						}
527						#
528						# conflict marker
529						#
530						elsif (/^<<<<<<<\s+(.*)/) {
531						CONFLICT:
532						{
533	0				0	my $c1 = $_;
	0				0
534
535	0			0	0	my @old = read_until { /^=======$/ } *DIFF;
	0				0
536	0		0		0	my $c2 = pop @old // do {
537	0				0	print $c1, @old;
538	0				0	last;
539						};
540
541	0			0	0	my @new = read_until { /^>>>>>>>\s+(.)/ } DIFF;
	0				0
542	0		0		0	my $c3 = pop @new // do {
543	0				0	print $c1, @old, $c2, @new;
544	0				0	last;
545						};
546
547	0			0	0	my $i = first { $old[$_] =~ /^\Q\|\|\|\|\|\|\|\E/ } keys @old;
	0				0
548	0	0			0	my @mrg = defined $i ? splice @old, $i : ();
549
550	0				0	my $old = join '', @old;
551	0				0	my $new = join '', @new;
552	0	0			0	$converter->('', $old, $new) if $converter;
553	0	0			0	compare(\$old, \$new) if $app->unit;
554
555	0	0			0	print $c1 if $app->mark;
556	0	0			0	print color("OTEXT", $old) if $app->show_old;
557	0	0			0	if (@mrg) {
558	0				0	my $c4 = shift @mrg;
559	0	0			0	print $c4 if $app->mark;
560	0				0	my $mrg = join '', @mrg;
561	0	0			0	$converter->('', $mrg) if $converter;
562	0	0			0	print color("MTEXT", $mrg) if $app->show_mrg;
563						}
564	0	0			0	println $c2 if $app->mark;
565	0	0			0	println color("NTEXT", $new) if $app->show_new;
566	0	0			0	println $c3 if $app->mark;
567						}
568						}
569						else {
570	74	50			1866	if ($app->unknown) {
571	74	100			431	if (my $f = $func{UNKNOWN}) {
572	72				121	print $f->($_);
573						} else {
574	2				6	print;
575						}
576						}
577						}
578						}
579						continue {
580	97				1865	$stdout->fh->flush;
581	97				1216	local *_ = $stdout->buffer;
582	97	50			535	if (length) {
583	97				492	$_ = decode 'utf8', $_;
584	97	50			2127	$effector->() if $effector;
585	97				492	STDOUT->printflush($_);
586	97				10582	$stdout->clear;
587						}
588						}
589	9				565	close DIFF;
590	9	50			121	my $exit = $DIFF =~ /\\|$/ ? $? >> 8 : 0;
591
592	9				43	select STDOUT;
593
594	9	50			83	if ($app->stat) {
595	0				0	select STDERR;
596
597	0				0	print("TOTAL: ");
598						printf("append=%d delete=%d change=%d\n",
599	0				0	$stat{'a'}, $stat{'d'}, $stat{'c'});
600	0				0	print("IGNORE WHITESPACE: ");
601						printf("append=%d delete=%d change=%d\n",
602						$stat{'anl'},
603						$stat{'dnl'},
604	0				0	$stat{'cnl'});
605	0				0	print("IGNORE WHITESPACE (bytes): ");
606						printf("append=%d delete=%d change=%d\n",
607						$stat{'anlb'},
608						$stat{'dnlb'},
609	0				0	$stat{'cnlb'});
610						}
611
612	9				166	exit($exit > 1);
613
614						######################################################################
615
616						sub compare_unified {
617	31			31	40	my $column = shift;
618	31	50			72	goto &compare_unified_3 if $column == 3;
619	31				104	goto &compare_unified_generic;
620						}
621
622						sub compare_unified_generic {
623	31			31	68	my($buf, $mark_re) = @_;
624	31				321	my $c = $buf->collect(qr/^[\t ]+$/);
625	31				112	my $o = $buf->collect(qr/[-]/);
626	31				86	my $n = $buf->collect(qr/[+]/);
627	31	50			95	compare(\$o, \$n, $mark_re) if $app->unit;
628	31	100			224	if ($app->color) {
629						map {
630	30				302	my($buf, $m, $t) = @$_;
	90				1011
631	90				483	$$buf =~ s{^($mark_re)(.*)}{
632	154	50			2644	($app->mark ? color($m, $1) : '') . color($t, $2)
633						}mge;
634						} ( [ \$c, 'UMARK', 'UTEXT' ],
635						[ \$o, 'OMARK', 'OTEXT' ],
636						[ \$n, 'NMARK', 'NTEXT' ] );
637						} else {
638	1	50			6	map { s/^$mark_re//mg } $c, $o, $n if not $app->mark;
	0				0
639						}
640	31				1077	($c, $o, $n);
641						}
642
643						sub compare_unified_3 {
644	0			0	0	my($buf, $mark_re) = @_;
645
646	0				0	my @mark = ( " ", "--", "- ", " -", "+ ", " +", "++" );
647	0				0	my %buf = map { $_ => scalar($buf->collect($_)) } @mark;
	0				0
648
649	0	0			0	goto SKIP unless $app->unit;
650
651	0				0	my @r;
652	0				0	$r[0] = compare(\@buf{q/--/, q/++/}, $mark_re);
653	0				0	$r[1] = compare(\@buf{q/- /, q/ -/}, $mark_re);
654	0	0			0	if ($r[1] == 0) {
655	0				0	$r[2] = compare(\@buf{q/- /, q/+ /}, $mark_re);
656	0				0	$r[3] = compare(\@buf{q/ -/, q/ +/}, $mark_re);
657						}
658	0	0			0	if (sum(@r) == 0) {
659	0				0	$r[4] = compare(\@buf{q/- /, q/++/}, $mark_re);
660	0	0			0	$r[5] = compare(\@buf{q/ -/, q/++/}, $mark_re) unless $r[4];
661	0				0	$r[6] = compare(\@buf{q/--/, q/+ /}, $mark_re);
662	0	0			0	$r[7] = compare(\@buf{q/--/, q/ +/}, $mark_re) unless $r[6];
663						}
664
665						SKIP:
666
667	0	0			0	if ($app->color) {
668	0	0			0	map { s/^$mark_re//mg } $buf{q/ /} if not $app->mark;
	0				0
669						map {
670	0				0	my($s, $m, $t) = @$_;
671	0	0			0	$$s =~ s{^($mark_re)(.*)}{
672	0	0			0	($app->mark ? color($m, $1) : '') . color($t, $2)
673						}mge if $$s;
674						} ( [ \$buf{q/--/}, 'CMARK', 'CTEXT' ],
675						[ \$buf{q/- /}, 'OMARK', 'OTEXT' ],
676						[ \$buf{q/ -/}, 'NMARK', 'NTEXT' ],
677						[ \$buf{q/+ /}, 'MMARK', 'MTEXT' ],
678						[ \$buf{q/ +/}, 'MMARK', 'MTEXT' ],
679	0				0	[ \$buf{q/++/}, 'MMARK', 'MTEXT' ] );
680						} else {
681	0	0			0	map { s/^$mark_re//mg } @buf{@mark} if not $app->mark;
	0				0
682						}
683
684	0				0	@buf{$buf->labels};
685						}
686
687						sub print_command {
688	23	50		23	94	return unless $app->command;
689	23				151	my $line = shift;
690	23	100			61	if ($app->color) {
691	22				163	$line = color($colormap{COMMAND}, $line);
692						}
693	23				5564	print $line;
694						}
695
696						sub compare {
697	43			43	402	my($old, $new) = splice @_, 0, 2;
698	43	100	33		410	return 0 unless $old && $new && $$old && $$new;
			66
			100
699
700						# skip word comparison for lines exceeding max length
701	25	50			131	if (my $max = $app->limit->{length}) {
702	25	100	66		563	return 0 if $$old =~ /^.{$max}/m or $$new =~ /^.{$max}/m;
703						}
704
705	24		50		61	my $omark_re = shift \|\| undef;
706	24		66		65	my $nmark_re = shift \|\| $omark_re;
707
708	24				29	my(@omark, @nmark);
709	24	50			53	if ($omark_re) {
710	24				239	$$old =~ s/^($omark_re)/push(@omark, $1), ""/mge;
	32				171
711	24				193	$$new =~ s/^($nmark_re)/push(@nmark, $1), ""/mge;
	32				121
712						}
713
714	24				103	($$old, $$new) = context($$old, $$new);
715
716	24	50			151	$$old =~ s/^/shift @omark/mge if @omark;
	32				156
717	24	50			109	$$new =~ s/^/shift @nmark/mge if @nmark;
	32				105
718
719	24				127	1;
720						}
721
722	9			9	68	use Getopt::EX::Colormap qw(colorize);
	9				13
	9				50824
723
724						sub debug_list {
725	0			0	0	my $i = 0;
726	0				0	my @cmap = qw( K/444 K/333 );
727	0				0	join "", map { colorize $cmap[$i++ % @cmap], $_ } @_;
	0				0
728						}
729
730						sub context {
731	24			24	58	my($old, $new) = @_;
732	24				31	local $_;
733
734	24	50			73	if ($debug{s}) {
735	0				0	print STDERR "****************************** Comparing ...\n";
736	0				0	print STDERR $old;
737	0				0	print STDERR "****************************** and\n";
738	0				0	print STDERR $new;
739	0				0	print STDERR "****************************** .\n";
740						}
741
742	24				75	my @owlist = wordlist($old);
743	24				52	my @nwlist = wordlist($new);
744
745	24	50			67	if ($debug{w}) {
746	0				0	print STDERR "****************************** Comparing ...\n";
747	0				0	print STDERR debug_list @owlist;
748	0				0	print STDERR "****************************** and\n";
749	0				0	print STDERR debug_list @nwlist;
750	0				0	print STDERR "****************************** .\n";
751						}
752
753	24				97	maketmp($T1, @owlist);
754	24				1634	maketmp($T2, @nwlist);
755
756	24				1579	my $diff = sprintf "$app->{subdiff} @sub_diffopts %s %s", $T1->path, $T2->path;
757	24	50			101481	open my $cdif, "$diff \|" or die "diff: $!\n";
758	24				1670	binmode $cdif, ":encoding(utf8)";
759	24				2982	my @command;
760	24				426	my %c = (a => 0, d => 0, c => 0);
761	24				22254	while (<$cdif>) {
762	320	50			1066	warn $_ if $debug{d};
763
764						## Quick hack to read `git diff` unified output
765	320	50			535	if (/^\@\@ \s+ \-(\d+)(?:,(\d+))? \s+ \+(\d+)(?:,(\d+))? \s+ \@\@/x) {
766	0		0		0	my($o, $ol, $n, $nl) = ($1, $2//1, $3, $4//1);
			0
767	0	0			0	my $cmd = ($ol == 0) ? "a" : ($nl == 0) ? "d" : "c";
		0
768	0	0			0	my $old = ($ol <= 1) ? $o : sprintf "%d,%d", $o, $o + $ol - 1;
769	0	0			0	my $new = ($nl <= 1) ? $n : sprintf "%d,%d", $n, $n + $nl - 1;
770	0				0	$_ = sprintf "%s%s%s\n", $old, $cmd, $new;
771						}
772
773	320	100			1319	if (/^[\d,]+([adc])[\d,]+$/) {
774	48				181	push @command, $_;
775	48				392	$c{$1}++;
776						}
777						}
778	24				1397	close $cdif;
779
780	24	50			184	if ($debug{d}) {
781	0				0	printf(STDERR "old=%d new=%d command=%d\n",
782						@owlist+0, @nwlist+0, @command+0);
783	0				0	printf(STDERR "append=$c{a} delete=$c{d} change=$c{c}\n");
784						}
785
786	24				286	my($obuf, $nbuf) = makebuf(\@owlist, \@nwlist, \@command);
787	24				238	my $status = $?>>8;
788	24	50			71	die "Unexpected status of subprocess ($status)\n" if $status > 1;
789
790	24				633	($obuf, $nbuf);
791						}
792
793						sub wordlist {
794	48			48	70	my $text = shift;
795	48	50	33		99	if ($app->unit eq 'mecab' and $text =~ /\P{ASCII}/) {
796	0				0	mecab_words($text);
797						} else {
798	48				268	normal_words($text);
799						}
800						}
801
802						sub normal_words {
803	48			48	54	my $text = shift;
804	48				77	my @words;
805	48				1700	while ($text =~ /\G($w_pattern)/g) {
806	768				2715	push @words, $1;
807						}
808	48				361	@words;
809						}
810
811						sub mecab_words {
812	0			0	0	my $text = shift;
813	0				0	state $mecab = App::cdif::Command::Mecab->new;
814	0				0	$mecab->wordlist($text);
815						}
816
817						sub maketmp {
818	48			48	240	my($tmpfile, @list) = @_;
819	48				294	$tmpfile->reset;
820	48				3978	for (@list) {
821	768	50	33		7541	s/[ \t]+// if $app->ignore_space_change \|\| $app->ignore_all_space;
822	768				5308	$tmpfile->write($_);
823	768	100			9179	$tmpfile->write("\n") unless /\n\z/;
824						}
825	48				97	$tmpfile->flush->rewind;
826						}
827
828						##
829						## @owlist: old word list
830						## @nwlist: new word list
831						## @command: how different these lists (`diff' result command lines)
832						##
833						sub makebuf {
834	24			24	182	my($ol, $nl, $c) = @_;
835	24				404	my @owlist = @$ol;
836	24				274	my @nwlist = @$nl;
837	24				79	my @command = @$c;
838
839	24				71	my($o, $n) = (0, 0); # pointers
840	24				58	my(@obuf, @nbuf);
841
842	24				66	for (@command) {
843	48	50			456	my($old, $cmd, $new) = /([\d,]+)([adc])([\d,]+)/ or do {
844	0				0	die "Panic! Unexpected diff output";
845						};
846	48				411	my($o1, $o2) = range($old);
847	48				97	my($n1, $n2) = range($new);
848	48				152	map { $_-- } $o1, $o2, $n1, $n2; # make them zero origined
	192				303
849
850	48	100			212	push(@obuf, @owlist[$o .. $o1 - 1]), $o = $o1 if $o < $o1;
851	48	100			218	push(@nbuf, @nwlist[$n .. $n1 - 1]), $n = $n1 if $n < $n1;
852
853	48				158	$stat{$cmd}++;
854
855	48	50			200	if ($cmd eq 'd') {
		50
		0
856	0				0	my $os = join('', @owlist[$o1 .. $o2]);
857	0	0			0	if ($owlist[$o2] =~ /\S/) {
858	0				0	$stat{'dnl'}++;
859	0				0	$stat{'dnlb'} += length($os);
860						}
861	0				0	push(@obuf, $func{DELETE}->($os));
862	0				0	$o = $o2 + 1;
863						}
864						elsif ($cmd eq 'c') {
865	48				238	my $os = join('', @owlist[$o1 .. $o2]);
866	48				143	my $ns = join('', @nwlist[$n1 .. $n2]);
867	48	50	33		269	if (($owlist[$o2] =~ /\S/) \|\| ($nwlist[$n2] =~ /\S/)) {
868	48				97	$stat{'cnl'}++;
869	48				147	$stat{'cnlb'} += length($os);
870	48				82	$stat{'cnlb'} += length($ns);
871						}
872	48				231	push(@obuf, $func{OLD}->($os));
873	48				9989	push(@nbuf, $func{NEW}->($ns));
874	48				2346	$o = $o2 + 1;
875	48				124	$n = $n2 + 1;
876						}
877						elsif ($cmd eq 'a') {
878	0				0	my $ns = join('', @nwlist[$n1 .. $n2]);
879	0	0			0	if ($nwlist[$n2] =~ /\S/) {
880	0				0	$stat{'anl'}++;
881	0				0	$stat{'anlb'} += length($ns);
882						}
883	0				0	push(@nbuf, $func{APPEND}->($ns));
884	0				0	$n = $n2 + 1;
885						}
886						}
887	24				125	push(@obuf, @owlist[$o .. $#owlist]);
888	24				144	push(@nbuf, @nwlist[$n .. $#nwlist]);
889
890	24				234	(join('', @obuf), join('', @nbuf));
891						}
892
893						sub read_diff {
894	22			22	78	my($FH, $c) = @_;
895	22				81	my @buf = ();
896	22				39	while ($c-- > 0) {
897	30				133	push @buf, scalar <$FH>;
898						}
899	22	50			125	wantarray ? @buf : join '', @buf;
900						}
901
902						sub read_diffc {
903	4			4	11	my($FH, $n) = @_;
904	4				4	my @buf;
905	4				5	local $_;
906	4				5	my $i = 0;
907	4		66		22	while ($n-- && ($_ = <$FH>)) {
908	56	100			86	$i++ if ($i % 2) != /^!/;
909	56				62	$buf[$i] .= $_;
910	56	50			142	last if /^--- /;
911						}
912	4		50		5	map { $_ // "" } @buf;
	28				76
913						}
914
915						sub ul {
916	0			0		local $_ = join '', @_;
917	0					s/(.)/["", "_\010", "__\010\010"]->[vwidth($1)].$1/ge;
	0
918	0					$_;
919						}
920						sub bd {
921	0			0		local $_ = join '', @_;
922	0					s/(\S)/$1.["", "\010", "\010\010"]->[vwidth($1)].$1/ge;
	0
923	0					$_;
924						}
925
926						sub wc_l {
927	0			0		my $file = shift;
928	0					my $line;
929	0					$file->rewind;
930	0					$line++ while $file->fh->getline;
931	0					$file->rewind;
932	0					$line;
933						}
934
935						sub eval {
936	0	0	0	0		print STDERR &unctrl($_[0]), "\n" x ($_[0] !~ /\n$/) if $_[1] \|\| $debug{e};
937	0					CORE::eval shift;
938	0	0				die sprintf("eval failed in file %s on line %s\n$@", (caller)[1,2]) if $@;
939						}
940
941						######################################################################
942
943						=head1 NAME
944
945						cdif - word context diff
946
947						=head1 VERSION
948
949						Version 4.4501
950
951						=head1 SYNOPSIS
952
953						cdif [option] file1 file2
954
955						cdif [option] [diff-data]
956
957						Options:
958
959						-c, -Cn context diff
960						-u, -Un unified diff
961						-i ignore case
962						-b ignore space change
963						-w ignore whitespace
964						-t expand tabs
965
966						--diff=command specify diff command
967						--subdiff=command specify backend diff command
968						--stat show statistical information
969						--colormap=s specify color map
970						--sdif sdif friendly option
971						--[no]color color or not (default true)
972						--[no]256 ANSI 256 color mode (default true)
973						--[no]cc color command line (default true)
974						--[no]mc color diff mark (default true)
975						--[no]tc color normal text (default true)
976						--[no]uc color unknown text (default true)
977						--[no]old print old text (default true)
978						--[no]new print new text (default true)
979						--[no]mrg print merged text (default true)
980						--[no]command print diff command line (default true)
981						--[no]unknown print unknown line (default true)
982						--[no]mark print mark or not (default true)
983						--[no]prefix read git --graph output (default true)
984						--unit=s word/letter/char/mecab (default word)
985						--[no]mecab use mecab tokenizer (default false)
986						--prefix-pattern prefix pattern
987						--visible char=? set visible attributes
988						--[no]lenience suppress unexpected input warning (default true)
989						--limit length=# skip word comparison for long lines (default 1000)
990						--lxl compare input data line-by-line
991						--style=diff print --lxl output in diff style
992						--version show version
993
994
995						=head1 DESCRIPTION
996
997						B is a post-processor of the Unix diff command. It highlights
998						deleted, changed and added words based on word context (B<--unit=word>
999						by default). If you want to compare text character-by-character, use
1000						option B<--unit=char>. Option B<--unit=mecab> tells to use external
1001						B command as a tokenizer for Japanese text.
1002
1003						If single or no file is specified, B reads that file or STDIN as
1004						an output from diff command. In addition to normal diff, context
1005						diff, and unified (combined) diff, the L-compatible conflict
1006						marker format is supported as input format.
1007
1008						Lines that don't look like diff output are simply ignored and
1009						printed.
1010
1011						=head2 STARTUP and MODULE
1012
1013						B utilizes Perl L module, and reads I<~/.cdifrc>
1014						file if available when starting up. You can define original and
1015						default option there. The next line enables B<--mecab> option and adds
1016						crossed-out effect for deleted words.
1017
1018						option default --mecab --cm DELETE=+X
1019
1020						Modules under B can be loaded by B<-M> option without
1021						prefix. The next command loads B module.
1022
1023						$ cdif -Mcolors
1024
1025						You can also define options in module file. Read `perldoc
1026						Getopt::EX::Module` for detail.
1027
1028						=head2 COLOR
1029
1030						Each line is displayed in different colors. Each text segment has
1031						own labels, and color for them can be specified by B<--colormap>
1032						option. Read `perldoc Getopt::EX::Colormap` for detail.
1033
1034						Standard module B<-Mcolors> is loaded by default, and defines several
1035						color maps for light and dark screen. If you want to use CMY colors in
1036						dark screen, place next line in your F<~/.cdifrc>.
1037
1038						option default --dark-cmy
1039
1040						Option B<--autocolor> is defined in B module to call
1041						L module. It sets B<--light> or B<--dark>
1042						option according to the brightness of the terminal screen. You can
1043						set preferred color in your F<~/.cdifrc> like:
1044
1045						option --light --cmy
1046						option --dark --dark-cmy
1047
1048						Automatic setting is done by L module and it
1049						works with macOS Terminal.app and iTerm.app, and other XTerm
1050						compatible terminals. This module accepts environment variable
1051						L as a terminal background color. For example, use
1052						C<000> or C<#000000> for black and C<555> or C<#FFFFFF> for white.
1053
1054						Option B<--autocolor> is set by default, so override it to do nothing
1055						to disable.
1056
1057						option --autocolor --nop
1058
1059						=head2 EXIT STATUS
1060
1061						B always exits with status zero unless an error occurs.
1062
1063						=head1 OPTIONS
1064
1065						=over 7
1066
1067						=item B<->[B]
1068
1069						Almost same as B command.
1070
1071						=begin deprecated
1072
1073						=item B<--rcs>, B<-r>I, B<-q>
1074
1075						Use rcsdiff instead of normal diff. Option B<--rcs> is not required
1076						when B<-r>I is supplied.
1077
1078						=end deprecated
1079
1080						=item B<-->B=[C,C,C,C,C<0>,C<>]
1081
1082						=item B<-->B=[C,C,C,C,C<0>,C<>]
1083
1084						Specify the comparing unit. Default is I and compare each line
1085						word-by-word. Specify C if you want to compare them
1086						character-by-character. Unit C is almost same as C but
1087						does not include underscore.
1088
1089						When C is given as an unit, B command is called as a
1090						tokenizer for non-ASCII text. ASCII text is compared word-by-word.
1091						External B command has to be installed.
1092
1093						If you give empty string like C<--unit=>, or C<0>, B does not
1094						compare text in any way. You'll still get colorization effect.
1095
1096						=item B<--mecab>
1097
1098						Shortcut for B<--unit=mecab>.
1099
1100						=item B<--diff>=I
1101
1102						Specify the diff command to use.
1103
1104						=item B<--subdiff>=I
1105
1106						Specify the backend diff command to get word differences. Accept
1107						normal and unified diff format.
1108
1109						If you want to use B command, don't forget to set I<-U0>
1110						option.
1111
1112						--subdiff="git diff -U0 --no-index --histogram"
1113
1114						=item B<-->[B]B
1115
1116						Use ANSI color escape sequence for output.
1117
1118						=item B<--colormap>=I, B<--cm>=I
1119
1120						Basic I format is :
1121
1122						FIELD=COLOR
1123
1124						where the FIELD is one from these :
1125
1126						COMMAND Command line
1127						OMARK Old mark
1128						NMARK New mark
1129						UTEXT Same text
1130						OTEXT Old text
1131						NTEXT New text
1132						OCHANGE Old change part
1133						NCHANGE New change part
1134						APPEND Appended part
1135						DELETE Deleted part
1136
1137						=for comment
1138						VISIBLE Visualized invisible chars
1139
1140						and additional I and I FIELDs for git-diff combined
1141						format.
1142
1143						CMARK Common mark
1144						CTEXT Common text
1145						MMARK Merged mark
1146						MTEXT Merged text
1147
1148						You can make multiple fields same color joining them by = :
1149
1150						FIELD1=FIELD2=...=COLOR
1151
1152						Also wildcard can be used for field name :
1153
1154						*CHANGE=BDw
1155
1156						Multiple fields can be specified by repeating options
1157
1158						--cm FIELD1=COLOR1 --cm FIELD2=COLOR2 ...
1159
1160						or combined with comma (,) :
1161
1162						--cm FIELD1=COLOR1,FIELD2=COLOR2, ...
1163
1164						Color specification is a combination of single uppercase character
1165						representing 8 colors :
1166
1167						R Red
1168						G Green
1169						B Blue
1170						C Cyan
1171						M Magenta
1172						Y Yellow
1173						K Black
1174						W White
1175
1176						and alternative (usually brighter) colors in lowercase :
1177
1178						r, g, b, c, m, y, k, w
1179
1180						or RGB values and 24 grey levels if using ANSI 256 or full color
1181						terminal :
1182
1183						(255,255,255) : 24bit decimal RGB colors
1184						#000000 .. #FFFFFF : 24bit hex RGB colors
1185						#000 .. #FFF : 12bit hex RGB 4096 colors
1186						000 .. 555 : 6x6x6 RGB 216 colors
1187						L00 .. L25 : Black (L00), 24 grey levels, White (L25)
1188
1189						or color names enclosed by angle bracket :
1190
1191
1192
1193
1194
1195						with other special effects :
1196
1197						D Double-struck (boldface)
1198						I Italic
1199						U Underline
1200						S Stand-out (reverse video)
1201
1202						Above color spec is simplified summary so if you want complete
1203						information, read L.
1204
1205						Defaults are :
1206
1207						COMMAND => "555/222E"
1208						OMARK => "CS"
1209						NMARK => "MS"
1210						UTEXT => ""
1211						OTEXT => "C"
1212						NTEXT => "M"
1213						OCHANGE => "K/445"
1214						NCHANGE => "K/445"
1215						DELETE => "K/544"
1216						APPEND => "K/544"
1217
1218						CMARK => "GS"
1219						MMARK => "YS"
1220						CTEXT => "G"
1221						MTEXT => "Y"
1222
1223						This is equivalent to :
1224
1225						cdif --cm 'COMMAND=555/222E,OMARK=CS,NMARK=MS' \
1226						--cm 'UTEXT=,OTEXT=C,NTEXT=M,*CHANGE=BD/445,DELETE=APPEND=RD/544' \
1227						--cm 'CMARK=GS,MMARK=YS,CTEXT=G,MTEXT=Y'
1228
1229						=item B<--colormap>=C<&func>
1230
1231						=item B<--colormap>=C
1232
1233						You can also set the name of perl subroutine name or definition to be
1234						called handling matched words. Target word is passed as variable
1235						C<$_>, and the return value of the subroutine will be displayed.
1236
1237						Next option produces L-like formatted output.
1238
1239						--cm '*'= \
1240						--cm DELETE=OCHANGE='sub{"[-$_-]"}' \
1241						--cm APPEND=NCHANGE='sub{"{+$_+}"}'
1242
1243						See L for detail.
1244
1245						=item B<-->[B]B, B<-->[B]B
1246
1247						=item B<-->[B]B, B<-->[B]B
1248
1249						=item B<-->[B]B, B<-->[B]B
1250
1251						=item B<-->[B]B, B<-->[B]B
1252
1253						Enable/Disable using color for the corresponding field.
1254
1255						=item B<--sdif>
1256
1257						Disable options appropriate to use for B's input:
1258						B<--commandcolor>, B<--markcolor>, B<--textcolor> and
1259						B<--unknowncolor>.
1260
1261						=item B<-->[B]B, B<-->[B]B, B<-->[B]B
1262
1263						Print or not old/new/mrg text in diff output.
1264
1265						=item B<-->[B]B
1266
1267						Print or not command lines preceding diff output.
1268
1269						=item B<-->[B]B
1270
1271						Print or not lines that do not look like diff output.
1272
1273						=item B<-->[B]B
1274
1275						Print or not marks at the top of diff output lines. At this point,
1276						this option is effective only for unified diff.
1277
1278						The next example produces output exactly the same as I except for
1279						visual effects.
1280
1281						cdif -U100 --no-mark --no-old --no-command --no-unknown old new
1282
1283						These options are prepared for watchdiff(1) command.
1284
1285						=item B<-->[B]B
1286
1287						Understand prefix for diff output including B B<--graph> option.
1288						True by default.
1289
1290						=item B<--prefix-pattern>=I
1291
1292						Specify prefix pattern in regex. Default pattern is:
1293
1294						(?:\\| )(?: )
1295
1296						This pattern matches B graph style and whitespace indented diff
1297						output.
1298
1299						=item B<--visible> I=[0,1]
1300
1301						Set visible attribute for specified characters. Visible character is
1302						converted to corresponding Unicode symbol character. Default visible:
1303						nul, bel, bs, vt, np, cr, esc, del, and all non-breaking/special
1304						spaces. Default invisible: ht, nl, sp.
1305
1306						NAME CODE Unicode NAME DEFAULT
1307						------ ------ -------------------------------- -------
1308						nul \000 SYMBOL FOR NULL YES
1309						soh \001 SYMBOL FOR SOH YES
1310						bel \007 SYMBOL FOR BELL YES
1311						bs \010 SYMBOL FOR BACKSPACE YES
1312						ht \011 SYMBOL FOR HORIZONTAL TABULATION NO
1313						nl \012 SYMBOL FOR NEWLINE NO
1314						vt \013 SYMBOL FOR VERTICAL TABULATION YES
1315						np \014 SYMBOL FOR FORM FEED YES
1316						cr \015 SYMBOL FOR CARRIAGE RETURN YES
1317						esc \033 SYMBOL FOR ESCAPE YES
1318						sp \040 SYMBOL FOR SPACE NO
1319						del \177 SYMBOL FOR DELETE YES
1320						nbsp \240 OPEN BOX (No-Break Space) YES
1321						nnbsp U+202F OPEN BOX (Narrow No-Break Space) YES
1322						ensp U+2002 OPEN BOX (En Space) YES
1323						emsp U+2003 OPEN BOX (Em Space) YES
1324						thinsp U+2009 OPEN BOX (Thin Space) YES
1325						hairsp U+200A OPEN BOX (Hair Space) YES
1326						zwsp U+200B OPEN BOX (Zero Width Space) YES
1327						idesp U+3000 OPEN BOX (Ideographic Space) YES
1328
1329						Since there are no dedicated Unicode symbols for nbsp and other
1330						special space characters, OPEN BOX is used instead.
1331
1332						Multiple characters can be specified at once, by assembling them by
1333						comma (C<,>) like C<--visible ht=1,sp=1>; or connecting them by equal
1334						sign (C<=>) like C<--visible ht=sp=1>. Character names accept
1335						wildcard; C<--visible '*=1'>.
1336
1337						=begin comment
1338
1339						Colormap label C is applied to those characters. Default
1340						setting is C~~, and visible characters are displayed in reverse~~
1341						video. Unlike other colormaps, only special effects can be set to
1342						this label. Effect C (double-struck) is exception (See
1343						L).
1344
1345						=end comment
1346
1347						B command also supports B<--visible> option for horizontal tab
1348						with better visibility.
1349
1350						=begin deprecated
1351
1352						=item B<-->[B]B
1353
1354						=item B<-->[B]B
1355
1356						Set CARRIAGE-RETURN and ESCAPE visible attributes. These options will
1357						be deprecated soon. Use B<--visible> option instead.
1358
1359						=end deprecated
1360
1361						=item B<--stat>
1362
1363						Print statistical information at the end of output. It shows number
1364						of total appended/deleted/changed words in the context of cdif. It's
1365						common to have many insertions and deletions of newlines because of
1366						text filling process. So normal information is followed by modified
1367						number which ignores insert/delete newlines.
1368
1369						=item B<-->[B]B
1370
1371						Suppress warning message for unexpected input from diff command. True
1372						by default.
1373
1374						=item B<--limit> I=I
1375
1376						Set resource limits. Available keys are:
1377
1378						=over 4
1379
1380						=item B=I<#>
1381
1382						Skip word comparison for lines longer than I<#> characters. Default
1383						is 1000. Extremely long lines, such as those found in SVG data, can
1384						make word-level diff processing very slow with no useful result. When
1385						this limit is exceeded, the section is still displayed with basic
1386						colorization but without fine-grained word highlighting. Set to 0 to
1387						disable this limit.
1388
1389						=back
1390
1391						=item B<--linebyline>, B<--lxl>
1392
1393						=item B<--style>=I