checkpatch.pl 98.5 KB
Newer Older
1
#!/usr/bin/perl -w
2
# (c) 2001, Dave Jones. (the file handling bit)
3
# (c) 2005, Joel Schopp <jschopp@austin.ibm.com> (the ugly bit)
Andy Whitcroft's avatar
Andy Whitcroft committed
4
# (c) 2007,2008, Andy Whitcroft <apw@uk.ibm.com> (new conditions, test suite)
5
# (c) 2008-2010 Andy Whitcroft <apw@canonical.com>
6
7
8
9
10
# Licensed under the terms of the GNU GPL License version 2

use strict;

my $P = $0;
11
$P =~ s@.*/@@g;
12

13
my $V = '0.32';
14
15
16
17
18
19
20

use Getopt::Long qw(:config no_auto_abbrev);

my $quiet = 0;
my $tree = 1;
my $chk_signoff = 1;
my $chk_patch = 1;
21
my $tst_only;
22
my $emacs = 0;
23
my $terse = 0;
24
25
my $file = 0;
my $check = 0;
26
27
my $summary = 1;
my $mailback = 0;
28
my $summary_file = 0;
29
my $show_types = 0;
30
my $root;
31
my %debug;
32
33
my %ignore_type = ();
my @ignore = ();
34
my $help = 0;
35
my $configuration_file = ".checkpatch.conf";
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52

sub help {
	my ($exitcode) = @_;

	print << "EOM";
Usage: $P [OPTION]... [FILE]...
Version: $V

Options:
  -q, --quiet                quiet
  --no-tree                  run without a kernel tree
  --no-signoff               do not check for 'Signed-off-by' line
  --patch                    treat FILE as patchfile (default)
  --emacs                    emacs compile window format
  --terse                    one line per report
  -f, --file                 treat FILE as regular source file
  --subjective, --strict     enable more subjective tests
53
54
  --ignore TYPE(,TYPE2...)   ignore various comma separated message types
  --show-types               show the message "types" in the output
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
  --root=PATH                PATH to the kernel tree root
  --no-summary               suppress the per-file summary
  --mailback                 only produce a report in case of warnings/errors
  --summary-file             include the filename in summary
  --debug KEY=[0|1]          turn on/off debugging of KEY, where KEY is one of
                             'values', 'possible', 'type', and 'attr' (default
                             is all off)
  --test-only=WORD           report only warnings/errors containing WORD
                             literally
  -h, --help, --version      display this help and exit

When FILE is - read standard input.
EOM

	exit($exitcode);
}

72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
my $conf = which_conf($configuration_file);
if (-f $conf) {
	my @conf_args;
	open(my $conffile, '<', "$conf")
	    or warn "$P: Can't find a readable $configuration_file file $!\n";

	while (<$conffile>) {
		my $line = $_;

		$line =~ s/\s*\n?$//g;
		$line =~ s/^\s*//g;
		$line =~ s/\s+/ /g;

		next if ($line =~ m/^\s*#/);
		next if ($line =~ m/^\s*$/);

		my @words = split(" ", $line);
		foreach my $word (@words) {
			last if ($word =~ m/^#/);
			push (@conf_args, $word);
		}
	}
	close($conffile);
	unshift(@ARGV, @conf_args) if @conf_args;
}

98
GetOptions(
99
	'q|quiet+'	=> \$quiet,
100
101
102
	'tree!'		=> \$tree,
	'signoff!'	=> \$chk_signoff,
	'patch!'	=> \$chk_patch,
103
	'emacs!'	=> \$emacs,
104
	'terse!'	=> \$terse,
105
	'f|file!'	=> \$file,
106
107
	'subjective!'	=> \$check,
	'strict!'	=> \$check,
108
109
	'ignore=s'	=> \@ignore,
	'show-types!'	=> \$show_types,
110
	'root=s'	=> \$root,
111
112
	'summary!'	=> \$summary,
	'mailback!'	=> \$mailback,
113
114
	'summary-file!'	=> \$summary_file,

115
	'debug=s'	=> \%debug,
116
	'test-only=s'	=> \$tst_only,
117
118
119
120
121
	'h|help'	=> \$help,
	'version'	=> \$help
) or help(1);

help(0) if ($help);
122
123
124
125

my $exit = 0;

if ($#ARGV < 0) {
126
	print "$P: no input files\n";
127
128
129
	exit(1);
}

130
131
132
133
134
135
136
137
138
139
140
141
142
@ignore = split(/,/, join(',',@ignore));
foreach my $word (@ignore) {
	$word =~ s/\s*\n?$//g;
	$word =~ s/^\s*//g;
	$word =~ s/\s+/ /g;
	$word =~ tr/[a-z]/[A-Z]/;

	next if ($word =~ m/^\s*#/);
	next if ($word =~ m/^\s*$/);

	$ignore_type{$word}++;
}

143
144
my $dbg_values = 0;
my $dbg_possible = 0;
145
my $dbg_type = 0;
146
my $dbg_attr = 0;
147
for my $key (keys %debug) {
148
149
150
	## no critic
	eval "\${dbg_$key} = '$debug{$key}';";
	die "$@" if ($@);
151
152
}

153
154
my $rpt_cleaners = 0;

155
156
157
158
159
if ($terse) {
	$emacs = 1;
	$quiet++;
}

160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
if ($tree) {
	if (defined $root) {
		if (!top_of_kernel_tree($root)) {
			die "$P: $root: --root does not point at a valid tree\n";
		}
	} else {
		if (top_of_kernel_tree('.')) {
			$root = '.';
		} elsif ($0 =~ m@(.*)/scripts/[^/]*$@ &&
						top_of_kernel_tree($1)) {
			$root = $1;
		}
	}

	if (!defined $root) {
		print "Must be run from the top-level dir. of a kernel tree\n";
		exit(2);
	}
178
179
}

180
181
my $emitted_corrupt = 0;

182
183
184
185
our $Ident	= qr{
			[A-Za-z_][A-Za-z\d_]*
			(?:\s*\#\#\s*[A-Za-z_][A-Za-z\d_]*)*
		}x;
186
187
188
189
190
191
192
193
our $Storage	= qr{extern|static|asmlinkage};
our $Sparse	= qr{
			__user|
			__kernel|
			__force|
			__iomem|
			__must_check|
			__init_refok|
194
			__kprobes|
195
196
			__ref|
			__rcu
197
		}x;
198
199
200

# Notes to $Attribute:
# We need \b after 'init' otherwise 'initconst' will cause a false positive in a check
201
202
our $Attribute	= qr{
			const|
203
204
205
206
207
208
209
210
211
212
213
214
215
216
			__percpu|
			__nocast|
			__safe|
			__bitwise__|
			__packed__|
			__packed2__|
			__naked|
			__maybe_unused|
			__always_unused|
			__noreturn|
			__used|
			__cold|
			__noclone|
			__deprecated|
217
218
			__read_mostly|
			__kprobes|
219
			__(?:mem|cpu|dev|)(?:initdata|initconst|init\b)|
220
221
			____cacheline_aligned|
			____cacheline_aligned_in_smp|
222
223
			____cacheline_internodealigned_in_smp|
			__weak
224
		  }x;
225
our $Modifier;
226
227
228
229
our $Inline	= qr{inline|__always_inline|noinline};
our $Member	= qr{->$Ident|\.$Ident|\[[^]]*\]};
our $Lval	= qr{$Ident(?:$Member)*};

230
our $Constant	= qr{(?i:(?:[0-9]+|0x[0-9a-f]+)[ul]*)};
231
our $Assignment	= qr{(?:\*\=|/=|%=|\+=|-=|<<=|>>=|&=|\^=|\|=|=)};
232
our $Compare    = qr{<=|>=|==|!=|<|>};
233
234
235
our $Operators	= qr{
			<=|>=|==|!=|
			=>|->|<<|>>|<|>|!|~|
236
			&&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|%
237
238
		  }x;

239
240
241
242
our $NonptrType;
our $Type;
our $Declare;

243
244
our $NON_ASCII_UTF8	= qr{
	[\xC2-\xDF][\x80-\xBF]               # non-overlong 2-byte
245
246
247
248
249
250
251
252
	|  \xE0[\xA0-\xBF][\x80-\xBF]        # excluding overlongs
	| [\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}  # straight 3-byte
	|  \xED[\x80-\x9F][\x80-\xBF]        # excluding surrogates
	|  \xF0[\x90-\xBF][\x80-\xBF]{2}     # planes 1-3
	| [\xF1-\xF3][\x80-\xBF]{3}          # planes 4-15
	|  \xF4[\x80-\x8F][\x80-\xBF]{2}     # plane 16
}x;

253
254
255
256
257
our $UTF8	= qr{
	[\x09\x0A\x0D\x20-\x7E]              # ASCII
	| $NON_ASCII_UTF8
}x;

258
our $typeTypedefs = qr{(?x:
259
	(?:__)?(?:u|s|be|le)(?:8|16|32|64)|
260
261
262
	atomic_t
)};

263
our $logFunctions = qr{(?x:
264
265
266
	printk(?:_ratelimited|_once|)|
	[a-z0-9]+_(?:printk|emerg|alert|crit|err|warning|warn|notice|info|debug|dbg|vdbg|devel|cont|WARN)(?:_ratelimited|_once|)|
	WARN(?:_RATELIMIT|_ONCE|)|
267
268
	panic|
	MODULE_[A-Z_]+
269
270
)};

271
272
273
274
275
276
277
278
279
280
our $signature_tags = qr{(?xi:
	Signed-off-by:|
	Acked-by:|
	Tested-by:|
	Reviewed-by:|
	Reported-by:|
	To:|
	Cc:
)};

281
282
our @typeList = (
	qr{void},
283
284
285
286
287
288
289
	qr{(?:unsigned\s+)?char},
	qr{(?:unsigned\s+)?short},
	qr{(?:unsigned\s+)?int},
	qr{(?:unsigned\s+)?long},
	qr{(?:unsigned\s+)?long\s+int},
	qr{(?:unsigned\s+)?long\s+long},
	qr{(?:unsigned\s+)?long\s+long\s+int},
290
291
292
293
294
295
296
297
298
299
300
	qr{unsigned},
	qr{float},
	qr{double},
	qr{bool},
	qr{struct\s+$Ident},
	qr{union\s+$Ident},
	qr{enum\s+$Ident},
	qr{${Ident}_t},
	qr{${Ident}_handler},
	qr{${Ident}_handler_fn},
);
301
302
303
our @modifierList = (
	qr{fastcall},
);
304

305
306
307
308
309
310
our $allowed_asm_includes = qr{(?x:
	irq|
	memory
)};
# memory.h: ARM has a custom one

311
sub build_types {
312
313
	my $mods = "(?x:  \n" . join("|\n  ", @modifierList) . "\n)";
	my $all = "(?x:  \n" . join("|\n  ", @typeList) . "\n)";
314
	$Modifier	= qr{(?:$Attribute|$Sparse|$mods)};
315
	$NonptrType	= qr{
316
			(?:$Modifier\s+|const\s+)*
317
			(?:
318
				(?:typeof|__typeof__)\s*\([^\)]*\)|
319
				(?:$typeTypedefs\b)|
320
				(?:${all}\b)
321
			)
322
			(?:\s+$Modifier|\s+const)*
323
324
		  }x;
	$Type	= qr{
325
			$NonptrType
326
			(?:(?:\s|\*|\[\])+\s*const|(?:\s|\*|\[\])+|(?:\s*\[\s*\])+)?
327
			(?:\s+$Inline|\s+$Modifier)*
328
329
330
331
		  }x;
	$Declare	= qr{(?:$Storage\s+)?$Type};
}
build_types();
332

333
334

our $Typecast	= qr{\s*(\(\s*$NonptrType\s*\)){0,1}\s*};
335
336
337
338
339
340
341

# Using $balanced_parens, $LvalOrFunc, or $FuncArg
# requires at least perl version v5.10.0
# Any use must be runtime checked with $^V

our $balanced_parens = qr/(\((?:[^\(\)]++|(?-1))*\))/;
our $LvalOrFunc	= qr{($Lval)\s*($balanced_parens{0,1})\s*};
342
our $FuncArg = qr{$Typecast{0,1}($LvalOrFunc|$Constant)};
343
344
345
346
347
348
349
350
351
352

sub deparenthesize {
	my ($string) = @_;
	return "" if (!defined($string));
	$string =~ s@^\s*\(\s*@@g;
	$string =~ s@\s*\)\s*$@@g;
	$string =~ s@\s+@ @g;
	return $string;
}

353
354
$chk_signoff = 0 if ($file);

355
356
my @dep_includes = ();
my @dep_functions = ();
357
358
my $removal = "Documentation/feature-removal-schedule.txt";
if ($tree && -f "$root/$removal") {
359
	open(my $REMOVE, '<', "$root/$removal") ||
360
				die "$P: $removal: open failed - $!\n";
361
	while (<$REMOVE>) {
362
363
364
		if (/^Check:\s+(.*\S)/) {
			for my $entry (split(/[, ]+/, $1)) {
				if ($entry =~ m@include/(.*)@) {
365
366
					push(@dep_includes, $1);

367
368
369
				} elsif ($entry !~ m@/@) {
					push(@dep_functions, $entry);
				}
370
			}
371
372
		}
	}
373
	close($REMOVE);
374
375
}

376
my @rawlines = ();
377
378
my @lines = ();
my $vname;
379
for my $filename (@ARGV) {
380
	my $FILE;
381
	if ($file) {
382
		open($FILE, '-|', "diff -u /dev/null $filename") ||
383
			die "$P: $filename: diff failed - $!\n";
384
385
	} elsif ($filename eq '-') {
		open($FILE, '<&STDIN');
386
	} else {
387
		open($FILE, '<', "$filename") ||
388
			die "$P: $filename: open failed - $!\n";
389
	}
390
391
392
393
394
	if ($filename eq '-') {
		$vname = 'Your patch';
	} else {
		$vname = $filename;
	}
395
	while (<$FILE>) {
396
397
398
		chomp;
		push(@rawlines, $_);
	}
399
	close($FILE);
400
	if (!process($filename)) {
401
402
403
		$exit = 1;
	}
	@rawlines = ();
404
	@lines = ();
405
406
407
408
409
}

exit($exit);

sub top_of_kernel_tree {
410
411
412
413
414
415
416
417
418
419
420
421
	my ($root) = @_;

	my @tree_check = (
		"COPYING", "CREDITS", "Kbuild", "MAINTAINERS", "Makefile",
		"README", "Documentation", "arch", "include", "drivers",
		"fs", "init", "ipc", "kernel", "lib", "scripts",
	);

	foreach my $check (@tree_check) {
		if (! -e $root . '/' . $check) {
			return 0;
		}
422
	}
423
	return 1;
424
}
425

426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
sub parse_email {
	my ($formatted_email) = @_;

	my $name = "";
	my $address = "";
	my $comment = "";

	if ($formatted_email =~ /^(.*)<(\S+\@\S+)>(.*)$/) {
		$name = $1;
		$address = $2;
		$comment = $3 if defined $3;
	} elsif ($formatted_email =~ /^\s*<(\S+\@\S+)>(.*)$/) {
		$address = $1;
		$comment = $2 if defined $2;
	} elsif ($formatted_email =~ /(\S+\@\S+)(.*)$/) {
		$address = $1;
		$comment = $2 if defined $2;
		$formatted_email =~ s/$address.*$//;
		$name = $formatted_email;
		$name =~ s/^\s+|\s+$//g;
		$name =~ s/^\"|\"$//g;
		# If there's a name left after stripping spaces and
		# leading quotes, and the address doesn't have both
		# leading and trailing angle brackets, the address
		# is invalid. ie:
		#   "joe smith joe@smith.com" bad
		#   "joe smith <joe@smith.com" bad
		if ($name ne "" && $address !~ /^<[^>]+>$/) {
			$name = "";
			$address = "";
			$comment = "";
		}
	}

	$name =~ s/^\s+|\s+$//g;
	$name =~ s/^\"|\"$//g;
	$address =~ s/^\s+|\s+$//g;
	$address =~ s/^\<|\>$//g;

	if ($name =~ /[^\w \-]/i) { ##has "must quote" chars
		$name =~ s/(?<!\\)"/\\"/g; ##escape quotes
		$name = "\"$name\"";
	}

	return ($name, $address, $comment);
}

sub format_email {
	my ($name, $address) = @_;

	my $formatted_email;

	$name =~ s/^\s+|\s+$//g;
	$name =~ s/^\"|\"$//g;
	$address =~ s/^\s+|\s+$//g;

	if ($name =~ /[^\w \-]/i) { ##has "must quote" chars
		$name =~ s/(?<!\\)"/\\"/g; ##escape quotes
		$name = "\"$name\"";
	}

	if ("$name" eq "") {
		$formatted_email = "$address";
	} else {
		$formatted_email = "$name <$address>";
	}

	return $formatted_email;
}

496
497
498
499
500
501
502
503
504
505
506
507
sub which_conf {
	my ($conf) = @_;

	foreach my $path (split(/:/, ".:$ENV{HOME}:.scripts")) {
		if (-e "$path/$conf") {
			return "$path/$conf";
		}
	}

	return "";
}

508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
sub expand_tabs {
	my ($str) = @_;

	my $res = '';
	my $n = 0;
	for my $c (split(//, $str)) {
		if ($c eq "\t") {
			$res .= ' ';
			$n++;
			for (; ($n % 8) != 0; $n++) {
				$res .= ' ';
			}
			next;
		}
		$res .= $c;
		$n++;
	}

	return $res;
}
528
sub copy_spacing {
529
	(my $res = shift) =~ tr/\t/ /c;
530
531
	return $res;
}
532

533
534
535
536
537
538
539
540
541
542
543
544
545
sub line_stats {
	my ($line) = @_;

	# Drop the diff line leader and expand tabs
	$line =~ s/^.//;
	$line = expand_tabs($line);

	# Pick the indent from the front of the line.
	my ($white) = ($line =~ /^(\s*)/);

	return (length($line), length($white));
}

546
547
548
549
550
551
552
553
554
555
556
my $sanitise_quote = '';

sub sanitise_line_reset {
	my ($in_comment) = @_;

	if ($in_comment) {
		$sanitise_quote = '*/';
	} else {
		$sanitise_quote = '';
	}
}
557
558
559
560
561
562
sub sanitise_line {
	my ($line) = @_;

	my $res = '';
	my $l = '';

563
	my $qlen = 0;
564
565
	my $off = 0;
	my $c;
566

567
568
569
570
571
572
573
574
575
576
577
578
579
580
	# Always copy over the diff marker.
	$res = substr($line, 0, 1);

	for ($off = 1; $off < length($line); $off++) {
		$c = substr($line, $off, 1);

		# Comments we are wacking completly including the begin
		# and end, all to $;.
		if ($sanitise_quote eq '' && substr($line, $off, 2) eq '/*') {
			$sanitise_quote = '*/';

			substr($res, $off, 2, "$;$;");
			$off++;
			next;
581
		}
582
		if ($sanitise_quote eq '*/' && substr($line, $off, 2) eq '*/') {
583
584
585
586
			$sanitise_quote = '';
			substr($res, $off, 2, "$;$;");
			$off++;
			next;
587
		}
588
589
590
591
592
593
594
		if ($sanitise_quote eq '' && substr($line, $off, 2) eq '//') {
			$sanitise_quote = '//';

			substr($res, $off, 2, $sanitise_quote);
			$off++;
			next;
		}
595
596
597
598
599
600
601

		# A \ in a string means ignore the next character.
		if (($sanitise_quote eq "'" || $sanitise_quote eq '"') &&
		    $c eq "\\") {
			substr($res, $off, 2, 'XX');
			$off++;
			next;
602
		}
603
604
605
606
		# Regular quotes.
		if ($c eq "'" || $c eq '"') {
			if ($sanitise_quote eq '') {
				$sanitise_quote = $c;
607

608
609
610
611
612
613
				substr($res, $off, 1, $c);
				next;
			} elsif ($sanitise_quote eq $c) {
				$sanitise_quote = '';
			}
		}
614

615
		#print "c<$c> SQ<$sanitise_quote>\n";
616
617
		if ($off != 0 && $sanitise_quote eq '*/' && $c ne "\t") {
			substr($res, $off, 1, $;);
618
619
		} elsif ($off != 0 && $sanitise_quote eq '//' && $c ne "\t") {
			substr($res, $off, 1, $;);
620
621
622
623
624
		} elsif ($off != 0 && $sanitise_quote && $c ne "\t") {
			substr($res, $off, 1, 'X');
		} else {
			substr($res, $off, 1, $c);
		}
625
626
	}

627
628
629
630
	if ($sanitise_quote eq '//') {
		$sanitise_quote = '';
	}

631
	# The pathname on a #include may be surrounded by '<' and '>'.
632
	if ($res =~ /^.\s*\#\s*include\s+\<(.*)\>/) {
633
634
635
636
		my $clean = 'X' x length($1);
		$res =~ s@\<.*\>@<$clean>@;

	# The whole of a #error is a string.
637
	} elsif ($res =~ /^.\s*\#\s*(?:error|warning)\s+(.*)\b/) {
638
		my $clean = 'X' x length($1);
639
		$res =~ s@(\#\s*(?:error|warning)\s+).*@$1$clean@;
640
641
	}

642
643
644
	return $res;
}

645
646
647
648
649
650
sub ctx_statement_block {
	my ($linenr, $remain, $off) = @_;
	my $line = $linenr - 1;
	my $blk = '';
	my $soff = $off;
	my $coff = $off - 1;
651
	my $coff_set = 0;
652

653
654
	my $loff = 0;

655
656
	my $type = '';
	my $level = 0;
657
	my @stack = ();
658
	my $p;
659
660
	my $c;
	my $len = 0;
661
662

	my $remainder;
663
	while (1) {
664
665
		@stack = (['', 0]) if ($#stack == -1);

666
		#warn "CSB: blk<$blk> remain<$remain>\n";
667
668
669
670
		# If we are about to drop off the end, pull in more
		# context.
		if ($off >= $len) {
			for (; $remain > 0; $line++) {
671
				last if (!defined $lines[$line]);
672
				next if ($lines[$line] =~ /^-/);
673
				$remain--;
674
				$loff = $len;
675
				$blk .= $lines[$line] . "\n";
676
677
678
679
680
681
				$len = length($blk);
				$line++;
				last;
			}
			# Bail if there is no further context.
			#warn "CSB: blk<$blk> off<$off> len<$len>\n";
682
			if ($off >= $len) {
683
684
				last;
			}
685
686
687
688
			if ($level == 0 && substr($blk, $off) =~ /^.\s*#\s*define/) {
				$level++;
				$type = '#';
			}
689
		}
690
		$p = $c;
691
		$c = substr($blk, $off, 1);
692
		$remainder = substr($blk, $off);
693

694
		#warn "CSB: c<$c> type<$type> level<$level> remainder<$remainder> coff_set<$coff_set>\n";
695
696
697
698
699
700
701
702
703
704

		# Handle nested #if/#else.
		if ($remainder =~ /^#\s*(?:ifndef|ifdef|if)\s/) {
			push(@stack, [ $type, $level ]);
		} elsif ($remainder =~ /^#\s*(?:else|elif)\b/) {
			($type, $level) = @{$stack[$#stack - 1]};
		} elsif ($remainder =~ /^#\s*endif\b/) {
			($type, $level) = @{pop(@stack)};
		}

705
706
707
708
709
710
		# Statement ends at the ';' or a close '}' at the
		# outermost level.
		if ($level == 0 && $c eq ';') {
			last;
		}

711
		# An else is really a conditional as long as its not else if
712
713
714
715
716
717
718
719
		if ($level == 0 && $coff_set == 0 &&
				(!defined($p) || $p =~ /(?:\s|\}|\+)/) &&
				$remainder =~ /^(else)(?:\s|{)/ &&
				$remainder !~ /^else\s+if\b/) {
			$coff = $off + length($1) - 1;
			$coff_set = 1;
			#warn "CSB: mark coff<$coff> soff<$soff> 1<$1>\n";
			#warn "[" . substr($blk, $soff, $coff - $soff + 1) . "]\n";
720
721
		}

722
723
724
725
726
727
728
729
730
731
		if (($type eq '' || $type eq '(') && $c eq '(') {
			$level++;
			$type = '(';
		}
		if ($type eq '(' && $c eq ')') {
			$level--;
			$type = ($level != 0)? '(' : '';

			if ($level == 0 && $coff < $soff) {
				$coff = $off;
732
733
				$coff_set = 1;
				#warn "CSB: mark coff<$coff>\n";
734
735
736
737
738
739
740
741
742
743
744
			}
		}
		if (($type eq '' || $type eq '{') && $c eq '{') {
			$level++;
			$type = '{';
		}
		if ($type eq '{' && $c eq '}') {
			$level--;
			$type = ($level != 0)? '{' : '';

			if ($level == 0) {
745
746
747
				if (substr($blk, $off + 1, 1) eq ';') {
					$off++;
				}
748
749
750
				last;
			}
		}
751
752
753
754
755
756
757
		# Preprocessor commands end at the newline unless escaped.
		if ($type eq '#' && $c eq "\n" && $p ne "\\") {
			$level--;
			$type = '';
			$off++;
			last;
		}
758
759
		$off++;
	}
760
	# We are truly at the end, so shuffle to the next line.
761
	if ($off == $len) {
762
		$loff = $len + 1;
763
764
765
		$line++;
		$remain--;
	}
766
767
768
769
770
771
772

	my $statement = substr($blk, $soff, $off - $soff + 1);
	my $condition = substr($blk, $soff, $coff - $soff + 1);

	#warn "STATEMENT<$statement>\n";
	#warn "CONDITION<$condition>\n";

773
	#print "coff<$coff> soff<$off> loff<$loff>\n";
774
775
776
777
778

	return ($statement, $condition,
			$line, $remain + 1, $off - $loff + 1, $level);
}

779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
sub statement_lines {
	my ($stmt) = @_;

	# Strip the diff line prefixes and rip blank lines at start and end.
	$stmt =~ s/(^|\n)./$1/g;
	$stmt =~ s/^\s*//;
	$stmt =~ s/\s*$//;

	my @stmt_lines = ($stmt =~ /\n/g);

	return $#stmt_lines + 2;
}

sub statement_rawlines {
	my ($stmt) = @_;

	my @stmt_lines = ($stmt =~ /\n/g);

	return $#stmt_lines + 2;
}

sub statement_block_size {
	my ($stmt) = @_;

	$stmt =~ s/(^|\n)./$1/g;
	$stmt =~ s/^\s*{//;
	$stmt =~ s/}\s*$//;
	$stmt =~ s/^\s*//;
	$stmt =~ s/\s*$//;

	my @stmt_lines = ($stmt =~ /\n/g);
	my @stmt_statements = ($stmt =~ /;/g);

	my $stmt_lines = $#stmt_lines + 2;
	my $stmt_statements = $#stmt_statements + 1;

	if ($stmt_lines > $stmt_statements) {
		return $stmt_lines;
	} else {
		return $stmt_statements;
	}
}

822
823
824
825
826
827
sub ctx_statement_full {
	my ($linenr, $remain, $off) = @_;
	my ($statement, $condition, $level);

	my (@chunks);

828
	# Grab the first conditional/block pair.
829
830
	($statement, $condition, $linenr, $remain, $off, $level) =
				ctx_statement_block($linenr, $remain, $off);
831
	#print "F: c<$condition> s<$statement> remain<$remain>\n";
832
833
834
835
836
837
838
	push(@chunks, [ $condition, $statement ]);
	if (!($remain > 0 && $condition =~ /^\s*(?:\n[+-])?\s*(?:if|else|do)\b/s)) {
		return ($level, $linenr, @chunks);
	}

	# Pull in the following conditional/block pairs and see if they
	# could continue the statement.
839
840
841
	for (;;) {
		($statement, $condition, $linenr, $remain, $off, $level) =
				ctx_statement_block($linenr, $remain, $off);
842
		#print "C: c<$condition> s<$statement> remain<$remain>\n";
843
		last if (!($remain > 0 && $condition =~ /^(?:\s*\n[+-])*\s*(?:else|do)\b/s));
844
845
		#print "C: push\n";
		push(@chunks, [ $condition, $statement ]);
846
847
848
	}

	return ($level, $linenr, @chunks);
849
850
}

851
sub ctx_block_get {
852
	my ($linenr, $remain, $outer, $open, $close, $off) = @_;
853
854
855
856
857
858
859
	my $line;
	my $start = $linenr - 1;
	my $blk = '';
	my @o;
	my @c;
	my @res = ();

860
	my $level = 0;
861
	my @stack = ($level);
862
863
864
865
866
	for ($line = $start; $remain > 0; $line++) {
		next if ($rawlines[$line] =~ /^-/);
		$remain--;

		$blk .= $rawlines[$line];
867
868

		# Handle nested #if/#else.
869
		if ($lines[$line] =~ /^.\s*#\s*(?:ifndef|ifdef|if)\s/) {
870
			push(@stack, $level);
871
		} elsif ($lines[$line] =~ /^.\s*#\s*(?:else|elif)\b/) {
872
			$level = $stack[$#stack - 1];
873
		} elsif ($lines[$line] =~ /^.\s*#\s*endif\b/) {
874
875
876
			$level = pop(@stack);
		}

877
		foreach my $c (split(//, $lines[$line])) {
878
879
880
881
882
			##print "C<$c>L<$level><$open$close>O<$off>\n";
			if ($off > 0) {
				$off--;
				next;
			}
883

884
885
886
887
888
889
890
			if ($c eq $close && $level > 0) {
				$level--;
				last if ($level == 0);
			} elsif ($c eq $open) {
				$level++;
			}
		}
891

892
		if (!$outer || $level <= 1) {
893
			push(@res, $rawlines[$line]);
894
895
		}

896
		last if ($level == 0);
897
898
	}

899
	return ($level, @res);
900
901
902
903
}
sub ctx_block_outer {
	my ($linenr, $remain) = @_;

904
905
	my ($level, @r) = ctx_block_get($linenr, $remain, 1, '{', '}', 0);
	return @r;
906
907
908
909
}
sub ctx_block {
	my ($linenr, $remain) = @_;

910
911
	my ($level, @r) = ctx_block_get($linenr, $remain, 0, '{', '}', 0);
	return @r;
912
913
}
sub ctx_statement {
914
915
916
917
918
919
	my ($linenr, $remain, $off) = @_;

	my ($level, @r) = ctx_block_get($linenr, $remain, 0, '(', ')', $off);
	return @r;
}
sub ctx_block_level {
920
921
	my ($linenr, $remain) = @_;

922
	return ctx_block_get($linenr, $remain, 0, '{', '}', 0);
923
}
924
925
926
927
928
sub ctx_statement_level {
	my ($linenr, $remain, $off) = @_;

	return ctx_block_get($linenr, $remain, 0, '(', ')', $off);
}
929
930
931
932
933

sub ctx_locate_comment {
	my ($first_line, $end_line) = @_;

	# Catch a comment on the end of the line itself.
934
	my ($current_comment) = ($rawlines[$end_line - 1] =~ m@.*(/\*.*\*/)\s*(?:\\\s*)?$@);
935
936
937
938
939
940
941
	return $current_comment if (defined $current_comment);

	# Look through the context and try and figure out if there is a
	# comment.
	my $in_comment = 0;
	$current_comment = '';
	for (my $linenr = $first_line; $linenr < $end_line; $linenr++) {
942
943
		my $line = $rawlines[$linenr - 1];
		#warn "           $line\n";
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
		if ($linenr == $first_line and $line =~ m@^.\s*\*@) {
			$in_comment = 1;
		}
		if ($line =~ m@/\*@) {
			$in_comment = 1;
		}
		if (!$in_comment && $current_comment ne '') {
			$current_comment = '';
		}
		$current_comment .= $line . "\n" if ($in_comment);
		if ($line =~ m@\*/@) {
			$in_comment = 0;
		}
	}

	chomp($current_comment);
	return($current_comment);
}
sub ctx_has_comment {
	my ($first_line, $end_line) = @_;
	my $cmt = ctx_locate_comment($first_line, $end_line);

966
	##print "LINE: $rawlines[$end_line - 1 ]\n";
967
968
969
970
971
	##print "CMMT: $cmt\n";

	return ($cmt ne '');
}

972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
sub raw_line {
	my ($linenr, $cnt) = @_;

	my $offset = $linenr - 1;
	$cnt++;

	my $line;
	while ($cnt) {
		$line = $rawlines[$offset++];
		next if (defined($line) && $line =~ /^-/);
		$cnt--;
	}

	return $line;
}

988
989
990
sub cat_vet {
	my ($vet) = @_;
	my ($res, $coded);
991

992
993
994
995
996
997
	$res = '';
	while ($vet =~ /([^[:cntrl:]]*)([[:cntrl:]]|$)/g) {
		$res .= $1;
		if ($2 ne '') {
			$coded = sprintf("^%c", unpack('C', $2) + 64);
			$res .= $coded;
998
999
		}
	}
1000
	$res =~ s/$/\$/;
1001

1002
	return $res;
1003
1004
}

1005
my $av_preprocessor = 0;
1006
my $av_pending;
1007
my @av_paren_type;
1008
my $av_pend_colon;
1009
1010
1011

sub annotate_reset {
	$av_preprocessor = 0;
1012
1013
	$av_pending = '_';
	@av_paren_type = ('E');
1014
	$av_pend_colon = 'O';
1015
1016
}

1017
1018
sub annotate_values {
	my ($stream, $type) = @_;
1019

1020
	my $res;
1021
	my $var = '_' x length($stream);
1022
1023
	my $cur = $stream;

1024
	print "$stream\n" if ($dbg_values > 1);
1025
1026

	while (length($cur)) {
1027
		@av_paren_type = ('E') if ($#av_paren_type < 0);
1028
		print " <" . join('', @av_paren_type) .
1029
				"> <$type> <$av_pending>" if ($dbg_values > 1);
1030
		if ($cur =~ /^(\s+)/o) {
1031
1032
			print "WS($1)\n" if ($dbg_values > 1);
			if ($1 =~ /\n/ && $av_preprocessor) {
1033
				$type = pop(@av_paren_type);
1034
				$av_preprocessor = 0;
1035
1036
			}

1037
		} elsif ($cur =~ /^(\(\s*$Type\s*)\)/ && $av_pending eq '_') {
1038
1039
			print "CAST($1)\n" if ($dbg_values > 1);
			push(@av_paren_type, $type);
1040
			$type = 'c';
1041

1042
		} elsif ($cur =~ /^($Type)\s*(?:$Ident|,|\)|\(|\s*$)/) {
1043
			print "DECLARE($1)\n" if ($dbg_values > 1);
1044
1045
			$type = 'T';

1046
1047
1048
1049
		} elsif ($cur =~ /^($Modifier)\s*/) {
			print "MODIFIER($1)\n" if ($dbg_values > 1);
			$type = 'T';

1050
		} elsif ($cur =~ /^(\#\s*define\s*$Ident)(\(?)/o) {
1051
			print "DEFINE($1,$2)\n" if ($dbg_values > 1);
1052
			$av_preprocessor = 1;
1053
1054
1055
1056
1057
1058
			push(@av_paren_type, $type);
			if ($2 ne '') {
				$av_pending = 'N';
			}
			$type = 'E';

1059
		} elsif ($cur =~ /^(\#\s*(?:undef\s*$Ident|include\b))/o) {
1060
1061
1062
			print "UNDEF($1)\n" if ($dbg_values > 1);
			$av_preprocessor = 1;
			push(@av_paren_type, $type);
1063

1064
		} elsif ($cur =~ /^(\#\s*(?:ifdef|ifndef|if))/o) {
1065
			print "PRE_START($1)\n" if ($dbg_values > 1);
1066
			$av_preprocessor = 1;
1067
1068
1069

			push(@av_paren_type, $type);
			push(@av_paren_type, $type);
1070
			$type = 'E';
1071

1072
		} elsif ($cur =~ /^(\#\s*(?:else|elif))/o) {
1073
1074
1075
1076
1077
			print "PRE_RESTART($1)\n" if ($dbg_values > 1);
			$av_preprocessor = 1;

			push(@av_paren_type, $av_paren_type[$#av_paren_type]);

1078
			$type = 'E';
1079

1080
		} elsif ($cur =~ /^(\#\s*(?:endif))/o) {
1081
1082
1083
1084
1085
1086
1087
1088
			print "PRE_END($1)\n" if ($dbg_values > 1);

			$av_preprocessor = 1;

			# Assume all arms of the conditional end as this
			# one does, and continue as if the #endif was not here.
			pop(@av_paren_type);
			push(@av_paren_type, $type);
1089
			$type = 'E';
1090
1091

		} elsif ($cur =~ /^(\\\n)/o) {
1092
			print "PRECONT($1)\n" if ($dbg_values > 1);
1093

1094
1095
1096
1097
1098
		} elsif ($cur =~ /^(__attribute__)\s*\(?/o) {
			print "ATTR($1)\n" if ($dbg_values > 1);
			$av_pending = $type;
			$type = 'N';

1099
		} elsif ($cur =~ /^(sizeof)\s*(\()?/o) {
1100
			print "SIZEOF($1)\n" if ($dbg_values > 1);
1101
			if (defined $2) {
1102
				$av_pending = 'V';
1103
1104
1105
			}
			$type = 'N';

1106
		} elsif ($cur =~ /^(if|while|for)\b/o) {
1107
			print "COND($1)\n" if ($dbg_values > 1);
1108
			$av_pending = 'E';
1109
1110
			$type = 'N';

1111
1112
1113
1114
1115
		} elsif ($cur =~/^(case)/o) {
			print "CASE($1)\n" if ($dbg_values > 1);
			$av_pend_colon = 'C';
			$type = 'N';

1116
		} elsif ($cur =~/^(return|else|goto|typeof|__typeof__)\b/o) {
1117
			print "KEYWORD($1)\n" if ($dbg_values > 1);
1118
1119
1120
			$type = 'N';

		} elsif ($cur =~ /^(\()/o) {
1121
			print "PAREN('$1')\n" if ($dbg_values > 1);
1122
1123
			push(@av_paren_type, $av_pending);
			$av_pending = '_';
1124
1125
1126
			$type = 'N';

		} elsif ($cur =~ /^(\))/o) {
1127
1128
1129
			my $new_type = pop(@av_paren_type);
			if ($new_type ne '_') {
				$type = $new_type;
1130
1131
				print "PAREN('$1') -> $type\n"
							if ($dbg_values > 1);
1132
			} else {
1133
				print "PAREN('$1')\n" if ($dbg_values > 1);
1134
1135
			}

1136
		} elsif ($cur =~ /^($Ident)\s*\(/o) {
1137
			print "FUNC($1)\n" if ($dbg_values > 1);
1138
			$type = 'V';
1139
			$av_pending = 'V';
1140

1141
1142
		} elsif ($cur =~ /^($Ident\s*):(?:\s*\d+\s*(,|=|;))?/) {
			if (defined $2 && $type eq 'C' || $type eq 'T') {
1143
				$av_pend_colon = 'B';
1144
1145
			} elsif ($type eq 'E') {
				$av_pend_colon = 'L';
1146
1147
1148
1149
			}
			print "IDENT_COLON($1,$type>$av_pend_colon)\n" if ($dbg_values > 1);
			$type = 'V';

1150
		} elsif ($cur =~ /^($Ident|$Constant)/o) {
1151
			print "IDENT($1)\n" if ($dbg_values > 1);
1152
1153
1154
			$type = 'V';

		} elsif ($cur =~ /^($Assignment)/o) {
1155
			print "ASSIGN($1)\n" if ($dbg_values > 1);
1156
1157
			$type = 'N';

1158
		} elsif ($cur =~/^(;|{|})/) {
1159
			print "END($1)\n" if ($dbg_values > 1);
1160
			$type = 'E';
1161
1162
			$av_pend_colon = 'O';

1163
1164
1165
1166
		} elsif ($cur =~/^(,)/) {
			print "COMMA($1)\n" if ($dbg_values > 1);
			$type = 'C';

1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
		} elsif ($cur =~ /^(\?)/o) {
			print "QUESTION($1)\n" if ($dbg_values > 1);
			$type = 'N';

		} elsif ($cur =~ /^(:)/o) {
			print "COLON($1,$av_pend_colon)\n" if ($dbg_values > 1);

			substr($var, length($res), 1, $av_pend_colon);
			if ($av_pend_colon eq 'C' || $av_pend_colon eq 'L') {
				$type = 'E';
			} else {
				$type = 'N';
			}
			$av_pend_colon = 'O';
1181

1182
		} elsif ($cur =~ /^(\[)/o) {
1183
			print "CLOSE($1)\n" if ($dbg_values > 1);
1184
1185
			$type = 'N';

1186
		} elsif ($cur =~ /^(-(?![->])|\+(?!\+)|\*|\&\&|\&)/o) {
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
			my $variant;

			print "OPV($1)\n" if ($dbg_values > 1);
			if ($type eq 'V') {
				$variant = 'B';
			} else {
				$variant = 'U';
			}

			substr($var, length($res), 1, $variant);
			$type = 'N';

1199
		} elsif ($cur =~ /^($Operators)/o) {
1200
			print "OP($1)\n" if ($dbg_values > 1);
1201
1202
1203
1204
1205
			if ($1 ne '++' && $1 ne '--') {
				$type = 'N';
			}

		} elsif ($cur =~ /(^.)/o) {
1206
			print "C($1)\n" if ($dbg_values > 1);
1207
1208
1209
1210
1211
		}
		if (defined $1) {
			$cur = substr($cur, length($1));
			$res .= $type x length($1);
		}