checkpatch.pl 99.8 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
my $max_line_length = 80;
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53

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
54
  --ignore TYPE(,TYPE2...)   ignore various comma separated message types
55
  --max-line-length=n        set the maximum line length, if exceeded, warn
56
  --show-types               show the message "types" in the output
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
  --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);
}

74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
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;
}

100
GetOptions(
101
	'q|quiet+'	=> \$quiet,
102
103
104
	'tree!'		=> \$tree,
	'signoff!'	=> \$chk_signoff,
	'patch!'	=> \$chk_patch,
105
	'emacs!'	=> \$emacs,
106
	'terse!'	=> \$terse,
107
	'f|file!'	=> \$file,
108
109
	'subjective!'	=> \$check,
	'strict!'	=> \$check,
110
111
	'ignore=s'	=> \@ignore,
	'show-types!'	=> \$show_types,
112
	'max-line-length=i' => \$max_line_length,
113
	'root=s'	=> \$root,
114
115
	'summary!'	=> \$summary,
	'mailback!'	=> \$mailback,
116
117
	'summary-file!'	=> \$summary_file,

118
	'debug=s'	=> \%debug,
119
	'test-only=s'	=> \$tst_only,
120
121
122
123
124
	'h|help'	=> \$help,
	'version'	=> \$help
) or help(1);

help(0) if ($help);
125
126
127
128

my $exit = 0;

if ($#ARGV < 0) {
129
	print "$P: no input files\n";
130
131
132
	exit(1);
}

133
134
135
136
137
138
139
140
141
142
143
144
145
@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}++;
}

146
147
my $dbg_values = 0;
my $dbg_possible = 0;
148
my $dbg_type = 0;
149
my $dbg_attr = 0;
150
for my $key (keys %debug) {
151
152
153
	## no critic
	eval "\${dbg_$key} = '$debug{$key}';";
	die "$@" if ($@);
154
155
}

156
157
my $rpt_cleaners = 0;

158
159
160
161
162
if ($terse) {
	$emacs = 1;
	$quiet++;
}

163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
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);
	}
181
182
}

183
184
my $emitted_corrupt = 0;

185
186
187
188
our $Ident	= qr{
			[A-Za-z_][A-Za-z\d_]*
			(?:\s*\#\#\s*[A-Za-z_][A-Za-z\d_]*)*
		}x;
189
190
191
192
193
194
195
196
our $Storage	= qr{extern|static|asmlinkage};
our $Sparse	= qr{
			__user|
			__kernel|
			__force|
			__iomem|
			__must_check|
			__init_refok|
197
			__kprobes|
198
199
			__ref|
			__rcu
200
		}x;
201
202
203

# Notes to $Attribute:
# We need \b after 'init' otherwise 'initconst' will cause a false positive in a check
204
205
our $Attribute	= qr{
			const|
206
207
208
209
210
211
212
213
214
215
216
217
218
219
			__percpu|
			__nocast|
			__safe|
			__bitwise__|
			__packed__|
			__packed2__|
			__naked|
			__maybe_unused|
			__always_unused|
			__noreturn|
			__used|
			__cold|
			__noclone|
			__deprecated|
220
221
			__read_mostly|
			__kprobes|
222
			__(?:mem|cpu|dev|)(?:initdata|initconst|init\b)|
223
224
			____cacheline_aligned|
			____cacheline_aligned_in_smp|
225
226
			____cacheline_internodealigned_in_smp|
			__weak
227
		  }x;
228
our $Modifier;
229
230
231
232
our $Inline	= qr{inline|__always_inline|noinline};
our $Member	= qr{->$Ident|\.$Ident|\[[^]]*\]};
our $Lval	= qr{$Ident(?:$Member)*};

233
234
235
our $Float_hex	= qr{(?i)0x[0-9a-f]+p-?[0-9]+[fl]?};
our $Float_dec	= qr{(?i)(?:[0-9]+\.[0-9]*|[0-9]*\.[0-9]+)(?:e-?[0-9]+)?[fl]?};
our $Float_int	= qr{(?i)[0-9]+e-?[0-9]+[fl]?};
236
our $Float	= qr{$Float_hex|$Float_dec|$Float_int};
237
238
our $Constant	= qr{$Float|(?i)(?:0x[0-9a-f]+|[0-9]+)[ul]*};
our $Assignment	= qr{\*\=|/=|%=|\+=|-=|<<=|>>=|&=|\^=|\|=|=};
239
our $Compare    = qr{<=|>=|==|!=|<|>};
240
241
242
our $Operators	= qr{
			<=|>=|==|!=|
			=>|->|<<|>>|<|>|!|~|
243
			&&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|%
244
245
		  }x;

246
247
248
249
our $NonptrType;
our $Type;
our $Declare;

250
251
our $NON_ASCII_UTF8	= qr{
	[\xC2-\xDF][\x80-\xBF]               # non-overlong 2-byte
252
253
254
255
256
257
258
259
	|  \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;

260
261
262
263
264
our $UTF8	= qr{
	[\x09\x0A\x0D\x20-\x7E]              # ASCII
	| $NON_ASCII_UTF8
}x;

265
our $typeTypedefs = qr{(?x:
266
	(?:__)?(?:u|s|be|le)(?:8|16|32|64)|
267
268
269
	atomic_t
)};

270
our $logFunctions = qr{(?x:
271
272
273
	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|)|
274
275
	panic|
	MODULE_[A-Z_]+
276
277
)};

278
279
280
281
282
283
284
285
286
287
our $signature_tags = qr{(?xi:
	Signed-off-by:|
	Acked-by:|
	Tested-by:|
	Reviewed-by:|
	Reported-by:|
	To:|
	Cc:
)};

288
289
our @typeList = (
	qr{void},
290
291
292
293
294
295
296
	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},
297
298
299
300
301
302
303
304
305
306
307
	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},
);
308
309
310
our @modifierList = (
	qr{fastcall},
);
311

312
313
314
315
316
317
our $allowed_asm_includes = qr{(?x:
	irq|
	memory
)};
# memory.h: ARM has a custom one

318
sub build_types {
319
320
	my $mods = "(?x:  \n" . join("|\n  ", @modifierList) . "\n)";
	my $all = "(?x:  \n" . join("|\n  ", @typeList) . "\n)";
321
	$Modifier	= qr{(?:$Attribute|$Sparse|$mods)};
322
	$NonptrType	= qr{
323
			(?:$Modifier\s+|const\s+)*
324
			(?:
325
				(?:typeof|__typeof__)\s*\([^\)]*\)|
326
				(?:$typeTypedefs\b)|
327
				(?:${all}\b)
328
			)
329
			(?:\s+$Modifier|\s+const)*
330
331
		  }x;
	$Type	= qr{
332
			$NonptrType
333
			(?:(?:\s|\*|\[\])+\s*const|(?:\s|\*|\[\])+|(?:\s*\[\s*\])+)?
334
			(?:\s+$Inline|\s+$Modifier)*
335
336
337
338
		  }x;
	$Declare	= qr{(?:$Storage\s+)?$Type};
}
build_types();
339

340
341

our $Typecast	= qr{\s*(\(\s*$NonptrType\s*\)){0,1}\s*};
342
343
344
345
346
347
348

# 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*};
349
our $FuncArg = qr{$Typecast{0,1}($LvalOrFunc|$Constant)};
350
351
352
353
354
355
356
357
358
359

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

360
361
$chk_signoff = 0 if ($file);

362
my @rawlines = ();
363
364
my @lines = ();
my $vname;
365
for my $filename (@ARGV) {
366
	my $FILE;
367
	if ($file) {
368
		open($FILE, '-|', "diff -u /dev/null $filename") ||
369
			die "$P: $filename: diff failed - $!\n";
370
371
	} elsif ($filename eq '-') {
		open($FILE, '<&STDIN');
372
	} else {
373
		open($FILE, '<', "$filename") ||
374
			die "$P: $filename: open failed - $!\n";
375
	}
376
377
378
379
380
	if ($filename eq '-') {
		$vname = 'Your patch';
	} else {
		$vname = $filename;
	}
381
	while (<$FILE>) {
382
383
384
		chomp;
		push(@rawlines, $_);
	}
385
	close($FILE);
386
	if (!process($filename)) {
387
388
389
		$exit = 1;
	}
	@rawlines = ();
390
	@lines = ();
391
392
393
394
395
}

exit($exit);

sub top_of_kernel_tree {
396
397
398
399
400
401
402
403
404
405
406
407
	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;
		}
408
	}
409
	return 1;
410
}
411

412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
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;
}

482
483
484
485
486
487
488
489
490
491
492
493
sub which_conf {
	my ($conf) = @_;

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

	return "";
}

494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
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;
}
514
sub copy_spacing {
515
	(my $res = shift) =~ tr/\t/ /c;
516
517
	return $res;
}
518

519
520
521
522
523
524
525
526
527
528
529
530
531
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));
}

532
533
534
535
536
537
538
539
540
541
542
my $sanitise_quote = '';

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

	if ($in_comment) {
		$sanitise_quote = '*/';
	} else {
		$sanitise_quote = '';
	}
}
543
544
545
546
547
548
sub sanitise_line {
	my ($line) = @_;

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

549
	my $qlen = 0;
550
551
	my $off = 0;
	my $c;
552

553
554
555
556
557
558
559
560
561
562
563
564
565
566
	# 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, $off, 2, "$;$;");
			$off++;
			next;
567
		}
568
		if ($sanitise_quote eq '*/' && substr($line, $off, 2) eq '*/') {
569
570
571
572
			$sanitise_quote = '';
			substr($res, $off, 2, "$;$;");
			$off++;
			next;
573
		}
574
575
576
577
578
579
580
		if ($sanitise_quote eq '' && substr($line, $off, 2) eq '//') {
			$sanitise_quote = '//';

			substr($res, $off, 2, $sanitise_quote);
			$off++;
			next;
		}
581
582
583
584
585
586
587

		# A \ in a string means ignore the next character.
		if (($sanitise_quote eq "'" || $sanitise_quote eq '"') &&
		    $c eq "\\") {
			substr($res, $off, 2, 'XX');
			$off++;
			next;
588
		}
589
590
591
592
		# Regular quotes.
		if ($c eq "'" || $c eq '"') {
			if ($sanitise_quote eq '') {
				$sanitise_quote = $c;
593

594
595
596
597
598
599
				substr($res, $off, 1, $c);
				next;
			} elsif ($sanitise_quote eq $c) {
				$sanitise_quote = '';
			}
		}
600

601
		#print "c<$c> SQ<$sanitise_quote>\n";
602
603
		if ($off != 0 && $sanitise_quote eq '*/' && $c ne "\t") {
			substr($res, $off, 1, $;);
604
605
		} elsif ($off != 0 && $sanitise_quote eq '//' && $c ne "\t") {
			substr($res, $off, 1, $;);
606
607
608
609
610
		} elsif ($off != 0 && $sanitise_quote && $c ne "\t") {
			substr($res, $off, 1, 'X');
		} else {
			substr($res, $off, 1, $c);
		}
611
612
	}

613
614
615
616
	if ($sanitise_quote eq '//') {
		$sanitise_quote = '';
	}

617
	# The pathname on a #include may be surrounded by '<' and '>'.
618
	if ($res =~ /^.\s*\#\s*include\s+\<(.*)\>/) {
619
620
621
622
		my $clean = 'X' x length($1);
		$res =~ s@\<.*\>@<$clean>@;

	# The whole of a #error is a string.
623
	} elsif ($res =~ /^.\s*\#\s*(?:error|warning)\s+(.*)\b/) {
624
		my $clean = 'X' x length($1);
625
		$res =~ s@(\#\s*(?:error|warning)\s+).*@$1$clean@;
626
627
	}

628
629
630
	return $res;
}

631
632
633
634
635
636
sub ctx_statement_block {
	my ($linenr, $remain, $off) = @_;
	my $line = $linenr - 1;
	my $blk = '';
	my $soff = $off;
	my $coff = $off - 1;
637
	my $coff_set = 0;
638

639
640
	my $loff = 0;

641
642
	my $type = '';
	my $level = 0;
643
	my @stack = ();
644
	my $p;
645
646
	my $c;
	my $len = 0;
647
648

	my $remainder;
649
	while (1) {
650
651
		@stack = (['', 0]) if ($#stack == -1);

652
		#warn "CSB: blk<$blk> remain<$remain>\n";
653
654
655
656
		# If we are about to drop off the end, pull in more
		# context.
		if ($off >= $len) {
			for (; $remain > 0; $line++) {
657
				last if (!defined $lines[$line]);
658
				next if ($lines[$line] =~ /^-/);
659
				$remain--;
660
				$loff = $len;
661
				$blk .= $lines[$line] . "\n";
662
663
664
665
666
667
				$len = length($blk);
				$line++;
				last;
			}
			# Bail if there is no further context.
			#warn "CSB: blk<$blk> off<$off> len<$len>\n";
668
			if ($off >= $len) {
669
670
				last;
			}
671
672
673
674
			if ($level == 0 && substr($blk, $off) =~ /^.\s*#\s*define/) {
				$level++;
				$type = '#';
			}
675
		}
676
		$p = $c;
677
		$c = substr($blk, $off, 1);
678
		$remainder = substr($blk, $off);
679

680
		#warn "CSB: c<$c> type<$type> level<$level> remainder<$remainder> coff_set<$coff_set>\n";
681
682
683
684
685
686
687
688
689
690

		# 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)};
		}

691
692
693
694
695
696
		# Statement ends at the ';' or a close '}' at the
		# outermost level.
		if ($level == 0 && $c eq ';') {
			last;
		}

697
		# An else is really a conditional as long as its not else if
698
699
700
701
702
703
704
705
		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";
706
707
		}

708
709
710
711
712
713
714
715
716
717
		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;
718
719
				$coff_set = 1;
				#warn "CSB: mark coff<$coff>\n";
720
721
722
723
724
725
726
727
728
729
730
			}
		}
		if (($type eq '' || $type eq '{') && $c eq '{') {
			$level++;
			$type = '{';
		}
		if ($type eq '{' && $c eq '}') {
			$level--;
			$type = ($level != 0)? '{' : '';

			if ($level == 0) {
731
732
733
				if (substr($blk, $off + 1, 1) eq ';') {
					$off++;
				}
734
735
736
				last;
			}
		}
737
738
739
740
741
742
743
		# Preprocessor commands end at the newline unless escaped.
		if ($type eq '#' && $c eq "\n" && $p ne "\\") {
			$level--;
			$type = '';
			$off++;
			last;
		}
744
745
		$off++;
	}
746
	# We are truly at the end, so shuffle to the next line.
747
	if ($off == $len) {
748
		$loff = $len + 1;
749
750
751
		$line++;
		$remain--;
	}
752
753
754
755
756
757
758

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

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

759
	#print "coff<$coff> soff<$off> loff<$loff>\n";
760
761
762
763
764

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

765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
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;
	}
}

808
809
810
811
812
813
sub ctx_statement_full {
	my ($linenr, $remain, $off) = @_;
	my ($statement, $condition, $level);

	my (@chunks);

814
	# Grab the first conditional/block pair.
815
816
	($statement, $condition, $linenr, $remain, $off, $level) =
				ctx_statement_block($linenr, $remain, $off);
817
	#print "F: c<$condition> s<$statement> remain<$remain>\n";
818
819
820
821
822
823
824
	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.
825
826
827
	for (;;) {
		($statement, $condition, $linenr, $remain, $off, $level) =
				ctx_statement_block($linenr, $remain, $off);
828
		#print "C: c<$condition> s<$statement> remain<$remain>\n";
829
		last if (!($remain > 0 && $condition =~ /^(?:\s*\n[+-])*\s*(?:else|do)\b/s));
830
831
		#print "C: push\n";
		push(@chunks, [ $condition, $statement ]);
832
833
834
	}

	return ($level, $linenr, @chunks);
835
836
}

837
sub ctx_block_get {
838
	my ($linenr, $remain, $outer, $open, $close, $off) = @_;
839
840
841
842
843
844
845
	my $line;
	my $start = $linenr - 1;
	my $blk = '';
	my @o;
	my @c;
	my @res = ();

846
	my $level = 0;
847
	my @stack = ($level);
848
849
850
851
852
	for ($line = $start; $remain > 0; $line++) {
		next if ($rawlines[$line] =~ /^-/);
		$remain--;

		$blk .= $rawlines[$line];
853
854

		# Handle nested #if/#else.
855
		if ($lines[$line] =~ /^.\s*#\s*(?:ifndef|ifdef|if)\s/) {
856
			push(@stack, $level);
857
		} elsif ($lines[$line] =~ /^.\s*#\s*(?:else|elif)\b/) {
858
			$level = $stack[$#stack - 1];
859
		} elsif ($lines[$line] =~ /^.\s*#\s*endif\b/) {
860
861
862
			$level = pop(@stack);
		}

863
		foreach my $c (split(//, $lines[$line])) {
864
865
866
867
868
			##print "C<$c>L<$level><$open$close>O<$off>\n";
			if ($off > 0) {
				$off--;
				next;
			}
869

870
871
872
873
874
875
876
			if ($c eq $close && $level > 0) {
				$level--;
				last if ($level == 0);
			} elsif ($c eq $open) {
				$level++;
			}
		}
877

878
		if (!$outer || $level <= 1) {
879
			push(@res, $rawlines[$line]);
880
881
		}

882
		last if ($level == 0);
883
884
	}

885
	return ($level, @res);
886
887
888
889
}
sub ctx_block_outer {
	my ($linenr, $remain) = @_;

890
891
	my ($level, @r) = ctx_block_get($linenr, $remain, 1, '{', '}', 0);
	return @r;
892
893
894
895
}
sub ctx_block {
	my ($linenr, $remain) = @_;

896
897
	my ($level, @r) = ctx_block_get($linenr, $remain, 0, '{', '}', 0);
	return @r;
898
899
}
sub ctx_statement {
900
901
902
903
904
905
	my ($linenr, $remain, $off) = @_;

	my ($level, @r) = ctx_block_get($linenr, $remain, 0, '(', ')', $off);
	return @r;
}
sub ctx_block_level {
906
907
	my ($linenr, $remain) = @_;

908
	return ctx_block_get($linenr, $remain, 0, '{', '}', 0);
909
}
910
911
912
913
914
sub ctx_statement_level {
	my ($linenr, $remain, $off) = @_;

	return ctx_block_get($linenr, $remain, 0, '(', ')', $off);
}
915
916
917
918
919

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

	# Catch a comment on the end of the line itself.
920
	my ($current_comment) = ($rawlines[$end_line - 1] =~ m@.*(/\*.*\*/)\s*(?:\\\s*)?$@);
921
922
923
924
925
926
927
	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++) {
928
929
		my $line = $rawlines[$linenr - 1];
		#warn "           $line\n";
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
		if ($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);

952
	##print "LINE: $rawlines[$end_line - 1 ]\n";
953
954
955
956
957
	##print "CMMT: $cmt\n";

	return ($cmt ne '');
}

958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
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;
}

974
975
976
sub cat_vet {
	my ($vet) = @_;
	my ($res, $coded);
977

978
979
980
981
982
983
	$res = '';
	while ($vet =~ /([^[:cntrl:]]*)([[:cntrl:]]|$)/g) {
		$res .= $1;
		if ($2 ne '') {
			$coded = sprintf("^%c", unpack('C', $2) + 64);
			$res .= $coded;
984
985
		}
	}
986
	$res =~ s/$/\$/;
987

988
	return $res;
989
990
}

991
my $av_preprocessor = 0;
992
my $av_pending;
993
my @av_paren_type;
994
my $av_pend_colon;
995
996
997

sub annotate_reset {
	$av_preprocessor = 0;
998
999
	$av_pending = '_';
	@av_paren_type = ('E');
1000
	$av_pend_colon = 'O';
1001
1002
}

1003
1004
sub annotate_values {
	my ($stream, $type) = @_;
1005

1006
	my $res;
1007
	my $var = '_' x length($stream);
1008
1009
	my $cur = $stream;

1010
	print "$stream\n" if ($dbg_values > 1);
1011
1012

	while (length($cur)) {
1013
		@av_paren_type = ('E') if ($#av_paren_type < 0);
1014
		print " <" . join('', @av_paren_type) .
1015
				"> <$type> <$av_pending>" if ($dbg_values > 1);
1016
		if ($cur =~ /^(\s+)/o) {
1017
1018
			print "WS($1)\n" if ($dbg_values > 1);
			if ($1 =~ /\n/ && $av_preprocessor) {
1019
				$type = pop(@av_paren_type);
1020
				$av_preprocessor = 0;
1021
1022
			}

1023
		} elsif ($cur =~ /^(\(\s*$Type\s*)\)/ && $av_pending eq '_') {
1024
1025
			print "CAST($1)\n" if ($dbg_values > 1);
			push(@av_paren_type, $type);
1026
			$type = 'c';
1027

1028
		} elsif ($cur =~ /^($Type)\s*(?:$Ident|,|\)|\(|\s*$)/) {
1029
			print "DECLARE($1)\n" if ($dbg_values > 1);
1030
1031
			$type = 'T';

1032
1033
1034
1035
		} elsif ($cur =~ /^($Modifier)\s*/) {
			print "MODIFIER($1)\n" if ($dbg_values > 1);
			$type = 'T';

1036
		} elsif ($cur =~ /^(\#\s*define\s*$Ident)(\(?)/o) {
1037
			print "DEFINE($1,$2)\n" if ($dbg_values > 1);
1038
			$av_preprocessor = 1;
1039
1040
1041
1042
1043
1044
			push(@av_paren_type, $type);
			if ($2 ne '') {
				$av_pending = 'N';
			}
			$type = 'E';

1045
		} elsif ($cur =~ /^(\#\s*(?:undef\s*$Ident|include\b))/o) {
1046
1047
1048
			print "UNDEF($1)\n" if ($dbg_values > 1);
			$av_preprocessor = 1;
			push(@av_paren_type, $type);
1049

1050
		} elsif ($cur =~ /^(\#\s*(?:ifdef|ifndef|if))/o) {
1051
			print "PRE_START($1)\n" if ($dbg_values > 1);
1052
			$av_preprocessor = 1;
1053
1054
1055

			push(@av_paren_type, $type);
			push(@av_paren_type, $type);
1056
			$type = 'E';
1057

1058
		} elsif ($cur =~ /^(\#\s*(?:else|elif))/o) {
1059
1060
1061
1062
1063
			print "PRE_RESTART($1)\n" if ($dbg_values > 1);
			$av_preprocessor = 1;

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

1064
			$type = 'E';
1065

1066
		} elsif ($cur =~ /^(\#\s*(?:endif))/o) {
1067
1068
1069
1070
1071
1072
1073
1074
			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);
1075
			$type = 'E';
1076
1077

		} elsif ($cur =~ /^(\\\n)/o) {
1078
			print "PRECONT($1)\n" if ($dbg_values > 1);
1079

1080
1081
1082
1083
1084
		} elsif ($cur =~ /^(__attribute__)\s*\(?/o) {
			print "ATTR($1)\n" if ($dbg_values > 1);
			$av_pending = $type;
			$type = 'N';

1085
		} elsif ($cur =~ /^(sizeof)\s*(\()?/o) {
1086
			print "SIZEOF($1)\n" if ($dbg_values > 1);
1087
			if (defined $2) {
1088
				$av_pending = 'V';
1089
1090
1091
			}
			$type = 'N';

1092
		} elsif ($cur =~ /^(if|while|for)\b/o) {
1093
			print "COND($1)\n" if ($dbg_values > 1);
1094
			$av_pending = 'E';
1095
1096
			$type = 'N';

1097
1098
1099
1100
1101
		} elsif ($cur =~/^(case)/o) {
			print "CASE($1)\n" if ($dbg_values > 1);
			$av_pend_colon = 'C';
			$type = 'N';

1102
		} elsif ($cur =~/^(return|else|goto|typeof|__typeof__)\b/o) {
1103
			print "KEYWORD($1)\n" if ($dbg_values > 1);
1104
1105
1106
			$type = 'N';

		} elsif ($cur =~ /^(\()/o) {
1107
			print "PAREN('$1')\n" if ($dbg_values > 1);
1108
1109
			push(@av_paren_type, $av_pending);
			$av_pending = '_';
1110
1111
1112
			$type = 'N';

		} elsif ($cur =~ /^(\))/o) {
1113
1114
1115
			my $new_type = pop(@av_paren_type);
			if ($new_type ne '_') {
				$type = $new_type;
1116
1117
				print "PAREN('$1') -> $type\n"
							if ($dbg_values > 1);
1118
			} else {
1119
				print "PAREN('$1')\n" if ($dbg_values > 1);
1120
1121
			}

1122
		} elsif ($cur =~ /^($Ident)\s*\(/o) {
1123
			print "FUNC($1)\n" if ($dbg_values > 1);
1124
			$type = 'V';
1125
			$av_pending = 'V';
1126

1127
1128
		} elsif ($cur =~ /^($Ident\s*):(?:\s*\d+\s*(,|=|;))?/) {
			if (defined $2 && $type eq 'C' || $type eq 'T') {
1129
				$av_pend_colon = 'B';
1130
1131
			} elsif ($type eq 'E') {
				$av_pend_colon = 'L';
1132
1133
1134
1135
			}
			print "IDENT_COLON($1,$type>$av_pend_colon)\n" if ($dbg_values > 1);
			$type = 'V';

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

		} elsif ($cur =~ /^($Assignment)/o) {
1141
			print "ASSIGN($1)\n" if ($dbg_values > 1);
1142
1143
			$type = 'N';

1144
		} elsif ($cur =~/^(;|{|})/) {
1145
			print "END($1)\n" if ($dbg_values > 1);
1146
			$type = 'E';
1147
1148
			$av_pend_colon = 'O';

1149
1150
1151
1152
		} elsif ($cur =~/^(,)/) {
			print "COMMA($1)\n" if ($dbg_values > 1);
			$type = 'C';

1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
		} 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';
1167

1168
		} elsif ($cur =~ /^(\[)/o) {
1169
			print "CLOSE($1)\n" if ($dbg_values > 1);
1170
1171
			$type = 'N';

1172
		} elsif ($cur =~ /^(-(?![->])|\+(?!\+)|\*|\&\&|\&)/o) {
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
			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';

1185
		} elsif ($cur =~ /^($Operators)/o) {
1186
			print "OP($1)\n" if ($dbg_values > 1);
1187
1188
1189
1190
1191
			if ($1 ne '++' && $1 ne '--') {
				$type = 'N';
			}

		} elsif ($cur =~ /(^.)/o) {
1192
			print "C($1)\n" if ($dbg_values > 1);
1193
1194
1195
1196
1197
		}
		if (defined $1) {
			$cur = substr($cur, length($1));
			$res .= $type x length($1);
		}
1198
	}
1199

1200
	return ($res, $var);
1201
1202
}

1203
sub possible {
1204
	my ($possible, $line) = @_;
1205
	my $notPermitted = qr{(?:
1206
1207
1208
1209
		^(?:
			$Modifier|
			$Storage|
			$Type|