summaryrefslogtreecommitdiffstats
path: root/src/pmdk/utils/cstyle
blob: 5221a2d56c4e35a6d5d0c0717362762a1eebc205 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
#!/usr/bin/env perl
#
# CDDL HEADER START
#
# The contents of this file are subject to the terms of the
# Common Development and Distribution License (the "License").
# You may not use this file except in compliance with the License.
#
# You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
# or http://www.opensolaris.org/os/licensing.
# See the License for the specific language governing permissions
# and limitations under the License.
#
# When distributing Covered Code, include this CDDL HEADER in each
# file and include the License file at usr/src/OPENSOLARIS.LICENSE.
# If applicable, add the following below this CDDL HEADER, with the
# fields enclosed by brackets "[]" replaced with your own identifying
# information: Portions Copyright [yyyy] [name of copyright owner]
#
# CDDL HEADER END
#
#
# Copyright 2008 Sun Microsystems, Inc.  All rights reserved.
# Use is subject to license terms.
#
# Portions copyright 2017, Intel Corporation.
#
# @(#)cstyle 1.58 98/09/09 (from shannon)
#ident	"%Z%%M%	%I%	%E% SMI"
#
# cstyle - check for some common stylistic errors.
#
#	cstyle is a sort of "lint" for C coding style.
#	It attempts to check for the style used in the
#	kernel, sometimes known as "Bill Joy Normal Form".
#
#	There's a lot this can't check for, like proper indentation
#	of code blocks.  There's also a lot more this could check for.
#
#	A note to the non perl literate:
#
#		perl regular expressions are pretty much like egrep
#		regular expressions, with the following special symbols
#
#		\s	any space character
#		\S	any non-space character
#		\w	any "word" character [a-zA-Z0-9_]
#		\W	any non-word character
#		\d	a digit [0-9]
#		\D	a non-digit
#		\b	word boundary (between \w and \W)
#		\B	non-word boundary
#

require 5.0;
use IO::File;
use Getopt::Std;
use strict;
use warnings;

my $usage =
"usage: cstyle [-chpvCP] [-o constructs] file ...
	-c	check continuation indentation inside functions
	-h	perform heuristic checks that are sometimes wrong
	-p	perform some of the more picky checks
	-v	verbose
	-C	don't check anything in header block comments
	-P	check for use of non-POSIX types
	-o constructs
		allow a comma-separated list of optional constructs:
		    doxygen	allow doxygen-style block comments (/** /*!)
		    splint	allow splint-style lint comments (/*@ ... @*/)
";

my %opts;

if (!getopts("cho:pvCP", \%opts)) {
	print $usage;
	exit 2;
}

my $check_continuation = $opts{'c'};
my $heuristic = $opts{'h'};
my $picky = $opts{'p'};
my $verbose = $opts{'v'};
my $ignore_hdr_comment = $opts{'C'};
my $check_posix_types = $opts{'P'};

my $doxygen_comments = 0;
my $splint_comments = 0;

if (defined($opts{'o'})) {
	for my $x (split /,/, $opts{'o'}) {
		if ($x eq "doxygen") {
			$doxygen_comments = 1;
		} elsif ($x eq "splint") {
			$splint_comments = 1;
		} else {
			print "cstyle: unrecognized construct \"$x\"\n";
			print $usage;
			exit 2;
		}
	}
}

my ($filename, $line, $prev);		# shared globals

my $fmt;
my $hdr_comment_start;

if ($verbose) {
	$fmt = "%s:%d: %s\n%s\n";
} else {
	$fmt = "%s:%d: %s\n";
}

if ($doxygen_comments) {
	# doxygen comments look like "/*!" or "/**"; allow them.
	$hdr_comment_start = qr/^\s*\/\*[\!\*]?$/;
} else {
	$hdr_comment_start = qr/^\s*\/\*$/;
}

# Note, following must be in single quotes so that \s and \w work right.
my $typename = '(int|char|short|long|unsigned|float|double' .
    '|\w+_t|struct\s+\w+|union\s+\w+|FILE|BOOL)';

# mapping of old types to POSIX compatible types
my %old2posix = (
	'unchar' => 'uchar_t',
	'ushort' => 'ushort_t',
	'uint' => 'uint_t',
	'ulong' => 'ulong_t',
	'u_int' => 'uint_t',
	'u_short' => 'ushort_t',
	'u_long' => 'ulong_t',
	'u_char' => 'uchar_t',
	'quad' => 'quad_t'
);

my $lint_re = qr/\/\*(?:
	ARGSUSED[0-9]*|NOTREACHED|LINTLIBRARY|VARARGS[0-9]*|
	CONSTCOND|CONSTANTCOND|CONSTANTCONDITION|EMPTY|
	FALLTHRU|FALLTHROUGH|LINTED.*?|PRINTFLIKE[0-9]*|
	PROTOLIB[0-9]*|SCANFLIKE[0-9]*|CSTYLED.*?
    )\*\//x;

my $splint_re = qr/\/\*@.*?@\*\//x;

my $warlock_re = qr/\/\*\s*(?:
	VARIABLES\ PROTECTED\ BY|
	MEMBERS\ PROTECTED\ BY|
	ALL\ MEMBERS\ PROTECTED\ BY|
	READ-ONLY\ VARIABLES:|
	READ-ONLY\ MEMBERS:|
	VARIABLES\ READABLE\ WITHOUT\ LOCK:|
	MEMBERS\ READABLE\ WITHOUT\ LOCK:|
	LOCKS\ COVERED\ BY|
	LOCK\ UNNEEDED\ BECAUSE|
	LOCK\ NEEDED:|
	LOCK\ HELD\ ON\ ENTRY:|
	READ\ LOCK\ HELD\ ON\ ENTRY:|
	WRITE\ LOCK\ HELD\ ON\ ENTRY:|
	LOCK\ ACQUIRED\ AS\ SIDE\ EFFECT:|
	READ\ LOCK\ ACQUIRED\ AS\ SIDE\ EFFECT:|
	WRITE\ LOCK\ ACQUIRED\ AS\ SIDE\ EFFECT:|
	LOCK\ RELEASED\ AS\ SIDE\ EFFECT:|
	LOCK\ UPGRADED\ AS\ SIDE\ EFFECT:|
	LOCK\ DOWNGRADED\ AS\ SIDE\ EFFECT:|
	FUNCTIONS\ CALLED\ THROUGH\ POINTER|
	FUNCTIONS\ CALLED\ THROUGH\ MEMBER|
	LOCK\ ORDER:
    )/x;

my $err_stat = 0;		# exit status

if ($#ARGV >= 0) {
	foreach my $arg (@ARGV) {
		my $fh = new IO::File $arg, "r";
		if (!defined($fh)) {
			printf "%s: can not open\n", $arg;
		} else {
			&cstyle($arg, $fh);
			close $fh;
		}
	}
} else {
	&cstyle("<stdin>", *STDIN);
}
exit $err_stat;

my $no_errs = 0;		# set for CSTYLED-protected lines

sub err($) {
	my ($error) = @_;
	unless ($no_errs) {
		if ($verbose) {
			printf $fmt, $filename, $., $error, $line;
		} else {
			printf $fmt, $filename, $., $error;
		}
		$err_stat = 1;
	}
}

sub err_prefix($$) {
	my ($prevline, $error) = @_;
	my $out = $prevline."\n".$line;
	unless ($no_errs) {
		printf $fmt, $filename, $., $error, $out;
		$err_stat = 1;
	}
}

sub err_prev($) {
	my ($error) = @_;
	unless ($no_errs) {
		printf $fmt, $filename, $. - 1, $error, $prev;
		$err_stat = 1;
	}
}

sub cstyle($$) {

my ($fn, $filehandle) = @_;
$filename = $fn;			# share it globally

my $in_cpp = 0;
my $next_in_cpp = 0;

my $in_comment = 0;
my $in_header_comment = 0;
my $comment_done = 0;
my $in_warlock_comment = 0;
my $in_function = 0;
my $in_function_header = 0;
my $in_declaration = 0;
my $note_level = 0;
my $nextok = 0;
my $nocheck = 0;

my $in_string = 0;

my ($okmsg, $comment_prefix);

$line = '';
$prev = '';
reset_indent();

line: while (<$filehandle>) {
	s/\r?\n$//;	# strip return and newline

	# save the original line, then remove all text from within
	# double or single quotes, we do not want to check such text.

	$line = $_;

	#
	# C allows strings to be continued with a backslash at the end of
	# the line.  We translate that into a quoted string on the previous
	# line followed by an initial quote on the next line.
	#
	# (we assume that no-one will use backslash-continuation with character
	# constants)
	#
	$_ = '"' . $_		if ($in_string && !$nocheck && !$in_comment);

	#
	# normal strings and characters
	#
	s/'([^\\']|\\[^xX0]|\\0[0-9]*|\\[xX][0-9a-fA-F]*)'/''/g;
	s/"([^\\"]|\\.)*"/\"\"/g;

	#
	# detect string continuation
	#
	if ($nocheck || $in_comment) {
		$in_string = 0;
	} else {
		#
		# Now that all full strings are replaced with "", we check
		# for unfinished strings continuing onto the next line.
		#
		$in_string =
		    (s/([^"](?:"")*)"([^\\"]|\\.)*\\$/$1""/ ||
		    s/^("")*"([^\\"]|\\.)*\\$/""/);
	}

	#
	# figure out if we are in a cpp directive
	#
	$in_cpp = $next_in_cpp || /^\s*#/;	# continued or started
	$next_in_cpp = $in_cpp && /\\$/;	# only if continued

	# strip off trailing backslashes, which appear in long macros
	s/\s*\\$//;

	# an /* END CSTYLED */ comment ends a no-check block.
	if ($nocheck) {
		if (/\/\* *END *CSTYLED *\*\//) {
			$nocheck = 0;
		} else {
			reset_indent();
			next line;
		}
	}

	# a /*CSTYLED*/ comment indicates that the next line is ok.
	if ($nextok) {
		if ($okmsg) {
			err($okmsg);
		}
		$nextok = 0;
		$okmsg = 0;
		if (/\/\* *CSTYLED.*\*\//) {
			/^.*\/\* *CSTYLED *(.*) *\*\/.*$/;
			$okmsg = $1;
			$nextok = 1;
		}
		$no_errs = 1;
	} elsif ($no_errs) {
		$no_errs = 0;
	}

	# check length of line.
	# first, a quick check to see if there is any chance of being too long.
	if (($line =~ tr/\t/\t/) * 7 + length($line) > 80) {
		# yes, there is a chance.
		# replace tabs with spaces and check again.
		my $eline = $line;
		1 while $eline =~
		    s/\t+/' ' x (length($&) * 8 - length($`) % 8)/e;
		if (length($eline) > 80) {
			# allow long line if it is user visible string
			# find if line start from " or L" and ends
			# with " + 2 optional characters
			# (these characters can be i.e. '");' '" \' or '",' etc...)
			if($eline =~ /^ *L?".*"[^"]{0,2}$/) {
				# check if entire line is one string literal
				$eline =~ s/^ *L?"//;
				$eline =~ s/"[^"]{0,2}$//;

				if($eline =~ /[^\\]"|[^\\](\\\\)+"/) {
					err("line > 80 characters");
				}
			} else {
				err("line > 80 characters");
			}
		}
	}

	# ignore NOTE(...) annotations (assumes NOTE is on lines by itself).
	if ($note_level || /\b_?NOTE\s*\(/) { # if in NOTE or this is NOTE
		s/[^()]//g;			  # eliminate all non-parens
		$note_level += s/\(//g - length;  # update paren nest level
		next;
	}

	# a /* BEGIN CSTYLED */ comment starts a no-check block.
	if (/\/\* *BEGIN *CSTYLED *\*\//) {
		$nocheck = 1;
	}

	# a /*CSTYLED*/ comment indicates that the next line is ok.
	if (/\/\* *CSTYLED.*\*\//) {
		/^.*\/\* *CSTYLED *(.*) *\*\/.*$/;
		$okmsg = $1;
		$nextok = 1;
	}
	if (/\/\/ *CSTYLED/) {
		/^.*\/\/ *CSTYLED *(.*)$/;
		$okmsg = $1;
		$nextok = 1;
	}

	# universal checks; apply to everything
	if (/\t +\t/) {
		err("spaces between tabs");
	}
	if (/ \t+ /) {
		err("tabs between spaces");
	}
	if (/\s$/) {
		err("space or tab at end of line");
	}
	if (/[^ \t(]\/\*/ && !/\w\(\/\*.*\*\/\);/) {
		err("comment preceded by non-blank");
	}

	# is this the beginning or ending of a function?
	# (not if "struct foo\n{\n")
	if (/^{$/ && $prev =~ /\)\s*(const\s*)?(\/\*.*\*\/\s*)?\\?$/) {
		$in_function = 1;
		$in_declaration = 1;
		$in_function_header = 0;
		$prev = $line;
		next line;
	}
	if (/^}\s*(\/\*.*\*\/\s*)*$/) {
		if ($prev =~ /^\s*return\s*;/) {
			err_prev("unneeded return at end of function");
		}
		$in_function = 0;
		reset_indent();		# we don't check between functions
		$prev = $line;
		next line;
	}
	if (/^\w*\($/) {
		$in_function_header = 1;
	}

	if ($in_warlock_comment && /\*\//) {
		$in_warlock_comment = 0;
		$prev = $line;
		next line;
	}

	# a blank line terminates the declarations within a function.
	# XXX - but still a problem in sub-blocks.
	if ($in_declaration && /^$/) {
		$in_declaration = 0;
	}

	if ($comment_done) {
		$in_comment = 0;
		$in_header_comment = 0;
		$comment_done = 0;
	}
	# does this looks like the start of a block comment?
	if (/$hdr_comment_start/) {
		if (!/^\t*\/\*/) {
			err("block comment not indented by tabs");
		}
		$in_comment = 1;
		/^(\s*)\//;
		$comment_prefix = $1;
		if ($comment_prefix eq "") {
			$in_header_comment = 1;
		}
		$prev = $line;
		next line;
	}
	# are we still in the block comment?
	if ($in_comment) {
		if (/^$comment_prefix \*\/$/) {
			$comment_done = 1;
		} elsif (/\*\//) {
			$comment_done = 1;
			err("improper block comment close")
			    unless ($ignore_hdr_comment && $in_header_comment);
		} elsif (!/^$comment_prefix \*[ \t]/ &&
		    !/^$comment_prefix \*$/) {
			err("improper block comment")
			    unless ($ignore_hdr_comment && $in_header_comment);
		}
	}

	if ($in_header_comment && $ignore_hdr_comment) {
		$prev = $line;
		next line;
	}

	# check for errors that might occur in comments and in code.

	# allow spaces to be used to draw pictures in header and block comments.
	if (/[^ ]     / && !/".*     .*"/ && !$in_header_comment && !$in_comment) {
		err("spaces instead of tabs");
	}
	if (/^ / && !/^ \*[ \t\/]/ && !/^ \*$/ &&
	    (!/^    \w/ || $in_function != 0)) {
		err("indent by spaces instead of tabs");
	}
	if (/^\t+ [^ \t\*]/ || /^\t+  \S/ || /^\t+   \S/) {
		err("continuation line not indented by 4 spaces");
	}
	if (/$warlock_re/ && !/\*\//) {
		$in_warlock_comment = 1;
		$prev = $line;
		next line;
	}
	if (/^\s*\/\*./ && !/^\s*\/\*.*\*\// && !/$hdr_comment_start/) {
		err("improper first line of block comment");
	}

	if ($in_comment) {	# still in comment, don't do further checks
		$prev = $line;
		next line;
	}

	if ((/[^(]\/\*\S/ || /^\/\*\S/) &&
	    !(/$lint_re/ || ($splint_comments && /$splint_re/))) {
		err("missing blank after open comment");
	}
	if (/\S\*\/[^)]|\S\*\/$/ &&
	    !(/$lint_re/ || ($splint_comments && /$splint_re/))) {
		err("missing blank before close comment");
	}
	if (/\/\/\S/) {		# C++ comments
		err("missing blank after start comment");
	}
	# check for unterminated single line comments, but allow them when
	# they are used to comment out the argument list of a function
	# declaration.
	if (/\S.*\/\*/ && !/\S.*\/\*.*\*\// && !/\(\/\*/) {
		err("unterminated single line comment");
	}

	if (/^(#else|#endif|#include)(.*)$/) {
		$prev = $line;
		if ($picky) {
			my $directive = $1;
			my $clause = $2;
			# Enforce ANSI rules for #else and #endif: no noncomment
			# identifiers are allowed after #endif or #else.  Allow
			# C++ comments since they seem to be a fact of life.
			if ((($1 eq "#endif") || ($1 eq "#else")) &&
			    ($clause ne "") &&
			    (!($clause =~ /^\s+\/\*.*\*\/$/)) &&
			    (!($clause =~ /^\s+\/\/.*$/))) {
				err("non-comment text following " .
				    "$directive (or malformed $directive " .
				    "directive)");
			}
		}
		next line;
	}

	#
	# delete any comments and check everything else.  Note that
	# ".*?" is a non-greedy match, so that we don't get confused by
	# multiple comments on the same line.
	#
	s/\/\*.*?\*\//\x01/g;
	s/\/\/.*$/\x01/;		# C++ comments

	# delete any trailing whitespace; we have already checked for that.
	s/\s*$//;

	# following checks do not apply to text in comments.

	if (/[^ \t\+]\+[^\+=]/ || /[^\+]\+[^ \+=]/) {
		err("missing space around + operator");
	}
	if (/[^ \t]\+=/ || /\+=[^ ]/) {
		err("missing space around += operator");
	}
	if (/[^ \t\-]\-[^\->]/ && !/\(\w+\)\-\w/ && !/[\(\[]\-[\w \t]+[\)\],]/) {
		err("missing space before - operator");
	}
	if (/[^\-]\-[^ \-=>]/ && !/\(\-\w+\)/ &&
		!/(return|case|=|>|<|\?|:|,|^[ \t]+)[ \t]+\-[\w\(]/ && !/(\([^\)]+\)|\[|\()\-[\w\(\]]/) {
		err("missing space after - operator");
	}
	if (/(return|case|=|\?|:|,|\[)[ \t]+\-[ \t]/ || /[\(\[]\-[ \t]/) {
		err("extra space after - operator");
	}
	if (/[ \t]\+\+ /) {
		err("extra space before or after ++ operator");
	}
	if (/[ \t]\-\- /) {
		err("extra space before or after -- operator");
	}
	if (/[^ \t]\-=/ || /\-=[^ ]/) {
		err("missing space around -= operator");
	}
	if (/[^ \t][\%\/]/ || /[\%\/][^ =]/ || /[\%\/]=[^ ]/) {
		err("missing space around one of operators: % %= / /=");
	}
	if (/[^ \t]\*=/ || /\*=[^ ]/) {
		err("missing space around *= operator");
	}
	if (/[^ \t\(\)\*\[]\*/) {
		err("missing space before * operator");
	}
	if (/\*[^ =\*\w\(,]/ && !/\(.+ \*+\)/ && !/\*\[\]/ &&
	    !/\*\-\-\w/ && !/\*\+\+\w/ && !/\*\)/) {
		err("missing space after * operator");
	}
	if (/[^<>\s][!<>=]=/ || /[^<>][!<>=]=[^\s,]/ ||
	    (/[^->]>[^,=>\s]/ && !/[^->]>$/) ||
	    (/[^<]<[^,=<\s]/ && !/[^<]<$/) ||
	    /[^<\s]<[^<]/ || /[^->\s]>[^>]/) {
		err("missing space around relational operator");
	}
	if (/\S>>=/ || /\S<<=/ || />>=\S/ || /<<=\S/ || /\S[-+*\/&|^%]=/ ||
	    (/[^-+*\/&|^%!<>=\s]=[^=]/ && !/[^-+*\/&|^%!<>=\s]=$/) ||
	    (/[^!<>=]=[^=\s]/ && !/[^!<>=]=$/)) {
		# XXX - should only check this for C++ code
		# XXX - there are probably other forms that should be allowed
		if (!/\soperator=/) {
			err("missing space around assignment operator");
		}
	}
	if (/[,;]\S/ && !/\bfor \(;;\)/) {
		err("comma or semicolon followed by non-blank");
	}
	# allow "for" statements to have empty "while" clauses
	if (/\s[,;]/ && !/^[\t]+;$/ && !/^\s*for \([^;]*; ;[^;]*\)/) {
		err("comma or semicolon preceded by blank");
	}
	if (/^\s*(&&|\|\|)/) {
		err("improper boolean continuation");
	}
	if (/\S   *(&&|\|\|)/ || /(&&|\|\|)   *\S/) {
		err("more than one space around boolean operator");
	}
	if (/\b(for|if|while|switch|return|case)\(/) {
		err("missing space between keyword and paren");
	}
	if (/(\b(for|if|while|switch|return)\b.*){2,}/ && !/^#define/) {
		# multiple "case" and "sizeof" allowed
		err("more than one keyword on line");
	}
	if (/\b(for|if|while|switch|return|case)\s\s+\(/ &&
	    !/^#if\s+\(/) {
		err("extra space between keyword and paren");
	}
	# try to detect "func (x)" but not "if (x)" or
	# "#define foo (x)" or "int (*func)();"
	if (/\w\s\(/) {
		my $s = $_;
		# strip off all keywords on the line
		s/\b(for|if|while|switch|return|case)\s\(/XXX(/g;
		s/\b(sizeof|typeof|__typeof__)\s*\(/XXX(/g;
		s/#elif\s\(/XXX(/g;
		s/^#define\s+\w+\s+\(/XXX(/;
		# do not match things like "void (*f)();"
		# or "typedef void (func_t)();"
		s/\w\s\(+\*/XXX(*/g;
		s/\b($typename|void)\s+\(+/XXX(/og;
		s/\btypedef\s($typename|void)\s+\(+/XXX(/og;
		# do not match "__attribute__ ((format (...)))"
		s/\b__attribute__\s*\(\(format\s*\(/__attribute__((XXX(/g;
		if (/\w\s\(/) {
			err("extra space between function name and left paren");
		}
		$_ = $s;
	}
	# try to detect "int foo(x)", but not "extern int foo(x);"
	# XXX - this still trips over too many legitimate things,
	# like "int foo(x,\n\ty);"
#		if (/^(\w+(\s|\*)+)+\w+\(/ && !/\)[;,](\s|\x01)*$/ &&
#		    !/^(extern|static)\b/) {
#			err("return type of function not on separate line");
#		}
	# this is a close approximation
	if (/^(\w+(\s|\*)+)+\w+\(.*\)(\s|\x01)*$/ &&
	    !/^(extern|static)\b/) {
		err("return type of function not on separate line");
	}
	if (/^#define\t/ || /^#ifdef\t/ || /^#ifndef\t/) {
		err("#define/ifdef/ifndef followed by tab instead of space");
	}
	if (/^#define\s\s+/ || /^#ifdef\s\s+/ || /^#ifndef\s\s+/) {
		err("#define/ifdef/ifndef followed by more than one space");
	}
	# AON C-style doesn't require this.
	#if (/^\s*return\W[^;]*;/ && !/^\s*return\s*\(.*\);/) {
	#	err("unparenthesized return expression");
	#}
	if (/\bsizeof\b/ && !/\bsizeof\s*\(.*\)/) {
		err("unparenthesized sizeof expression");
	}
	if (/\b(sizeof|typeof)\b/ && /\b(sizeof|typeof)\s+\(.*\)/) {
		err("spaces between sizeof/typeof expression and paren");
	}
	if (/\(\s/) {
		err("whitespace after left paren");
	}
	# allow "for" statements to have empty "continue" clauses
	if (/\s\)/ && !/^\s*for \([^;]*;[^;]*; \)/) {
		err("whitespace before right paren");
	}
	if (/^\s*\(void\)[^ ]/) {
		err("missing space after (void) cast");
	}
	if (/\S\{/ && !/\{\{/ && !/\(struct \w+\)\{/) {
		err("missing space before left brace");
	}
	if ($in_function && /^\s+{/ &&
	    ($prev =~ /\)\s*$/ || $prev =~ /\bstruct\s+\w+$/)) {
		err("left brace starting a line");
	}
	if (/}(else|while)/) {
		err("missing space after right brace");
	}
	if (/}\s\s+(else|while)/) {
		err("extra space after right brace");
	}
	if (/\b_VOID\b|\bVOID\b|\bSTATIC\b/) {
		err("obsolete use of VOID or STATIC");
	}
	if (/\b($typename|void)\*/o) {
		err("missing space between type name and *");
	}
	if (/^\s+#/) {
		err("preprocessor statement not in column 1");
	}
	if (/^#\s/) {
		err("blank after preprocessor #");
	}
	if (/!\s*(strcmp|strncmp|bcmp)\s*\(/) {
		err("don't use boolean ! with comparison functions");
	}
	if (/^\S+\([\S\s]*\)\s*{/) {
		err("brace of function definition not at beginning of line");
	}
	if (/static\s+\S+\s*=\s*(0|NULL)\s*;/) {
		err("static variable initialized with 0 or NULL");
	}
	if (/typedef[\S\s]+\*\s*\w+\s*;/) {
		err("typedefed pointer type");
	}
	if (/unsigned\s+int\s/) {
		err("'unsigned int' instead of just 'unsigned'");
	}
	if (/long\s+long\s+int\s/) {
		err("'long long int' instead of just 'long long'");
	} elsif (/long\s+int\s/) {
		err("'long int' instead of just 'long'");
	}

	#
	# We completely ignore, for purposes of indentation:
	#  * lines outside of functions
	#  * preprocessor lines
	#
	if ($check_continuation && $in_function && !$in_cpp) {
		process_indent($_);
	}
	if ($picky) {
		# try to detect spaces after casts, but allow (e.g.)
		# "sizeof (int) + 1", "void (*funcptr)(int) = foo;", and
		# "int foo(int) __NORETURN;"
		if ((/^\($typename( \*+)?\)\s/o ||
		    /\W\($typename( \*+)?\)\s/o) &&
		    !/sizeof\($typename( \*)?\)\s/o &&
		    !/\($typename( \*+)?\)\s+=[^=]/o) {
			err("space after cast");
		}
		if (/\b($typename|void)\s*\*\s/o &&
		    !/\b($typename|void)\s*\*\s+const\b/o) {
			err("unary * followed by space");
		}
	}
	if ($check_posix_types) {
		# try to detect old non-POSIX types.
		# POSIX requires all non-standard typedefs to end in _t,
		# but historically these have been used.
		if (/\b(unchar|ushort|uint|ulong|u_int|u_short|u_long|u_char|quad)\b/) {
			err("non-POSIX typedef $1 used: use $old2posix{$1} instead");
		}
	}
	if ($heuristic) {
		# cannot check this everywhere due to "struct {\n...\n} foo;"
		if ($in_function && !$in_declaration &&
		    /}./ && !/}\s+=/ && !/{.*}[;,]$/ && !/}(\s|\x01)*$/ &&
		    !/} (else|while)/ && !/}}/) {
			err("possible bad text following right brace");
		}
		# cannot check this because sub-blocks in
		# the middle of code are ok
		if ($in_function && /^\s+{/) {
			err("possible left brace starting a line");
		}
	}
	if (/^\s*else\W/) {
		if ($prev =~ /^\s*}$/) {
			err_prefix($prev,
			    "else and right brace should be on same line");
		}
	}
	$prev = $line;
}

if ($prev eq "") {
	err("last line in file is blank");
}

}

#
# Continuation-line checking
#
# The rest of this file contains the code for the continuation checking
# engine.  It's a pretty simple state machine which tracks the expression
# depth (unmatched '('s and '['s).
#
# Keep in mind that the argument to process_indent() has already been heavily
# processed; all comments have been replaced by control-A, and the contents of
# strings and character constants have been elided.
#

my $cont_in;		# currently inside of a continuation
my $cont_off;		# skipping an initializer or definition
my $cont_noerr;		# suppress cascading errors
my $cont_start;		# the line being continued
my $cont_base;		# the base indentation
my $cont_first;		# this is the first line of a statement
my $cont_multiseg;	# this continuation has multiple segments

my $cont_special;	# this is a C statement (if, for, etc.)
my $cont_macro;		# this is a macro
my $cont_case;		# this is a multi-line case

my @cont_paren;		# the stack of unmatched ( and [s we've seen

sub
reset_indent()
{
	$cont_in = 0;
	$cont_off = 0;
}

sub
delabel($)
{
	#
	# replace labels with tabs.  Note that there may be multiple
	# labels on a line.
	#
	local $_ = $_[0];

	while (/^(\t*)( *(?:(?:\w+\s*)|(?:case\b[^:]*)): *)(.*)$/) {
		my ($pre_tabs, $label, $rest) = ($1, $2, $3);
		$_ = $pre_tabs;
		while ($label =~ s/^([^\t]*)(\t+)//) {
			$_ .= "\t" x (length($2) + length($1) / 8);
		}
		$_ .= ("\t" x (length($label) / 8)).$rest;
	}

	return ($_);
}

sub
process_indent($)
{
	require strict;
	local $_ = $_[0];			# preserve the global $_

	s/\x01//g;	# No comments
	s/\s+$//;	# Strip trailing whitespace

	return			if (/^$/);	# skip empty lines

	# regexps used below; keywords taking (), macros, and continued cases
	my $special = '(?:(?:\}\s*)?else\s+)?(?:if|for|while|switch)\b';
	my $macro = '[A-Z_][A-Z_0-9]*\(';
	my $case = 'case\b[^:]*$';

	# skip over enumerations, array definitions, initializers, etc.
	if ($cont_off <= 0 && !/^\s*$special/ &&
	    (/(?:(?:\b(?:enum|struct|union)\s*[^\{]*)|(?:\s+=\s*))\{/ ||
	    (/^\s*{/ && $prev =~ /=\s*(?:\/\*.*\*\/\s*)*$/))) {
		$cont_in = 0;
		$cont_off = tr/{/{/ - tr/}/}/;
		return;
	}
	if ($cont_off) {
		$cont_off += tr/{/{/ - tr/}/}/;
		return;
	}

	if (!$cont_in) {
		$cont_start = $line;

		if (/^\t* /) {
			err("non-continuation indented 4 spaces");
			$cont_noerr = 1;		# stop reporting
		}
		$_ = delabel($_);	# replace labels with tabs

		# check if the statement is complete
		return		if (/^\s*\}?$/);
		return		if (/^\s*\}?\s*else\s*\{?$/);
		return		if (/^\s*do\s*\{?$/);
		return		if (/{$/);
		return		if (/}[,;]?$/);

		# Allow macros on their own lines
		return		if (/^\s*[A-Z_][A-Z_0-9]*$/);

		# cases we don't deal with, generally non-kosher
		if (/{/) {
			err("stuff after {");
			return;
		}

		# Get the base line, and set up the state machine
		/^(\t*)/;
		$cont_base = $1;
		$cont_in = 1;
		@cont_paren = ();
		$cont_first = 1;
		$cont_multiseg = 0;

		# certain things need special processing
		$cont_special = /^\s*$special/? 1 : 0;
		$cont_macro = /^\s*$macro/? 1 : 0;
		$cont_case = /^\s*$case/? 1 : 0;
	} else {
		$cont_first = 0;

		# Strings may be pulled back to an earlier (half-)tabstop
		unless ($cont_noerr || /^$cont_base    / ||
		    (/^\t*(?:    )?(?:gettext\()?\"/ && !/^$cont_base\t/)) {
			err_prefix($cont_start,
			    "continuation should be indented 4 spaces");
		}
	}

	my $rest = $_;			# keeps the remainder of the line

	#
	# The split matches 0 characters, so that each 'special' character
	# is processed separately.  Parens and brackets are pushed and
	# popped off the @cont_paren stack.  For normal processing, we wait
	# until a ; or { terminates the statement.  "special" processing
	# (if/for/while/switch) is allowed to stop when the stack empties,
	# as is macro processing.  Case statements are terminated with a :
	# and an empty paren stack.
	#
	foreach $_ (split /[^\(\)\[\]\{\}\;\:]*/) {
		next		if (length($_) == 0);

		# rest contains the remainder of the line
		my $rxp = "[^\Q$_\E]*\Q$_\E";
		$rest =~ s/^$rxp//;

		if (/\(/ || /\[/) {
			push @cont_paren, $_;
		} elsif (/\)/ || /\]/) {
			my $cur = $_;
			tr/\)\]/\(\[/;

			my $old = (pop @cont_paren);
			if (!defined($old)) {
				err("unexpected '$cur'");
				$cont_in = 0;
				last;
			} elsif ($old ne $_) {
				err("'$cur' mismatched with '$old'");
				$cont_in = 0;
				last;
			}

			#
			# If the stack is now empty, do special processing
			# for if/for/while/switch and macro statements.
			#
			next		if (@cont_paren != 0);
			if ($cont_special) {
				if ($rest =~ /^\s*{?$/) {
					$cont_in = 0;
					last;
				}
				if ($rest =~ /^\s*;$/) {
					err("empty if/for/while body ".
					    "not on its own line");
					$cont_in = 0;
					last;
				}
				if (!$cont_first && $cont_multiseg == 1) {
					err_prefix($cont_start,
					    "multiple statements continued ".
					    "over multiple lines");
					$cont_multiseg = 2;
				} elsif ($cont_multiseg == 0) {
					$cont_multiseg = 1;
				}
				# We've finished this section, start
				# processing the next.
				goto section_ended;
			}
			if ($cont_macro) {
				if ($rest =~ /^$/) {
					$cont_in = 0;
					last;
				}
			}
		} elsif (/\;/) {
			if ($cont_case) {
				err("unexpected ;");
			} elsif (!$cont_special) {
				err("unexpected ;")	if (@cont_paren != 0);
				if (!$cont_first && $cont_multiseg == 1) {
					err_prefix($cont_start,
					    "multiple statements continued ".
					    "over multiple lines");
					$cont_multiseg = 2;
				} elsif ($cont_multiseg == 0) {
					$cont_multiseg = 1;
				}
				if ($rest =~ /^$/) {
					$cont_in = 0;
					last;
				}
				if ($rest =~ /^\s*special/) {
					err("if/for/while/switch not started ".
					    "on its own line");
				}
				goto section_ended;
			}
		} elsif (/\{/) {
			err("{ while in parens/brackets") if (@cont_paren != 0);
			err("stuff after {")		if ($rest =~ /[^\s}]/);
			$cont_in = 0;
			last;
		} elsif (/\}/) {
			err("} while in parens/brackets") if (@cont_paren != 0);
			if (!$cont_special && $rest !~ /^\s*(while|else)\b/) {
				if ($rest =~ /^$/) {
					err("unexpected }");
				} else {
					err("stuff after }");
				}
				$cont_in = 0;
				last;
			}
		} elsif (/\:/ && $cont_case && @cont_paren == 0) {
			err("stuff after multi-line case") if ($rest !~ /$^/);
			$cont_in = 0;
			last;
		}
		next;
section_ended:
		# End of a statement or if/while/for loop.  Reset
		# cont_special and cont_macro based on the rest of the
		# line.
		$cont_special = ($rest =~ /^\s*$special/)? 1 : 0;
		$cont_macro = ($rest =~ /^\s*$macro/)? 1 : 0;
		$cont_case = 0;
		next;
	}
	$cont_noerr = 0			if (!$cont_in);
}