summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rwxr-xr-xscripts/checkpatch.pl294
1 files changed, 198 insertions, 96 deletions
diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 579f50fa838..545471a99ee 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -9,7 +9,7 @@ use strict;
my $P = $0;
$P =~ s@.*/@@g;
-my $V = '0.12';
+my $V = '0.13';
use Getopt::Long qw(:config no_auto_abbrev);
@@ -25,6 +25,7 @@ my $check = 0;
my $summary = 1;
my $mailback = 0;
my $root;
+my %debug;
GetOptions(
'q|quiet+' => \$quiet,
'tree!' => \$tree,
@@ -39,6 +40,7 @@ GetOptions(
'root=s' => \$root,
'summary!' => \$summary,
'mailback!' => \$mailback,
+ 'debug=s' => \%debug,
) or exit;
my $exit = 0;
@@ -56,6 +58,12 @@ if ($#ARGV < 0) {
exit(1);
}
+my $dbg_values = 0;
+my $dbg_possible = 0;
+for my $key (keys %debug) {
+ eval "\${dbg_$key} = '$debug{$key}';"
+}
+
if ($terse) {
$emacs = 1;
$quiet++;
@@ -110,7 +118,7 @@ our $Assignment = qr{(?:\*\=|/=|%=|\+=|-=|<<=|>>=|&=|\^=|\|=|=)};
our $Operators = qr{
<=|>=|==|!=|
=>|->|<<|>>|<|>|!|~|
- &&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/
+ &&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|%
}x;
our $NonptrType;
@@ -152,7 +160,7 @@ sub build_types {
$Type = qr{
\b$NonptrType\b
(?:\s*\*+\s*const|\s*\*+|(?:\s*\[\s*\])+)?
- (?:\s+$Sparse|\s+$Attribute)*
+ (?:\s+$Inline|\s+$Sparse|\s+$Attribute)*
}x;
$Declare = qr{(?:$Storage\s+)?$Type};
}
@@ -181,6 +189,8 @@ if ($tree && -f "$root/$removal") {
}
my @rawlines = ();
+my @lines = ();
+my $vname;
for my $filename (@ARGV) {
if ($file) {
open(FILE, "diff -u /dev/null $filename|") ||
@@ -189,12 +199,17 @@ for my $filename (@ARGV) {
open(FILE, "<$filename") ||
die "$P: $filename: open failed - $!\n";
}
+ if ($filename eq '-') {
+ $vname = 'Your patch';
+ } else {
+ $vname = $filename;
+ }
while (<FILE>) {
chomp;
push(@rawlines, $_);
}
close(FILE);
- if (!process($filename, @rawlines)) {
+ if (!process($filename)) {
$exit = 1;
}
@rawlines = ();
@@ -274,20 +289,30 @@ sub sanitise_line {
my $l = '';
my $quote = '';
+ my $qlen = 0;
foreach my $c (split(//, $line)) {
+ # The second backslash of a pair is not a "quote".
+ if ($l eq "\\" && $c eq "\\") {
+ $c = 'X';
+ }
if ($l ne "\\" && ($c eq "'" || $c eq '"')) {
if ($quote eq '') {
$quote = $c;
$res .= $c;
$l = $c;
+ $qlen = 0;
next;
} elsif ($quote eq $c) {
$quote = '';
}
}
+ if ($quote eq "'" && $qlen > 1) {
+ $quote = '';
+ }
if ($quote && $c ne "\t") {
$res .= "X";
+ $qlen++;
} else {
$res .= $c;
}
@@ -295,6 +320,28 @@ sub sanitise_line {
$l = $c;
}
+ # Clear out the comments.
+ while ($res =~ m@(/\*.*?\*/)@) {
+ substr($res, $-[1], $+[1] - $-[1]) = ' ' x ($+[1] - $-[1]);
+ }
+ if ($res =~ m@(/\*.*)@) {
+ substr($res, $-[1], $+[1] - $-[1]) = ' ' x ($+[1] - $-[1]);
+ }
+ if ($res =~ m@^.(.*\*/)@) {
+ substr($res, $-[1], $+[1] - $-[1]) = ' ' x ($+[1] - $-[1]);
+ }
+
+ # The pathname on a #include may be surrounded by '<' and '>'.
+ if ($res =~ /^.#\s*include\s+\<(.*)\>/) {
+ my $clean = 'X' x length($1);
+ $res =~ s@\<.*\>@<$clean>@;
+
+ # The whole of a #error is a string.
+ } elsif ($res =~ /^.#\s*(?:error|warning)\s+(.*)\b/) {
+ my $clean = 'X' x length($1);
+ $res =~ s@(#\s*(?:error|warning)\s+).*@$1$clean@;
+ }
+
return $res;
}
@@ -315,9 +362,9 @@ sub ctx_statement_block {
# context.
if ($off >= $len) {
for (; $remain > 0; $line++) {
- next if ($rawlines[$line] =~ /^-/);
+ next if ($lines[$line] =~ /^-/);
$remain--;
- $blk .= sanitise_line($rawlines[$line]) . "\n";
+ $blk .= $lines[$line] . "\n";
$len = length($blk);
$line++;
last;
@@ -500,103 +547,106 @@ sub cat_vet {
return $res;
}
+my $av_preprocessor = 0;
+my $av_paren = 0;
+my @av_paren_type;
+
+sub annotate_reset {
+ $av_preprocessor = 0;
+ $av_paren = 0;
+ @av_paren_type = ();
+}
+
sub annotate_values {
my ($stream, $type) = @_;
my $res;
my $cur = $stream;
- my $debug = 0;
-
- print "$stream\n" if ($debug);
-
- ##my $type = 'N';
- my $pos = 0;
- my $preprocessor = 0;
- my $paren = 0;
- my @paren_type;
+ print "$stream\n" if ($dbg_values > 1);
while (length($cur)) {
- print " <$type> " if ($debug);
+ print " <$type> " if ($dbg_values > 1);
if ($cur =~ /^(\s+)/o) {
- print "WS($1)\n" if ($debug);
- if ($1 =~ /\n/ && $preprocessor) {
- $preprocessor = 0;
+ print "WS($1)\n" if ($dbg_values > 1);
+ if ($1 =~ /\n/ && $av_preprocessor) {
+ $av_preprocessor = 0;
$type = 'N';
}
} elsif ($cur =~ /^($Type)/) {
- print "DECLARE($1)\n" if ($debug);
+ print "DECLARE($1)\n" if ($dbg_values > 1);
$type = 'T';
} elsif ($cur =~ /^(#\s*define\s*$Ident)(\(?)/o) {
- print "DEFINE($1)\n" if ($debug);
- $preprocessor = 1;
- $paren_type[$paren] = 'N';
+ print "DEFINE($1)\n" if ($dbg_values > 1);
+ $av_preprocessor = 1;
+ $av_paren_type[$av_paren] = 'N';
- } elsif ($cur =~ /^(#\s*(?:ifdef|ifndef|if|else|endif))/o) {
- print "PRE($1)\n" if ($debug);
- $preprocessor = 1;
+ } elsif ($cur =~ /^(#\s*(?:ifdef|ifndef|if|else|elif|endif))/o) {
+ print "PRE($1)\n" if ($dbg_values > 1);
+ $av_preprocessor = 1;
$type = 'N';
} elsif ($cur =~ /^(\\\n)/o) {
- print "PRECONT($1)\n" if ($debug);
+ print "PRECONT($1)\n" if ($dbg_values > 1);
} elsif ($cur =~ /^(sizeof)\s*(\()?/o) {
- print "SIZEOF($1)\n" if ($debug);
+ print "SIZEOF($1)\n" if ($dbg_values > 1);
if (defined $2) {
- $paren_type[$paren] = 'V';
+ $av_paren_type[$av_paren] = 'V';
}
$type = 'N';
} elsif ($cur =~ /^(if|while|typeof|for)\b/o) {
- print "COND($1)\n" if ($debug);
- $paren_type[$paren] = 'N';
+ print "COND($1)\n" if ($dbg_values > 1);
+ $av_paren_type[$av_paren] = 'N';
$type = 'N';
} elsif ($cur =~/^(return|case|else)/o) {
- print "KEYWORD($1)\n" if ($debug);
+ print "KEYWORD($1)\n" if ($dbg_values > 1);
$type = 'N';
} elsif ($cur =~ /^(\()/o) {
- print "PAREN('$1')\n" if ($debug);
- $paren++;
+ print "PAREN('$1')\n" if ($dbg_values > 1);
+ $av_paren++;
$type = 'N';
} elsif ($cur =~ /^(\))/o) {
- $paren-- if ($paren > 0);
- if (defined $paren_type[$paren]) {
- $type = $paren_type[$paren];
- undef $paren_type[$paren];
- print "PAREN('$1') -> $type\n" if ($debug);
+ $av_paren-- if ($av_paren > 0);
+ if (defined $av_paren_type[$av_paren]) {
+ $type = $av_paren_type[$av_paren];
+ undef $av_paren_type[$av_paren];
+ print "PAREN('$1') -> $type\n"
+ if ($dbg_values > 1);
} else {
- print "PAREN('$1')\n" if ($debug);
+ print "PAREN('$1')\n" if ($dbg_values > 1);
}
} elsif ($cur =~ /^($Ident)\(/o) {
- print "FUNC($1)\n" if ($debug);
- $paren_type[$paren] = 'V';
+ print "FUNC($1)\n" if ($dbg_values > 1);
+ $av_paren_type[$av_paren] = 'V';
} elsif ($cur =~ /^($Ident|$Constant)/o) {
- print "IDENT($1)\n" if ($debug);
+ print "IDENT($1)\n" if ($dbg_values > 1);
$type = 'V';
} elsif ($cur =~ /^($Assignment)/o) {
- print "ASSIGN($1)\n" if ($debug);
+ print "ASSIGN($1)\n" if ($dbg_values > 1);
$type = 'N';
} elsif ($cur =~ /^(;|{|}|\?|:|\[)/o) {
- print "END($1)\n" if ($debug);
+ print "END($1)\n" if ($dbg_values > 1);
$type = 'N';
} elsif ($cur =~ /^($Operators)/o) {
- print "OP($1)\n" if ($debug);
+ print "OP($1)\n" if ($dbg_values > 1);
if ($1 ne '++' && $1 ne '--') {
$type = 'N';
}
} elsif ($cur =~ /(^.)/o) {
- print "C($1)\n" if ($debug);
+ print "C($1)\n" if ($dbg_values > 1);
}
if (defined $1) {
$cur = substr($cur, length($1));
@@ -616,7 +666,7 @@ sub possible {
$possible ne 'struct' && $possible ne 'enum' &&
$possible ne 'case' && $possible ne 'else' &&
$possible ne 'typedef') {
- #print "POSSIBLE<$possible>\n";
+ warn "POSSIBLE: $possible\n" if ($dbg_possible);
push(@typeList, $possible);
build_types();
}
@@ -655,11 +705,12 @@ sub CHK {
sub process {
my $filename = shift;
- my @lines = @_;
my $linenr=0;
my $prevline="";
+ my $prevrawline="";
my $stashline="";
+ my $stashrawline="";
my $length;
my $indent;
@@ -681,14 +732,26 @@ sub process {
my $realcnt = 0;
my $here = '';
my $in_comment = 0;
+ my $comment_edge = 0;
my $first_line = 0;
my $prev_values = 'N';
+ # Pre-scan the patch sanitizing the lines.
# Pre-scan the patch looking for any __setup documentation.
+ #
my @setup_docs = ();
my $setup_docs = 0;
- foreach my $line (@lines) {
+ my $line;
+ foreach my $rawline (@rawlines) {
+ # Standardise the strings and chars within the input to
+ # simplify matching.
+ $line = sanitise_line($rawline);
+ push(@lines, $line);
+
+ ##print "==>$rawline\n";
+ ##print "-->$line\n";
+
if ($line=~/^\+\+\+\s+(\S+)/) {
$setup_docs = 0;
if ($1 =~ m@Documentation/kernel-parameters.txt$@) {
@@ -707,8 +770,7 @@ sub process {
foreach my $line (@lines) {
$linenr++;
- my $rawline = $line;
-
+ my $rawline = $rawlines[$linenr - 1];
#extract the filename as it passes
if ($line=~/^\+\+\+\s+(\S+)/) {
@@ -728,6 +790,7 @@ sub process {
} else {
$realcnt=1+1;
}
+ annotate_reset();
$prev_values = 'N';
next;
}
@@ -746,7 +809,7 @@ sub process {
if ($linenr == $first_line) {
my $edge;
for (my $ln = $first_line; $ln < ($linenr + $realcnt); $ln++) {
- ($edge) = ($lines[$ln - 1] =~ m@(/\*|\*/)@);
+ ($edge) = ($rawlines[$ln - 1] =~ m@(/\*|\*/)@);
last if (defined $edge);
}
if (defined $edge && $edge eq '*/') {
@@ -757,25 +820,30 @@ sub process {
# Guestimate if this is a continuing comment. If this
# is the start of a diff block and this line starts
# ' *' then it is very likely a comment.
- if ($linenr == $first_line and $line =~ m@^.\s*\*@) {
+ if ($linenr == $first_line and $rawline =~ m@^.\s* \*(?:\s|$)@) {
$in_comment = 1;
}
# Find the last comment edge on _this_ line.
- while (($line =~ m@(/\*|\*/)@g)) {
+ $comment_edge = 0;
+ while (($rawline =~ m@(/\*|\*/)@g)) {
if ($1 eq '/*') {
$in_comment = 1;
} else {
$in_comment = 0;
}
+ $comment_edge = 1;
}
# Measure the line length and indent.
- ($length, $indent) = line_stats($line);
+ ($length, $indent) = line_stats($rawline);
# Track the previous line.
($prevline, $stashline) = ($stashline, $line);
($previndent, $stashindent) = ($stashindent, $indent);
+ ($prevrawline, $stashrawline) = ($stashrawline, $rawline);
+
+ #warn "ic<$in_comment> ce<$comment_edge> line<$line>\n";
} elsif ($realcnt == 1) {
$realcnt--;
@@ -786,9 +854,9 @@ sub process {
$here = "#$realline: " if ($file);
$here .= "FILE: $realfile:$realline:" if ($realcnt != 0);
- my $hereline = "$here\n$line\n";
- my $herecurr = "$here\n$line\n";
- my $hereprev = "$here\n$prevline\n$line\n";
+ my $hereline = "$here\n$rawline\n";
+ my $herecurr = "$here\n$rawline\n";
+ my $hereprev = "$here\n$prevrawline\n$rawline\n";
$prefix = "$filename:$realline: " if ($emacs && $file);
$prefix = "$filename:$linenr: " if ($emacs && !$file);
@@ -816,7 +884,7 @@ sub process {
# UTF-8 regex found at http://www.w3.org/International/questions/qa-forms-utf-8.en.php
if (($realfile =~ /^$/ || $line =~ /^\+/) &&
- !($line =~ m/^(
+ !($rawline =~ m/^(
[\x09\x0A\x0D\x20-\x7E] # ASCII
| [\xC2-\xDF][\x80-\xBF] # non-overlong 2-byte
| \xE0[\xA0-\xBF][\x80-\xBF] # excluding overlongs
@@ -826,7 +894,7 @@ sub process {
| [\xF1-\xF3][\x80-\xBF]{3} # planes 4-15
| \xF4[\x80-\x8F][\x80-\xBF]{2} # plane 16
)*$/x )) {
- ERROR("Invalid UTF-8\n" . $herecurr);
+ ERROR("Invalid UTF-8, patch and commit message should be encoded in UTF-8\n" . $herecurr);
}
#ignore lines being removed
@@ -837,15 +905,15 @@ sub process {
#trailing whitespace
if ($line =~ /^\+.*\015/) {
- my $herevet = "$here\n" . cat_vet($line) . "\n";
+ my $herevet = "$here\n" . cat_vet($rawline) . "\n";
ERROR("DOS line endings\n" . $herevet);
- } elsif ($line =~ /^\+.*\S\s+$/ || $line =~ /^\+\s+$/) {
- my $herevet = "$here\n" . cat_vet($line) . "\n";
+ } elsif ($rawline =~ /^\+.*\S\s+$/ || $rawline =~ /^\+\s+$/) {
+ my $herevet = "$here\n" . cat_vet($rawline) . "\n";
ERROR("trailing whitespace\n" . $herevet);
}
#80 column limit
- if ($line =~ /^\+/ && !($prevline=~/\/\*\*/) && $length > 80) {
+ if ($line =~ /^\+/ && !($prevrawline=~/\/\*\*/) && $length > 80) {
WARN("line over 80 characters\n" . $herecurr);
}
@@ -859,46 +927,48 @@ sub process {
# at the beginning of a line any tabs must come first and anything
# more than 8 must use tabs.
- if ($line=~/^\+\s* \t\s*\S/ or $line=~/^\+\s* \s*/) {
- my $herevet = "$here\n" . cat_vet($line) . "\n";
+ if ($rawline =~ /^\+\s* \t\s*\S/ ||
+ $rawline =~ /^\+\s* \s*/) {
+ my $herevet = "$here\n" . cat_vet($rawline) . "\n";
ERROR("use tabs not spaces\n" . $herevet);
}
-# Remove comments from the line before processing.
- my $comment_edge = ($line =~ s@/\*.*\*/@@g) +
- ($line =~ s@/\*.*@@) +
- ($line =~ s@^(.).*\*/@$1@);
+# check for RCS/CVS revision markers
+ if ($rawline =~ /\$(Revision|Log|Id)(?:\$|)/) {
+ WARN("CVS style keyword markers, these will _not_ be updated\n". $herecurr);
+ }
# The rest of our checks refer specifically to C style
# only apply those _outside_ comments. Only skip
# lines in the middle of comments.
next if (!$comment_edge && $in_comment);
-# Standardise the strings and chars within the input to simplify matching.
- $line = sanitise_line($line);
-
# Check for potential 'bare' types
- if ($realcnt &&
- $line !~ /$Ident:\s*$/ &&
- ($line =~ /^.\s*$Ident\s*\(\*+\s*$Ident\)\s*\(/ ||
- $line !~ /^.\s*$Ident\s*\(/)) {
+ if ($realcnt) {
+ # Ignore goto labels.
+ if ($line =~ /$Ident:\*$/) {
+
+ # Ignore functions being called
+ } elsif ($line =~ /^.\s*$Ident\s*\(/) {
+
# definitions in global scope can only start with types
- if ($line =~ /^.(?:$Storage\s+)?(?:$Inline\s+)?(?:const\s+)?($Ident)\b/) {
+ } elsif ($line =~ /^.(?:$Storage\s+)?(?:$Inline\s+)?(?:const\s+)?($Ident)\b/) {
possible($1);
# declarations always start with types
- } elsif ($prev_values eq 'N' && $line =~ /^.\s*(?:$Storage\s+)?($Ident)\b\s*\**\s*$Ident\s*(?:;|=)/) {
+ } elsif ($prev_values eq 'N' && $line =~ /^.\s*(?:$Storage\s+)?(?:const\s+)?($Ident)\b(:?\s+$Sparse)?\s*\**\s*$Ident\s*(?:;|=)/) {
possible($1);
+ }
# any (foo ... *) is a pointer cast, and foo is a type
- } elsif ($line =~ /\(($Ident)(?:\s+$Sparse)*\s*\*+\s*\)/) {
+ while ($line =~ /\(($Ident)(?:\s+$Sparse)*\s*\*+\s*\)/g) {
possible($1);
}
# Check for any sort of function declaration.
# int foo(something bar, other baz);
# void (*store_gdt)(x86_descr_ptr *);
- if ($prev_values eq 'N' && $line =~ /^(.(?:(?:$Storage|$Inline)\s*)*\s*$Type\s*(?:\b$Ident|\(\*\s*$Ident\))\s*)\(/) {
+ if ($prev_values eq 'N' && $line =~ /^(.(?:typedef\s*)?(?:(?:$Storage|$Inline)\s*)*\s*$Type\s*(?:\b$Ident|\(\*\s*$Ident\))\s*)\(/) {
my ($name_len) = length($1);
my ($level, @ctx) = ctx_statement_level($linenr, $realcnt, $name_len);
my $ctx = join("\n", @ctx);
@@ -974,8 +1044,11 @@ sub process {
my $opline = $line; $opline =~ s/^./ /;
my $curr_values = annotate_values($opline . "\n", $prev_values);
$curr_values = $prev_values . $curr_values;
- #warn "--> $opline\n";
- #warn "--> $curr_values ($prev_values)\n";
+ if ($dbg_values) {
+ my $outline = $opline; $outline =~ s/\t/ /g;
+ warn "--> .$outline\n";
+ warn "--> $curr_values\n";
+ }
$prev_values = substr($curr_values, -1);
#ignore lines not being added
@@ -1004,9 +1077,6 @@ sub process {
ERROR("malformed #include filename\n" .
$herecurr);
}
- # Sanitise this special form of string.
- $path = 'X' x length($path);
- $line =~ s{\<.*\>}{<$path>};
}
# no C99 // comments
@@ -1074,7 +1144,7 @@ sub process {
# }
if ($line =~ /\bLINUX_VERSION_CODE\b/) {
- WARN("LINUX_VERSION_CODE should be avoided, code should be for the version to which it is merged" . $herecurr);
+ WARN("LINUX_VERSION_CODE should be avoided, code should be for the version to which it is merged\n" . $herecurr);
}
# printk should use KERN_* levels. Note that follow on printk's on the
@@ -1102,7 +1172,7 @@ sub process {
# function brace can't be on same line, except for #defines of do while,
# or if closed on same line
- if (($line=~/$Type\s*[A-Za-z\d_]+\(.*\).* {/) and
+ if (($line=~/$Type\s*[A-Za-z\d_]+\(.*\).*\s{/) and
!($line=~/\#define.*do\s{/) and !($line=~/}/)) {
ERROR("open brace '{' following function declarations go on the next line\n" . $herecurr);
}
@@ -1115,8 +1185,22 @@ sub process {
# check for spaces between functions and their parentheses.
while ($line =~ /($Ident)\s+\(/g) {
- if ($1 !~ /^(?:if|for|while|switch|return|volatile|__volatile__|__attribute__|format|__extension__|Copyright|case)$/ &&
- $line !~ /$Type\s+\(/ && $line !~ /^.\#\s*define\b/) {
+ my $name = $1;
+ my $ctx = substr($line, 0, $-[1]);
+
+ # Ignore those directives where spaces _are_ permitted.
+ if ($name =~ /^(?:if|for|while|switch|return|volatile|__volatile__|__attribute__|format|__extension__|Copyright|case)$/) {
+
+ # cpp #define statements have non-optional spaces, ie
+ # if there is a space between the name and the open
+ # parenthesis it is simply not a parameter group.
+ } elsif ($ctx =~ /^.\#\s*define\s*$/) {
+
+ # If this whole things ends with a type its most
+ # likely a typedef for a function.
+ } elsif ("$ctx$name" =~ /$Type$/) {
+
+ } else {
WARN("no space between function name and open parenthesis '('\n" . $herecurr);
}
}
@@ -1126,7 +1210,7 @@ sub process {
<<=|>>=|<=|>=|==|!=|
\+=|-=|\*=|\/=|%=|\^=|\|=|&=|
=>|->|<<|>>|<|>|=|!|~|
- &&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/
+ &&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|%
}x;
my @elements = split(/($ops|;)/, $opline);
my $off = 0;
@@ -1239,7 +1323,8 @@ sub process {
} elsif ($op eq '<<' or $op eq '>>' or
$op eq '&' or $op eq '^' or $op eq '|' or
$op eq '+' or $op eq '-' or
- $op eq '*' or $op eq '/')
+ $op eq '*' or $op eq '/' or
+ $op eq '%')
{
if ($ctx !~ /VxV|WxW|VxE|WxE|VxO/) {
ERROR("need consistent spacing around '$op' $at\n" .
@@ -1324,7 +1409,7 @@ sub process {
my ($s, $c) = ctx_statement_block($linenr, $realcnt, 0);
if ($c =~ /\bif\s*\(.*[^<>!=]=[^=].*/) {
- ERROR("do not use assignment in if condition ($c)\n" . $herecurr);
+ ERROR("do not use assignment in if condition\n" . $herecurr);
}
# Find out what is on the end of the line after the
@@ -1350,6 +1435,20 @@ sub process {
ERROR("else should follow close brace '}'\n" . $hereprev);
}
+ if ($prevline=~/}\s*$/ and $line=~/^.\s*while\s*/ and
+ $previndent == $indent) {
+ my ($s, $c) = ctx_statement_block($linenr, $realcnt, 0);
+
+ # Find out what is on the end of the line after the
+ # conditional.
+ substr($s, 0, length($c)) = '';
+ $s =~ s/\n.*//g;
+
+ if ($s =~ /^\s*;/) {
+ ERROR("while should follow close brace '}'\n" . $hereprev);
+ }
+ }
+
#studly caps, commented out until figure out how to distinguish between use of existing and adding new
# if (($line=~/[\w_][a-z\d]+[A-Z]/) and !($line=~/print/)) {
# print "No studly caps, use _\n";
@@ -1447,8 +1546,11 @@ sub process {
# Count the newlines, if there is only one
# then the block should not have {}'s.
my @lines = ($stmt =~ /\n/g);
+ my @statements = ($stmt =~ /;/g);
#print "lines<" . scalar(@lines) . ">\n";
+ #print "statements<" . scalar(@statements) . ">\n";
if ($lvl == 0 && scalar(@lines) == 0 &&
+ scalar(@statements) < 2 &&
$stmt !~ /{/ && $stmt !~ /\bif\b/ &&
$before !~ /}/ && $after !~ /{/) {
my $herectx = "$here\n" . join("\n", @control, @block[1 .. $#block]) . "\n";
@@ -1587,10 +1689,10 @@ sub process {
}
if ($clean == 1 && $quiet == 0) {
- print "Your patch has no obvious style problems and is ready for submission.\n"
+ print "$vname has no obvious style problems and is ready for submission.\n"
}
if ($clean == 0 && $quiet == 0) {
- print "Your patch has style problems, please review. If any of these errors\n";
+ print "$vname has style problems, please review. If any of these errors\n";
print "are false positives report them to the maintainer, see\n";
print "CHECKPATCH in MAINTAINERS.\n";
}