in scripts/checkpatch.pl [1287:1634]
sub process {
my $filename = shift;
my $linenr=0;
my $prevline="";
my $prevrawline="";
my $stashline="";
my $stashrawline="";
my $length;
my $indent;
my $previndent=0;
my $stashindent=0;
our $clean = 1;
my $signoff = 0;
my $is_patch = 0;
my $in_header_lines = $file ? 0 : 1;
my $in_commit_log = 0; #Scanning lines before patch
my $reported_maintainer_file = 0;
my $non_utf8_charset = 0;
our @report = ();
our $cnt_lines = 0;
our $cnt_error = 0;
our $cnt_warn = 0;
our $cnt_chk = 0;
# Trace the real file/line as we go.
my $realfile = '';
my $realline = 0;
my $realcnt = 0;
my $here = '';
my $in_comment = 0;
my $comment_edge = 0;
my $first_line = 0;
my $p1_prefix = '';
my $prev_values = 'E';
# suppression flags
my %suppress_ifbraces;
my %suppress_whiletrailers;
my %suppress_export;
my $acpi_testexpected;
my $acpi_nontestexpected;
# Pre-scan the patch sanitizing the lines.
sanitise_line_reset();
my $line;
foreach my $rawline (@rawlines) {
$linenr++;
$line = $rawline;
if ($rawline=~/^\@\@ -\d+(?:,\d+)? \+(\d+)(,(\d+))? \@\@/) {
$realline=$1-1;
if (defined $2) {
$realcnt=$3+1;
} else {
$realcnt=1+1;
}
$in_comment = 0;
# Guestimate if this is a continuing comment. Run
# the context looking for a comment "edge". If this
# edge is a close comment then we must be in a comment
# at context start.
my $edge;
my $cnt = $realcnt;
for (my $ln = $linenr + 1; $cnt > 0; $ln++) {
next if (defined $rawlines[$ln - 1] &&
$rawlines[$ln - 1] =~ /^-/);
$cnt--;
#print "RAW<$rawlines[$ln - 1]>\n";
last if (!defined $rawlines[$ln - 1]);
if ($rawlines[$ln - 1] =~ m@(/\*|\*/)@ &&
$rawlines[$ln - 1] !~ m@"[^"]*(?:/\*|\*/)[^"]*"@) {
($edge) = $1;
last;
}
}
if (defined $edge && $edge eq '*/') {
$in_comment = 1;
}
# 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 (!defined $edge &&
$rawlines[$linenr] =~ m@^.\s*(?:\*\*+| \*)(?:\s|$)@)
{
$in_comment = 1;
}
##print "COMMENT:$in_comment edge<$edge> $rawline\n";
sanitise_line_reset($in_comment);
} elsif ($realcnt && $rawline =~ /^(?:\+| |$)/) {
# Standardise the strings and chars within the input to
# simplify matching -- only bother with positive lines.
$line = sanitise_line($rawline);
}
push(@lines, $line);
if ($realcnt > 1) {
$realcnt-- if ($line =~ /^(?:\+| |$)/);
} else {
$realcnt = 0;
}
#print "==>$rawline\n";
#print "-->$line\n";
}
$prefix = '';
$realcnt = 0;
$linenr = 0;
foreach my $line (@lines) {
$linenr++;
my $rawline = $rawlines[$linenr - 1];
#extract the line range in the file after the patch is applied
if ($line=~/^\@\@ -\d+(?:,\d+)? \+(\d+)(,(\d+))? \@\@/) {
$is_patch = 1;
$first_line = $linenr + 1;
$realline=$1-1;
if (defined $2) {
$realcnt=$3+1;
} else {
$realcnt=1+1;
}
annotate_reset();
$prev_values = 'E';
%suppress_ifbraces = ();
%suppress_whiletrailers = ();
%suppress_export = ();
next;
# track the line number as we move through the hunk, note that
# new versions of GNU diff omit the leading space on completely
# blank context lines so we need to count that too.
} elsif ($line =~ /^( |\+|$)/) {
$realline++;
$realcnt-- if ($realcnt != 0);
# Measure the line length and indent.
($length, $indent) = line_stats($rawline);
# Track the previous line.
($prevline, $stashline) = ($stashline, $line);
($previndent, $stashindent) = ($stashindent, $indent);
($prevrawline, $stashrawline) = ($stashrawline, $rawline);
#warn "line<$line>\n";
} elsif ($realcnt == 1) {
$realcnt--;
}
my $hunk_line = ($realcnt != 0);
#make up the handle for any error we report on this line
$prefix = "$filename:$realline: " if ($emacs && $file);
$prefix = "$filename:$linenr: " if ($emacs && !$file);
$here = "#$linenr: " if (!$file);
$here = "#$realline: " if ($file);
# extract the filename as it passes
if ($line =~ /^diff --git.*?(\S+)$/) {
$realfile = $1;
$realfile =~ s@^([^/]*)/@@ if (!$file);
checkfilename($realfile, \$acpi_testexpected, \$acpi_nontestexpected);
} elsif ($line =~ /^\+\+\+\s+(\S+)/) {
$realfile = $1;
$realfile =~ s@^([^/]*)/@@ if (!$file);
checkfilename($realfile, \$acpi_testexpected, \$acpi_nontestexpected);
$p1_prefix = $1;
if (!$file && $tree && $p1_prefix ne '' &&
-e "$root/$p1_prefix") {
WARN("patch prefix '$p1_prefix' exists, appears to be a -p0 patch\n");
}
next;
}
$here .= "FILE: $realfile:$realline:" if ($realcnt != 0);
my $hereline = "$here\n$rawline\n";
my $herecurr = "$here\n$rawline\n";
my $hereprev = "$here\n$prevrawline\n$rawline\n";
$cnt_lines++ if ($realcnt != 0);
# Check for incorrect file permissions
if ($line =~ /^new (file )?mode.*[7531]\d{0,2}$/) {
my $permhere = $here . "FILE: $realfile\n";
if ($realfile =~ /(\bMakefile(?:\.objs)?|\.c|\.cc|\.cpp|\.h|\.mak|\.[sS])$/) {
ERROR("do not set execute permissions for source files\n" . $permhere);
}
}
# Only allow Python 3 interpreter
if ($realline == 1 &&
$line =~ /^\+#!\ *\/usr\/bin\/(?:env )?python$/) {
ERROR("please use python3 interpreter\n" . $herecurr);
}
# Accept git diff extended headers as valid patches
if ($line =~ /^(?:rename|copy) (?:from|to) [\w\/\.\-]+\s*$/) {
$is_patch = 1;
}
if ($line =~ /^(Author|From): .* via .*<qemu-devel\@nongnu.org>/) {
ERROR("Author email address is mangled by the mailing list\n" . $herecurr);
}
#check the patch for a signoff:
if ($line =~ /^\s*signed-off-by:/i) {
# This is a signoff, if ugly, so do not double report.
$signoff++;
$in_commit_log = 0;
if (!($line =~ /^\s*Signed-off-by:/)) {
ERROR("The correct form is \"Signed-off-by\"\n" .
$herecurr);
}
if ($line =~ /^\s*signed-off-by:\S/i) {
ERROR("space required after Signed-off-by:\n" .
$herecurr);
}
}
# Check if MAINTAINERS is being updated. If so, there's probably no need to
# emit the "does MAINTAINERS need updating?" message on file add/move/delete
if ($line =~ /^\s*MAINTAINERS\s*\|/) {
$reported_maintainer_file = 1;
}
# Check for added, moved or deleted files
if (!$reported_maintainer_file && !$in_commit_log &&
($line =~ /^(?:new|deleted) file mode\s*\d+\s*$/ ||
$line =~ /^rename (?:from|to) [\w\/\.\-]+\s*$/ ||
($line =~ /\{\s*([\w\/\.\-]*)\s*\=\>\s*([\w\/\.\-]*)\s*\}/ &&
(defined($1) || defined($2)))) &&
!(($realfile ne '') &&
defined($acpi_testexpected) &&
($realfile eq $acpi_testexpected))) {
$reported_maintainer_file = 1;
WARN("added, moved or deleted file(s), does MAINTAINERS need updating?\n" . $herecurr);
}
# Check for wrappage within a valid hunk of the file
if ($realcnt != 0 && $line !~ m{^(?:\+|-| |\\ No newline|$)}) {
ERROR("patch seems to be corrupt (line wrapped?)\n" .
$herecurr) if (!$emitted_corrupt++);
}
# UTF-8 regex found at http://www.w3.org/International/questions/qa-forms-utf-8.en.php
if (($realfile =~ /^$/ || $line =~ /^\+/) &&
$rawline !~ m/^$UTF8*$/) {
my ($utf8_prefix) = ($rawline =~ /^($UTF8*)/);
my $blank = copy_spacing($rawline);
my $ptr = substr($blank, 0, length($utf8_prefix)) . "^";
my $hereptr = "$hereline$ptr\n";
ERROR("Invalid UTF-8, patch and commit message should be encoded in UTF-8\n" . $hereptr);
}
if ($rawline =~ m/$UTF8_MOJIBAKE/) {
ERROR("Doubly-encoded UTF-8\n" . $herecurr);
}
# Check if it's the start of a commit log
# (not a header line and we haven't seen the patch filename)
if ($in_header_lines && $realfile =~ /^$/ &&
!($rawline =~ /^\s+\S/ ||
$rawline =~ /^(commit\b|from\b|[\w-]+:).*$/i)) {
$in_header_lines = 0;
$in_commit_log = 1;
}
# Check if there is UTF-8 in a commit log when a mail header has explicitly
# declined it, i.e defined some charset where it is missing.
if ($in_header_lines &&
$rawline =~ /^Content-Type:.+charset="(.+)".*$/ &&
$1 !~ /utf-8/i) {
$non_utf8_charset = 1;
}
if ($in_commit_log && $non_utf8_charset && $realfile =~ /^$/ &&
$rawline =~ /$NON_ASCII_UTF8/) {
WARN("8-bit UTF-8 used in possible commit log\n" . $herecurr);
}
# ignore non-hunk lines and lines being removed
next if (!$hunk_line || $line =~ /^-/);
# ignore files that are being periodically imported from Linux
next if ($realfile =~ /^(linux-headers|include\/standard-headers)\//);
#trailing whitespace
if ($line =~ /^\+.*\015/) {
my $herevet = "$here\n" . cat_vet($rawline) . "\n";
ERROR("DOS line endings\n" . $herevet);
} elsif ($realfile =~ /^docs\/.+\.txt/ ||
$realfile =~ /^docs\/.+\.md/) {
if ($rawline =~ /^\+\s+$/ && $rawline !~ /^\+ {4}$/) {
# TODO: properly check we're in a code block
# (surrounding text is 4-column aligned)
my $herevet = "$here\n" . cat_vet($rawline) . "\n";
ERROR("code blocks in documentation should have " .
"empty lines with exactly 4 columns of " .
"whitespace\n" . $herevet);
}
} elsif ($rawline =~ /^\+.*\S\s+$/ || $rawline =~ /^\+\s+$/) {
my $herevet = "$here\n" . cat_vet($rawline) . "\n";
ERROR("trailing whitespace\n" . $herevet);
$rpt_cleaners = 1;
}
# checks for trace-events files
if ($realfile =~ /trace-events$/ && $line =~ /^\+/) {
if ($rawline =~ /%[-+ 0]*#/) {
ERROR("Don't use '#' flag of printf format ('%#') in " .
"trace-events, use '0x' prefix instead\n" . $herecurr);
} else {
my $hex =
qr/%[-+ *.0-9]*([hljztL]|ll|hh)?(x|X|"\s*PRI[xX][^"]*"?)/;
# don't consider groups splitted by [.:/ ], like 2A.20:12ab
my $tmpline = $rawline;
$tmpline =~ s/($hex[.:\/ ])+$hex//g;
if ($tmpline =~ /(?<!0x)$hex/) {
ERROR("Hex numbers must be prefixed with '0x'\n" .
$herecurr);
}
}
}