checkpatch: add a --strict check for utf-8 in commit logs

Some find using utf-8 in commit logs inappropriate. Some patch commit logs contain unintended utf-8 characters when doing things like copy/pasting compilation output. Look for the start of any commit log by skipping initial lines that look like email headers and "From: " lines. Stop looking for utf-8 at the first signature line. Signed-off-by: Joe Perches <joe@perches.com> Suggested-by: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Whitcroft <apw@shadowen.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

checkpatch: add a --strict check for utf-8 in commit logs
Some find using utf-8 in commit logs inappropriate. Some patch commit logs contain unintended utf-8 characters when doing things like copy/pasting compilation output. Look for the start of any commit log by skipping initial lines that look like email headers and "From: " lines. Stop looking for utf-8 at the first signature line. Signed-off-by: Joe Perches <joe@perches.com> Suggested-by: Andrew Morton <akpm@linux-foundation.org> Cc: Andy Whitcroft <apw@shadowen.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Joe Perches · Linus Torvalds
1 parent 67d0a07544
Showing 1 changed file with 26 additions and 4 deletions Side-by-side Diff
scripts/checkpatch.pl
@@ -240,9 +240,8 @@
 our $Type;
 our $Declare;
  
-our $UTF8	= qr {
-	[\x09\x0A\x0D\x20-\x7E]              # ASCII
-	| [\xC2-\xDF][\x80-\xBF]             # non-overlong 2-byte
+our $NON_ASCII_UTF8	= qr{
+	[\xC2-\xDF][\x80-\xBF]               # non-overlong 2-byte
 	|  \xE0[\xA0-\xBF][\x80-\xBF]        # excluding overlongs
 	| [\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}  # straight 3-byte
 	|  \xED[\x80-\x9F][\x80-\xBF]        # excluding surrogates
@@ -251,6 +250,11 @@
 	|  \xF4[\x80-\x8F][\x80-\xBF]{2}     # plane 16
 }x;
  
+our $UTF8	= qr{
+	[\x09\x0A\x0D\x20-\x7E]              # ASCII
+	| $NON_ASCII_UTF8
+}x;
+
 our $typeTypedefs = qr{(?x:
 	(?:__)?(?:u|s|be|le)(?:8|16|32|64)|
 	atomic_t
@@ -1330,6 +1334,9 @@
 	my $signoff = 0;
 	my $is_patch = 0;
  
+	my $in_header_lines = 1;
+	my $in_commit_log = 0;		#Scanning lines before patch
+
 	our @report = ();
 	our $cnt_lines = 0;
 	our $cnt_error = 0;
@@ -1497,7 +1504,6 @@
 		if ($line =~ /^diff --git.*?(\S+)$/) {
 			$realfile = $1;
 			$realfile =~ s@^([^/]*)/@@;
-
 		} elsif ($line =~ /^\+\+\+\s+(\S+)/) {
 			$realfile = $1;
 			$realfile =~ s@^([^/]*)/@@;
@@ -1536,6 +1542,7 @@
 # Check the patch for a signoff:
 		if ($line =~ /^\s*signed-off-by:/i) {
 			$signoff++;
+			$in_commit_log = 0;
 		}
  
 # Check signature styles
@@ -1611,6 +1618,21 @@
  
 			CHK("INVALID_UTF8",
 			    "Invalid UTF-8, patch and commit message should be encoded in UTF-8\n" . $hereptr);
+		}
+
+# Check if it's the start of a commit log
+# (not a header line and we haven't seen the patch filename)
+		if ($in_header_lines && $realfile =~ /^$/ &&
+		    $rawline !~ /^(commit\b|from\b|\w+:).+$/i) {
+			$in_header_lines = 0;
+			$in_commit_log = 1;
+		}
+
+# Still not yet in a patch, check for any UTF-8
+		if ($in_commit_log && $realfile =~ /^$/ &&
+		    $rawline =~ /$NON_ASCII_UTF8/) {
+			CHK("UTF8_BEFORE_PATCH",
+			    "8-bit UTF-8 used in possible commit log\n" . $herecurr);
 		}
  
 # ignore non-hunk lines and lines being removed
...	...	@@ -240,9 +240,8 @@
240	240	our $Type;
241	241	our $Declare;
242	242
243		-our $UTF8 = qr {
244		- [\x09\x0A\x0D\x20-\x7E] # ASCII
245		- \| [\xC2-\xDF][\x80-\xBF] # non-overlong 2-byte
	243	+our $NON_ASCII_UTF8 = qr{
	244	+ [\xC2-\xDF][\x80-\xBF] # non-overlong 2-byte
246	245	\| \xE0[\xA0-\xBF][\x80-\xBF] # excluding overlongs
247	246	\| [\xE1-\xEC\xEE\xEF][\x80-\xBF]{2} # straight 3-byte
248	247	\| \xED[\x80-\x9F][\x80-\xBF] # excluding surrogates
...	...	@@ -251,6 +250,11 @@
251	250	\| \xF4[\x80-\x8F][\x80-\xBF]{2} # plane 16
252	251	}x;
253	252
	253	+our $UTF8 = qr{
	254	+ [\x09\x0A\x0D\x20-\x7E] # ASCII
	255	+ \| $NON_ASCII_UTF8
	256	+}x;
	257	+
254	258	our $typeTypedefs = qr{(?x:
255	259	(?:__)?(?:u\|s\|be\|le)(?:8\|16\|32\|64)\|
256	260	atomic_t
...	...	@@ -1330,6 +1334,9 @@
1330	1334	my $signoff = 0;
1331	1335	my $is_patch = 0;
1332	1336
	1337	+ my $in_header_lines = 1;
	1338	+ my $in_commit_log = 0; #Scanning lines before patch
	1339	+
1333	1340	our @report = ();
1334	1341	our $cnt_lines = 0;
1335	1342	our $cnt_error = 0;
...	...	@@ -1497,7 +1504,6 @@
1497	1504	if ($line =~ /^diff --git.*?(\S+)$/) {
1498	1505	$realfile = $1;
1499	1506	$realfile =~ s@^([^/]*)/@@;
1500		-
1501	1507	} elsif ($line =~ /^\+\+\+\s+(\S+)/) {
1502	1508	$realfile = $1;
1503	1509	$realfile =~ s@^([^/]*)/@@;
...	...	@@ -1536,6 +1542,7 @@
1536	1542	# Check the patch for a signoff:
1537	1543	if ($line =~ /^\s*signed-off-by:/i) {
1538	1544	$signoff++;
	1545	+ $in_commit_log = 0;
1539	1546	}
1540	1547
1541	1548	# Check signature styles
...	...	@@ -1611,6 +1618,21 @@
1611	1618
1612	1619	CHK("INVALID_UTF8",
1613	1620	"Invalid UTF-8, patch and commit message should be encoded in UTF-8\n" . $hereptr);
	1621	+ }
	1622	+
	1623	+# Check if it's the start of a commit log
	1624	+# (not a header line and we haven't seen the patch filename)
	1625	+ if ($in_header_lines && $realfile =~ /^$/ &&
	1626	+ $rawline !~ /^(commit\b\|from\b\|\w+:).+$/i) {
	1627	+ $in_header_lines = 0;
	1628	+ $in_commit_log = 1;
	1629	+ }
	1630	+
	1631	+# Still not yet in a patch, check for any UTF-8
	1632	+ if ($in_commit_log && $realfile =~ /^$/ &&
	1633	+ $rawline =~ /$NON_ASCII_UTF8/) {
	1634	+ CHK("UTF8_BEFORE_PATCH",
	1635	+ "8-bit UTF-8 used in possible commit log\n" . $herecurr);
1614	1636	}
1615	1637
1616	1638	# ignore non-hunk lines and lines being removed