util/lint/checkpatch.pl: Update the check of repeated words

Update to v5.18-2 version.

Signed-off-by: Elyes Haouas <ehaouas@noos.fr>
Change-Id: I7f5e597bb76e1b9feeb2d6ea290626f45e9fe6c0
Reviewed-on: https://review.coreboot.org/c/coreboot/+/63577
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Martin L Roth <martinroth@google.com>

1 files changed, 62 insertions, 0 deletions

diff --git a/util/lint/checkpatch.pl b/util/lint/checkpatch.pl
index 6defb99ba55f..d75d6559cec9 100755
--- a/util/lint/checkpatch.pl
+++ b/util/lint/checkpatch.pl
@@ -708,6 +708,8 @@ our @mode_permission_funcs = (
 	["__ATTR", 2],
 );
 
+my $word_pattern = '\b[A-Z]?[a-z]{2,}\b';
+
 #Create a search pattern for all these functions to speed up a loop below
 our $mode_perms_search = "";
 foreach my $entry (@mode_permission_funcs) {
@@ -943,6 +945,13 @@ our $declaration_macros = qr{(?x:
 	(?:SKCIPHER_REQUEST|SHASH_DESC|AHASH_REQUEST)_ON_STACK\s*\(
 )};
 
+our %allow_repeated_words = (
+	add => '',
+	added => '',
+	bad => '',
+	be => '',
+);
+
 sub deparenthesize {
 	my ($string) = @_;
 	return "" if (!defined($string));
@@ -3000,6 +3009,59 @@ sub process {
 				}
 			}
 		}
+# check for repeated words separated by a single space
+# avoid false positive from list command eg, '-rw-r--r-- 1 root root'
+		if (($rawline =~ /^\+/ || $in_commit_log) &&
+		    $rawline !~ /[bcCdDlMnpPs\?-][rwxsStT-]{9}/) {
+			pos($rawline) = 1 if (!$in_commit_log);
+			while ($rawline =~ /\b($word_pattern) (?=($word_pattern))/g) {
+
+				my $first = $1;
+				my $second = $2;
+				my $start_pos = $-[1];
+				my $end_pos = $+[2];
+				if ($first =~ /(?:struct|union|enum)/) {
+					pos($rawline) += length($first) + length($second) + 1;
+					next;
+				}
+
+				next if (lc($first) ne lc($second));
+				next if ($first eq 'long');
+
+				# check for character before and after the word matches
+				my $start_char = '';
+				my $end_char = '';
+				$start_char = substr($rawline, $start_pos - 1, 1) if ($start_pos > ($in_commit_log ? 0 : 1));
+				$end_char = substr($rawline, $end_pos, 1) if ($end_pos < length($rawline));
+
+				next if ($start_char =~ /^\S$/);
+				next if (index(" \t.,;?!", $end_char) == -1);
+
+				# avoid repeating hex occurrences like 'ff ff fe 09 ...'
+				if ($first =~ /\b[0-9a-f]{2,}\b/i) {
+					next if (!exists($allow_repeated_words{lc($first)}));
+				}
+
+				if (WARN("REPEATED_WORD",
+					 "Possible repeated word: '$first'\n" . $herecurr) &&
+				    $fix) {
+					$fixed[$fixlinenr] =~ s/\b$first $second\b/$first/;
+				}
+			}
+
+			# if it's a repeated word on consecutive lines in a comment block
+			if ($prevline =~ /$;+\s*$/ &&
+			    $prevrawline =~ /($word_pattern)\s*$/) {
+				my $last_word = $1;
+				if ($rawline =~ /^\+\s*\*\s*$last_word /) {
+					if (WARN("REPEATED_WORD",
+						 "Possible repeated word: '$last_word'\n" . $hereprev) &&
+					    $fix) {
+						$fixed[$fixlinenr] =~ s/(\+\s*\*\s*)$last_word /$1/;
+					}
+				}
+			}
+		}
 
 # ignore non-hunk lines and lines being removed
 		next if (!$hunk_line || $line =~ /^-/);