Re: [PATCH] scripts: warn about invalid MAINTAINERS patterns

[Joe Perches <joe@perches.com>]

On Tue, 2017-10-31 at 16:37 -0500, Tom Saeger wrote:
> Add "--pattern-checks" option to get_maintainer.pl to warn about invalid
> "F" and "X" patterns found in MAINTAINERS file(s).

Hey Tom.

I've come around to this addition, but I think a few
changes are useful.

o Change --pattern-checks to --self-test so future checks
  can be added (valid email address, .mailmap uses, existence
  of git trees, etc...)
o Do not require an unnecessary argument with --self-test
o Validate --self-test if it is the only command line argument
o Use emacs filename:line: style output for easier linking
o --self-test emits to STDOUT not STDERR

---

 scripts/get_maintainer.pl | 59 +++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 57 insertions(+), 2 deletions(-)

diff --git a/scripts/get_maintainer.pl b/scripts/get_maintainer.pl
index bc443201d3ef..128bc90f7034 100755
--- a/scripts/get_maintainer.pl
+++ b/scripts/get_maintainer.pl
@@ -57,6 +57,7 @@ my $sections = 0;
 my $file_emails = 0;
 my $from_filename = 0;
 my $pattern_depth = 0;
+my $self_test = 0;
 my $version = 0;
 my $help = 0;
 my $find_maintainer_files = 0;
@@ -138,6 +139,7 @@ my %VCS_cmds_git = (
     "subject_pattern" => "^GitSubject: (.*)",
     "stat_pattern" => "^(\\d+)\\t(\\d+)\\t\$file\$",
     "file_exists_cmd" => "git ls-files \$file",
+    "list_files_cmd" => "git ls-files \$file",
 );
 
 my %VCS_cmds_hg = (
@@ -167,6 +169,7 @@ my %VCS_cmds_hg = (
     "subject_pattern" => "^HgSubject: (.*)",
     "stat_pattern" => "^(\\d+)\t(\\d+)\t\$file\$",
     "file_exists_cmd" => "hg files \$file",
+    "list_files_cmd" => "hg files \$file",
 );
 
 my $conf = which_conf(".get_maintainer.conf");
@@ -216,6 +219,14 @@ if (-f $ignore_file) {
     close($ignore);
 }
 
+if ($#ARGV > 0) {
+    foreach (@ARGV) {
+	if ($_ eq "-self-test" || $_ eq "--self-test") {
+	    die "$P: using --self-test does not allow any other option or argument\n";
+	}
+    }
+}
+
 if (!GetOptions(
 		'email!' => \$email,
 		'git!' => \$email_git,
@@ -252,6 +263,7 @@ if (!GetOptions(
 		'fe|file-emails!' => \$file_emails,
 		'f|file' => \$from_filename,
 		'find-maintainer-files' => \$find_maintainer_files,
+		'self-test' => \$self_test,
 		'v|version' => \$version,
 		'h|help|usage' => \$help,
 		)) {
@@ -268,7 +280,7 @@ if ($version != 0) {
     exit 0;
 }
 
-if (-t STDIN && !@ARGV) {
+if (!$self_test && -t STDIN && !@ARGV) {
     # We're talking to a terminal, but have no command line arguments.
     die "$P: missing patchfile or -f file - use --help if necessary\n";
 }
@@ -311,12 +323,14 @@ if (!top_of_kernel_tree($lk_path)) {
 my @typevalue = ();
 my %keyword_hash;
 my @mfiles = ();
+my @check_pattern_info = ();
 
 sub read_maintainer_file {
     my ($file) = @_;
 
     open (my $maint, '<', "$file")
 	or die "$P: Can't open MAINTAINERS file '$file': $!\n";
+    my $i = 1;
     while (<$maint>) {
 	my $line = $_;
 
@@ -333,6 +347,9 @@ sub read_maintainer_file {
 		if ((-d $value)) {
 		    $value =~ s@([^/])$@$1/@;
 		}
+		if ($self_test) {
+			push(@check_pattern_info, {file=>$file, linenr=>$i, line=>$line, pat=>$value});
+		}
 	    } elsif ($type eq "K") {
 		$keyword_hash{@typevalue} = $value;
 	    }
@@ -341,6 +358,7 @@ sub read_maintainer_file {
 	    $line =~ s/\n$//g;
 	    push(@typevalue, $line);
 	}
+	$i++;
     }
     close($maint);
 }
@@ -464,7 +482,7 @@ my @range = ();
 my @keyword_tvi = ();
 my @file_emails = ();
 
-if (!@ARGV) {
+if (!$self_test && !@ARGV) {
     push(@ARGV, "&STDIN");
 }
 
@@ -543,6 +561,11 @@ foreach my $file (@ARGV) {
     }
 }
 
+if ($self_test) {
+    check_maintainers_patterns();
+    exit 0;
+}
+
 @file_emails = uniq(@file_emails);
 
 my %email_hash_name;
@@ -586,6 +609,20 @@ if ($web) {
 
 exit($exit);
 
+sub check_maintainers_patterns {
+    my @lsfiles = ();
+
+    @lsfiles = vcs_list_files($lk_path);
+
+    for my $x (@check_pattern_info) {
+        if (!grep(m@^$x->{pat}@, @lsfiles)) {
+            my $line = $x->{line};
+            chomp($line);
+            print("$x->{file}:$x->{linenr}:\twarning: no matches\t$line\n");
+        }
+    }
+}
+
 sub ignore_email_address {
     my ($address) = @_;
 
@@ -863,6 +900,7 @@ Other options:
   --sections => print all of the subsystem sections with pattern matches
   --letters => print all matching 'letter' types from all matching sections
   --mailmap => use .mailmap file (default: $email_use_mailmap)
+  --self-test => show possible invalid MAINTAINERS file content
   --version => show version
   --help => show this help information
 
@@ -2192,6 +2230,23 @@ sub vcs_file_exists {
     return $exists;
 }
 
+sub vcs_list_files {
+    my ($file) = @_;
+
+    my @lsfiles = ();
+
+    my $vcs_used = vcs_exists();
+    return 0 if (!$vcs_used);
+
+    my $cmd = $VCS_cmds{"list_files_cmd"};
+    $cmd =~ s/(\$\w+)/$1/eeg;   # interpolate $cmd
+    @lsfiles = &{$VCS_cmds{"execute_cmd"}}($cmd);
+
+    return () if ($? != 0);
+
+    return @lsfiles;
+}
+
 sub uniq {
     my (@parms) = @_;