document count option (and imply unique)

[git-grep-footer.git] / git-grep-footer

diff --git a/git-grep-footer b/git-grep-footer
index fa417c114bb41bf8ad9604a7665eb56ae2e0235d..b676ab8aef983e241ffb59921849d94075dd889d 100755 (executable)
--- a/git-grep-footer
+++ b/git-grep-footer
@@ -23,7 +23,7 @@ GetOptions(\my %opt,
 local $| = 1;
 local $/ = "\0";
 
 local $| = 1;
 local $/ = "\0";
 

-my $HEADERMATCH = qr/ [a-z]+ (?: (?:-\w+)+ | \ by ) /ix;
+my $HEADERMATCH = qr/ [a-z]+ (?: (?:-\w+)+ | \ by ) | cc | reference /ix;

 
 my (%headercount, @headercache);
 
 
 my (%headercount, @headercache);
 


@@ -40,12 +40,12 @@ while (readline) {
        # if invalid, assume it's latin1
               $_ = decode(cp1252 => $_) if $@;
 
        # if invalid, assume it's latin1
               $_ = decode(cp1252 => $_) if $@;
 

-       my $prefix = 0;

        my %attr;
 
        BLOCK:
        for (reverse split /\n\n/) {
                my @headers;
        my %attr;
 
        BLOCK:
        for (reverse split /\n\n/) {
                my @headers;

+               my $prefix = 0;

 
                LINE:
                for (split /\n/) {
 
                LINE:
                for (split /\n/) {


@@ -85,7 +85,7 @@ while (readline) {
                                        }{<...>}imsx;
                                }
                                when (['var', 'vars', '']) {
                                        }{<...>}imsx;
                                }
                                when (['var', 'vars', '']) {

-                                       when ($header[0] =~ /[ _-] (?: by | to ) $/imsx) {
+                                       when ($header[0] =~ /[ _-] (?: by | to ) $ | ^cc$/imsx) {

                                                $header[1] = undef;
                                        }
                                        for ($header[1]) {
                                                $header[1] = undef;
                                        }
                                        for ($header[1]) {


@@ -125,7 +125,7 @@ while (readline) {
                        my $line = $_->[2] // join(': ', @$_);
                        $line =~ s/^/$hash / if defined $hash;
 
                        my $line = $_->[2] // join(': ', @$_);
                        $line =~ s/^/$hash / if defined $hash;
 

-                       if (defined $opt{min} or $opt{max}) {
+                       if (defined $opt{min} or $opt{max} or $opt{count}) {

                                my $counter = \$headercount{ $_->[0] }->{ $_->[1] // '' };
                                my $excess = $$counter++ - ($opt{min} // 0);
                                next if $excess >= ($opt{max} || 1);
                                my $counter = \$headercount{ $_->[0] }->{ $_->[1] // '' };
                                my $excess = $$counter++ - ($opt{min} // 0);
                                next if $excess >= ($opt{max} || 1);


@@ -163,6 +163,7 @@ a common convention to list custom metadata such as
 C<Signed-off-by> and C<Acked-by>.
 
 Sections are identified by at least one leading keyword containing a dash
 C<Signed-off-by> and C<Acked-by>.
 
 Sections are identified by at least one leading keyword containing a dash

+(or exceptionally recognised)

 followed by a colon.
 
 =head1 OPTIONS
 followed by a colon.
 
 =head1 OPTIONS


@@ -210,6 +211,11 @@ The original line is given for each match,
 but simplifications still apply for duplicate determination.
 Additional samples are optionally given upto the given maximum.
 
 but simplifications still apply for duplicate determination.
 Additional samples are optionally given upto the given maximum.
 

+=item -c, --count
+
+Prefixes (unique) lines by the number of occurrences.
+Causes output to be buffered until all input has been read (obviously).
+

 =back
 
 =head1 AUTHOR
 =back
 
 =head1 AUTHOR