Added a filetype of "--text", which matches any text file. This

is off by default, just as --binary is. Added a filetype of "--ignored", which matches any file that ack ignores, like core dumps and backup files. It will NOT include files that are in ignored directories like blib/ and .svn/
beyondgrep · Jun 13, 2007 · 4484844 · 4484844
1 parent 1ebff69
commit 4484844
Show file tree

Hide file tree

Showing 9 changed files with 145 additions and 61 deletions.
diff --git a/Ack.pm b/Ack.pm
@@ -38,6 +38,7 @@ BEGIN {
         elisp       => [qw( el )],
         haskell     => [qw( hs lhs )],
         html        => [qw( htm html shtml )],
+        ignored     => q{Files, but not directories, normally ignored by ack (default: off)},
         lisp        => [qw( lisp )],
         java        => [qw( java properties )],
         js          => [qw( js )],
@@ -55,6 +56,7 @@ BEGIN {
         sql         => [qw( sql ctl )],
         tcl         => [qw( tcl )],
         tex         => [qw( tex cls sty )],
+        text        => q{Text files, as defined by Perl's -T op (default: off)},
         tt          => [qw( tt tt2 ttml )],
         vb          => [qw( bas cls frm ctl vb resx )],
         vim         => [qw( vim )],
@@ -105,22 +107,24 @@ F<foo.pod> could be "perl" or "parrot".
 The filetype will be C<undef> if we can't determine it.  This could
 be if the file doesn't exist, or it can't be read.
 
-It will be '-ignore' if it's something that ack should always ignore,
+It will be 'ignored' if it's something that ack should always ignore,
 even under -a.
 
 =cut
 
+use constant TEXT => 'text';
+
 sub filetypes {
     my $filename = shift;
 
-    return '-ignore' unless is_searchable( $filename );
+    return 'ignored' unless is_searchable( $filename );
 
-    return 'make' if $filename =~ m{$path_sep?Makefile$}io;
+    return ('make',TEXT) if $filename =~ m{$path_sep?Makefile$}io;
 
     # If there's an extension, look it up
     if ( $filename =~ m{\.([^\.$path_sep]+)$}o ) {
         my $ref = $types{lc $1};
-        return @{$ref} if $ref;
+        return (@{$ref},TEXT) if $ref;
     }
 
     # At this point, we can't tell from just the name.  Now we have to
@@ -155,12 +159,12 @@ sub filetypes {
     }
 
     if ( $header =~ /^#!/ ) {
-        return $1       if $header =~ /\b(ruby|p(erl|hp|ython))\b/;
-        return 'shell'  if $header =~ /\b(?:ba|c|k|z)?sh\b/;
+        return ($1,TEXT)       if $header =~ /\b(ruby|p(erl|hp|ython))\b/;
+        return ('shell','text')  if $header =~ /\b(?:ba|c|k|z)?sh\b/;
     }
-    return 'xml' if $header =~ /<\?xml /;
+    return ('xml',TEXT) if $header =~ /<\?xml /;
 
-    return;
+    return (TEXT);
 }
 
 =head2 is_searchable( $filename )

diff --git a/Changes b/Changes
@@ -7,6 +7,15 @@ NEXT
     set this useful environment variable, your tests would fail.
     Thanks, Michael Hendricks.
 
+    [ENHANCEMENTS]
+    Added a filetype of "--text", which matches any text file.  This
+    is off by default, just as --binary is.
+
+    Added a filetype of "--ignored", which matches any file that
+    ack ignores, like core dumps and backup files.  It will NOT
+    include files that are in ignored directories like blib/ and
+    .svn/
+
 1.63_01
     [THINGS THAT MAY BREAK FOR YOU]
     The changes I made to detect if ack is outputting to the screen

diff --git a/MANIFEST b/MANIFEST
@@ -13,6 +13,7 @@ t/00-load.t
 t/ack-a.t
 t/ack-binary.t
 t/ack-c.t
+t/ack-text.t
 t/ack-type.t
 t/ack-v.t
 t/filetypes.t

diff --git a/ack b/ack
@@ -136,7 +136,7 @@ MAIN: {
     # If anyone says --no-whatever, we assume all other types must be on.
     if ( !$filetypes_supported_set ) {
         for my $i ( keys %type_wanted ) {
-            $type_wanted{$i} = 1 unless ( defined( $type_wanted{$i} ) || $i eq 'binary' || $i eq 'text' );
+            $type_wanted{$i} = 1 unless ( defined( $type_wanted{$i} ) || $i eq 'binary' || $i eq 'text' || $i eq 'ignored' );
         }
     }
 

diff --git a/ack-help-types.txt b/ack-help-types.txt
@@ -16,6 +16,7 @@ Note that some extensions may appear in multiple types.  For example,
     --[no]elisp     .el
     --[no]haskell   .hs .lhs
     --[no]html      .htm .html .shtml
+    --[no]ignored   Files, but not directories, normally ignored by ack (default: off)
     --[no]java      .java .properties
     --[no]js        .js
     --[no]jsp       .jsp .jspx .jhtm .jhtml
@@ -33,6 +34,7 @@ Note that some extensions may appear in multiple types.  For example,
     --[no]sql       .sql .ctl
     --[no]tcl       .tcl
     --[no]tex       .tex .cls .sty
+    --[no]text      Text files, as defined by Perl's -T op (default: off)
     --[no]tt        .tt .tt2 .ttml
     --[no]vb        .bas .cls .frm .ctl .vb .resx
     --[no]vim       .vim

diff --git a/ack-standalone b/ack-standalone
@@ -64,6 +64,7 @@ MAIN: {
         'o|output:s'            => \$opt{o},
         'Q|literal'             => \$opt{Q},
         'sort-files'            => \$opt{sort_files},
+        'text'                  => \$opt{text},
         'v|invert-match'        => \$opt{v},
         'w|word-regexp'         => \$opt{w},
 
@@ -133,7 +134,7 @@ MAIN: {
     # If anyone says --no-whatever, we assume all other types must be on.
     if ( !$filetypes_supported_set ) {
         for my $i ( keys %type_wanted ) {
-            $type_wanted{$i} = 1 unless ( defined( $type_wanted{$i} ) || $i eq 'binary' );
+            $type_wanted{$i} = 1 unless ( defined( $type_wanted{$i} ) || $i eq 'binary' || $i eq 'text' || $i eq 'ignored' );
         }
     }
 
@@ -905,6 +906,7 @@ BEGIN {
         elisp       => [qw( el )],
         haskell     => [qw( hs lhs )],
         html        => [qw( htm html shtml )],
+        ignored     => q{Files, but not directories, normally ignored by ack (default: off)},
         lisp        => [qw( lisp )],
         java        => [qw( java properties )],
         js          => [qw( js )],
@@ -922,6 +924,7 @@ BEGIN {
         sql         => [qw( sql ctl )],
         tcl         => [qw( tcl )],
         tex         => [qw( tex cls sty )],
+        text        => q{Text files, as defined by Perl's -T op (default: off)},
         tt          => [qw( tt tt2 ttml )],
         vb          => [qw( bas cls frm ctl vb resx )],
         vim         => [qw( vim )],
@@ -950,17 +953,19 @@ sub skipdir_filter {
 }
 
 
+use constant TEXT => 'text';
+
 sub filetypes {
     my $filename = shift;
 
-    return '-ignore' unless is_searchable( $filename );
+    return 'ignored' unless is_searchable( $filename );
 
-    return 'make' if $filename =~ m{$path_sep?Makefile$}io;
+    return ('make',TEXT) if $filename =~ m{$path_sep?Makefile$}io;
 
     # If there's an extension, look it up
     if ( $filename =~ m{\.([^\.$path_sep]+)$}o ) {
         my $ref = $types{lc $1};
-        return @{$ref} if $ref;
+        return (@{$ref},TEXT) if $ref;
     }
 
     # At this point, we can't tell from just the name.  Now we have to
@@ -995,12 +1000,12 @@ sub filetypes {
     }
 
     if ( $header =~ /^#!/ ) {
-        return $1       if $header =~ /\b(ruby|p(erl|hp|ython))\b/;
-        return 'shell'  if $header =~ /\b(?:ba|c|k|z)?sh\b/;
+        return ($1,TEXT)       if $header =~ /\b(ruby|p(erl|hp|ython))\b/;
+        return ('shell','text')  if $header =~ /\b(?:ba|c|k|z)?sh\b/;
     }
-    return 'xml' if $header =~ /<\?xml /;
+    return ('xml',TEXT) if $header =~ /<\?xml /;
 
-    return;
+    return (TEXT);
 }
 
 

diff --git a/t/ack-text.t b/t/ack-text.t
@@ -0,0 +1,72 @@
+#!perl
+
+use warnings;
+use strict;
+
+use Test::More tests => 1;
+use App::Ack ();
+use File::Next ();
+delete $ENV{ACK_OPTIONS};
+
+use lib 't';
+use Util;
+
+
+ACK_F_TEXT: {
+    my @expected = qw(
+        t/00-load.t
+        t/ack-a.t
+        t/ack-binary.t
+        t/ack-c.t
+        t/ack-text.t
+        t/ack-type.t
+        t/ack-v.t
+        t/etc/buttonhook.html.xxx
+        t/etc/buttonhook.noxml.xxx
+        t/etc/buttonhook.rfc.xxx
+        t/etc/buttonhook.rss.xxx
+        t/etc/buttonhook.xml.xxx
+        t/etc/shebang.foobar.xxx
+        t/etc/shebang.php.xxx
+        t/etc/shebang.pl.xxx
+        t/etc/shebang.py.xxx
+        t/etc/shebang.rb.xxx
+        t/etc/shebang.sh.xxx
+        t/filetypes.t
+        t/interesting.t
+        t/longopts.t
+        t/pod-coverage.t
+        t/pod.t
+        t/standalone.t
+        t/swamp/0
+        t/swamp/c-header.h
+        t/swamp/c-source.c
+        t/swamp/html.htm
+        t/swamp/html.html
+        t/swamp/javascript.js
+        t/swamp/Makefile
+        t/swamp/Makefile.PL
+        t/swamp/options.pl
+        t/swamp/parrot.pir
+        t/swamp/perl-test.t
+        t/swamp/perl-without-extension
+        t/swamp/perl.cgi
+        t/swamp/perl.pl
+        t/swamp/perl.pm
+        t/swamp/perl.pod
+        t/text/boy-named-sue.txt
+        t/text/freedom-of-choice.txt
+        t/text/science-of-myth.txt
+        t/text/shut-up-be-happy.txt
+        t/Util.pm
+        t/zero.t
+    );
+
+    my @files = qw( t );
+    my @args = qw( -f --text );
+    my $cmd = "$^X ./ack-standalone @args @files";
+    my @results = `$cmd`;
+    chomp @results;
+
+    sets_match( \@results, \@expected, 'Looking for text files' );
+}
diff --git a/t/ack-type.t b/t/ack-type.t
@@ -21,6 +21,7 @@ my $perl = [qw(
     t/ack-a.t
     t/ack-binary.t
     t/ack-c.t
+    t/ack-text.t
     t/ack-type.t
     t/ack-v.t
     t/etc/shebang.pl.xxx

diff --git a/t/filetypes.t b/t/filetypes.t
@@ -3,8 +3,7 @@
 use warnings;
 use strict;
 
-use Test::More tests => 28;
-use Data::Dumper;
+use Test::More tests => 29;
 delete $ENV{ACK_OPTIONS};
 
 use lib 't';
@@ -14,64 +13,55 @@ BEGIN {
     use_ok( 'App::Ack' );
 }
 
-my @foo_pod_types = App::Ack::filetypes( 'foo.pod' ); # 5.6.1 doesn't like to sort(filetypes())
-sets_match( [@foo_pod_types], [qw( parrot perl )], 'foo.pod can be multiple things' );
-sets_match( [App::Ack::filetypes( 'Bongo.pm' )], [qw( perl )], 'Bongo.pm' );
-sets_match( [App::Ack::filetypes( 'Makefile.PL' )], [qw( perl )], 'Makefile.PL' );
+sets_match( [App::Ack::filetypes( 'foo.pod' )], [qw( parrot perl text )], 'foo.pod can be multiple things' );
+sets_match( [App::Ack::filetypes( 'Bongo.pm' )], [qw( perl text )], 'Bongo.pm' );
+sets_match( [App::Ack::filetypes( 'Makefile.PL' )], [qw( perl text )], 'Makefile.PL' );
 sets_match( [App::Ack::filetypes( 'Unknown.wango' )], [], 'Unknown' );
 
 ok(  is_filetype( 'foo.pod', 'perl' ), 'foo.pod can be perl' );
 ok(  is_filetype( 'foo.pod', 'parrot' ), 'foo.pod can be parrot' );
+ok(  is_filetype( 'foo.pod', 'text' ), 'foo.pod can be parrot' );
 ok( !is_filetype( 'foo.pod', 'ruby' ), 'foo.pod cannot be ruby' );
 ok(  is_filetype( 'foo.handler.pod', 'perl' ), 'foo.handler.pod can be perl' );
 ok(  is_filetype( '/tmp/wango/foo.pod', 'perl' ), '/tmp/wango/foo.pod can be perl' );
 ok(  is_filetype( '/tmp/wango/foo.handler.pod', 'perl' ), '/tmp/wango/foo.handler.pod can be perl' );
 ok(  is_filetype( '/tmp/blongo/makefile', 'make' ), '/tmp/blongo/makefile is a makefile' );
 ok(  is_filetype( 'Makefile', 'make' ), 'Makefile is a makefile' );
 
-is(App::Ack::filetypes('foo.pod~'), '-ignore',
+is(App::Ack::filetypes('foo.pod~'), 'ignored',
     'correctly ignore backup file');
 
-is(App::Ack::filetypes('#some.pod#'), '-ignore',
+is(App::Ack::filetypes('#some.pod#'), 'ignored',
     'correctly ignore files starting and ending with hash mark');
 
-is(App::Ack::filetypes('core.987654321'), '-ignore',
+is(App::Ack::filetypes('core.987654321'), 'ignored',
     'correctly ignore files named core.NNNN');
 
-is(App::Ack::filetypes('t/etc/shebang.pl.xxx'), 'perl',
-    'file identified as Perl from shebang line');
-
-is(App::Ack::filetypes('t/etc/shebang.php.xxx'), 'php',
-    'file identified as PHP from shebang line');
-
-is(App::Ack::filetypes('t/etc/shebang.py.xxx'), 'python',
-    'file identified as Python from shebang line');
-
-is(App::Ack::filetypes('t/etc/shebang.rb.xxx'), 'ruby',
-    'file identified as Ruby from shebang line');
-
-is(App::Ack::filetypes('t/etc/shebang.sh.xxx'), 'shell',
-    'file identified as shell from shebang line');
-
-ok(! defined App::Ack::filetypes('t/etc/shebang.foobar.xxx'),
-    'file could not be identified from shebang line');
-
-is(App::Ack::filetypes('t/etc/shebang.empty.xxx'), 'binary', 
-    'empty file returns "binary"');
-
-## Tests documenting current behavior in 1.50
-is(App::Ack::filetypes('t/etc/buttonhook.xml.xxx'), 'xml',
-    'file identified as xml from <?xml line');
-
-ok(! defined App::Ack::filetypes('t/etc/buttonhook.noxml.xxx'),
-    'no <?xml> found, so no filetype');
-
-
-is(App::Ack::filetypes('t/etc/buttonhook.xml.xxx'),'xml',
-   'filetype by <?xml>');
+MATCH_VIA_CONTENT: {
+    my %lookups = (
+        't/swamp/Makefile'          => 'make',
+        't/swamp/Makefile.PL'       => 'perl',
+        't/swamp/buttonhook.xml'    => 'xml',
+        't/etc/shebang.php.xxx'     => 'php',
+        't/etc/shebang.pl.xxx'      => 'perl',
+        't/etc/shebang.py.xxx'      => 'python',
+        't/etc/shebang.rb.xxx'      => 'ruby',
+        't/etc/shebang.sh.xxx'      => 'shell',
+        't/etc/buttonhook.xml.xxx'  => 'xml',
+    );
+    for my $filename ( sort keys %lookups ) {
+        my $type = $lookups{$filename};
+        sets_match( [App::Ack::filetypes( $filename )], [ $type, 'text' ], "Checking $filename" );
+    }
+
+    is(App::Ack::filetypes('t/etc/shebang.empty.xxx'), 'binary',
+        'empty file returns "binary"');
+}
 
-sets_match([App::Ack::filetypes('t/swamp/buttonhook.xml')], ['xml'],
-    'file identified as xml ');
+FAIL_MATCHING_VIA_CONTENT: {
+    is( App::Ack::filetypes('t/etc/shebang.foobar.xxx'), 'text',
+        'file could not be identified from shebang line');
 
-ok(! defined App::Ack::filetypes('t/etc/x.html.xxx'),
-   '<!DOCTYPE not yet supported so no filetype');
+    is( App::Ack::filetypes('t/etc/buttonhook.noxml.xxx'), 'text',
+        'no <?xml> found, so no filetype');
+}