From: Frank Lichtenheld <frank@lichtenheld.de>
Date: Sun, 17 Jun 2007 17:13:13 +0000 (+0200)
Subject: Remove obsolete descriptions.txt previously used for fulltext search
X-Git-Url: https://git.deb.at/w?a=commitdiff_plain;h=cdae3720732a12007e363f01c24de74c2620d969;p=deb%2Fpackages.git

Remove obsolete descriptions.txt previously used for fulltext search

I will definetly stick with xapian for now.
---

diff --git a/BACKEND b/BACKEND
index 445a45d..11f1409 100644
--- a/BACKEND
+++ b/BACKEND
@@ -41,11 +41,6 @@ Generated by means of Packages.gz files:
 |  key: "packagename version arch"
 |  value: a unique description id, did
  
-| descriptions.txt:
-|  on each line:
-|    description with strange characters mangled for proper substring
-|    searching, linenumber being the did
-
 | descriptions.db:
 |  key: did
 |  value: description, first line being short, the rest being long [no
diff --git a/TODO b/TODO
index d705607..0c45640 100644
--- a/TODO
+++ b/TODO
@@ -19,10 +19,6 @@ search_packages:
 	matches" and abce: 90 matches"
 
 - fulltext search:
-  - Max 100 results
-  - Better exact=1 performance by indexing per word?
-  - drop case-sensitive from options, descriptions.txt all lowercase and without
-	punctuation, such that instead of =~ //, indexof can be used
   - in results, show full descriptions, so one sees what's being matched?
 
 - backend:
diff --git a/bin/parse-packages b/bin/parse-packages
index 79fb6d9..1536b81 100755
--- a/bin/parse-packages
+++ b/bin/parse-packages
@@ -246,7 +246,6 @@ my %descriptions_db;
 tie %descriptions_db, "DB_File", "$DBDIR/descriptions.db.new",
 	O_RDWR|O_CREAT, 0666, $DB_BTREE
 	or die "Error creating DB: $!";
-open DESCR, ">", "$DBDIR/descriptions.txt" or die "Error creating descriptions textfile";
 print "Index $#descriptions descriptions\n";
 for (my $i=1; $i<= $#descriptions; $i++) {
 	my $plain_description = $descriptions[$i];
@@ -259,7 +258,6 @@ for (my $i=1; $i<= $#descriptions; $i++) {
 	$plain_description = " $plain_description ";
 	$plain_description =~ s/[(),.-]+//og;
 	$plain_description =~ s#[^a-z0-9_/+]+# #og;
-	print DESCR "$plain_description\n";
 
 	#XAPIAN
 	eval {
@@ -283,7 +281,6 @@ for (my $i=1; $i<= $#descriptions; $i++) {
 
 	$descriptions_db{$i} = $only_desc;
 }
-close DESCR;
 untie %descriptions_db;
 $xapian_db->flush;
 undef $xapian_db;
@@ -332,6 +329,5 @@ rename("$DBDIR/packages_descriptions.db.new",
        "$DBDIR/packages_descriptions.db");
 rename("$DBDIR/descriptions_packages.db.new",
        "$DBDIR/descriptions_packages.db");
-rename("$DBDIR/descriptions.txt.new", "$DBDIR/descriptions.txt");
 rename("$DBDIR/descriptions.db.new", "$DBDIR/descriptions.db");
 rename("$DBDIR/package_postfixes.db.new", "$DBDIR/package_postfixes.db");
diff --git a/lib/Packages/DoSearch.pm b/lib/Packages/DoSearch.pm
index a94dd7b..5b0b4bb 100644
--- a/lib/Packages/DoSearch.pm
+++ b/lib/Packages/DoSearch.pm
@@ -53,21 +53,14 @@ sub do_search {
 	    do_names_search( [ @keywords ], \%packages, $p_obj,
 			     \&read_entry_all, $opts,
 			     \@results, \@non_results );
-#	    my $fts0 = new Benchmark;
-#	    do_fulltext_search( [ @keywords ], "$DBDIR/descriptions.txt",
-#				\%did2pkg, \%packages,
-#				\&read_entry_all, $opts,
-#				\@results, \@non_results );
 	    my $fts1 = new Benchmark;
 	    do_xapian_search( [ @keywords ], "$DBDIR/xapian/",
 				\%did2pkg, \%packages,
 				\&read_entry_all, $opts,
 				\@results, \@non_results );
 	    my $fts2 = new Benchmark;
-#	    my $fts_grep = timediff($fts1,$fts0);
 	    my $fts_xapian = timediff($fts2,$fts1);
-#	    debug( "Fulltext search took ".timestr($fts_grep)." (grep)" ) if DEBUG;
-	    debug( "Fulltext search took ".timestr($fts_xapian)." (Xapian)" )
+	    debug( "Fulltext search took ".timestr($fts_xapian) )
 		if DEBUG;
 	}
     }
diff --git a/lib/Packages/Search.pm b/lib/Packages/Search.pm
index fdd3a3a..671e340 100644
--- a/lib/Packages/Search.pm
+++ b/lib/Packages/Search.pm
@@ -196,55 +196,6 @@ sub do_names_search {
 	&$read_entry( $packages, $pkg, $results, $non_results, $opts );
     }
 }
-sub do_fulltext_search {
-    my ($keywords, $file, $did2pkg, $packages, $read_entry, $opts,
-	$results, $non_results) = @_;
-
-# NOTE: this needs to correspond with parse-packages!
-    my @tmp;
-    foreach my $keyword (@$keywords) {
-	$keyword =~ tr [A-Z] [a-z];
-	if ($opts->{exact}) {
-	    $keyword = " $keyword ";
-	}
-	$keyword =~ s/[(),.-]+//og;
-	$keyword =~ s;[^a-z0-9_/+]+; ;og;
-	push @tmp, $keyword;
-    }
-    my $first_keyword = shift @tmp;
-    @$keywords = @tmp;
-
-    my $numres = 0;
-    my %tmp_results;
-    # fgrep is seriously faster than using perl
-    open DESC, '-|', 'fgrep', '-n', '--', $first_keyword, $file
-	or die "couldn't open $file: $!";
-  LINE:
-    while (<DESC>) {
-	foreach my $k (@$keywords) {
-	    next LINE unless /\Q$k\E/;
-	}
-	/^(\d+)/;
-	my $nr = $1;
-	debug( "Matched line $_", 2) if DEBUG;
-	my $result = $did2pkg->{$nr};
-	foreach (split /\000/o, $result) {
-	    my @data = split /\s/, $_, 3;
-#	    debug ("Considering $data[0], arch = $data[2]", 3) if DEBUG;
-#	    next unless $data[2] eq 'all' || $opts->{h_archs}{$data[2]};
-#	    debug ("Ok", 3) if DEBUG;
-	    $numres++ unless $tmp_results{$data[0]}++;
-	}
-	last if $numres > 100;
-    }
-    close DESC;
-    $too_many_hits++ if $numres > 100;
-
-    my @results;
-    foreach my $pkg (keys %tmp_results) {
-	&$read_entry( $packages, $pkg, $results, $non_results, $opts );
-    }
- }
 
 sub do_xapian_search {
     my ($keywords, $db, $did2pkg, $packages, $read_entry, $opts,