2 # Convert Contents.gz files into Sleepycat db files for efficient usage of
5 # Copyright (C) 2006 Jeroen van Wolffelaar <jeroen@wolffelaar.nl>
6 # This program is free software; you can redistribute it and/or modify
7 # it under the terms of the GNU General Public License as published by
8 # the Free Software Foundation; either version 2 of the License, or
9 # (at your option) any later version.
11 # This program is distributed in the hope that it will be useful,
12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 # GNU General Public License for more details.
16 # You should have received a copy of the GNU General Public License
17 # along with this program; if not, write to the Free Software
18 # Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
26 # Important, we want sorting and such to happen like in the C locale: binary,
27 # without any fancy collation. FIXME: is this actually adequate?
30 my $what = $ARGV[0] ? "head -10000|" : "";
32 # More RAM vs more disk I/O tradeoff parameters, does not change
33 # functionality. True will always use more RAM at the benefit of less
34 # temporary files, and is adviced when possible
35 my $SORT_REVERSE_CONCURRENTLY = 1;
42 use Packages::CommonCode qw(:all);
43 use Packages::Config qw( $TOPDIR $DBDIR @ARCHIVES @SUITES @ARCHITECTURES );
44 &Packages::Config::init( './' );
46 my @archives = @ARCHIVES;
48 my @archs = @ARCHITECTURES;
50 $DBDIR .= "/contents";
53 for my $suite (@suites) {
54 for my $arch (@archs) {
56 my $filelist_db = "$DBDIR/filelists_${suite}_${arch}.db";
57 my $dbtime = (stat $filelist_db)[9];
58 my %packages_contents = ();
59 my %packages_contents_nr = ();
60 my %packages_contents_lastword = ();
63 $extra = "|sort" if $SORT_REVERSE_CONCURRENTLY;
65 open REVERSED, "$extra>$DBDIR/reverse.tmp"
66 or die "Failed to open output reverse file: $!";
69 for my $archive (@archives) {
71 my $filename = "$TOPDIR/archive/$archive/$suite/Contents-$arch.gz";
72 next unless -f $filename;
73 # Note: ctime, because mtime is set back via rsync
74 my $ftime = (stat $filename)[10];
75 next if defined $dbtime and $dbtime > $ftime;
76 print "$archive/$suite/$arch needs update\n";
80 for my $archive (@archives) {
82 my $filename = "$TOPDIR/archive/$archive/$suite/Contents-$arch.gz";
83 next unless -f $filename;
84 print "Reading $archive/$suite/$arch...\n";
86 open CONT, "zcat $filename|$what"
88 while (<CONT>) { last if /^FILE/mo; }
89 if (eof(CONT)) { # no header found
90 close CONT; # explicit close to reset $.
91 open CONT, "zcat $filename|$what";
97 print "Doing line ".($NR/1000)."k (out of approx 2.0M)\n"
100 my ($file, $value) = ($1, $2);
101 $value =~ s#[^,/]+/##og;
102 my @packages = split m/,/, $value;
104 $packages_contents_nr{$_}++;
105 my $lw = $packages_contents_lastword{$_} || "\0";
107 while (substr($file,$i,1) eq substr($lw,$i++,1)) {}
109 $i = 255 if $i > 255;
110 $packages_contents{$_} .= pack "CC/a*", ($i, substr($file, $i));
111 $packages_contents_lastword{$_} = "$file\0";
113 # Searches are case-insensitive
114 (my $nocase = $file) =~ tr [A-Z] [a-z];
115 my $case = ($nocase eq $file) ? '-' : $file;
117 print REVERSED (reverse $nocase)."\0".$case."\0".
118 (join ":$arch\0", @packages).":$arch\n";
125 print "Sorting reverse list if needed\n";
126 system("cd $DBDIR && sort reverse.tmp > reverse.sorted &&".
127 " mv reverse.{sorted,tmp}") == 0
128 or die "Failed to sort reverse"
129 unless $SORT_REVERSE_CONCURRENTLY;
131 print "Writing filelist db\n";
132 tie my %packages_contents_db, "DB_File", "$filelist_db.new",
133 O_RDWR|O_CREAT, 0666, $DB_BTREE
134 or die "Error creating DB: $!";
135 while (my ($k, $v) = each(%packages_contents)) {
136 $packages_contents_db{$k} = (pack "L", $packages_contents_nr{$k})
139 untie %packages_contents_db;
141 rename("$DBDIR/reverse.tmp", "$DBDIR/reverse_${suite}_${arch}.txt");
143 activate($filelist_db);
144 #FIXME: hardcoded archs. (debports has no contrib/non-free)
145 if ($arch ne 'avr32' and $arch ne 'm68k') {
146 system("ln", "-sf", basename($filelist_db),
147 "$DBDIR/filelists_${suite}_all.db") == 0
154 my $suite_mtime = (stat "$DBDIR/reverse_$suite.db")[9];
155 for my $file (glob "$DBDIR/reverse_${suite}_*.txt") {
156 $go = 1 if not defined $suite_mtime
157 or $suite_mtime < (stat $file)[9];
161 print "Merging reverse path lists for ${suite}...\n";
163 open MERGED, "-|", "sort -m $DBDIR/reverse_${suite}_*.txt"
164 or die "Failed to open merged list";
165 open FILENAMES, ">", "$DBDIR/filenames_$suite.txt.new"
166 or die "Failed to open filenames list";
167 tie my %reverse_path_db, "DB_File", "$DBDIR/reverse_${suite}.db.new",
168 O_RDWR|O_CREAT, 0666, $DB_BTREE
169 or die "Error creating DB: $!";
171 my $lastpath = my $lastcasepath = my $lastfile = "";
174 print "Doing line ".($NR/1000000)."M (out of approx. 20M)\n"
175 if $NR % 1000000 == 0;
177 my @line = split m/\0/o, $_;
178 my $revpath = shift @line;
179 my $casepath = shift @line;
180 if ($revpath ne $lastpath) {
181 # Wrap: Do useful stuff with this ($lastpath, @matches)
182 if ($lastpath ne "") {
184 while (my ($k, $v) = each %matches) {
185 push @matches, join("\0", $k, @$v);
187 $reverse_path_db{$lastpath} = join "\1", @matches;
190 $lastpath =~ s,/.*,,o;
191 if ($lastfile ne $lastpath) {
192 $lastfile = $lastpath;
193 print FILENAMES (reverse $lastfile)."\n";
196 $lastpath = $revpath;
197 $lastcasepath = $casepath;
198 $matches{$casepath} = \@line;
200 # } elsif ($lastcasepath ne "" and $casepath ne $lastcasepath) {
201 # warn reverse($revpath)." has more than one casepath: $casepath $lastcasepath\n";
203 push @{$matches{$casepath}}, @line;
205 # Note: do useful stuff here too, for out last entry. Maybe prevent this by
206 # adding a fake ultimate entry?
209 while (my ($k, $v) = each %matches) {
210 push @matches, join("\0", $k, @$v);
212 $reverse_path_db{$lastpath} = join "\1", @matches;
215 untie %reverse_path_db;
219 activate("$DBDIR/filenames_$suite.txt");
220 activate("$DBDIR/reverse_$suite.db");