gfx/thebes/test/process-textruns.pl

   1 #!/usr/bin/perl -w
   2
   3 # ***** BEGIN LICENSE BLOCK *****
   4 # Version: MPL 1.1/GPL 2.0/LGPL 2.1
   5 #
   6 # The contents of this file are subject to the Mozilla Public License Version
   7 # 1.1 (the "License"); you may not use this file except in compliance with
   8 # the License. You may obtain a copy of the License at
   9 # http://www.mozilla.org/MPL/
  10 #
  11 # Software distributed under the License is distributed on an "AS IS" basis,
  12 # WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
  13 # for the specific language governing rights and limitations under the
  14 # License.
  15 #
  16 # The Original Code is Mozilla Corportation code.
  17 #
  18 # The Initial Developer of the Original Code is Mozilla Corporation.
  19 # Portions created by the Initial Developer are Copyright (C) 2007
  20 # the Initial Developer. All Rights Reserved.
  21 #
  22 # Contributor(s):
  23 #   Robert O'Callahan <robert@ocallahan.org>
  24 #
  25 # Alternatively, the contents of this file may be used under the terms of
  26 # either the GNU General Public License Version 2 or later (the "GPL"), or
  27 # the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
  28 # in which case the provisions of the GPL or the LGPL are applicable instead
  29 # of those above. If you wish to allow use of your version of this file only
  30 # under the terms of either the GPL or the LGPL, and not to allow others to
  31 # use your version of this file under the terms of the MPL, indicate your
  32 # decision by deleting the provisions above and replace them with the notice
  33 # and other provisions required by the GPL or the LGPL. If you do not delete
  34 # the provisions above, a recipient may use your version of this file under
  35 # the terms of any one of the MPL, the GPL or the LGPL.
  36 #
  37 # ***** END LICENSE BLOCK *****
  38
  39 # This script is a bunch of utilities for computing statistics about the textruns
  40 # created during a Gecko run.
  41 #
  42 # Usage:
  43 # 1) Uncomment #define DUMP_TEXT_RUNS in gfxAtsuiFonts.cpp
  44 # 2) Build
  45 # 3) Run over some test set, redirecting stdout to a file
  46 # 4) Pipe that file through this script
  47
  48 # --exclude-spaces-only: ignore all textruns that consistent of zero or more
  49 # spaces
  50 my $exclude_spaces = grep(/^--exclude-spaces-only$/, @ARGV);
  51
  52 # --dump-runs: process textruns into a format that can be used by
  53 # gfxTextRunPerfTest, print that on standard output, and do nothing else
  54 my $dump_runs = grep(/^--dump-runs$/, @ARGV);
  55
  56 # --obfuscate: ROTL13 the textrun text
  57 my $obfuscate = grep(/^--obfuscate$/, @ARGV);
  58
  59 my @textruns = ();
  60
  61 while (<STDIN>) {
  62   if (/^0x(\w+)\((.*)\) TEXTRUN "(.*)" ENDTEXTRUN$/) {
  63     my %tr = ( fontgroup => $1,
  64                families => $2,
  65                text => $3 );
  66     push(@textruns, \%tr);
  67   } elsif (/^0x(\w+)\((.*)\) TEXTRUN "(.*)$/) {
  68     my %tr = ( fontgroup => $1,
  69                families => $2 );
  70     my $text = $3."\n";
  71     while (<STDIN>) {
  72       if (/^(.*)" ENDTEXTRUN$/) {
  73         $text .= $1;
  74         last;
  75       }
  76       $text .= $_;
  77     }
  78     $tr{text} = $text;
  79     push(@textruns, \%tr);
  80   }
  81 }
  82
  83 my %quote = ( "\\" => 1, "\"" => 1 );
  84
  85 sub quote_str {
  86     my ($text) = @_;
  87     my @chars = split(//, $text);
  88     my @strs = ();
  89     foreach my $c (@chars) {
  90       if (ord($c) >= 0x80) {
  91         $c = "\\x".sprintf("%x",ord($c)).'""';
  92       } elsif ($quote{$c}) {
  93         $c = "\\$c";
  94       } elsif ($c eq "\n") {
  95         $c = " ";
  96       }
  97       push(@strs, $c);
  98     }
  99     return '"'.join("", @strs).'"';
 100 }
 101
 102 if ($dump_runs) {
 103   foreach my $tr (@textruns) {
 104     print "{ ", &quote_str($tr->{families}), ",\n";
 105     my $text = $tr->{text};
 106     if ($obfuscate) {
 107         $text =~ tr/a-mA-Mn-zN-Z/n-zN-Za-mA-M/;
 108     }
 109     print "  ", &quote_str($text), " },\n";
 110   }
 111   exit(0);
 112 }
 113
 114 my %trs_by_text = ();
 115 my %trs_by_text_and_fontgroup = ();
 116 my %trs_by_trimmed_text_and_fontgroup = ();
 117 my @tr_lengths = ();
 118
 119 $trs_by_text{" "} = [];
 120 $trs_by_text{""} = [];
 121
 122 sub trim {
 123   my ($s) = @_;
 124   $s =~ s/^ *//g;
 125   $s =~ s/ *$//g;
 126   return $s;
 127 }
 128
 129 my $total_textruns = 0;
 130
 131 foreach my $tr (@textruns) {
 132   if ($exclude_spaces && $tr->{text} =~ /^ *$/) {
 133     next;
 134   }
 135   ++$total_textruns;
 136   push(@{$trs_by_text{$tr->{text}}}, $tr);
 137   push(@{$trs_by_text_and_fontgroup{$tr->{fontgroup}.$tr->{text}}}, $tr);
 138   push(@{$trs_by_trimmed_text_and_fontgroup{$tr->{fontgroup}.&trim($tr->{text})}}, $tr);
 139   if (1 < scalar(@{$trs_by_trimmed_text_and_fontgroup{$tr->{fontgroup}.&trim($tr->{text})}})) {
 140     $tr_lengths[length($tr->{text})]++;
 141   }
 142 }
 143
 144 print "Number of textruns:\t$total_textruns\n";
 145 print "Number of textruns which are one space:\t", scalar(@{$trs_by_text{" "}}), "\n";
 146 print "Number of textruns which are empty:\t", scalar(@{$trs_by_text{""}}), "\n";
 147
 148 my $count = 0;
 149 foreach my $k (keys(%trs_by_text)) {
 150   if ($k =~ /^ *$/) {
 151     $count += @{$trs_by_text{$k}};
 152   }
 153 }
 154 print "Number of textruns which are zero or more spaces:\t$count\n";
 155
 156 print "Number of unique textruns by text and fontgroup:\t", scalar(keys(%trs_by_text_and_fontgroup)), "\n";
 157 print "Number of unique textruns by trimmed text and fontgroup:\t", scalar(keys(%trs_by_trimmed_text_and_fontgroup)), "\n";
 158
 159 my $sum = 0;
 160 my $weighted_sum = 0;
 161 if (1) {
 162   print "Textrun length distribution:\n";
 163   for my $i (0..(scalar(@tr_lengths)-1)) {
 164     my $amount = defined($tr_lengths[$i])?$tr_lengths[$i]:0;
 165     $sum += $amount;
 166     $weighted_sum += $i*$amount;
 167     print "$i\t$sum\t$weighted_sum\n";
 168   }
 169 }