android_webview/tools/find_copyrights.pl

   1 #!/usr/bin/perl -w
   2 # Copyright 2013 The Chromium Authors. All rights reserved.
   3 # Use of this source code is governed by a BSD-style license that can be
   4 # found in the LICENSE file.
   5
   6 # Use: echo filename1.cc ... | find_copyrights.pl
   7 #  or: find_copyrights.pl list_file
   8 #  or: find_files.pl ... | find_copyrights.pl
   9
  10 use strict;
  11 use warnings;
  12 use File::Basename;
  13
  14 sub check_is_generated_file($);
  15 sub start_copyright_parsing();
  16
  17 my $progname = basename($0);
  18
  19 my $generated_file_scan_boundary = 25;
  20 while (<>) {
  21     chomp;
  22     my $file = $_;
  23     my $file_header = '';
  24     my %copyrights;
  25     open (F, "<$file") or die "$progname: Unable to access $file\n";
  26     my $parse_copyright = start_copyright_parsing();
  27     while (<F>) {
  28         $file_header .= $_ unless $. > $generated_file_scan_boundary;
  29         my $copyright_match = $parse_copyright->($_, $.);
  30         if ($copyright_match) {
  31             $copyrights{lc("$copyright_match")} = "$copyright_match";
  32         }
  33     }
  34     close(F);
  35     my $copyright = join(" / ", sort values %copyrights);
  36     print "$file\t";
  37     if (check_is_generated_file($file_header)) {
  38         print "GENERATED FILE";
  39     } else {
  40         print ($copyright or "*No copyright*");
  41     }
  42     print "\n";
  43 }
  44
  45 sub check_is_generated_file($) {
  46     my $license = uc($_[0]);
  47     # Remove Python multiline comments to avoid false positives
  48     if (index($license, '"""') != -1) {
  49         $license =~ s/"""[^"]*(?:"""|$)//mg;
  50     }
  51     if (index($license, "'''") != -1) {
  52         $license =~ s/'''[^']*(?:'''|$)//mg;
  53     }
  54     # Quick checks using index.
  55     if (index($license, 'ALL CHANGES MADE IN THIS FILE WILL BE LOST') != -1) {
  56         return 1;
  57     }
  58     if (index($license, 'DO NOT EDIT') != -1 ||
  59         index($license, 'DO NOT DELETE') != -1 ||
  60         index($license, 'GENERATED') != -1) {
  61         return ($license =~ /(All changes made in this file will be lost' .
  62             'DO NOT (EDIT|delete this file)|Generated (at|automatically|data)' .
  63             '|Automatically generated|\Wgenerated\s+(?:\w+\s+)*file\W)/i);
  64     }
  65     return 0;
  66 }
  67
  68 sub are_within_increasing_progression($$$) {
  69     my $delta = $_[0] - $_[1];
  70     return $delta >= 0 && $delta <= $_[2];
  71 }
  72
  73 sub start_copyright_parsing() {
  74     my $max_line_numbers_proximity = 3;
  75     # Set up the defaults the way that proximity checks will not succeed.
  76     my $last_a_item_line_number = -200;
  77     my $last_b_item_line_number = -100;
  78
  79     return sub {
  80         my $line = $_[0];
  81         my $line_number = $_[1];
  82
  83         # Remove C / C++ strings to avoid false positives.
  84         if (index($line, '"') != -1) {
  85             $line =~ s/"[^"\\]*(?:\\.[^"\\]*)*"//g;
  86         }
  87
  88         my $uc_line = uc($line);
  89
  90         # Record '(a)' and '(b)' last occurences in C++ comments.
  91         my $cpp_comment_idx = index($uc_line, '//');
  92         if ($cpp_comment_idx != -1) {
  93             if (index($uc_line, '(A)') > $cpp_comment_idx) {
  94                 $last_a_item_line_number = $line_number;
  95             }
  96             if (index($uc_line, '(B)') > $cpp_comment_idx) {
  97                 $last_b_item_line_number = $line_number;
  98             }
  99         }
 100
 101         # Fast bailout, uses the same patterns as the regexp.
 102         if (index($uc_line, 'COPYRIGHT') == -1 &&
 103             index($uc_line, 'COPR.') == -1 &&
 104             index($uc_line, '\x{00a9}') == -1 &&
 105             index($uc_line, '\xc2\xa9') == -1) {
 106
 107             my $c_item_index = index($uc_line, '(C)');
 108             return '' if ($c_item_index == -1);
 109             # Filter out 'c' used as a list item inside C++ comments.
 110             # E.g. "// blah-blah (a) blah\n// blah-blah (b) and (c) blah"
 111             if ($c_item_index > $cpp_comment_idx &&
 112                 are_within_increasing_progression(
 113                     $line_number,
 114                     $last_b_item_line_number,
 115                     $max_line_numbers_proximity) &&
 116                 are_within_increasing_progression(
 117                     $last_b_item_line_number,
 118                     $last_a_item_line_number,
 119                     $max_line_numbers_proximity)) {
 120                 return '';
 121             }
 122         }
 123
 124         my $copyright_indicator_regex =
 125             '(?:copyright|copr\.|\x{00a9}|\xc2\xa9|\(c\))';
 126         my $full_copyright_indicator_regex =
 127             sprintf '(?:\W|^)%s(?::\s*|\s+)(\w.*)$', $copyright_indicator_regex;
 128         my $copyright_disindicator_regex =
 129             '\b(?:info(?:rmation)?|notice|and|or)\b';
 130
 131         my $copyright = '';
 132         if ($line =~ m%$full_copyright_indicator_regex%i) {
 133             my $match = $1;
 134             if ($match !~ m%^\s*$copyright_disindicator_regex%i) {
 135                 $match =~ s/([,.])?\s*$//;
 136                 $match =~ s/$copyright_indicator_regex//ig;
 137                 $match =~ s/^\s+//;
 138                 $match =~ s/\s{2,}/ /g;
 139                 $match =~ s/\\@/@/g;
 140                 $copyright = $match;
 141             }
 142         }
 143
 144         return $copyright;
 145     }
 146 }