Daily bump.
[gcc.git] / libjava / scripts / encodings.pl
1 # encodings.pl - Download IANA text and compute alias list.
2 # Assumes you are running this program from gnu/gcj/convert/.
3 # Output suitable for direct inclusion in IOConverter.java.
4
5 # Map IANA canonical names onto our canonical names.
6 %map = (
7 'ANSI_X3.4-1968' => 'ASCII',
8 'ISO_8859-1:1987' => '8859_1',
9 'UTF-8' => 'UTF8',
10 'Shift_JIS' => 'SJIS',
11 'Extended_UNIX_Code_Packed_Format_for_Japanese' => 'EUCJIS',
12 'UTF16-LE' => 'UnicodeLittle',
13 'UTF16-BE' => 'UnicodeBig'
14 );
15
16 if ($ARGV[0] eq '')
17 {
18 $file = 'character-sets';
19 if (! -f $file)
20 {
21 # Too painful to figure out how to get Perl to do it.
22 system 'wget -o .wget-log http://www.iana.org/assignments/character-sets';
23 }
24 }
25 else
26 {
27 $file = $ARGV[0];
28 }
29
30 # Include canonical names in the output.
31 foreach $key (keys %map)
32 {
33 $output{lc ($key)} = $map{$key};
34 }
35
36 open (INPUT, "< $file") || die "couldn't open $file: $!";
37
38 $body = 0;
39 $current = '';
40 while (<INPUT>)
41 {
42 chop;
43 $body = 1 if /^Name:/;
44 next unless $body;
45
46 if (/^$/)
47 {
48 $current = '';
49 next;
50 }
51
52 ($type, $name) = split (/\s+/);
53 # Encoding names are case-insensitive. We do all processing on
54 # the lower-case form.
55 my $lower = lc ($name);
56 if ($type eq 'Name:')
57 {
58 $current = $map{$name};
59 if ($current)
60 {
61 $output{$lower} = $current;
62 }
63 }
64 elsif ($type eq 'Alias:')
65 {
66 # The IANA list has some ugliness.
67 if ($name ne '' && $lower ne 'none' && $current)
68 {
69 $output{$lower} = $current;
70 }
71 }
72 }
73
74 close (INPUT);
75
76 foreach $key (sort keys %output)
77 {
78 print " hash.put (\"$key\", \"$output{$key}\");\n";
79 }