Commit | Line | Data |
3ef515df |
1 | #!./perl |
51e4e64d |
2 | # $Id: piconv,v 2.3 2007/04/06 12:53:41 dankogai Exp dankogai $ |
67d7b5ef |
3 | # |
9160fdbd |
4 | use 5.8.0; |
67d7b5ef |
5 | use strict; |
6 | use Encode ; |
7 | use Encode::Alias; |
8 | my %Scheme = map {$_ => 1} qw(from_to decode_encode perlio); |
9 | |
bedba681 |
10 | use File::Basename; |
11 | my $name = basename($0); |
12 | |
c14d784c |
13 | use Getopt::Long qw(:config no_ignore_case); |
bedba681 |
14 | |
15 | my %Opt; |
16 | |
17 | help() |
18 | unless |
19 | GetOptions(\%Opt, |
d1256cb1 |
20 | 'from|f=s', |
21 | 'to|t=s', |
22 | 'list|l', |
23 | 'string|s=s', |
24 | 'check|C=i', |
25 | 'c', |
26 | 'perlqq|p', |
27 | 'htmlcref', |
28 | 'xmlcref', |
29 | 'debug|D', |
30 | 'scheme|S=s', |
31 | 'resolve|r=s', |
32 | 'help', |
33 | ); |
bedba681 |
34 | |
35 | $Opt{help} and help(); |
36 | $Opt{list} and list_encodings(); |
67d7b5ef |
37 | my $locale = $ENV{LC_CTYPE} || $ENV{LC_ALL} || $ENV{LANG}; |
bedba681 |
38 | defined $Opt{resolve} and resolve_encoding($Opt{resolve}); |
39 | $Opt{from} || $Opt{to} || help(); |
40 | my $from = $Opt{from} || $locale or help("from_encoding unspecified"); |
41 | my $to = $Opt{to} || $locale or help("to_encoding unspecified"); |
42 | $Opt{string} and Encode::from_to($Opt{string}, $from, $to) and print $Opt{string} and exit; |
51e4e64d |
43 | my $scheme = exists $Scheme{$Opt{scheme}} ? $Opt{scheme} : 'from_to'; |
bedba681 |
44 | $Opt{check} ||= $Opt{c}; |
d1256cb1 |
45 | $Opt{perlqq} and $Opt{check} = Encode::PERLQQ; |
46 | $Opt{htmlcref} and $Opt{check} = Encode::HTMLCREF; |
47 | $Opt{xmlcref} and $Opt{check} = Encode::XMLCREF; |
bedba681 |
48 | |
49 | if ($Opt{debug}){ |
67d7b5ef |
50 | my $cfrom = Encode->getEncoding($from)->name; |
51 | my $cto = Encode->getEncoding($to)->name; |
ce912cd4 |
52 | print <<"EOT"; |
67d7b5ef |
53 | Scheme: $scheme |
54 | From: $from => $cfrom |
55 | To: $to => $cto |
56 | EOT |
57 | } |
58 | |
8f1ed24a |
59 | # we do not use <> (or ARGV) for the sake of binmode() |
d1256cb1 |
60 | @ARGV or push @ARGV, \*STDIN; |
8f1ed24a |
61 | |
d1256cb1 |
62 | unless ( $scheme eq 'perlio' ) { |
8f1ed24a |
63 | binmode STDOUT; |
d1256cb1 |
64 | for my $argv (@ARGV) { |
65 | my $ifh = ref $argv ? $argv : undef; |
66 | $ifh or open $ifh, "<", $argv or next; |
67 | binmode $ifh; |
68 | if ( $scheme eq 'from_to' ) { # default |
69 | while (<$ifh>) { |
70 | Encode::from_to( $_, $from, $to, $Opt{check} ); |
71 | print; |
72 | } |
73 | } |
74 | elsif ( $scheme eq 'decode_encode' ) { # step-by-step |
75 | while (<$ifh>) { |
76 | my $decoded = decode( $from, $_, $Opt{check} ); |
77 | my $encoded = encode( $to, $decoded ); |
78 | print $encoded; |
79 | } |
80 | } |
81 | else { # won't reach |
82 | die "$name: unknown scheme: $scheme"; |
83 | } |
8f1ed24a |
84 | } |
d1256cb1 |
85 | } |
86 | else { |
87 | |
8f1ed24a |
88 | # NI-S favorite |
89 | binmode STDOUT => "raw:encoding($to)"; |
d1256cb1 |
90 | for my $argv (@ARGV) { |
91 | my $ifh = ref $argv ? $argv : undef; |
92 | $ifh or open $ifh, "<", $argv or next; |
93 | binmode $ifh => "raw:encoding($from)"; |
94 | print while (<$ifh>); |
8f1ed24a |
95 | } |
67d7b5ef |
96 | } |
97 | |
d1256cb1 |
98 | sub list_encodings { |
99 | print join( "\n", Encode->encodings(":all") ), "\n"; |
bedba681 |
100 | exit 0; |
101 | } |
102 | |
103 | sub resolve_encoding { |
d1256cb1 |
104 | if ( my $alias = Encode::resolve_alias( $_[0] ) ) { |
105 | print $alias, "\n"; |
106 | exit 0; |
107 | } |
108 | else { |
109 | warn "$name: $_[0] is not known to Encode\n"; |
110 | exit 1; |
bedba681 |
111 | } |
67d7b5ef |
112 | } |
113 | |
d1256cb1 |
114 | sub help { |
67d7b5ef |
115 | my $message = shift; |
67d7b5ef |
116 | $message and print STDERR "$name error: $message\n"; |
117 | print STDERR <<"EOT"; |
118 | $name [-f from_encoding] [-t to_encoding] [-s string] [files...] |
119 | $name -l |
bedba681 |
120 | $name -r encoding_alias |
121 | -l,--list |
122 | lists all available encodings |
123 | -r,--resolve encoding_alias |
124 | resolve encoding to its (Encode) canonical name |
125 | -f,--from from_encoding |
126 | when omitted, the current locale will be used |
127 | -t,--to to_encoding |
128 | when omitted, the current locale will be used |
129 | -s,--string string |
130 | "string" will be the input instead of STDIN or files |
131 | The following are mainly of interest to Encode hackers: |
132 | -D,--debug show debug information |
d1256cb1 |
133 | -C N | -c check the validity of the input |
bedba681 |
134 | -S,--scheme scheme use the scheme for conversion |
d1256cb1 |
135 | Those are handy when you can only see ascii characters: |
136 | -p,--perlqq |
137 | --htmlcref |
138 | --xmlcref |
67d7b5ef |
139 | EOT |
d1256cb1 |
140 | exit; |
67d7b5ef |
141 | } |
142 | |
143 | __END__ |
144 | |
145 | =head1 NAME |
146 | |
147 | piconv -- iconv(1), reinvented in perl |
148 | |
149 | =head1 SYNOPSIS |
150 | |
151 | piconv [-f from_encoding] [-t to_encoding] [-s string] [files...] |
152 | piconv -l |
bedba681 |
153 | piconv [-C N|-c|-p] |
154 | piconv -S scheme ... |
155 | piconv -r encoding |
156 | piconv -D ... |
157 | piconv -h |
67d7b5ef |
158 | |
159 | =head1 DESCRIPTION |
160 | |
0ab8f81e |
161 | B<piconv> is perl version of B<iconv>, a character encoding converter |
162 | widely available for various Unixen today. This script was primarily |
163 | a technology demonstrator for Perl 5.8.0, but you can use piconv in the |
164 | place of iconv for virtually any case. |
67d7b5ef |
165 | |
0ab8f81e |
166 | piconv converts the character encoding of either STDIN or files |
167 | specified in the argument and prints out to STDOUT. |
67d7b5ef |
168 | |
bedba681 |
169 | Here is the list of options. Each option can be in short format (-f) |
170 | or long (--from). |
67d7b5ef |
171 | |
172 | =over 4 |
173 | |
bedba681 |
174 | =item -f,--from from_encoding |
67d7b5ef |
175 | |
0ab8f81e |
176 | Specifies the encoding you are converting from. Unlike B<iconv>, |
177 | this option can be omitted. In such cases, the current locale is used. |
67d7b5ef |
178 | |
bedba681 |
179 | =item -t,--to to_encoding |
67d7b5ef |
180 | |
0ab8f81e |
181 | Specifies the encoding you are converting to. Unlike B<iconv>, |
182 | this option can be omitted. In such cases, the current locale is used. |
67d7b5ef |
183 | |
0ab8f81e |
184 | Therefore, when both -f and -t are omitted, B<piconv> just acts |
185 | like B<cat>. |
67d7b5ef |
186 | |
bedba681 |
187 | =item -s,--string I<string> |
67d7b5ef |
188 | |
bedba681 |
189 | uses I<string> instead of file for the source of text. |
67d7b5ef |
190 | |
bedba681 |
191 | =item -l,--list |
67d7b5ef |
192 | |
ce912cd4 |
193 | Lists all available encodings, one per line, in case-insensitive |
0ab8f81e |
194 | order. Note that only the canonical names are listed; many aliases |
ce912cd4 |
195 | exist. For example, the names are case-insensitive, and many standard |
0ab8f81e |
196 | and common aliases work, such as "latin1" for "ISO-8859-1", or "ibm850" |
ce912cd4 |
197 | instead of "cp850", or "winlatin1" for "cp1252". See L<Encode::Supported> |
0ab8f81e |
198 | for a full discussion. |
67d7b5ef |
199 | |
bedba681 |
200 | =item -C,--check I<N> |
b2704119 |
201 | |
202 | Check the validity of the stream if I<N> = 1. When I<N> = -1, something |
203 | interesting happens when it encounters an invalid character. |
204 | |
205 | =item -c |
206 | |
207 | Same as C<-C 1>. |
208 | |
bedba681 |
209 | =item -p,--perlqq |
b2704119 |
210 | |
d1256cb1 |
211 | =item --htmlcref |
212 | |
213 | =item --xmlcref |
214 | |
215 | Applies PERLQQ, HTMLCREF, XMLCREF, respectively. Try |
216 | |
217 | piconv -f utf8 -t ascii --perlqq |
218 | |
219 | To see what it does. |
b2704119 |
220 | |
bedba681 |
221 | =item -h,--help |
d31fa6c4 |
222 | |
223 | Show usage. |
224 | |
bedba681 |
225 | =item -D,--debug |
67d7b5ef |
226 | |
7748829a |
227 | Invokes debugging mode. Primarily for Encode hackers. |
67d7b5ef |
228 | |
bedba681 |
229 | =item -S,--scheme scheme |
67d7b5ef |
230 | |
231 | Selects which scheme is to be used for conversion. Available schemes |
0ab8f81e |
232 | are as follows: |
67d7b5ef |
233 | |
234 | =over 4 |
235 | |
236 | =item from_to |
237 | |
238 | Uses Encode::from_to for conversion. This is the default. |
239 | |
240 | =item decode_encode |
241 | |
7748829a |
242 | Input strings are decode()d then encode()d. A straight two-step |
67d7b5ef |
243 | implementation. |
244 | |
245 | =item perlio |
246 | |
7748829a |
247 | The new perlIO layer is used. NI-S' favorite. |
67d7b5ef |
248 | |
51e4e64d |
249 | You should use this option if you are using UTF-16 and others which |
250 | linefeed is not $/. |
251 | |
67d7b5ef |
252 | =back |
253 | |
0ab8f81e |
254 | Like the I<-D> option, this is also for Encode hackers. |
67d7b5ef |
255 | |
256 | =back |
257 | |
258 | =head1 SEE ALSO |
259 | |
bedba681 |
260 | L<iconv/1> |
261 | L<locale/3> |
67d7b5ef |
262 | L<Encode> |
ce912cd4 |
263 | L<Encode::Supported> |
264 | L<Encode::Alias> |
67d7b5ef |
265 | L<PerlIO> |
266 | |
267 | =cut |