1 | ###########################################################################
|
---|
2 | #
|
---|
3 | # multiread.pm --
|
---|
4 | #
|
---|
5 | # Copyright (C) 1999 DigiLib Systems Limited, NZ
|
---|
6 | # Copyright (C) 2005 New Zealand Digital Library project
|
---|
7 | #
|
---|
8 | # This program is free software; you can redistribute it and/or modify
|
---|
9 | # it under the terms of the GNU General Public License as published by
|
---|
10 | # the Free Software Foundation; either version 2 of the License, or
|
---|
11 | # (at your option) any later version.
|
---|
12 | #
|
---|
13 | # This program is distributed in the hope that it will be useful,
|
---|
14 | # but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
15 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
16 | # GNU General Public License for more details.
|
---|
17 | #
|
---|
18 | # You should have received a copy of the GNU General Public License
|
---|
19 | # along with this program; if not, write to the Free Software
|
---|
20 | # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
---|
21 | #
|
---|
22 | ###########################################################################
|
---|
23 |
|
---|
24 | # the multiread object will read in a number of encodings,
|
---|
25 | # the results are always returned in the utf-8 format
|
---|
26 |
|
---|
27 | # encodings currently supported are
|
---|
28 | #
|
---|
29 | # utf8 - either utf8 or unicode (automatically detected)
|
---|
30 | # unicode - 2-byte UCS (does endian detection)
|
---|
31 | #
|
---|
32 | # plus all encodings in the "encodings" package
|
---|
33 |
|
---|
34 | package multiread;
|
---|
35 |
|
---|
36 | eval {require bytes};
|
---|
37 |
|
---|
38 | use strict;
|
---|
39 | no strict 'refs'; # allow filehandles to be variables and viceversa
|
---|
40 |
|
---|
41 | use unicode;
|
---|
42 |
|
---|
43 | sub new {
|
---|
44 | my ($class) = @_;
|
---|
45 |
|
---|
46 | my $self = {'handle' => "",
|
---|
47 | 'first' => 1,
|
---|
48 | 'encoding' => "utf8",
|
---|
49 | 'bigendian' => 1};
|
---|
50 |
|
---|
51 | return bless $self, $class;
|
---|
52 | }
|
---|
53 |
|
---|
54 | # set_handle expects the file to be already open but
|
---|
55 | # not read yet
|
---|
56 | sub set_handle {
|
---|
57 | my $self = shift;
|
---|
58 | $self->{'handle'} = shift;
|
---|
59 | binmode( $self->{'handle'} );
|
---|
60 | $self->{'first'} = 1;
|
---|
61 | $self->{'encoding'} = "utf8";
|
---|
62 | $self->{'bigendian'} = 1;
|
---|
63 | }
|
---|
64 |
|
---|
65 | # set_encoding should be called after set_handle
|
---|
66 | sub set_encoding {
|
---|
67 | my $self = shift;
|
---|
68 | $self->{'encoding'} = shift;
|
---|
69 | }
|
---|
70 |
|
---|
71 | sub get_encoding {
|
---|
72 | my $self = shift (@_);
|
---|
73 | return $self->{'encoding'};
|
---|
74 | }
|
---|
75 |
|
---|
76 | # undef will be returned if the eof has been reached
|
---|
77 | # the result will always be returned in utf-8
|
---|
78 |
|
---|
79 | sub read_unicode_char {
|
---|
80 | my $self = shift (@_);
|
---|
81 |
|
---|
82 | # make sure we have a file handle
|
---|
83 | return undef if ($self->{'handle'} eq "");
|
---|
84 | my $handle = $self->{'handle'};
|
---|
85 |
|
---|
86 | if ($self->{'encoding'} eq "utf8") {
|
---|
87 | # utf-8 text, how many characters we get depends
|
---|
88 | # on what we find
|
---|
89 | my $c1 = "";
|
---|
90 | my $c2 = "";
|
---|
91 | my $c3 = "";
|
---|
92 |
|
---|
93 | while (!eof ($handle)) {
|
---|
94 | $c1 = ord (getc ($handle));
|
---|
95 |
|
---|
96 | if ($c1 <= 0x7f) {
|
---|
97 | # one byte character
|
---|
98 | return chr ($c1);
|
---|
99 |
|
---|
100 | } elsif ($c1 >= 0xc0 && $c1 <= 0xdf) {
|
---|
101 | # two byte character
|
---|
102 | $c2 = getc ($handle) if (!eof ($handle));
|
---|
103 | return chr ($c1) . $c2;
|
---|
104 |
|
---|
105 | } elsif ($c1 >= 0xe0 && $c1 <= 0xef) {
|
---|
106 | # three byte character
|
---|
107 | $c2 = getc ($handle) if (!eof ($handle));
|
---|
108 | $c3 = getc ($handle) if (!eof ($handle));
|
---|
109 | return chr ($c1) . $c2 . $c3;
|
---|
110 | }
|
---|
111 |
|
---|
112 | # if we get here there was an error in the file, we should
|
---|
113 | # be able to recover from it however, maybe the file is in
|
---|
114 | # another encoding
|
---|
115 | }
|
---|
116 |
|
---|
117 | return undef if (eof ($handle));
|
---|
118 | }
|
---|
119 |
|
---|
120 | if ($self->{'encoding'} eq "unicode") {
|
---|
121 | # unicode text, get the next two characters
|
---|
122 | return undef if (eof ($handle));
|
---|
123 | my $c1 = ord (getc ($handle));
|
---|
124 | return undef if (eof ($handle));
|
---|
125 | my $c2 = ord (getc ($handle));
|
---|
126 |
|
---|
127 | return &unicode::unicode2utf8 ([(($self->{'bigendian'}) ? ($c1*256+$c2) : ($c2*256+$c1))]);
|
---|
128 | }
|
---|
129 |
|
---|
130 | return undef;
|
---|
131 | }
|
---|
132 |
|
---|
133 |
|
---|
134 | # undef will be returned if the eof has been reached
|
---|
135 | # the result will always be returned in utf-8
|
---|
136 | sub read_line {
|
---|
137 | my $self = shift (@_);
|
---|
138 |
|
---|
139 | # make sure we have a file handle
|
---|
140 | return undef if ($self->{'handle'} eq "");
|
---|
141 |
|
---|
142 | my $handle = $self->{'handle'};
|
---|
143 |
|
---|
144 | if ($self->{'encoding'} eq "utf8") {
|
---|
145 | # utf-8 line
|
---|
146 | return <$handle>;
|
---|
147 | }
|
---|
148 |
|
---|
149 | if ($self->{'encoding'} eq "unicode") {
|
---|
150 | # unicode line
|
---|
151 | my $c = "";
|
---|
152 | my ($c1, $c2) = ("", "");
|
---|
153 | my $out = "";
|
---|
154 | while (read ($handle, $c, 2) == 2) {
|
---|
155 | $c1 = ord (substr ($c, 0, 1));
|
---|
156 | $c2 = ord (substr ($c, 1, 1));
|
---|
157 | $c = &unicode::unicode2utf8([(($self->{'bigendian'}) ? ($c1*256+$c2) : ($c2*256+$c1))]);
|
---|
158 | $out .= $c;
|
---|
159 | last if ($c eq "\n");
|
---|
160 | }
|
---|
161 |
|
---|
162 | return $out if (length ($out) > 0);
|
---|
163 | return undef;
|
---|
164 | }
|
---|
165 |
|
---|
166 | if ($self->{'encoding'} eq "iso_8859_1") {
|
---|
167 | # we'll use ascii2utf8() for this as it's faster than going
|
---|
168 | # through convert2unicode()
|
---|
169 | my $line = "";
|
---|
170 | if (defined ($line = <$handle>)) {
|
---|
171 | return &unicode::ascii2utf8 (\$line);
|
---|
172 | }
|
---|
173 | }
|
---|
174 |
|
---|
175 | # everything else uses unicode::convert2unicode
|
---|
176 | my $line = "";
|
---|
177 | if (defined ($line = <$handle>)) {
|
---|
178 | return &unicode::unicode2utf8 (&unicode::convert2unicode ($self->{'encoding'}, \$line));
|
---|
179 | }
|
---|
180 |
|
---|
181 | return undef;
|
---|
182 | }
|
---|
183 |
|
---|
184 |
|
---|
185 |
|
---|
186 | # this will look for a Byte Order Marker at the start of the file, and
|
---|
187 | # set the encoding appropriately if there is one, returning any
|
---|
188 | # non-marker text on the first line (or returns undef).
|
---|
189 | sub find_unicode_bom {
|
---|
190 | my $self=shift;
|
---|
191 |
|
---|
192 | my $non_bom_text=""; # to return if we read in 'real' text
|
---|
193 |
|
---|
194 | if ($self->{'first'} == 0) { return }
|
---|
195 |
|
---|
196 | # make sure we have a file handle
|
---|
197 | return if ($self->{'handle'} eq "");
|
---|
198 | my $handle = $self->{'handle'};
|
---|
199 |
|
---|
200 | $self->{'first'} = 0;
|
---|
201 |
|
---|
202 | my $b1 = ord(getc ($handle));
|
---|
203 | my $b2;
|
---|
204 | my $b3;
|
---|
205 |
|
---|
206 | if ($b1 == 0xfe || $b1 == 0xff) {
|
---|
207 | $b2 = ord (getc ($handle)) if (!eof ($handle));
|
---|
208 | if ($b1 == 0xff && $b2 == 0xfe) {
|
---|
209 | $self->{'encoding'} = "unicode";
|
---|
210 | $self->{'bigendian'} = 0;
|
---|
211 | return;
|
---|
212 | } elsif ($b1 == 0xfe && $b2 == 0xff) {
|
---|
213 | $self->{'encoding'} = "unicode";
|
---|
214 | $self->{'bigendian'} = 1;
|
---|
215 | return;
|
---|
216 | } elsif ($b1 == 0xef && $b2 == 0xbb) {
|
---|
217 | $b3 = ord(getc($handle));
|
---|
218 | if ($b3 == 0xbf) {
|
---|
219 | $self->{'encoding'} = "utf8";
|
---|
220 | $self->{'bigendian'} = 1;
|
---|
221 | return;
|
---|
222 | }
|
---|
223 | }
|
---|
224 | } else { # $b1 != fe or ff
|
---|
225 | $handle->ungetc($b1); return;
|
---|
226 | }
|
---|
227 | # if here, we have removed some chars and they aren't a BOM
|
---|
228 | if ($self->{'encoding'} eq "unicode") { # return the 2byte char
|
---|
229 | if (defined ($b3)) { # we looked at this... return it
|
---|
230 | $handle->ungetc($b3);
|
---|
231 | }
|
---|
232 | return &unicode::unicode2utf8([$self->{'bigendian'}?
|
---|
233 | ($b1*256+$b2) : ($b2*256+$b1)]);
|
---|
234 | }
|
---|
235 | # if here, it's utf-8
|
---|
236 | if ($b2 < 0x80) {
|
---|
237 | if (defined ($b3)) { # we grabbed this, but don't need it now
|
---|
238 | $handle->ungetc($b3);
|
---|
239 | }
|
---|
240 | return ($b1 . $b2);
|
---|
241 | }
|
---|
242 | # if here, we have taken part of a multi-byte char. we need to make
|
---|
243 | # sure we return the entire character
|
---|
244 | if (defined($b3) && $b3 < 0x80) { # we have all we need
|
---|
245 | $handle->ungetc($b3);
|
---|
246 | return ($b1 . $b2);
|
---|
247 | }
|
---|
248 | my $c=$b1.$b2.$b3;
|
---|
249 | my $b4=$handle->getc();
|
---|
250 | while ($b4 > 0x7f) { # note - this will return consecutive mb utf8 chars
|
---|
251 | $c .= $b4;
|
---|
252 | $b4=$handle->getc();
|
---|
253 | if (eof($handle)) { last }
|
---|
254 | }
|
---|
255 | if (! eof($handle)) {
|
---|
256 | $handle->ungetc($b4); # this byte is an ascii byte
|
---|
257 | }
|
---|
258 | return $c;
|
---|
259 | }
|
---|
260 |
|
---|
261 |
|
---|
262 |
|
---|
263 | # will convert entire contents of file to utf8 and append result to $outputref
|
---|
264 | # this may be a slightly faster way to get the contents of a file than by
|
---|
265 | # recursively calling read_line()
|
---|
266 | sub read_file {
|
---|
267 | my $self = shift (@_);
|
---|
268 | my ($outputref) = @_;
|
---|
269 |
|
---|
270 | # make sure we have a file handle
|
---|
271 | return if ($self->{'handle'} eq "");
|
---|
272 |
|
---|
273 | my $handle = $self->{'handle'};
|
---|
274 |
|
---|
275 | # if encoding is set to utf8 or unicode, sniff to see if there is a
|
---|
276 | # byte order marker
|
---|
277 | if ($self->{'first'} &&
|
---|
278 | ($self->{'encoding'} eq "utf8" || $self->{'encoding'} eq 'unicode')) {
|
---|
279 | # this will change $self's encoding if there is a BOM
|
---|
280 | my $read_text = $self->find_unicode_bom();
|
---|
281 | $$outputref .= $read_text if (defined($read_text));
|
---|
282 | }
|
---|
283 |
|
---|
284 | if ($self->{'encoding'} eq "utf8") {
|
---|
285 | undef $/;
|
---|
286 | $$outputref .= <$handle>;
|
---|
287 | $/ = "\n";
|
---|
288 | return;
|
---|
289 | }
|
---|
290 |
|
---|
291 | if ($self->{'encoding'} eq "unicode") {
|
---|
292 | my $line = "";
|
---|
293 | while (defined ($line = $self->read_line())) {
|
---|
294 | $$outputref .= $line;
|
---|
295 | }
|
---|
296 | return;
|
---|
297 | }
|
---|
298 |
|
---|
299 | if ($self->{'encoding'} eq "iso_8859_1" || $self->{'encoding'} eq "ascii") {
|
---|
300 | # we'll use ascii2utf8() for this as it's faster than going
|
---|
301 | # through convert2unicode()
|
---|
302 | undef $/;
|
---|
303 | my $text = <$handle>;
|
---|
304 | $/ = "\n";
|
---|
305 | $$outputref .= &unicode::ascii2utf8 (\$text);
|
---|
306 | return;
|
---|
307 | }
|
---|
308 |
|
---|
309 | # everything else uses unicode::convert2unicode
|
---|
310 | undef $/;
|
---|
311 | my $text = <$handle>;
|
---|
312 | $/ = "\n";
|
---|
313 | $$outputref .= &unicode::unicode2utf8 (&unicode::convert2unicode ($self->{'encoding'}, \$text));
|
---|
314 | }
|
---|
315 |
|
---|
316 | 1;
|
---|