3 # Copyright (C) 2008 LibLime
5 # This file is part of Koha.
7 # Koha is free software; you can redistribute it and/or modify it under the
8 # terms of the GNU General Public License as published by the Free Software
9 # Foundation; either version 2 of the License, or (at your option) any later
12 # Koha is distributed in the hope that it will be useful, but WITHOUT ANY
13 # WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
14 # A PARTICULAR PURPOSE. See the GNU General Public License for more details.
16 # You should have received a copy of the GNU General Public License along with
17 # Koha; if not, write to the Free Software Foundation, Inc., 59 Temple Place,
18 # Suite 330, Boston, MA 02111-1307 USA
21 use MARC::Charset qw/marc8_to_utf8/;
24 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS);
27 # set the version for version checking
41 C4::Charset - utilities for handling character set conversions.
49 This module contains routines for dealing with character set
50 conversions, particularly for MARC records.
52 A variety of character encodings are in use by various MARC
53 standards, and even more character encodings are used by
54 non-standard MARC records. The various MARC formats generally
55 do not do a good job of advertising a given record's character
56 encoding, and even when a record does advertise its encoding,
57 e.g., via the Leader/09, experience has shown that one cannot
60 Ultimately, all MARC records are stored in Koha in UTF-8 and
61 must be converted from whatever the source character encoding is.
62 The goal of this module is to ensure that these conversions
63 take place accurately. When a character conversion cannot take
64 place, or at least not accurately, the module was provide
65 enough information to allow user-facing code to inform the user
66 on how to deal with the situation.
72 =head2 IsStringUTF8ish
76 my $is_utf8 = IsStringUTF8ish($str);
80 Determines if C<$str> is valid UTF-8. This can mean
87 The Perl UTF-8 flag is set and the string contains valid UTF-8.
91 The Perl UTF-8 flag is B<not> set, but the octets contain
96 The function is named C<IsStringUTF8ish> instead of C<IsStringUTF8>
97 because in one could be presented with a MARC blob that is
98 not actually in UTF-8 but whose sequence of octets appears to be
99 valid UTF-8. The rest of the MARC character conversion functions
100 will assume that this situation occur does not very often.
104 sub IsStringUTF8ish {
107 return 1 if utf8::is_utf8($str);
108 return utf8::decode($str);
111 =head2 MarcToUTF8Record
115 ($marc_record, $converted_from, $errors_arrayref) = MarcToUTF8Record($marc_blob, $marc_flavour, [, $source_encoding]);
119 Given a MARC blob or a C<MARC::Record>, the MARC flavour, and an
120 optional source encoding, return a C<MARC::Record> that is
123 The returned C<$marc_record> is guaranteed to be in valid UTF-8, but
124 is not guaranteed to have been converted correctly. Specifically,
125 if C<$converted_from> is 'failed', the MARC record returned failed
126 character conversion and had each of its non-ASCII octets changed
127 to the Unicode replacement character.
129 If the source encoding was not specified, this routine will
130 try to guess it; the character encoding used for a successful
131 conversion is returned in C<$converted_from>.
135 sub MarcToUTF8Record {
137 my $marc_flavour = shift;
138 my $source_encoding = shift;
141 my $marc_blob_is_utf8 = 0;
142 if (ref($marc) eq 'MARC::Record') {
143 my $marc_blob = $marc->as_usmarc();
144 $marc_blob_is_utf8 = IsStringUTF8ish($marc_blob);
145 $marc_record = $marc;
147 # dealing with a MARC blob
149 # remove any ersatz whitespace from the beginning and
150 # end of the MARC blob -- these can creep into MARC
151 # files produced by several sources -- caller really
152 # should be doing this, however
155 $marc_blob_is_utf8 = IsStringUTF8ish($marc);
156 $marc_record = MARC::Record->new_from_usmarc($marc);
159 # If we do not know the source encoding, try some guesses
161 # 1. Record is UTF-8 already.
162 # 2. If MARC flavor is MARC21, then
163 # a. record is MARC-8
164 # b. record is ISO-8859-1
165 # 3. If MARC flavor is UNIMARC, then
166 if (not defined $source_encoding) {
167 if ($marc_blob_is_utf8) {
168 # note that for MARC21 we are not bothering to check
169 # if the Leader/09 is set to 'a' or not -- because
170 # of problems with various ILSs (including Koha in the
171 # past, alas), this just is not trustworthy.
172 SetMarcUnicodeFlag($marc_record, $marc_flavour);
173 return $marc_record, 'UTF-8', [];
175 if ($marc_flavour eq 'MARC21') {
176 return _default_marc21_charconv_to_utf8($marc_record, $marc_flavour);
177 } elsif ($marc_flavour eq 'UNIMARC') {
178 return _default_unimarc_charconv_to_utf8($marc_record, $marc_flavour);
180 return _default_marc21_charconv_to_utf8($marc_record, $marc_flavour);
184 # caller knows the character encoding
185 my $original_marc_record = $marc_record->clone();
187 if ($source_encoding =~ /utf-?8/i) {
188 if ($marc_blob_is_utf8) {
189 SetMarcUnicodeFlag($marc_record, $marc_flavour);
190 return $marc_record, 'UTF-8', [];
192 push @errors, 'specified UTF-8 => UTF-8 conversion, but record is not in UTF-8';
194 } elsif ($source_encoding =~ /marc-?8/i) {
195 @errors = _marc_marc8_to_utf8($marc_record, $marc_flavour);
196 } elsif ($source_encoding =~ /5426/) {
197 @errors = _marc_iso5426_to_utf8($marc_record, $marc_flavour);
199 # assume any other character encoding is for Text::Iconv
200 @errors = _marc_to_utf8_via_text_iconv($marc_record, $marc_flavour, 'iso-8859-1');
204 _marc_to_utf8_replacement_char($original_marc_record, $marc_flavour);
205 return $original_marc_record, 'failed', \@errors;
207 return $marc_record, $source_encoding, [];
213 =head2 SetMarcUnicodeFlag
217 SetMarcUnicodeFlag($marc_record, $marc_flavour);
221 Set both the internal MARC::Record encoding flag
222 and the appropriate Leader/09 (MARC21) or
223 100/26-29 (UNIMARC) to indicate that the record
224 is in UTF-8. Note that this does B<not> do
225 any actual character conversion.
229 sub SetMarcUnicodeFlag {
230 my $marc_record = shift;
231 my $marc_flavour = shift;
233 $marc_record->encoding('UTF-8');
234 if ($marc_flavour eq 'MARC21') {
235 my $leader = $marc_record->leader();
236 substr($leader, 9, 1) = 'a';
237 $marc_record->leader($leader);
238 } elsif ($marc_flavour eq "UNIMARC") {
239 if (my $field = $marc_record->field('100')) {
240 my $sfa = $field->subfield('a');
241 substr($sfa, 26, 4) = '5050';
242 $field->update('a' => $sfa);
247 =head2 StripNonXmlChars
251 my $new_str = StripNonXmlChars($old_str);
255 Given a string, return a copy with the
256 characters that are illegal in XML
259 This function exists to work around a problem
260 that can occur with badly-encoded MARC records.
261 Specifically, if a UTF-8 MARC record also
262 has excape (\x1b) characters, MARC::File::XML
263 will let the escape characters pass through
264 when as_xml() or as_xml_record() is called. The
265 problem is that the escape character is not
266 legal in well-formed XML documents, so when
267 MARC::File::XML attempts to parse such a record,
268 the XML parser will fail.
270 Stripping such characters will allow a
271 MARC::Record->new_from_xml()
272 to work, at the possible risk of some data loss.
276 sub StripNonXmlChars {
278 $str =~ s/[^\x09\x0A\x0D\x{0020}-\x{D7FF}\x{E000}-\x{FFFD}\x{10000}-\x{10FFFF}]//g;
282 =head1 INTERNAL FUNCTIONS
284 =head2 _default_marc21_charconv_to_utf8
288 my ($new_marc_record, $guessed_charset) = _default_marc21_charconv_to_utf8($marc_record);
292 Converts a C<MARC::Record> of unknown character set to UTF-8,
293 first by trying a MARC-8 to UTF-8 conversion, then ISO-8859-1
294 to UTF-8, then a default conversion that replaces each non-ASCII
295 character with the replacement character.
297 The C<$guessed_charset> return value contains the character set
298 that resulted in a conversion to valid UTF-8; note that
299 if the MARC-8 and ISO-8859-1 conversions failed, the value of
304 sub _default_marc21_charconv_to_utf8 {
305 my $marc_record = shift;
306 my $marc_flavour = shift;
308 my $trial_marc8 = $marc_record->clone();
310 my @errors = _marc_marc8_to_utf8($trial_marc8, $marc_flavour);
312 return $trial_marc8, 'MARC-8', [];
314 push @all_errors, @errors;
316 my $trial_8859_1 = $marc_record->clone();
317 @errors = _marc_to_utf8_via_text_iconv($trial_8859_1, $marc_flavour, 'iso-8859-1');
319 return $trial_8859_1, 'iso-8859-1', []; # note -- we could return \@all_errors
320 # instead if we wanted to report details
321 # of the failed attempt at MARC-8 => UTF-8
323 push @all_errors, @errors;
325 my $default_converted = $marc_record->clone();
326 _marc_to_utf8_replacement_char($default_converted, $marc_flavour);
327 return $default_converted, 'failed', \@all_errors;
330 =head2 _default_unimarc_charconv_to_utf8
334 my ($new_marc_record, $guessed_charset) = _default_unimarc_charconv_to_utf8($marc_record);
338 Converts a C<MARC::Record> of unknown character set to UTF-8,
339 first by trying a ISO-5426 to UTF-8 conversion, then ISO-8859-1
340 to UTF-8, then a default conversion that replaces each non-ASCII
341 character with the replacement character.
343 The C<$guessed_charset> return value contains the character set
344 that resulted in a conversion to valid UTF-8; note that
345 if the MARC-8 and ISO-8859-1 conversions failed, the value of
350 sub _default_unimarc_charconv_to_utf8 {
351 my $marc_record = shift;
352 my $marc_flavour = shift;
354 my $trial_marc8 = $marc_record->clone();
356 my @errors = _marc_iso5426_to_utf8($trial_marc8, $marc_flavour);
358 return $trial_marc8, 'iso-5426';
360 push @all_errors, @errors;
362 my $trial_8859_1 = $marc_record->clone();
363 @errors = _marc_to_utf8_via_text_iconv($trial_8859_1, $marc_flavour, 'iso-8859-1');
365 return $trial_8859_1, 'iso-8859-1';
367 push @all_errors, @errors;
369 my $default_converted = $marc_record->clone();
370 _marc_to_utf8_replacement_char($default_converted, $marc_flavour);
371 return $default_converted, 'failed', \@all_errors;
374 =head2 _marc_marc8_to_utf8
378 my @errors = _marc_marc8_to_utf8($marc_record, $marc_flavour, $source_encoding);
382 Convert a C<MARC::Record> to UTF-8 in-place from MARC-8.
383 If the conversion fails for some reason, an
384 appropriate messages will be placed in the returned
389 sub _marc_marc8_to_utf8 {
390 my $marc_record = shift;
391 my $marc_flavour = shift;
393 my $prev_ignore = MARC::Charset->ignore_errors();
394 MARC::Charset->ignore_errors(1);
396 # trap warnings raised by MARC::Charset
398 local $SIG{__WARN__} = sub {
400 if ($msg =~ /MARC.Charset/) {
401 # FIXME - purpose of this regexp is to strip out the
402 # line reference to MARC/Charset.pm, but as it
403 # exists probably won't work quite on Windows --
404 # some sort of minimal-bunch back-tracking RE
405 # would be helpful here
406 $msg =~ s/at [\/].*?.MARC.Charset\.pm line \d+\.\n$//;
409 # if warning doesn't come from MARC::Charset, just
415 foreach my $field ($marc_record->fields()) {
416 if ($field->is_control_field()) {
417 ; # do nothing -- control fields should not contain non-ASCII characters
419 my @converted_subfields;
420 foreach my $subfield ($field->subfields()) {
421 my $utf8sf = MARC::Charset::marc8_to_utf8($subfield->[1]);
422 unless (IsStringUTF8ish($utf8sf)) {
423 # Because of a bug in MARC::Charset 0.98, if the string
424 # has (a) one or more diacritics that (b) are only in character positions
425 # 128 to 255 inclusive, the resulting converted string is not in
426 # UTF-8, but the legacy 8-bit encoding (e.g., ISO-8859-1). If that
427 # occurs, upgrade the string in place. Moral of the story seems to be
428 # that pack("U", ...) is better than chr(...) if you need to guarantee
429 # that the resulting string is UTF-8.
430 utf8::upgrade($utf8sf);
432 push @converted_subfields, $subfield->[0], $utf8sf;
435 $field->replace_with(MARC::Field->new(
436 $field->tag(), $field->indicator(1), $field->indicator(2),
437 @converted_subfields)
442 MARC::Charset->ignore_errors($prev_ignore);
444 SetMarcUnicodeFlag($marc_record, $marc_flavour);
449 =head2 _marc_iso5426_to_utf8
453 my @errors = _marc_iso5426_to_utf8($marc_record, $marc_flavour, $source_encoding);
457 Convert a C<MARC::Record> to UTF-8 in-place from ISO-5426.
458 If the conversion fails for some reason, an
459 appropriate messages will be placed in the returned
462 FIXME - is ISO-5426 equivalent enough to MARC-8
463 that C<MARC::Charset> can be used instead?
467 sub _marc_iso5426_to_utf8 {
468 my $marc_record = shift;
469 my $marc_flavour = shift;
473 foreach my $field ($marc_record->fields()) {
474 if ($field->is_control_field()) {
475 ; # do nothing -- control fields should not contain non-ASCII characters
477 my @converted_subfields;
478 foreach my $subfield ($field->subfields()) {
479 my $utf8sf = char_decode5426($subfield->[1]);
480 push @converted_subfields, $subfield->[0], $utf8sf;
483 $field->replace_with(MARC::Field->new(
484 $field->tag(), $field->indicator(1), $field->indicator(2),
485 @converted_subfields)
490 SetMarcUnicodeFlag($marc_record, $marc_flavour);
495 =head2 _marc_to_utf8_via_text_iconv
499 my @errors = _marc_to_utf8_via_text_iconv($marc_record, $marc_flavour, $source_encoding);
503 Convert a C<MARC::Record> to UTF-8 in-place using the
504 C<Text::Iconv> CPAN module. Any source encoding accepted
505 by the user's iconv installation should work. If
506 the source encoding is not recognized on the user's
507 server or the conversion fails for some reason,
508 appropriate messages will be placed in the returned
513 sub _marc_to_utf8_via_text_iconv {
514 my $marc_record = shift;
515 my $marc_flavour = shift;
516 my $source_encoding = shift;
520 eval { $decoder = Text::Iconv->new($source_encoding, 'utf8'); };
522 push @errors, "Could not initialze $source_encoding => utf8 converter: $@";
526 my $prev_raise_error = Text::Iconv->raise_error();
527 Text::Iconv->raise_error(1);
529 foreach my $field ($marc_record->fields()) {
530 if ($field->is_control_field()) {
531 ; # do nothing -- control fields should not contain non-ASCII characters
533 my @converted_subfields;
534 foreach my $subfield ($field->subfields()) {
536 my $conversion_ok = 1;
537 eval { $converted_value = $decoder->convert($subfield->[1]); };
541 } elsif (not defined $converted_value) {
543 push @errors, "Text::Iconv conversion failed - retval is " . $decoder->retval();
546 if ($conversion_ok) {
547 push @converted_subfields, $subfield->[0], $converted_value;
549 $converted_value = $subfield->[1];
550 $converted_value =~ s/[\200-\377]/\xef\xbf\xbd/g;
551 push @converted_subfields, $subfield->[0], $converted_value;
555 $field->replace_with(MARC::Field->new(
556 $field->tag(), $field->indicator(1), $field->indicator(2),
557 @converted_subfields)
562 SetMarcUnicodeFlag($marc_record, $marc_flavour);
563 Text::Iconv->raise_error($prev_raise_error);
568 =head2 _marc_to_utf8_replacement_char
572 _marc_to_utf8_replacement_char($marc_record, $marc_flavour);
576 Convert a C<MARC::Record> to UTF-8 in-place, adopting the
577 unsatisfactory method of replacing all non-ASCII (e.g.,
578 where the eight bit is set) octet with the Unicode
579 replacement character. This is meant as a last-ditch
580 method, and would be best used as part of a UI that
581 lets a cataloguer pick various character conversions
582 until he or she finds the right one.
586 sub _marc_to_utf8_replacement_char {
587 my $marc_record = shift;
588 my $marc_flavour = shift;
590 foreach my $field ($marc_record->fields()) {
591 if ($field->is_control_field()) {
592 ; # do nothing -- control fields should not contain non-ASCII characters
594 my @converted_subfields;
595 foreach my $subfield ($field->subfields()) {
596 my $value = $subfield->[1];
597 $value =~ s/[\200-\377]/\xef\xbf\xbd/g;
598 push @converted_subfields, $subfield->[0], $value;
601 $field->replace_with(MARC::Field->new(
602 $field->tag(), $field->indicator(1), $field->indicator(2),
603 @converted_subfields)
608 SetMarcUnicodeFlag($marc_record, $marc_flavour);
611 =head2 char_decode5426
615 my $utf8string = char_decode5426($iso_5426_string);
619 Converts a string from ISO-5426 to UTF-8.
623 sub char_decode5426 {
627 $chars{0xb0}=0x0101;#3/0ayn[ain]
628 $chars{0xb1}=0x0623;#3/1alif/hamzah[alefwithhamzaabove]
629 #$chars{0xb2}=0x00e0;#'Ã ';
630 $chars{0xb2}=0x00e0;#3/2leftlowsinglequotationmark
631 #$chars{0xb3}=0x00e7;#'ç';
632 $chars{0xb3}=0x00e7;#3/2leftlowsinglequotationmark
637 $chars{0x97}=0x003c;#3/2leftlowsinglequotationmark
638 $chars{0x98}=0x003e;#3/2leftlowsinglequotationmark
639 $chars{0xfa}=0x0153;#oe
640 $chars{0x81d1}=0x00b0;
643 ## combined characters iso5426
645 $chars{0xc041}=0x1ea2; # capital a with hook above
646 $chars{0xc045}=0x1eba; # capital e with hook above
647 $chars{0xc049}=0x1ec8; # capital i with hook above
648 $chars{0xc04f}=0x1ece; # capital o with hook above
649 $chars{0xc055}=0x1ee6; # capital u with hook above
650 $chars{0xc059}=0x1ef6; # capital y with hook above
651 $chars{0xc061}=0x1ea3; # small a with hook above
652 $chars{0xc065}=0x1ebb; # small e with hook above
653 $chars{0xc069}=0x1ec9; # small i with hook above
654 $chars{0xc06f}=0x1ecf; # small o with hook above
655 $chars{0xc075}=0x1ee7; # small u with hook above
656 $chars{0xc079}=0x1ef7; # small y with hook above
659 $chars{0xc141}=0x00c0; # capital a with grave accent
660 $chars{0xc145}=0x00c8; # capital e with grave accent
661 $chars{0xc149}=0x00cc; # capital i with grave accent
662 $chars{0xc14f}=0x00d2; # capital o with grave accent
663 $chars{0xc155}=0x00d9; # capital u with grave accent
664 $chars{0xc157}=0x1e80; # capital w with grave
665 $chars{0xc159}=0x1ef2; # capital y with grave
666 $chars{0xc161}=0x00e0; # small a with grave accent
667 $chars{0xc165}=0x00e8; # small e with grave accent
668 $chars{0xc169}=0x00ec; # small i with grave accent
669 $chars{0xc16f}=0x00f2; # small o with grave accent
670 $chars{0xc175}=0x00f9; # small u with grave accent
671 $chars{0xc177}=0x1e81; # small w with grave
672 $chars{0xc179}=0x1ef3; # small y with grave
674 $chars{0xc241}=0x00c1; # capital a with acute accent
675 $chars{0xc243}=0x0106; # capital c with acute accent
676 $chars{0xc245}=0x00c9; # capital e with acute accent
677 $chars{0xc247}=0x01f4; # capital g with acute
678 $chars{0xc249}=0x00cd; # capital i with acute accent
679 $chars{0xc24b}=0x1e30; # capital k with acute
680 $chars{0xc24c}=0x0139; # capital l with acute accent
681 $chars{0xc24d}=0x1e3e; # capital m with acute
682 $chars{0xc24e}=0x0143; # capital n with acute accent
683 $chars{0xc24f}=0x00d3; # capital o with acute accent
684 $chars{0xc250}=0x1e54; # capital p with acute
685 $chars{0xc252}=0x0154; # capital r with acute accent
686 $chars{0xc253}=0x015a; # capital s with acute accent
687 $chars{0xc255}=0x00da; # capital u with acute accent
688 $chars{0xc257}=0x1e82; # capital w with acute
689 $chars{0xc259}=0x00dd; # capital y with acute accent
690 $chars{0xc25a}=0x0179; # capital z with acute accent
691 $chars{0xc261}=0x00e1; # small a with acute accent
692 $chars{0xc263}=0x0107; # small c with acute accent
693 $chars{0xc265}=0x00e9; # small e with acute accent
694 $chars{0xc267}=0x01f5; # small g with acute
695 $chars{0xc269}=0x00ed; # small i with acute accent
696 $chars{0xc26b}=0x1e31; # small k with acute
697 $chars{0xc26c}=0x013a; # small l with acute accent
698 $chars{0xc26d}=0x1e3f; # small m with acute
699 $chars{0xc26e}=0x0144; # small n with acute accent
700 $chars{0xc26f}=0x00f3; # small o with acute accent
701 $chars{0xc270}=0x1e55; # small p with acute
702 $chars{0xc272}=0x0155; # small r with acute accent
703 $chars{0xc273}=0x015b; # small s with acute accent
704 $chars{0xc275}=0x00fa; # small u with acute accent
705 $chars{0xc277}=0x1e83; # small w with acute
706 $chars{0xc279}=0x00fd; # small y with acute accent
707 $chars{0xc27a}=0x017a; # small z with acute accent
708 $chars{0xc2e1}=0x01fc; # capital ae with acute
709 $chars{0xc2f1}=0x01fd; # small ae with acute
710 # 4/3 circumflex accent
711 $chars{0xc341}=0x00c2; # capital a with circumflex accent
712 $chars{0xc343}=0x0108; # capital c with circumflex
713 $chars{0xc345}=0x00ca; # capital e with circumflex accent
714 $chars{0xc347}=0x011c; # capital g with circumflex
715 $chars{0xc348}=0x0124; # capital h with circumflex
716 $chars{0xc349}=0x00ce; # capital i with circumflex accent
717 $chars{0xc34a}=0x0134; # capital j with circumflex
718 $chars{0xc34f}=0x00d4; # capital o with circumflex accent
719 $chars{0xc353}=0x015c; # capital s with circumflex
720 $chars{0xc355}=0x00db; # capital u with circumflex
721 $chars{0xc357}=0x0174; # capital w with circumflex
722 $chars{0xc359}=0x0176; # capital y with circumflex
723 $chars{0xc35a}=0x1e90; # capital z with circumflex
724 $chars{0xc361}=0x00e2; # small a with circumflex accent
725 $chars{0xc363}=0x0109; # small c with circumflex
726 $chars{0xc365}=0x00ea; # small e with circumflex accent
727 $chars{0xc367}=0x011d; # small g with circumflex
728 $chars{0xc368}=0x0125; # small h with circumflex
729 $chars{0xc369}=0x00ee; # small i with circumflex accent
730 $chars{0xc36a}=0x0135; # small j with circumflex
731 $chars{0xc36e}=0x00f1; # small n with tilde
732 $chars{0xc36f}=0x00f4; # small o with circumflex accent
733 $chars{0xc373}=0x015d; # small s with circumflex
734 $chars{0xc375}=0x00fb; # small u with circumflex
735 $chars{0xc377}=0x0175; # small w with circumflex
736 $chars{0xc379}=0x0177; # small y with circumflex
737 $chars{0xc37a}=0x1e91; # small z with circumflex
739 $chars{0xc441}=0x00c3; # capital a with tilde
740 $chars{0xc445}=0x1ebc; # capital e with tilde
741 $chars{0xc449}=0x0128; # capital i with tilde
742 $chars{0xc44e}=0x00d1; # capital n with tilde
743 $chars{0xc44f}=0x00d5; # capital o with tilde
744 $chars{0xc455}=0x0168; # capital u with tilde
745 $chars{0xc456}=0x1e7c; # capital v with tilde
746 $chars{0xc459}=0x1ef8; # capital y with tilde
747 $chars{0xc461}=0x00e3; # small a with tilde
748 $chars{0xc465}=0x1ebd; # small e with tilde
749 $chars{0xc469}=0x0129; # small i with tilde
750 $chars{0xc46e}=0x00f1; # small n with tilde
751 $chars{0xc46f}=0x00f5; # small o with tilde
752 $chars{0xc475}=0x0169; # small u with tilde
753 $chars{0xc476}=0x1e7d; # small v with tilde
754 $chars{0xc479}=0x1ef9; # small y with tilde
756 $chars{0xc541}=0x0100; # capital a with macron
757 $chars{0xc545}=0x0112; # capital e with macron
758 $chars{0xc547}=0x1e20; # capital g with macron
759 $chars{0xc549}=0x012a; # capital i with macron
760 $chars{0xc54f}=0x014c; # capital o with macron
761 $chars{0xc555}=0x016a; # capital u with macron
762 $chars{0xc561}=0x0101; # small a with macron
763 $chars{0xc565}=0x0113; # small e with macron
764 $chars{0xc567}=0x1e21; # small g with macron
765 $chars{0xc569}=0x012b; # small i with macron
766 $chars{0xc56f}=0x014d; # small o with macron
767 $chars{0xc575}=0x016b; # small u with macron
768 $chars{0xc572}=0x0159; # small r with macron
769 $chars{0xc5e1}=0x01e2; # capital ae with macron
770 $chars{0xc5f1}=0x01e3; # small ae with macron
772 $chars{0xc641}=0x0102; # capital a with breve
773 $chars{0xc645}=0x0114; # capital e with breve
774 $chars{0xc647}=0x011e; # capital g with breve
775 $chars{0xc649}=0x012c; # capital i with breve
776 $chars{0xc64f}=0x014e; # capital o with breve
777 $chars{0xc655}=0x016c; # capital u with breve
778 $chars{0xc661}=0x0103; # small a with breve
779 $chars{0xc665}=0x0115; # small e with breve
780 $chars{0xc667}=0x011f; # small g with breve
781 $chars{0xc669}=0x012d; # small i with breve
782 $chars{0xc66f}=0x014f; # small o with breve
783 $chars{0xc675}=0x016d; # small u with breve
785 $chars{0xc7b0}=0x01e1; # Ain with dot above
786 $chars{0xc742}=0x1e02; # capital b with dot above
787 $chars{0xc743}=0x010a; # capital c with dot above
788 $chars{0xc744}=0x1e0a; # capital d with dot above
789 $chars{0xc745}=0x0116; # capital e with dot above
790 $chars{0xc746}=0x1e1e; # capital f with dot above
791 $chars{0xc747}=0x0120; # capital g with dot above
792 $chars{0xc748}=0x1e22; # capital h with dot above
793 $chars{0xc749}=0x0130; # capital i with dot above
794 $chars{0xc74d}=0x1e40; # capital m with dot above
795 $chars{0xc74e}=0x1e44; # capital n with dot above
796 $chars{0xc750}=0x1e56; # capital p with dot above
797 $chars{0xc752}=0x1e58; # capital r with dot above
798 $chars{0xc753}=0x1e60; # capital s with dot above
799 $chars{0xc754}=0x1e6a; # capital t with dot above
800 $chars{0xc757}=0x1e86; # capital w with dot above
801 $chars{0xc758}=0x1e8a; # capital x with dot above
802 $chars{0xc759}=0x1e8e; # capital y with dot above
803 $chars{0xc75a}=0x017b; # capital z with dot above
804 $chars{0xc761}=0x0227; # small b with dot above
805 $chars{0xc762}=0x1e03; # small b with dot above
806 $chars{0xc763}=0x010b; # small c with dot above
807 $chars{0xc764}=0x1e0b; # small d with dot above
808 $chars{0xc765}=0x0117; # small e with dot above
809 $chars{0xc766}=0x1e1f; # small f with dot above
810 $chars{0xc767}=0x0121; # small g with dot above
811 $chars{0xc768}=0x1e23; # small h with dot above
812 $chars{0xc76d}=0x1e41; # small m with dot above
813 $chars{0xc76e}=0x1e45; # small n with dot above
814 $chars{0xc770}=0x1e57; # small p with dot above
815 $chars{0xc772}=0x1e59; # small r with dot above
816 $chars{0xc773}=0x1e61; # small s with dot above
817 $chars{0xc774}=0x1e6b; # small t with dot above
818 $chars{0xc777}=0x1e87; # small w with dot above
819 $chars{0xc778}=0x1e8b; # small x with dot above
820 $chars{0xc779}=0x1e8f; # small y with dot above
821 $chars{0xc77a}=0x017c; # small z with dot above
822 # 4/8 trema, diaresis
823 $chars{0xc820}=0x00a8; # diaeresis
824 $chars{0xc841}=0x00c4; # capital a with diaeresis
825 $chars{0xc845}=0x00cb; # capital e with diaeresis
826 $chars{0xc848}=0x1e26; # capital h with diaeresis
827 $chars{0xc849}=0x00cf; # capital i with diaeresis
828 $chars{0xc84f}=0x00d6; # capital o with diaeresis
829 $chars{0xc855}=0x00dc; # capital u with diaeresis
830 $chars{0xc857}=0x1e84; # capital w with diaeresis
831 $chars{0xc858}=0x1e8c; # capital x with diaeresis
832 $chars{0xc859}=0x0178; # capital y with diaeresis
833 $chars{0xc861}=0x00e4; # small a with diaeresis
834 $chars{0xc865}=0x00eb; # small e with diaeresis
835 $chars{0xc868}=0x1e27; # small h with diaeresis
836 $chars{0xc869}=0x00ef; # small i with diaeresis
837 $chars{0xc86f}=0x00f6; # small o with diaeresis
838 $chars{0xc874}=0x1e97; # small t with diaeresis
839 $chars{0xc875}=0x00fc; # small u with diaeresis
840 $chars{0xc877}=0x1e85; # small w with diaeresis
841 $chars{0xc878}=0x1e8d; # small x with diaeresis
842 $chars{0xc879}=0x00ff; # small y with diaeresis
844 $chars{0xc920}=0x00a8; # [diaeresis]
845 $chars{0xc961}=0x00e4; # a with umlaut
846 $chars{0xc965}=0x00eb; # e with umlaut
847 $chars{0xc969}=0x00ef; # i with umlaut
848 $chars{0xc96f}=0x00f6; # o with umlaut
849 $chars{0xc975}=0x00fc; # u with umlaut
851 $chars{0xca41}=0x00c5; # capital a with ring above
852 $chars{0xcaad}=0x016e; # capital u with ring above
853 $chars{0xca61}=0x00e5; # small a with ring above
854 $chars{0xca75}=0x016f; # small u with ring above
855 $chars{0xca77}=0x1e98; # small w with ring above
856 $chars{0xca79}=0x1e99; # small y with ring above
857 # 4/11 high comma off centre
858 # 4/12 inverted high comma centred
859 # 4/13 double acute accent
860 $chars{0xcd4f}=0x0150; # capital o with double acute
861 $chars{0xcd55}=0x0170; # capital u with double acute
862 $chars{0xcd6f}=0x0151; # small o with double acute
863 $chars{0xcd75}=0x0171; # small u with double acute
865 $chars{0xce54}=0x01a0; # latin capital letter o with horn
866 $chars{0xce55}=0x01af; # latin capital letter u with horn
867 $chars{0xce74}=0x01a1; # latin small letter o with horn
868 $chars{0xce75}=0x01b0; # latin small letter u with horn
870 $chars{0xcf41}=0x01cd; # capital a with caron
871 $chars{0xcf43}=0x010c; # capital c with caron
872 $chars{0xcf44}=0x010e; # capital d with caron
873 $chars{0xcf45}=0x011a; # capital e with caron
874 $chars{0xcf47}=0x01e6; # capital g with caron
875 $chars{0xcf49}=0x01cf; # capital i with caron
876 $chars{0xcf4b}=0x01e8; # capital k with caron
877 $chars{0xcf4c}=0x013d; # capital l with caron
878 $chars{0xcf4e}=0x0147; # capital n with caron
879 $chars{0xcf4f}=0x01d1; # capital o with caron
880 $chars{0xcf52}=0x0158; # capital r with caron
881 $chars{0xcf53}=0x0160; # capital s with caron
882 $chars{0xcf54}=0x0164; # capital t with caron
883 $chars{0xcf55}=0x01d3; # capital u with caron
884 $chars{0xcf5a}=0x017d; # capital z with caron
885 $chars{0xcf61}=0x01ce; # small a with caron
886 $chars{0xcf63}=0x010d; # small c with caron
887 $chars{0xcf64}=0x010f; # small d with caron
888 $chars{0xcf65}=0x011b; # small e with caron
889 $chars{0xcf67}=0x01e7; # small g with caron
890 $chars{0xcf69}=0x01d0; # small i with caron
891 $chars{0xcf6a}=0x01f0; # small j with caron
892 $chars{0xcf6b}=0x01e9; # small k with caron
893 $chars{0xcf6c}=0x013e; # small l with caron
894 $chars{0xcf6e}=0x0148; # small n with caron
895 $chars{0xcf6f}=0x01d2; # small o with caron
896 $chars{0xcf72}=0x0159; # small r with caron
897 $chars{0xcf73}=0x0161; # small s with caron
898 $chars{0xcf74}=0x0165; # small t with caron
899 $chars{0xcf75}=0x01d4; # small u with caron
900 $chars{0xcf7a}=0x017e; # small z with caron
902 $chars{0xd020}=0x00b8; # cedilla
903 $chars{0xd043}=0x00c7; # capital c with cedilla
904 $chars{0xd044}=0x1e10; # capital d with cedilla
905 $chars{0xd047}=0x0122; # capital g with cedilla
906 $chars{0xd048}=0x1e28; # capital h with cedilla
907 $chars{0xd04b}=0x0136; # capital k with cedilla
908 $chars{0xd04c}=0x013b; # capital l with cedilla
909 $chars{0xd04e}=0x0145; # capital n with cedilla
910 $chars{0xd052}=0x0156; # capital r with cedilla
911 $chars{0xd053}=0x015e; # capital s with cedilla
912 $chars{0xd054}=0x0162; # capital t with cedilla
913 $chars{0xd063}=0x00e7; # small c with cedilla
914 $chars{0xd064}=0x1e11; # small d with cedilla
915 $chars{0xd065}=0x0119; # small e with cedilla
916 $chars{0xd067}=0x0123; # small g with cedilla
917 $chars{0xd068}=0x1e29; # small h with cedilla
918 $chars{0xd06b}=0x0137; # small k with cedilla
919 $chars{0xd06c}=0x013c; # small l with cedilla
920 $chars{0xd06e}=0x0146; # small n with cedilla
921 $chars{0xd072}=0x0157; # small r with cedilla
922 $chars{0xd073}=0x015f; # small s with cedilla
923 $chars{0xd074}=0x0163; # small t with cedilla
926 # 5/3 ogonek (hook to right
927 $chars{0xd320}=0x02db; # ogonek
928 $chars{0xd341}=0x0104; # capital a with ogonek
929 $chars{0xd345}=0x0118; # capital e with ogonek
930 $chars{0xd349}=0x012e; # capital i with ogonek
931 $chars{0xd34f}=0x01ea; # capital o with ogonek
932 $chars{0xd355}=0x0172; # capital u with ogonek
933 $chars{0xd361}=0x0105; # small a with ogonek
934 $chars{0xd365}=0x0119; # small e with ogonek
935 $chars{0xd369}=0x012f; # small i with ogonek
936 $chars{0xd36f}=0x01eb; # small o with ogonek
937 $chars{0xd375}=0x0173; # small u with ogonek
939 $chars{0xd441}=0x1e00; # capital a with ring below
940 $chars{0xd461}=0x1e01; # small a with ring below
941 # 5/5 half circle below
942 $chars{0xf948}=0x1e2a; # capital h with breve below
943 $chars{0xf968}=0x1e2b; # small h with breve below
945 $chars{0xd641}=0x1ea0; # capital a with dot below
946 $chars{0xd642}=0x1e04; # capital b with dot below
947 $chars{0xd644}=0x1e0c; # capital d with dot below
948 $chars{0xd645}=0x1eb8; # capital e with dot below
949 $chars{0xd648}=0x1e24; # capital h with dot below
950 $chars{0xd649}=0x1eca; # capital i with dot below
951 $chars{0xd64b}=0x1e32; # capital k with dot below
952 $chars{0xd64c}=0x1e36; # capital l with dot below
953 $chars{0xd64d}=0x1e42; # capital m with dot below
954 $chars{0xd64e}=0x1e46; # capital n with dot below
955 $chars{0xd64f}=0x1ecc; # capital o with dot below
956 $chars{0xd652}=0x1e5a; # capital r with dot below
957 $chars{0xd653}=0x1e62; # capital s with dot below
958 $chars{0xd654}=0x1e6c; # capital t with dot below
959 $chars{0xd655}=0x1ee4; # capital u with dot below
960 $chars{0xd656}=0x1e7e; # capital v with dot below
961 $chars{0xd657}=0x1e88; # capital w with dot below
962 $chars{0xd659}=0x1ef4; # capital y with dot below
963 $chars{0xd65a}=0x1e92; # capital z with dot below
964 $chars{0xd661}=0x1ea1; # small a with dot below
965 $chars{0xd662}=0x1e05; # small b with dot below
966 $chars{0xd664}=0x1e0d; # small d with dot below
967 $chars{0xd665}=0x1eb9; # small e with dot below
968 $chars{0xd668}=0x1e25; # small h with dot below
969 $chars{0xd669}=0x1ecb; # small i with dot below
970 $chars{0xd66b}=0x1e33; # small k with dot below
971 $chars{0xd66c}=0x1e37; # small l with dot below
972 $chars{0xd66d}=0x1e43; # small m with dot below
973 $chars{0xd66e}=0x1e47; # small n with dot below
974 $chars{0xd66f}=0x1ecd; # small o with dot below
975 $chars{0xd672}=0x1e5b; # small r with dot below
976 $chars{0xd673}=0x1e63; # small s with dot below
977 $chars{0xd674}=0x1e6d; # small t with dot below
978 $chars{0xd675}=0x1ee5; # small u with dot below
979 $chars{0xd676}=0x1e7f; # small v with dot below
980 $chars{0xd677}=0x1e89; # small w with dot below
981 $chars{0xd679}=0x1ef5; # small y with dot below
982 $chars{0xd67a}=0x1e93; # small z with dot below
983 # 5/7 double dot below
984 $chars{0xd755}=0x1e72; # capital u with diaeresis below
985 $chars{0xd775}=0x1e73; # small u with diaeresis below
987 $chars{0xd820}=0x005f; # underline
988 # 5/9 double underline
989 $chars{0xd920}=0x2017; # double underline
990 # 5/10 small low vertical bar
991 $chars{0xda20}=0x02cc; #
992 # 5/11 circumflex below
993 # 5/12 (this position shall not be used)
994 # 5/13 left half of ligature sign and of double tilde
995 # 5/14 right half of ligature sign
996 # 5/15 right half of double tilde
997 # map {printf "%x :%x\n",$_,$chars{$_};}keys %chars;
998 my @data = unpack("C*", $string);
1000 my $length=scalar(@data);
1001 for (my $i = 0; $i < scalar(@data); $i++) {
1002 my $char= $data[$i];
1003 if ($char >= 0x00 && $char <= 0x7F){
1006 push @characters,$char unless ($char<0x02 ||$char== 0x0F);
1007 }elsif (($char >= 0xC0 && $char <= 0xDF)) {
1010 if ($chars{$char*256+$data[$i+1]}) {
1011 $convchar= $chars{$char * 256 + $data[$i+1]};
1013 # printf "char %x $char, char to convert %x , converted %x\n",$char,$char * 256 + $data[$i - 1],$convchar;
1014 } elsif ($chars{$char}) {
1015 $convchar= $chars{$char};
1016 # printf "0xC char %x, converted %x\n",$char,$chars{$char};
1020 push @characters,$convchar;
1023 if ($chars{$char}) {
1024 $convchar= $chars{$char};
1025 # printf "char %x, converted %x\n",$char,$chars{$char};
1027 # printf "char %x $char\n",$char;
1030 push @characters,$convchar;
1033 $result=pack "U*",@characters;
1034 # $result=~s/\x01//;
1035 # $result=~s/\x00//;
1039 $result=~s/\x1b\x5b//;
1040 # map{printf "%x",$_} @characters;
1050 Koha Development Team <info@koha.org>
1052 Galen Charlton <galen.charlton@liblime.com>