1Unicode::UCD(3pm) Perl Programmers Reference Guide Unicode::UCD(3pm)
2
6 Unicode::UCD - Unicode character database
7
9 use Unicode::UCD 'charinfo';
10 my $charinfo = charinfo($codepoint);
11 use Unicode::UCD 'charprop';
13 my $value = charprop($codepoint, $property);
14 use Unicode::UCD 'charprops_all';
16 my $all_values_hash_ref = charprops_all($codepoint);
17 use Unicode::UCD 'casefold';
19 my $casefold = casefold($codepoint);
20 use Unicode::UCD 'all_casefolds';
22 my $all_casefolds_ref = all_casefolds();
23 use Unicode::UCD 'casespec';
25 my $casespec = casespec($codepoint);
26 use Unicode::UCD 'charblock';
28 my $charblock = charblock($codepoint);
29 use Unicode::UCD 'charscript';
31 my $charscript = charscript($codepoint);
32 use Unicode::UCD 'charblocks';
34 my $charblocks = charblocks();
35 use Unicode::UCD 'charscripts';
37 my $charscripts = charscripts();
38 use Unicode::UCD qw(charscript charinrange);
40 my $range = charscript($script);
41 print "looks like $script\n" if charinrange($range, $codepoint);
42 use Unicode::UCD qw(general_categories bidi_types);
44 my $categories = general_categories();
45 my $types = bidi_types();
46 use Unicode::UCD 'prop_aliases';
48 my @space_names = prop_aliases("space");
49 use Unicode::UCD 'prop_value_aliases';
51 my @gc_punct_names = prop_value_aliases("Gc", "Punct");
52 use Unicode::UCD 'prop_values';
54 my @all_EA_short_names = prop_values("East_Asian_Width");
55 use Unicode::UCD 'prop_invlist';
57 my @puncts = prop_invlist("gc=punctuation");
58 use Unicode::UCD 'prop_invmap';
60 my ($list_ref, $map_ref, $format, $missing)
61 = prop_invmap("General Category");
62 use Unicode::UCD 'search_invlist';
64 my $index = search_invlist(\@invlist, $code_point);
65 # The following function should be used only internally in
67 # implementations of the Unicode Normalization Algorithm, and there
68 # are better choices than it.
69 use Unicode::UCD 'compexcl';
70 my $compexcl = compexcl($codepoint);
71 use Unicode::UCD 'namedseq';
73 my $namedseq = namedseq($named_sequence_name);
74 my $unicode_version = Unicode::UCD::UnicodeVersion();
76 my $convert_to_numeric =
78 Unicode::UCD::num("\N{RUMI DIGIT ONE}\N{RUMI DIGIT TWO}");
79
81 The Unicode::UCD module offers a series of functions that provide a
82 simple interface to the Unicode Character Database.
83 code point argument
85 Some of the functions are called with a code point argument, which is
86 either a decimal or a hexadecimal scalar designating a code point in
87 the platform's native character set (extended to Unicode), or a string
88 containing "U+" followed by hexadecimals designating a Unicode code
89 point. A leading 0 will force a hexadecimal interpretation, as will a
90 hexadecimal digit that isn't a decimal digit.
91 Examples:
93 223 # Decimal 223 in native character set
95 0223 # Hexadecimal 223, native (= 547 decimal)
96 0xDF # Hexadecimal DF, native (= 223 decimal)
97 '0xDF' # String form of hexadecimal (= 223 decimal)
98 'U+DF' # Hexadecimal DF, in Unicode's character set
99 (= LATIN SMALL LETTER SHARP S)
100 Note that the largest code point in Unicode is U+10FFFF.
102 charinfo()
104 use Unicode::UCD 'charinfo';
105 my $charinfo = charinfo(0x41);
107 This returns information about the input "code point argument" as a
109 reference to a hash of fields as defined by the Unicode standard. If
110 the "code point argument" is not assigned in the standard (i.e., has
111 the general category "Cn" meaning "Unassigned") or is a non-character
112 (meaning it is guaranteed to never be assigned in the standard),
113 "undef" is returned.
114 Fields that aren't applicable to the particular code point argument
116 exist in the returned hash, and are empty.
117 For results that are less "raw" than this function returns, or to get
119 the values for any property, not just the few covered by this function,
120 use the "charprop()" function.
121 The keys in the hash with the meanings of their values are:
123 code
125 the input native "code point argument" expressed in hexadecimal,
126 with leading zeros added if necessary to make it contain at least
127 four hexdigits
128 name
130 name of code, all IN UPPER CASE. Some control-type code points do
131 not have names. This field will be empty for "Surrogate" and
132 "Private Use" code points, and for the others without a name, it
133 will contain a description enclosed in angle brackets, like
134 "<control>".
135 category
137 The short name of the general category of code. This will match
138 one of the keys in the hash returned by "general_categories()".
139 The "prop_value_aliases()" function can be used to get all the
141 synonyms of the category name.
142 combining
144 the combining class number for code used in the Canonical Ordering
145 Algorithm. For Unicode 5.1, this is described in Section 3.11
146 "Canonical Ordering Behavior" available at
147 <http://www.unicode.org/versions/Unicode5.1.0/>
148 The "prop_value_aliases()" function can be used to get all the
150 synonyms of the combining class number.
151 bidi
153 bidirectional type of code. This will match one of the keys in the
154 hash returned by "bidi_types()".
155 The "prop_value_aliases()" function can be used to get all the
157 synonyms of the bidi type name.
158 decomposition
160 is empty if code has no decomposition; or is one or more codes
161 (separated by spaces) that, taken in order, represent a
162 decomposition for code. Each has at least four hexdigits. The
163 codes may be preceded by a word enclosed in angle brackets, then a
164 space, like "<compat> ", giving the type of decomposition
165 This decomposition may be an intermediate one whose components are
167 also decomposable. Use Unicode::Normalize to get the final
168 decomposition in one step.
169 decimal
171 if code represents a decimal digit this is its integer numeric
172 value
173 digit
175 if code represents some other digit-like number, this is its
176 integer numeric value
177 numeric
179 if code represents a whole or rational number, this is its numeric
180 value. Rational values are expressed as a string like "1/4".
181 mirrored
183 "Y" or "N" designating if code is mirrored in bidirectional text
184 unicode10
186 name of code in the Unicode 1.0 standard if one existed for this
187 code point and is different from the current name
188 comment
190 As of Unicode 6.0, this is always empty.
191 upper
193 is, if non-empty, the uppercase mapping for code expressed as at
194 least four hexdigits. This indicates that the full uppercase
195 mapping is a single character, and is identical to the simple
196 (single-character only) mapping. When this field is empty, it
197 means that the simple uppercase mapping is code itself; you'll need
198 some other means, (like "charprop()" or "casespec()" to get the
199 full mapping.
200 lower
202 is, if non-empty, the lowercase mapping for code expressed as at
203 least four hexdigits. This indicates that the full lowercase
204 mapping is a single character, and is identical to the simple
205 (single-character only) mapping. When this field is empty, it
206 means that the simple lowercase mapping is code itself; you'll need
207 some other means, (like "charprop()" or "casespec()" to get the
208 full mapping.
209 title
211 is, if non-empty, the titlecase mapping for code expressed as at
212 least four hexdigits. This indicates that the full titlecase
213 mapping is a single character, and is identical to the simple
214 (single-character only) mapping. When this field is empty, it
215 means that the simple titlecase mapping is code itself; you'll need
216 some other means, (like "charprop()" or "casespec()" to get the
217 full mapping.
218 block
220 the block code belongs to (used in "\p{Blk=...}"). The
221 "prop_value_aliases()" function can be used to get all the synonyms
222 of the block name.
223 See "Blocks versus Scripts".
225 script
227 the script code belongs to. The "prop_value_aliases()" function
228 can be used to get all the synonyms of the script name. Note that
229 this is the older "Script" property value, and not the improved
230 "Script_Extensions" value.
231 See "Blocks versus Scripts".
233 Note that you cannot do (de)composition and casing based solely on the
235 decomposition, combining, lower, upper, and title fields; you will need
236 also the "casespec()" function and the "Composition_Exclusion"
237 property. (Or you could just use the lc(), uc(), and ucfirst()
238 functions, and the Unicode::Normalize module.)
239 charprop()
241 use Unicode::UCD 'charprop';
242 print charprop(0x41, "Gc"), "\n";
244 print charprop(0x61, "General_Category"), "\n";
245 prints
247 Lu
248 Ll
249 This returns the value of the Unicode property given by the second
251 parameter for the "code point argument" given by the first.
252 The passed-in property may be specified as any of the synonyms returned
254 by "prop_aliases()".
255 The return value is always a scalar, either a string or a number. For
257 properties where there are synonyms for the values, the synonym
258 returned by this function is the longest, most descriptive form, the
259 one returned by "prop_value_aliases()" when called in a scalar context.
260 Of course, you can call "prop_value_aliases()" on the result to get
261 other synonyms.
262 The return values are more "cooked" than the "charinfo()" ones. For
264 example, the "uc" property value is the actual string containing the
265 full uppercase mapping of the input code point. You have to go to
266 extra trouble with "charinfo" to get this value from its "upper" hash
267 element when the full mapping differs from the simple one.
268 Special note should be made of the return values for a few properties:
270 Block
272 The value returned is the new-style (see "Old-style versus new-
273 style block names").
274 Decomposition_Mapping
276 Like "charinfo()", the result may be an intermediate decomposition
277 whose components are also decomposable. Use Unicode::Normalize to
278 get the final decomposition in one step.
279 Unlike "charinfo()", this does not include the decomposition type.
281 Use the "Decomposition_Type" property to get that.
282 Name_Alias
284 If the input code point's name has more than one synonym, they are
285 returned joined into a single comma-separated string.
286 Numeric_Value
288 If the result is a fraction, it is converted into a floating point
289 number to the accuracy of your platform.
290 Script_Extensions
292 If the result is multiple script names, they are returned joined
293 into a single comma-separated string.
294 When called with a property that is a Perl extension that isn't
296 expressible in a compound form, this function currently returns
297 "undef", as the only two possible values are true or false (1 or 0 I
298 suppose). This behavior may change in the future, so don't write code
299 that relies on it. "Present_In" is a Perl extension that is
300 expressible in a bipartite or compound form (for example,
301 "\p{Present_In=4.0}"), so "charprop" accepts it. But "Any" is a Perl
302 extension that isn't expressible that way, so "charprop" returns
303 "undef" for it. Also "charprop" returns "undef" for all Perl
304 extensions that are internal-only.
305 charprops_all()
307 use Unicode::UCD 'charprops_all';
308 my $%properties_of_A_hash_ref = charprops_all("U+41");
310 This returns a reference to a hash whose keys are all the distinct
312 Unicode (no Perl extension) properties, and whose values are the
313 respective values for those properties for the input "code point
314 argument".
315 Each key is the property name in its longest, most descriptive form.
317 The values are what "charprop()" would return.
318 This function is expensive in time and memory.
320 charblock()
322 use Unicode::UCD 'charblock';
323 my $charblock = charblock(0x41);
325 my $charblock = charblock(1234);
326 my $charblock = charblock(0x263a);
327 my $charblock = charblock("U+263a");
328 my $range = charblock('Armenian');
330 With a "code point argument" charblock() returns the block the code
332 point belongs to, e.g. "Basic Latin". The old-style block name is
333 returned (see "Old-style versus new-style block names"). The
334 "prop_value_aliases()" function can be used to get all the synonyms of
335 the block name.
336 If the code point is unassigned, this returns the block it would belong
338 to if it were assigned. (If the Unicode version being used is so early
339 as to not have blocks, all code points are considered to be in
340 "No_Block".)
341 See also "Blocks versus Scripts".
343 If supplied with an argument that can't be a code point, charblock()
345 tries to do the opposite and interpret the argument as an old-style
346 block name. On an ASCII platform, the return value is a range set with
347 one range: an anonymous array with a single element that consists of
348 another anonymous array whose first element is the first code point in
349 the block, and whose second element is the final code point in the
350 block. On an EBCDIC platform, the first two Unicode blocks are not
351 contiguous. Their range sets are lists containing start-of-range, end-
352 of-range code point pairs. You can test whether a code point is in a
353 range set using the "charinrange()" function. (To be precise, each
354 range set contains a third array element, after the range boundary
355 ones: the old_style block name.)
356 If the argument to charblock() is not a known block, "undef" is
358 returned.
359 charscript()
361 use Unicode::UCD 'charscript';
362 my $charscript = charscript(0x41);
364 my $charscript = charscript(1234);
365 my $charscript = charscript("U+263a");
366 my $range = charscript('Thai');
368 With a "code point argument", charscript() returns the script the code
370 point belongs to, e.g., "Latin", "Greek", "Han". If the code point is
371 unassigned or the Unicode version being used is so early that it
372 doesn't have scripts, this function returns "Unknown". The
373 "prop_value_aliases()" function can be used to get all the synonyms of
374 the script name.
375 Note that the Script_Extensions property is an improved version of the
377 Script property, and you should probably be using that instead, with
378 the "charprop()" function.
379 If supplied with an argument that can't be a code point, charscript()
381 tries to do the opposite and interpret the argument as a script name.
382 The return value is a range set: an anonymous array of arrays that
383 contain start-of-range, end-of-range code point pairs. You can test
384 whether a code point is in a range set using the "charinrange()"
385 function. (To be precise, each range set contains a third array
386 element, after the range boundary ones: the script name.)
387 If the charscript() argument is not a known script, "undef" is
389 returned.
390 See also "Blocks versus Scripts".
392 charblocks()
394 use Unicode::UCD 'charblocks';
395 my $charblocks = charblocks();
397 charblocks() returns a reference to a hash with the known block names
399 as the keys, and the code point ranges (see "charblock()") as the
400 values.
401 The names are in the old-style (see "Old-style versus new-style block
403 names").
404 prop_invmap("block") can be used to get this same data in a different
406 type of data structure.
407 prop_values("Block") can be used to get all the known new-style block
409 names as a list, without the code point ranges.
410 See also "Blocks versus Scripts".
412 charscripts()
414 use Unicode::UCD 'charscripts';
415 my $charscripts = charscripts();
417 charscripts() returns a reference to a hash with the known script names
419 as the keys, and the code point ranges (see "charscript()") as the
420 values.
421 prop_invmap("script") can be used to get this same data in a different
423 type of data structure. Since the Script_Extensions property is an
424 improved version of the Script property, you should instead use
425 prop_invmap("scx").
426 prop_values("Script") can be used to get all the known script names as
428 a list, without the code point ranges.
429 See also "Blocks versus Scripts".
431 charinrange()
433 In addition to using the "\p{Blk=...}" and "\P{Blk=...}" constructs,
434 you can also test whether a code point is in the range as returned by
435 "charblock()" and "charscript()" or as the values of the hash returned
436 by "charblocks()" and "charscripts()" by using charinrange():
437 use Unicode::UCD qw(charscript charinrange);
439 $range = charscript('Hiragana');
441 print "looks like hiragana\n" if charinrange($range, $codepoint);
442 general_categories()
444 use Unicode::UCD 'general_categories';
445 my $categories = general_categories();
447 This returns a reference to a hash which has short general category
449 names (such as "Lu", "Nd", "Zs", "S") as keys and long names (such as
450 "UppercaseLetter", "DecimalNumber", "SpaceSeparator", "Symbol") as
451 values. The hash is reversible in case you need to go from the long
452 names to the short names. The general category is the one returned
453 from "charinfo()" under the "category" key.
454 The "prop_values()" and "prop_value_aliases()" functions can be used as
456 an alternative to this function; the first returning a simple list of
457 the short category names; and the second gets all the synonyms of a
458 given category name.
459 bidi_types()
461 use Unicode::UCD 'bidi_types';
462 my $categories = bidi_types();
464 This returns a reference to a hash which has the short bidi
466 (bidirectional) type names (such as "L", "R") as keys and long names
467 (such as "Left-to-Right", "Right-to-Left") as values. The hash is
468 reversible in case you need to go from the long names to the short
469 names. The bidi type is the one returned from "charinfo()" under the
470 "bidi" key. For the exact meaning of the various bidi classes the
471 Unicode TR9 is recommended reading:
472 <http://www.unicode.org/reports/tr9/> (as of Unicode 5.0.0)
473 The "prop_values()" and "prop_value_aliases()" functions can be used as
475 an alternative to this function; the first returning a simple list of
476 the short bidi type names; and the second gets all the synonyms of a
477 given bidi type name.
478 compexcl()
480 WARNING: Unicode discourages the use of this function or any of the
481 alternative mechanisms listed in this section (the documentation of
482 compexcl()), except internally in implementations of the Unicode
483 Normalization Algorithm. You should be using Unicode::Normalize
484 directly instead of these. Using these will likely lead to half-baked
485 results.
486 use Unicode::UCD 'compexcl';
488 my $compexcl = compexcl(0x09dc);
490 This routine returns "undef" if the Unicode version being used is so
492 early that it doesn't have this property.
493 compexcl() is included for backwards compatibility, but as of Perl 5.12
495 and more modern Unicode versions, for most purposes it is probably more
496 convenient to use one of the following instead:
497 my $compexcl = chr(0x09dc) =~ /\p{Comp_Ex};
499 my $compexcl = chr(0x09dc) =~ /\p{Full_Composition_Exclusion};
500 or even
502 my $compexcl = chr(0x09dc) =~ /\p{CE};
504 my $compexcl = chr(0x09dc) =~ /\p{Composition_Exclusion};
505 The first two forms return true if the "code point argument" should not
507 be produced by composition normalization. For the final two forms to
508 return true, it is additionally required that this fact not otherwise
509 be determinable from the Unicode data base.
510 This routine behaves identically to the final two forms. That is, it
512 does not return true if the code point has a decomposition consisting
513 of another single code point, nor if its decomposition starts with a
514 code point whose combining class is non-zero. Code points that meet
515 either of these conditions should also not be produced by composition
516 normalization, which is probably why you should use the
517 "Full_Composition_Exclusion" property instead, as shown above.
518 The routine returns false otherwise.
520 casefold()
522 use Unicode::UCD 'casefold';
523 my $casefold = casefold(0xDF);
525 if (defined $casefold) {
526 my @full_fold_hex = split / /, $casefold->{'full'};
527 my $full_fold_string =
528 join "", map {chr(hex($_))} @full_fold_hex;
529 my @turkic_fold_hex =
530 split / /, ($casefold->{'turkic'} ne "")
531 ? $casefold->{'turkic'}
532 : $casefold->{'full'};
533 my $turkic_fold_string =
534 join "", map {chr(hex($_))} @turkic_fold_hex;
535 }
536 if (defined $casefold && $casefold->{'simple'} ne "") {
537 my $simple_fold_hex = $casefold->{'simple'};
538 my $simple_fold_string = chr(hex($simple_fold_hex));
539 }
540 This returns the (almost) locale-independent case folding of the
542 character specified by the "code point argument". (Starting in Perl
543 v5.16, the core function fc() returns the "full" mapping (described
544 below) faster than this does, and for entire strings.)
545 If there is no case folding for the input code point, "undef" is
547 returned.
548 If there is a case folding for that code point, a reference to a hash
550 with the following fields is returned:
551 code
553 the input native "code point argument" expressed in hexadecimal,
554 with leading zeros added if necessary to make it contain at least
555 four hexdigits
556 full
558 one or more codes (separated by spaces) that, taken in order, give
559 the code points for the case folding for code. Each has at least
560 four hexdigits.
561 simple
563 is empty, or is exactly one code with at least four hexdigits which
564 can be used as an alternative case folding when the calling program
565 cannot cope with the fold being a sequence of multiple code points.
566 If full is just one code point, then simple equals full. If there
567 is no single code point folding defined for code, then simple is
568 the empty string. Otherwise, it is an inferior, but still better-
569 than-nothing alternative folding to full.
570 mapping
572 is the same as simple if simple is not empty, and it is the same as
573 full otherwise. It can be considered to be the simplest possible
574 folding for code. It is defined primarily for backwards
575 compatibility.
576 status
578 is "C" (for "common") if the best possible fold is a single code
579 point (simple equals full equals mapping). It is "S" if there are
580 distinct folds, simple and full (mapping equals simple). And it is
581 "F" if there is only a full fold (mapping equals full; simple is
582 empty). Note that this describes the contents of mapping. It is
583 defined primarily for backwards compatibility.
584 For Unicode versions between 3.1 and 3.1.1 inclusive, status can
586 also be "I" which is the same as "C" but is a special case for
587 dotted uppercase I and dotless lowercase i:
588 * If you use this "I" mapping
590 the result is case-insensitive, but dotless and dotted I's are
591 not distinguished
592 * If you exclude this "I" mapping
594 the result is not fully case-insensitive, but dotless and
595 dotted I's are distinguished
596 turkic
598 contains any special folding for Turkic languages. For versions of
599 Unicode starting with 3.2, this field is empty unless code has a
600 different folding in Turkic languages, in which case it is one or
601 more codes (separated by spaces) that, taken in order, give the
602 code points for the case folding for code in those languages. Each
603 code has at least four hexdigits. Note that this folding does not
604 maintain canonical equivalence without additional processing.
605 For Unicode versions between 3.1 and 3.1.1 inclusive, this field is
607 empty unless there is a special folding for Turkic languages, in
608 which case status is "I", and mapping, full, simple, and turkic are
609 all equal.
610 Programs that want complete generality and the best folding results
612 should use the folding contained in the full field. But note that the
613 fold for some code points will be a sequence of multiple code points.
614 Programs that can't cope with the fold mapping being multiple code
616 points can use the folding contained in the simple field, with the loss
617 of some generality. In Unicode 5.1, about 7% of the defined foldings
618 have no single code point folding.
619 The mapping and status fields are provided for backwards compatibility
621 for existing programs. They contain the same values as in previous
622 versions of this function.
623 Locale is not completely independent. The turkic field contains
625 results to use when the locale is a Turkic language.
626 For more information about case mappings see
628 <http://www.unicode.org/reports/tr21>
629 all_casefolds()
631 use Unicode::UCD 'all_casefolds';
632 my $all_folds_ref = all_casefolds();
634 foreach my $char_with_casefold (sort { $a <=> $b }
635 keys %$all_folds_ref)
636 {
637 printf "%04X:", $char_with_casefold;
638 my $casefold = $all_folds_ref->{$char_with_casefold};
639 # Get folds for $char_with_casefold
641 my @full_fold_hex = split / /, $casefold->{'full'};
643 my $full_fold_string =
644 join "", map {chr(hex($_))} @full_fold_hex;
645 print " full=", join " ", @full_fold_hex;
646 my @turkic_fold_hex =
647 split / /, ($casefold->{'turkic'} ne "")
648 ? $casefold->{'turkic'}
649 : $casefold->{'full'};
650 my $turkic_fold_string =
651 join "", map {chr(hex($_))} @turkic_fold_hex;
652 print "; turkic=", join " ", @turkic_fold_hex;
653 if (defined $casefold && $casefold->{'simple'} ne "") {
654 my $simple_fold_hex = $casefold->{'simple'};
655 my $simple_fold_string = chr(hex($simple_fold_hex));
656 print "; simple=$simple_fold_hex";
657 }
658 print "\n";
659 }
660 This returns all the case foldings in the current version of Unicode in
662 the form of a reference to a hash. Each key to the hash is the decimal
663 representation of a Unicode character that has a casefold to other than
664 itself. The casefold of a semi-colon is itself, so it isn't in the
665 hash; likewise for a lowercase "a", but there is an entry for a capital
666 "A". The hash value for each key is another hash, identical to what is
667 returned by "casefold()" if called with that code point as its
668 argument. So the value "all_casefolds()->{ord("A")}'" is equivalent to
669 "casefold(ord("A"))";
670 casespec()
672 use Unicode::UCD 'casespec';
673 my $casespec = casespec(0xFB00);
675 This returns the potentially locale-dependent case mappings of the
677 "code point argument". The mappings may be longer than a single code
678 point (which the basic Unicode case mappings as returned by
679 "charinfo()" never are).
680 If there are no case mappings for the "code point argument", or if all
682 three possible mappings (lower, title and upper) result in single code
683 points and are locale independent and unconditional, "undef" is
684 returned (which means that the case mappings, if any, for the code
685 point are those returned by "charinfo()").
686 Otherwise, a reference to a hash giving the mappings (or a reference to
688 a hash of such hashes, explained below) is returned with the following
689 keys and their meanings:
690 The keys in the bottom layer hash with the meanings of their values
692 are:
693 code
695 the input native "code point argument" expressed in hexadecimal,
696 with leading zeros added if necessary to make it contain at least
697 four hexdigits
698 lower
700 one or more codes (separated by spaces) that, taken in order, give
701 the code points for the lower case of code. Each has at least four
702 hexdigits.
703 title
705 one or more codes (separated by spaces) that, taken in order, give
706 the code points for the title case of code. Each has at least four
707 hexdigits.
708 upper
710 one or more codes (separated by spaces) that, taken in order, give
711 the code points for the upper case of code. Each has at least four
712 hexdigits.
713 condition
715 the conditions for the mappings to be valid. If "undef", the
716 mappings are always valid. When defined, this field is a list of
717 conditions, all of which must be true for the mappings to be valid.
718 The list consists of one or more locales (see below) and/or
719 contexts (explained in the next paragraph), separated by spaces.
720 (Other than as used to separate elements, spaces are to be
721 ignored.) Case distinctions in the condition list are not
722 significant. Conditions preceded by "NON_" represent the negation
723 of the condition.
724 A context is one of those defined in the Unicode standard. For
726 Unicode 5.1, they are defined in Section 3.13 "Default Case
727 Operations" available at
728 <http://www.unicode.org/versions/Unicode5.1.0/>. These are for
729 context-sensitive casing.
730 The hash described above is returned for locale-independent casing,
732 where at least one of the mappings has length longer than one. If
733 "undef" is returned, the code point may have mappings, but if so, all
734 are length one, and are returned by "charinfo()". Note that when this
735 function does return a value, it will be for the complete set of
736 mappings for a code point, even those whose length is one.
737 If there are additional casing rules that apply only in certain
739 locales, an additional key for each will be defined in the returned
740 hash. Each such key will be its locale name, defined as a 2-letter ISO
741 3166 country code, possibly followed by a "_" and a 2-letter ISO
742 language code (possibly followed by a "_" and a variant code). You can
743 find the lists of all possible locales, see Locale::Country and
744 Locale::Language. (In Unicode 6.0, the only locales returned by this
745 function are "lt", "tr", and "az".)
746 Each locale key is a reference to a hash that has the form above, and
748 gives the casing rules for that particular locale, which take
749 precedence over the locale-independent ones when in that locale.
750 If the only casing for a code point is locale-dependent, then the
752 returned hash will not have any of the base keys, like "code", "upper",
753 etc., but will contain only locale keys.
754 For more information about case mappings see
756 <http://www.unicode.org/reports/tr21/>
757 namedseq()
759 use Unicode::UCD 'namedseq';
760 my $namedseq = namedseq("KATAKANA LETTER AINU P");
762 my @namedseq = namedseq("KATAKANA LETTER AINU P");
763 my %namedseq = namedseq();
764 If used with a single argument in a scalar context, returns the string
766 consisting of the code points of the named sequence, or "undef" if no
767 named sequence by that name exists. If used with a single argument in
768 a list context, it returns the list of the ordinals of the code points.
769 If used with no arguments in a list context, it returns a hash with the
771 names of all the named sequences as the keys and their sequences as
772 strings as the values. Otherwise, it returns "undef" or an empty list
773 depending on the context.
774 This function only operates on officially approved (not provisional)
776 named sequences.
777 Note that as of Perl 5.14, "\N{KATAKANA LETTER AINU P}" will insert the
779 named sequence into double-quoted strings, and
780 "charnames::string_vianame("KATAKANA LETTER AINU P")" will return the
781 same string this function does, but will also operate on character
782 names that aren't named sequences, without you having to know which are
783 which. See charnames.
784 num()
786 use Unicode::UCD 'num';
787 my $val = num("123");
789 my $one_quarter = num("\N{VULGAR FRACTION ONE QUARTER}");
790 my $val = num("12a", \$valid_length); # $valid_length contains 2
791 num() returns the numeric value of the input Unicode string; or "undef"
793 if it doesn't think the entire string has a completely valid, safe
794 numeric value. If called with an optional second parameter, a
795 reference to a scalar, num() will set the scalar to the length of any
796 valid initial substring; or to 0 if none.
797 If the string is just one character in length, the Unicode numeric
799 value is returned if it has one, or "undef" otherwise. If the optional
800 scalar ref is passed, it would be set to 1 if the return is valid; or 0
801 if the return is "undef". Note that the numeric value returned need
802 not be a whole number. "num("\N{TIBETAN DIGIT HALF ZERO}")", for
803 example returns -0.5.
804 If the string is more than one character, "undef" is returned unless
806 all its characters are decimal digits (that is, they would match
807 "\d+"), from the same script. For example if you have an ASCII '0' and
808 a Bengali '3', mixed together, they aren't considered a valid number,
809 and "undef" is returned. A further restriction is that the digits all
810 have to be of the same form. A half-width digit mixed with a full-
811 width one will return "undef". The Arabic script has two sets of
812 digits; "num" will return "undef" unless all the digits in the string
813 come from the same set. In all cases, the optional scalar ref
814 parameter is set to how long any valid initial substring of digits is;
815 hence it will be set to the entire string length if the main return
816 value is not "undef".
817 "num" errs on the side of safety, and there may be valid strings of
819 decimal digits that it doesn't recognize. Note that Unicode defines a
820 number of "digit" characters that aren't "decimal digit" characters.
821 "Decimal digits" have the property that they have a positional value,
822 i.e., there is a units position, a 10's position, a 100's, etc, AND
823 they are arranged in Unicode in blocks of 10 contiguous code points.
824 The Chinese digits, for example, are not in such a contiguous block,
825 and so Unicode doesn't view them as decimal digits, but merely digits,
826 and so "\d" will not match them. A single-character string containing
827 one of these digits will have its decimal value returned by "num", but
828 any longer string containing only these digits will return "undef".
829 Strings of multiple sub- and superscripts are not recognized as
831 numbers. You can use either of the compatibility decompositions in
832 Unicode::Normalize to change these into digits, and then call "num" on
833 the result.
834 prop_aliases()
836 use Unicode::UCD 'prop_aliases';
837 my ($short_name, $full_name, @other_names) = prop_aliases("space");
839 my $same_full_name = prop_aliases("Space"); # Scalar context
840 my ($same_short_name) = prop_aliases("Space"); # gets 0th element
841 print "The full name is $full_name\n";
842 print "The short name is $short_name\n";
843 print "The other aliases are: ", join(", ", @other_names), "\n";
844 prints:
846 The full name is White_Space
847 The short name is WSpace
848 The other aliases are: Space
849 Most Unicode properties have several synonymous names. Typically,
851 there is at least a short name, convenient to type, and a long name
852 that more fully describes the property, and hence is more easily
853 understood.
854 If you know one name for a Unicode property, you can use "prop_aliases"
856 to find either the long name (when called in scalar context), or a list
857 of all of the names, somewhat ordered so that the short name is in the
858 0th element, the long name in the next element, and any other synonyms
859 are in the remaining elements, in no particular order.
860 The long name is returned in a form nicely capitalized, suitable for
862 printing.
863 The input parameter name is loosely matched, which means that white
865 space, hyphens, and underscores are ignored (except for the trailing
866 underscore in the old_form grandfathered-in "L_", which is better
867 written as "LC", and both of which mean "General_Category=Cased
868 Letter").
869 If the name is unknown, "undef" is returned (or an empty list in list
871 context). Note that Perl typically recognizes property names in
872 regular expressions with an optional ""Is_"" (with or without the
873 underscore) prefixed to them, such as "\p{isgc=punct}". This function
874 does not recognize those in the input, returning "undef". Nor are they
875 included in the output as possible synonyms.
876 "prop_aliases" does know about the Perl extensions to Unicode
878 properties, such as "Any" and "XPosixAlpha", and the single form
879 equivalents to Unicode properties such as "XDigit", "Greek",
880 "In_Greek", and "Is_Greek". The final example demonstrates that the
881 "Is_" prefix is recognized for these extensions; it is needed to
882 resolve ambiguities. For example, prop_aliases('lc') returns the list
883 "(lc, Lowercase_Mapping)", but prop_aliases('islc') returns "(Is_LC,
884 Cased_Letter)". This is because "islc" is a Perl extension which is
885 short for "General_Category=Cased Letter". The lists returned for the
886 Perl extensions will not include the "Is_" prefix (whether or not the
887 input had it) unless needed to resolve ambiguities, as shown in the
888 "islc" example, where the returned list had one element containing
889 "Is_", and the other without.
890 It is also possible for the reverse to happen: prop_aliases('isc')
892 returns the list "(isc, ISO_Comment)"; whereas prop_aliases('c')
893 returns "(C, Other)" (the latter being a Perl extension meaning
894 "General_Category=Other". "Properties accessible through Unicode::UCD"
895 in perluniprops lists the available forms, including which ones are
896 discouraged from use.
897 Those discouraged forms are accepted as input to "prop_aliases", but
899 are not returned in the lists. prop_aliases('isL&') and
900 prop_aliases('isL_'), which are old synonyms for "Is_LC" and should not
901 be used in new code, are examples of this. These both return "(Is_LC,
902 Cased_Letter)". Thus this function allows you to take a discouraged
903 form, and find its acceptable alternatives. The same goes with single-
904 form Block property equivalences. Only the forms that begin with "In_"
905 are not discouraged; if you pass "prop_aliases" a discouraged form, you
906 will get back the equivalent ones that begin with "In_". It will
907 otherwise look like a new-style block name (see. "Old-style versus
908 new-style block names").
909 "prop_aliases" does not know about any user-defined properties, and
911 will return "undef" if called with one of those. Likewise for Perl
912 internal properties, with the exception of "Perl_Decimal_Digit" which
913 it does know about (and which is documented below in "prop_invmap()").
914 prop_values()
916 use Unicode::UCD 'prop_values';
917 print "AHex values are: ", join(", ", prop_values("AHex")),
919 "\n";
920 prints:
921 AHex values are: N, Y
922 Some Unicode properties have a restricted set of legal values. For
924 example, all binary properties are restricted to just "true" or
925 "false"; and there are only a few dozen possible General Categories.
926 Use "prop_values" to find out if a given property is one such, and if
927 so, to get a list of the values:
928 print join ", ", prop_values("NFC_Quick_Check");
930 prints:
931 M, N, Y
932 If the property doesn't have such a restricted set, "undef" is
934 returned.
935 There are usually several synonyms for each possible value. Use
937 "prop_value_aliases()" to access those.
938 Case, white space, hyphens, and underscores are ignored in the input
940 property name (except for the trailing underscore in the old-form
941 grandfathered-in general category property value "L_", which is better
942 written as "LC").
943 If the property name is unknown, "undef" is returned. Note that Perl
945 typically recognizes property names in regular expressions with an
946 optional ""Is_"" (with or without the underscore) prefixed to them,
947 such as "\p{isgc=punct}". This function does not recognize those in
948 the property parameter, returning "undef".
949 For the block property, new-style block names are returned (see "Old-
951 style versus new-style block names").
952 "prop_values" does not know about any user-defined properties, and will
954 return "undef" if called with one of those.
955 prop_value_aliases()
957 use Unicode::UCD 'prop_value_aliases';
958 my ($short_name, $full_name, @other_names)
960 = prop_value_aliases("Gc", "Punct");
961 my $same_full_name = prop_value_aliases("Gc", "P"); # Scalar cntxt
962 my ($same_short_name) = prop_value_aliases("Gc", "P"); # gets 0th
963 # element
964 print "The full name is $full_name\n";
965 print "The short name is $short_name\n";
966 print "The other aliases are: ", join(", ", @other_names), "\n";
967 prints:
969 The full name is Punctuation
970 The short name is P
971 The other aliases are: Punct
972 Some Unicode properties have a restricted set of legal values. For
974 example, all binary properties are restricted to just "true" or
975 "false"; and there are only a few dozen possible General Categories.
976 You can use "prop_values()" to find out if a given property is one
978 which has a restricted set of values, and if so, what those values are.
979 But usually each value actually has several synonyms. For example, in
980 Unicode binary properties, truth can be represented by any of the
981 strings "Y", "Yes", "T", or "True"; and the General Category
982 "Punctuation" by that string, or "Punct", or simply "P".
983 Like property names, there is typically at least a short name for each
985 such property-value, and a long name. If you know any name of the
986 property-value (which you can get by "prop_values()", you can use
987 "prop_value_aliases"() to get the long name (when called in scalar
988 context), or a list of all the names, with the short name in the 0th
989 element, the long name in the next element, and any other synonyms in
990 the remaining elements, in no particular order, except that any all-
991 numeric synonyms will be last.
992 The long name is returned in a form nicely capitalized, suitable for
994 printing.
995 Case, white space, hyphens, and underscores are ignored in the input
997 parameters (except for the trailing underscore in the old-form
998 grandfathered-in general category property value "L_", which is better
999 written as "LC").
1000 If either name is unknown, "undef" is returned. Note that Perl
1002 typically recognizes property names in regular expressions with an
1003 optional ""Is_"" (with or without the underscore) prefixed to them,
1004 such as "\p{isgc=punct}". This function does not recognize those in
1005 the property parameter, returning "undef".
1006 If called with a property that doesn't have synonyms for its values, it
1008 returns the input value, possibly normalized with capitalization and
1009 underscores, but not necessarily checking that the input value is
1010 valid.
1011 For the block property, new-style block names are returned (see "Old-
1013 style versus new-style block names").
1014 To find the synonyms for single-forms, such as "\p{Any}", use
1016 "prop_aliases()" instead.
1017 "prop_value_aliases" does not know about any user-defined properties,
1019 and will return "undef" if called with one of those.
1020 prop_invlist()
1022 "prop_invlist" returns an inversion list (described below) that defines
1023 all the code points for the binary Unicode property (or
1024 "property=value" pair) given by the input parameter string:
1025 use feature 'say';
1027 use Unicode::UCD 'prop_invlist';
1028 say join ", ", prop_invlist("Any");
1029 prints:
1031 0, 1114112
1032 If the input is unknown "undef" is returned in scalar context; an
1034 empty-list in list context. If the input is known, the number of
1035 elements in the list is returned if called in scalar context.
1036 perluniprops gives the list of properties that this function accepts,
1038 as well as all the possible forms for them (including with the optional
1039 "Is_" prefixes). (Except this function doesn't accept any Perl-
1040 internal properties, some of which are listed there.) This function
1041 uses the same loose or tighter matching rules for resolving the input
1042 property's name as is done for regular expressions. These are also
1043 specified in perluniprops. Examples of using the "property=value" form
1044 are:
1045 say join ", ", prop_invlist("Script_Extensions=Shavian");
1047 prints:
1049 66640, 66688
1050 say join ", ", prop_invlist("ASCII_Hex_Digit=No");
1052 prints:
1054 0, 48, 58, 65, 71, 97, 103
1055 say join ", ", prop_invlist("ASCII_Hex_Digit=Yes");
1057 prints:
1059 48, 58, 65, 71, 97, 103
1060 Inversion lists are a compact way of specifying Unicode property-value
1062 definitions. The 0th item in the list is the lowest code point that
1063 has the property-value. The next item (item [1]) is the lowest code
1064 point beyond that one that does NOT have the property-value. And the
1065 next item beyond that ([2]) is the lowest code point beyond that one
1066 that does have the property-value, and so on. Put another way, each
1067 element in the list gives the beginning of a range that has the
1068 property-value (for even numbered elements), or doesn't have the
1069 property-value (for odd numbered elements). The name for this data
1070 structure stems from the fact that each element in the list toggles (or
1071 inverts) whether the corresponding range is or isn't on the list.
1072 In the final example above, the first ASCII Hex digit is code point 48,
1074 the character "0", and all code points from it through 57 (a "9") are
1075 ASCII hex digits. Code points 58 through 64 aren't, but 65 (an "A")
1076 through 70 (an "F") are, as are 97 ("a") through 102 ("f"). 103 starts
1077 a range of code points that aren't ASCII hex digits. That range
1078 extends to infinity, which on your computer can be found in the
1079 variable $Unicode::UCD::MAX_CP. (This variable is as close to infinity
1080 as Perl can get on your platform, and may be too high for some
1081 operations to work; you may wish to use a smaller number for your
1082 purposes.)
1083 Note that the inversion lists returned by this function can possibly
1085 include non-Unicode code points, that is anything above 0x10FFFF.
1086 Unicode properties are not defined on such code points. You might wish
1087 to change the output to not include these. Simply add 0x110000 at the
1088 end of the non-empty returned list if it isn't already that value; and
1089 pop that value if it is; like:
1090 my @list = prop_invlist("foo");
1092 if (@list) {
1093 if ($list[-1] == 0x110000) {
1094 pop @list; # Defeat the turning on for above Unicode
1095 }
1096 else {
1097 push @list, 0x110000; # Turn off for above Unicode
1098 }
1099 }
1100 It is a simple matter to expand out an inversion list to a full list of
1102 all code points that have the property-value:
1103 my @invlist = prop_invlist($property_name);
1105 die "empty" unless @invlist;
1106 my @full_list;
1107 for (my $i = 0; $i < @invlist; $i += 2) {
1108 my $upper = ($i + 1) < @invlist
1109 ? $invlist[$i+1] - 1 # In range
1110 : $Unicode::UCD::MAX_CP; # To infinity.
1111 for my $j ($invlist[$i] .. $upper) {
1112 push @full_list, $j;
1113 }
1114 }
1115 "prop_invlist" does not know about any user-defined nor Perl internal-
1117 only properties, and will return "undef" if called with one of those.
1118 The "search_invlist()" function is provided for finding a code point
1120 within an inversion list.
1121 prop_invmap()
1123 use Unicode::UCD 'prop_invmap';
1124 my ($list_ref, $map_ref, $format, $default)
1125 = prop_invmap("General Category");
1126 "prop_invmap" is used to get the complete mapping definition for a
1128 property, in the form of an inversion map. An inversion map consists
1129 of two parallel arrays. One is an ordered list of code points that
1130 mark range beginnings, and the other gives the value (or mapping) that
1131 all code points in the corresponding range have.
1132 "prop_invmap" is called with the name of the desired property. The
1134 name is loosely matched, meaning that differences in case, white-space,
1135 hyphens, and underscores are not meaningful (except for the trailing
1136 underscore in the old-form grandfathered-in property "L_", which is
1137 better written as "LC", or even better, "Gc=LC").
1138 Many Unicode properties have more than one name (or alias).
1140 "prop_invmap" understands all of these, including Perl extensions to
1141 them. Ambiguities are resolved as described above for "prop_aliases()"
1142 (except if a property has both a complete mapping, and a binary "Y"/"N"
1143 mapping, then specifying the property name prefixed by "is" causes the
1144 binary one to be returned). The Perl internal property
1145 "Perl_Decimal_Digit, described below, is also accepted. An empty list
1146 is returned if the property name is unknown. See "Properties
1147 accessible through Unicode::UCD" in perluniprops for the properties
1148 acceptable as inputs to this function.
1149 It is a fatal error to call this function except in list context.
1151 In addition to the two arrays that form the inversion map,
1153 "prop_invmap" returns two other values; one is a scalar that gives some
1154 details as to the format of the entries of the map array; the other is
1155 a default value, useful in maps whose format name begins with the
1156 letter "a", as described below in its subsection; and for specialized
1157 purposes, such as converting to another data structure, described at
1158 the end of this main section.
1159 This means that "prop_invmap" returns a 4 element list. For example,
1161 my ($blocks_ranges_ref, $blocks_maps_ref, $format, $default)
1163 = prop_invmap("Block");
1164 In this call, the two arrays will be populated as shown below (for
1166 Unicode 6.0):
1167 Index @blocks_ranges @blocks_maps
1169 0 0x0000 Basic Latin
1170 1 0x0080 Latin-1 Supplement
1171 2 0x0100 Latin Extended-A
1172 3 0x0180 Latin Extended-B
1173 4 0x0250 IPA Extensions
1174 5 0x02B0 Spacing Modifier Letters
1175 6 0x0300 Combining Diacritical Marks
1176 7 0x0370 Greek and Coptic
1177 8 0x0400 Cyrillic
1178 ...
1179 233 0x2B820 No_Block
1180 234 0x2F800 CJK Compatibility Ideographs Supplement
1181 235 0x2FA20 No_Block
1182 236 0xE0000 Tags
1183 237 0xE0080 No_Block
1184 238 0xE0100 Variation Selectors Supplement
1185 239 0xE01F0 No_Block
1186 240 0xF0000 Supplementary Private Use Area-A
1187 241 0x100000 Supplementary Private Use Area-B
1188 242 0x110000 No_Block
1189 The first line (with Index [0]) means that the value for code point 0
1191 is "Basic Latin". The entry "0x0080" in the @blocks_ranges column in
1192 the second line means that the value from the first line, "Basic
1193 Latin", extends to all code points in the range from 0 up to but not
1194 including 0x0080, that is, through 127. In other words, the code
1195 points from 0 to 127 are all in the "Basic Latin" block. Similarly,
1196 all code points in the range from 0x0080 up to (but not including)
1197 0x0100 are in the block named "Latin-1 Supplement", etc. (Notice that
1198 the return is the old-style block names; see "Old-style versus new-
1199 style block names").
1200 The final line (with Index [242]) means that the value for all code
1202 points above the legal Unicode maximum code point have the value
1203 "No_Block", which is the term Unicode uses for a non-existing block.
1204 The arrays completely specify the mappings for all possible code
1206 points. The final element in an inversion map returned by this
1207 function will always be for the range that consists of all the code
1208 points that aren't legal Unicode, but that are expressible on the
1209 platform. (That is, it starts with code point 0x110000, the first code
1210 point above the legal Unicode maximum, and extends to infinity.) The
1211 value for that range will be the same that any typical unassigned code
1212 point has for the specified property. (Certain unassigned code points
1213 are not "typical"; for example the non-character code points, or those
1214 in blocks that are to be written right-to-left. The above-Unicode
1215 range's value is not based on these atypical code points.) It could be
1216 argued that, instead of treating these as unassigned Unicode code
1217 points, the value for this range should be "undef". If you wish, you
1218 can change the returned arrays accordingly.
1219 The maps for almost all properties are simple scalars that should be
1221 interpreted as-is. These values are those given in the Unicode-
1222 supplied data files, which may be inconsistent as to capitalization and
1223 as to which synonym for a property-value is given. The results may be
1224 normalized by using the "prop_value_aliases()" function.
1225 There are exceptions to the simple scalar maps. Some properties have
1227 some elements in their map list that are themselves lists of scalars;
1228 and some special strings are returned that are not to be interpreted
1229 as-is. Element [2] (placed into $format in the example above) of the
1230 returned four element list tells you if the map has any of these
1231 special elements or not, as follows:
1232 "s" means all the elements of the map array are simple scalars, with no
1234 special elements. Almost all properties are like this, like the
1235 "block" example above.
1236 "sl"
1238 means that some of the map array elements have the form given by
1239 "s", and the rest are lists of scalars. For example, here is a
1240 portion of the output of calling "prop_invmap"() with the "Script
1241 Extensions" property:
1242 @scripts_ranges @scripts_maps
1244 ...
1245 0x0953 Devanagari
1246 0x0964 [ Bengali, Devanagari, Gurumukhi, Oriya ]
1247 0x0966 Devanagari
1248 0x0970 Common
1249 Here, the code points 0x964 and 0x965 are both used in Bengali,
1251 Devanagari, Gurmukhi, and Oriya, but no other scripts.
1252 The Name_Alias property is also of this form. But each scalar
1254 consists of two components: 1) the name, and 2) the type of alias
1255 this is. They are separated by a colon and a space. In Unicode
1256 6.1, there are several alias types:
1257 "correction"
1259 indicates that the name is a corrected form for the original
1260 name (which remains valid) for the same code point.
1261 "control"
1263 adds a new name for a control character.
1264 "alternate"
1266 is an alternate name for a character
1267 "figment"
1269 is a name for a character that has been documented but was
1270 never in any actual standard.
1271 "abbreviation"
1273 is a common abbreviation for a character
1274 The lists are ordered (roughly) so the most preferred names come
1276 before less preferred ones.
1277 For example,
1279 @aliases_ranges @alias_maps
1281 ...
1282 0x009E [ 'PRIVACY MESSAGE: control', 'PM: abbreviation' ]
1283 0x009F [ 'APPLICATION PROGRAM COMMAND: control',
1284 'APC: abbreviation'
1285 ]
1286 0x00A0 'NBSP: abbreviation'
1287 0x00A1 ""
1288 0x00AD 'SHY: abbreviation'
1289 0x00AE ""
1290 0x01A2 'LATIN CAPITAL LETTER GHA: correction'
1291 0x01A3 'LATIN SMALL LETTER GHA: correction'
1292 0x01A4 ""
1293 ...
1294 A map to the empty string means that there is no alias defined for
1296 the code point.
1297 "a" is like "s" in that all the map array elements are scalars, but
1299 here they are restricted to all being integers, and some have to be
1300 adjusted (hence the name "a") to get the correct result. For
1301 example, in:
1302 my ($uppers_ranges_ref, $uppers_maps_ref, $format, $default)
1304 = prop_invmap("Simple_Uppercase_Mapping");
1305 the returned arrays look like this:
1307 @$uppers_ranges_ref @$uppers_maps_ref Note
1309 0 0
1310 97 65 'a' maps to 'A', b => B ...
1311 123 0
1312 181 924 MICRO SIGN => Greek Cap MU
1313 182 0
1314 ...
1315 and $default is 0.
1317 Let's start with the second line. It says that the uppercase of
1319 code point 97 is 65; or uc("a") == "A". But the line is for the
1320 entire range of code points 97 through 122. To get the mapping for
1321 any code point in this range, you take the offset it has from the
1322 beginning code point of the range, and add that to the mapping for
1323 that first code point. So, the mapping for 122 ("z") is derived by
1324 taking the offset of 122 from 97 (=25) and adding that to 65,
1325 yielding 90 ("Z"). Likewise for everything in between.
1326 Requiring this simple adjustment allows the returned arrays to be
1328 significantly smaller than otherwise, up to a factor of 10,
1329 speeding up searching through them.
1330 Ranges that map to $default, "0", behave somewhat differently. For
1332 these, each code point maps to itself. So, in the first line in
1333 the example, "ord(uc(chr(0)))" is 0, "ord(uc(chr(1)))" is 1, ..
1334 "ord(uc(chr(96)))" is 96.
1335 "al"
1337 means that some of the map array elements have the form given by
1338 "a", and the rest are ordered lists of code points. For example,
1339 in:
1340 my ($uppers_ranges_ref, $uppers_maps_ref, $format, $default)
1342 = prop_invmap("Uppercase_Mapping");
1343 the returned arrays look like this:
1345 @$uppers_ranges_ref @$uppers_maps_ref
1347 0 0
1348 97 65
1349 123 0
1350 181 924
1351 182 0
1352 ...
1353 0x0149 [ 0x02BC 0x004E ]
1354 0x014A 0
1355 0x014B 330
1356 ...
1357 This is the full Uppercase_Mapping property (as opposed to the
1359 Simple_Uppercase_Mapping given in the example for format "a"). The
1360 only difference between the two in the ranges shown is that the
1361 code point at 0x0149 (LATIN SMALL LETTER N PRECEDED BY APOSTROPHE)
1362 maps to a string of two characters, 0x02BC (MODIFIER LETTER
1363 APOSTROPHE) followed by 0x004E (LATIN CAPITAL LETTER N).
1364 No adjustments are needed to entries that are references to arrays;
1366 each such entry will have exactly one element in its range, so the
1367 offset is always 0.
1368 The fourth (index [3]) element ($default) in the list returned for
1370 this format is 0.
1371 "ae"
1373 This is like "a", but some elements are the empty string, and
1374 should not be adjusted. The one internal Perl property accessible
1375 by "prop_invmap" is of this type: "Perl_Decimal_Digit" returns an
1376 inversion map which gives the numeric values that are represented
1377 by the Unicode decimal digit characters. Characters that don't
1378 represent decimal digits map to the empty string, like so:
1379 @digits @values
1381 0x0000 ""
1382 0x0030 0
1383 0x003A: ""
1384 0x0660: 0
1385 0x066A: ""
1386 0x06F0: 0
1387 0x06FA: ""
1388 0x07C0: 0
1389 0x07CA: ""
1390 0x0966: 0
1391 ...
1392 This means that the code points from 0 to 0x2F do not represent
1394 decimal digits; the code point 0x30 (DIGIT ZERO) represents 0;
1395 code point 0x31, (DIGIT ONE), represents 0+1-0 = 1; ... code point
1396 0x39, (DIGIT NINE), represents 0+9-0 = 9; ... code points 0x3A
1397 through 0x65F do not represent decimal digits; 0x660 (ARABIC-INDIC
1398 DIGIT ZERO), represents 0; ... 0x07C1 (NKO DIGIT ONE), represents
1399 0+1-0 = 1 ...
1400 The fourth (index [3]) element ($default) in the list returned for
1402 this format is the empty string.
1403 "ale"
1405 is a combination of the "al" type and the "ae" type. Some of the
1406 map array elements have the forms given by "al", and the rest are
1407 the empty string. The property "NFKC_Casefold" has this form. An
1408 example slice is:
1409 @$ranges_ref @$maps_ref Note
1411 ...
1412 0x00AA 97 FEMININE ORDINAL INDICATOR => 'a'
1413 0x00AB 0
1414 0x00AD SOFT HYPHEN => ""
1415 0x00AE 0
1416 0x00AF [ 0x0020, 0x0304 ] MACRON => SPACE . COMBINING MACRON
1417 0x00B0 0
1418 ...
1419 The fourth (index [3]) element ($default) in the list returned for
1421 this format is 0.
1422 "ar"
1424 means that all the elements of the map array are either rational
1425 numbers or the string "NaN", meaning "Not a Number". A rational
1426 number is either an integer, or two integers separated by a solidus
1427 ("/"). The second integer represents the denominator of the
1428 division implied by the solidus, and is actually always positive,
1429 so it is guaranteed not to be 0 and to not be signed. When the
1430 element is a plain integer (without the solidus), it may need to be
1431 adjusted to get the correct value by adding the offset, just as
1432 other "a" properties. No adjustment is needed for fractions, as
1433 the range is guaranteed to have just a single element, and so the
1434 offset is always 0.
1435 If you want to convert the returned map to entirely scalar numbers,
1437 you can use something like this:
1438 my ($invlist_ref, $invmap_ref, $format) = prop_invmap($property);
1440 if ($format && $format eq "ar") {
1441 map { $_ = eval $_ if $_ ne 'NaN' } @$map_ref;
1442 }
1443 Here's some entries from the output of the property "Nv", which has
1445 format "ar".
1446 @numerics_ranges @numerics_maps Note
1448 0x00 "NaN"
1449 0x30 0 DIGIT 0 .. DIGIT 9
1450 0x3A "NaN"
1451 0xB2 2 SUPERSCRIPTs 2 and 3
1452 0xB4 "NaN"
1453 0xB9 1 SUPERSCRIPT 1
1454 0xBA "NaN"
1455 0xBC 1/4 VULGAR FRACTION 1/4
1456 0xBD 1/2 VULGAR FRACTION 1/2
1457 0xBE 3/4 VULGAR FRACTION 3/4
1458 0xBF "NaN"
1459 0x660 0 ARABIC-INDIC DIGIT ZERO .. NINE
1460 0x66A "NaN"
1461 The fourth (index [3]) element ($default) in the list returned for
1463 this format is "NaN".
1464 "n" means the Name property. All the elements of the map array are
1466 simple scalars, but some of them contain special strings that
1467 require more work to get the actual name.
1468 Entries such as:
1470 CJK UNIFIED IDEOGRAPH-<code point>
1472 mean that the name for the code point is "CJK UNIFIED IDEOGRAPH-"
1474 with the code point (expressed in hexadecimal) appended to it, like
1475 "CJK UNIFIED IDEOGRAPH-3403" (similarly for
1476 "CJK COMPATIBILITY IDEOGRAPH-<code point>").
1477 Also, entries like
1479 <hangul syllable>
1481 means that the name is algorithmically calculated. This is easily
1483 done by the function "charnames::viacode(code)" in charnames.
1484 Note that for control characters ("Gc=cc"), Unicode's data files
1486 have the string ""<control>"", but the real name of each of these
1487 characters is the empty string. This function returns that real
1488 name, the empty string. (There are names for these characters, but
1489 they are considered aliases, not the Name property name, and are
1490 contained in the "Name_Alias" property.)
1491 "ad"
1493 means the Decomposition_Mapping property. This property is like
1494 "al" properties, except that one of the scalar elements is of the
1495 form:
1496 <hangul syllable>
1498 This signifies that this entry should be replaced by the
1500 decompositions for all the code points whose decomposition is
1501 algorithmically calculated. (All of them are currently in one
1502 range and no others outside the range are likely to ever be added
1503 to Unicode; the "n" format has this same entry.) These can be
1504 generated via the function Unicode::Normalize::NFD().
1505 Note that the mapping is the one that is specified in the Unicode
1507 data files, and to get the final decomposition, it may need to be
1508 applied recursively. Unicode in fact discourages use of this
1509 property except internally in implementations of the Unicode
1510 Normalization Algorithm.
1511 The fourth (index [3]) element ($default) in the list returned for
1513 this format is 0.
1514 Note that a format begins with the letter "a" if and only the property
1516 it is for requires adjustments by adding the offsets in multi-element
1517 ranges. For all these properties, an entry should be adjusted only if
1518 the map is a scalar which is an integer. That is, it must match the
1519 regular expression:
1520 / ^ -? \d+ $ /xa
1522 Further, the first element in a range never needs adjustment, as the
1524 adjustment would be just adding 0.
1525 A binary search such as that provided by "search_invlist()", can be
1527 used to quickly find a code point in the inversion list, and hence its
1528 corresponding mapping.
1529 The final, fourth element (index [3], assigned to $default in the
1531 "block" example) in the four element list returned by this function is
1532 used with the "a" format types; it may also be useful for applications
1533 that wish to convert the returned inversion map data structure into
1534 some other, such as a hash. It gives the mapping that most code points
1535 map to under the property. If you establish the convention that any
1536 code point not explicitly listed in your data structure maps to this
1537 value, you can potentially make your data structure much smaller. As
1538 you construct your data structure from the one returned by this
1539 function, simply ignore those ranges that map to this value. For
1540 example, to convert to the data structure searchable by
1541 "charinrange()", you can follow this recipe for properties that don't
1542 require adjustments:
1543 my ($list_ref, $map_ref, $format, $default) = prop_invmap($property);
1545 my @range_list;
1546 # Look at each element in the list, but the -2 is needed because we
1548 # look at $i+1 in the loop, and the final element is guaranteed to map
1549 # to $default by prop_invmap(), so we would skip it anyway.
1550 for my $i (0 .. @$list_ref - 2) {
1551 next if $map_ref->[$i] eq $default;
1552 push @range_list, [ $list_ref->[$i],
1553 $list_ref->[$i+1],
1554 $map_ref->[$i]
1555 ];
1556 }
1557 print charinrange(\@range_list, $code_point), "\n";
1559 With this, charinrange() will return "undef" if its input code point
1561 maps to $default. You can avoid this by omitting the "next" statement,
1562 and adding a line after the loop to handle the final element of the
1563 inversion map.
1564 Similarly, this recipe can be used for properties that do require
1566 adjustments:
1567 for my $i (0 .. @$list_ref - 2) {
1569 next if $map_ref->[$i] eq $default;
1570 # prop_invmap() guarantees that if the mapping is to an array, the
1572 # range has just one element, so no need to worry about adjustments.
1573 if (ref $map_ref->[$i]) {
1574 push @range_list,
1575 [ $list_ref->[$i], $list_ref->[$i], $map_ref->[$i] ];
1576 }
1577 else { # Otherwise each element is actually mapped to a separate
1578 # value, so the range has to be split into single code point
1579 # ranges.
1580 my $adjustment = 0;
1582 # For each code point that gets mapped to something...
1584 for my $j ($list_ref->[$i] .. $list_ref->[$i+1] -1 ) {
1585 # ... add a range consisting of just it mapping to the
1587 # original plus the adjustment, which is incremented for the
1588 # next time through the loop, as the offset increases by 1
1589 # for each element in the range
1590 push @range_list,
1591 [ $j, $j, $map_ref->[$i] + $adjustment++ ];
1592 }
1593 }
1594 }
1595 Note that the inversion maps returned for the "Case_Folding" and
1597 "Simple_Case_Folding" properties do not include the Turkic-locale
1598 mappings. Use "casefold()" for these.
1599 "prop_invmap" does not know about any user-defined properties, and will
1601 return "undef" if called with one of those.
1602 The returned values for the Perl extension properties, such as "Any"
1604 and "Greek" are somewhat misleading. The values are either "Y" or
1605 ""N"". All Unicode properties are bipartite, so you can actually use
1606 the "Y" or ""N"" in a Perl regular expression for these, like
1607 "qr/\p{ID_Start=Y/}" or "qr/\p{Upper=N/}". But the Perl extensions
1608 aren't specified this way, only like "/qr/\p{Any}", etc. You can't
1609 actually use the "Y" and ""N"" in them.
1610 Getting every available name
1612 Instead of reading the Unicode Database directly from files, as you
1614 were able to do for a long time, you are encouraged to use the supplied
1615 functions. So, instead of reading "Name.pl" directly, which changed
1616 formats in 5.32, and may do so again without notice in the future or
1617 even disappear, you ought to use "prop_invmap()" like this:
1618 my (%name, %cp, %cps, $n);
1620 # All codepoints
1621 foreach my $cat (qw( Name Name_Alias )) {
1622 my ($codepoints, $names, $format, $default) = prop_invmap($cat);
1623 # $format => "n", $default => ""
1624 foreach my $i (0 .. @$codepoints - 2) {
1625 my ($cp, $n) = ($codepoints->[$i], $names->[$i]);
1626 # If $n is a ref, the same codepoint has multiple names
1627 foreach my $name (ref $n ? @$n : $n) {
1628 $name{$cp} //= $name;
1629 $cp{$name} //= $cp;
1630 }
1631 }
1632 }
1633 # Named sequences
1634 { my %ns = namedseq();
1635 foreach my $name (sort { $ns{$a} cmp $ns{$b} } keys %ns) {
1636 $cp{$name} //= [ map { ord } split "" => $ns{$name} ];
1637 }
1638 }
1639 search_invlist()
1641 use Unicode::UCD qw(prop_invmap prop_invlist);
1642 use Unicode::UCD 'search_invlist';
1643 my @invlist = prop_invlist($property_name);
1645 print $code_point, ((search_invlist(\@invlist, $code_point) // -1) % 2)
1646 ? " isn't"
1647 : " is",
1648 " in $property_name\n";
1649 my ($blocks_ranges_ref, $blocks_map_ref) = prop_invmap("Block");
1651 my $index = search_invlist($blocks_ranges_ref, $code_point);
1652 print "$code_point is in block ", $blocks_map_ref->[$index], "\n";
1653 "search_invlist" is used to search an inversion list returned by
1655 "prop_invlist" or "prop_invmap" for a particular "code point argument".
1656 "undef" is returned if the code point is not found in the inversion
1657 list (this happens only when it is not a legal "code point argument",
1658 or is less than the list's first element). A warning is raised in the
1659 first instance.
1660 Otherwise, it returns the index into the list of the range that
1662 contains the code point.; that is, find "i" such that
1663 list[i]<= code_point < list[i+1].
1665 As explained in "prop_invlist()", whether a code point is in the list
1667 or not depends on if the index is even (in) or odd (not in). And as
1668 explained in "prop_invmap()", the index is used with the returned
1669 parallel array to find the mapping.
1670 Unicode::UCD::UnicodeVersion
1672 This returns the version of the Unicode Character Database, in other
1673 words, the version of the Unicode standard the database implements.
1674 The version is a string of numbers delimited by dots ('.').
1675 Blocks versus Scripts
1677 The difference between a block and a script is that scripts are closer
1678 to the linguistic notion of a set of code points required to represent
1679 languages, while block is more of an artifact of the Unicode code point
1680 numbering and separation into blocks of consecutive code points (so far
1681 the size of a block is some multiple of 16, like 128 or 256).
1682 For example the Latin script is spread over several blocks, such as
1684 "Basic Latin", "Latin 1 Supplement", "Latin Extended-A", and "Latin
1685 Extended-B". On the other hand, the Latin script does not contain all
1686 the characters of the "Basic Latin" block (also known as ASCII): it
1687 includes only the letters, and not, for example, the digits nor the
1688 punctuation.
1689 For blocks see <http://www.unicode.org/Public/UNIDATA/Blocks.txt>
1691 For scripts see UTR #24: <http://www.unicode.org/reports/tr24/>
1693 Matching Scripts and Blocks
1695 Scripts are matched with the regular-expression construct "\p{...}"
1696 (e.g. "\p{Tibetan}" matches characters of the Tibetan script), while
1697 "\p{Blk=...}" is used for blocks (e.g. "\p{Blk=Tibetan}" matches any of
1698 the 256 code points in the Tibetan block).
1699 Old-style versus new-style block names
1701 Unicode publishes the names of blocks in two different styles, though
1702 the two are equivalent under Unicode's loose matching rules.
1703 The original style uses blanks and hyphens in the block names (except
1705 for "No_Block"), like so:
1706 Miscellaneous Mathematical Symbols-B
1708 The newer style replaces these with underscores, like this:
1710 Miscellaneous_Mathematical_Symbols_B
1712 This newer style is consistent with the values of other Unicode
1714 properties. To preserve backward compatibility, all the functions in
1715 Unicode::UCD that return block names (except as noted) return the old-
1716 style ones. "prop_value_aliases()" returns the new-style and can be
1717 used to convert from old-style to new-style:
1718 my $new_style = prop_values_aliases("block", $old_style);
1720 Perl also has single-form extensions that refer to blocks,
1722 "In_Cyrillic", meaning "Block=Cyrillic". These have always been
1723 written in the new style.
1724 To convert from new-style to old-style, follow this recipe:
1726 $old_style = charblock((prop_invlist("block=$new_style"))[0]);
1728 (which finds the range of code points in the block using
1730 "prop_invlist", gets the lower end of the range (0th element) and then
1731 looks up the old name for its block using "charblock").
1732 Note that starting in Unicode 6.1, many of the block names have shorter
1734 synonyms. These are always given in the new style.
1735 Use with older Unicode versions
1737 The functions in this module work as well as can be expected when used
1738 on earlier Unicode versions. But, obviously, they use the available
1739 data from that Unicode version. For example, if the Unicode version
1740 predates the definition of the script property (Unicode 3.1), then any
1741 function that deals with scripts is going to return "undef" for the
1742 script portion of the return value.
1743
1745 Jarkko Hietaniemi. Now maintained by perl5 porters.
1746perl v5.38.2 2023-11-30 Unicode::UCD(3pm)