1PERLUNICODE(1) Perl Programmers Reference Guide PERLUNICODE(1)
2
6 perlunicode - Unicode support in Perl
7
9 If you haven't already, before reading this document, you should become
10 familiar with both perlunitut and perluniintro.
11 Unicode aims to UNI-fy the en-CODE-ings of all the world's character
13 sets into a single Standard. For quite a few of the various coding
14 standards that existed when Unicode was first created, converting from
15 each to Unicode essentially meant adding a constant to each code point
16 in the original standard, and converting back meant just subtracting
17 that same constant. For ASCII and ISO-8859-1, the constant is 0. For
18 ISO-8859-5, (Cyrillic) the constant is 864; for Hebrew (ISO-8859-8),
19 it's 1488; Thai (ISO-8859-11), 3424; and so forth. This made it easy
20 to do the conversions, and facilitated the adoption of Unicode.
21 And it worked; nowadays, those legacy standards are rarely used. Most
23 everyone uses Unicode.
24 Unicode is a comprehensive standard. It specifies many things outside
26 the scope of Perl, such as how to display sequences of characters. For
27 a full discussion of all aspects of Unicode, see
28 <http://www.unicode.org>.
29 Important Caveats
31 Even though some of this section may not be understandable to you on
32 first reading, we think it's important enough to highlight some of the
33 gotchas before delving further, so here goes:
34 Unicode support is an extensive requirement. While Perl does not
36 implement the Unicode standard or the accompanying technical reports
37 from cover to cover, Perl does support many Unicode features.
38 Also, the use of Unicode may present security issues that aren't
40 obvious, see "Security Implications of Unicode" below.
41 Safest if you "use feature 'unicode_strings'"
43 In order to preserve backward compatibility, Perl does not turn on
44 full internal Unicode support unless the pragma
45 "use feature 'unicode_strings'" is specified. (This is
46 automatically selected if you "use 5.012" or higher.) Failure to
47 do this can trigger unexpected surprises. See "The "Unicode Bug""
48 below.
49 This pragma doesn't affect I/O. Nor does it change the internal
51 representation of strings, only their interpretation. There are
52 still several places where Unicode isn't fully supported, such as
53 in filenames.
54 Input and Output Layers
56 Use the ":encoding(...)" layer to read from and write to
57 filehandles using the specified encoding. (See open.)
58 You must convert your non-ASCII, non-UTF-8 Perl scripts to be UTF-8.
60 The encoding module has been deprecated since perl 5.18 and the
61 perl internals it requires have been removed with perl 5.26.
62 "use utf8" still needed to enable UTF-8 in scripts
64 If your Perl script is itself encoded in UTF-8, the "use utf8"
65 pragma must be explicitly included to enable recognition of that
66 (in string or regular expression literals, or in identifier names).
67 This is the only time when an explicit "use utf8" is needed. (See
68 utf8).
69 If a Perl script begins with the bytes that form the UTF-8 encoding
71 of the Unicode BYTE ORDER MARK ("BOM", see "Unicode Encodings"),
72 those bytes are completely ignored.
73 UTF-16 scripts autodetected
75 If a Perl script begins with the Unicode "BOM" (UTF-16LE,
76 UTF16-BE), or if the script looks like non-"BOM"-marked UTF-16 of
77 either endianness, Perl will correctly read in the script as the
78 appropriate Unicode encoding.
79 Byte and Character Semantics
81 Before Unicode, most encodings used 8 bits (a single byte) to encode
82 each character. Thus a character was a byte, and a byte was a
83 character, and there could be only 256 or fewer possible characters.
84 "Byte Semantics" in the title of this section refers to this behavior.
85 There was no need to distinguish between "Byte" and "Character".
86 Then along comes Unicode which has room for over a million characters
88 (and Perl allows for even more). This means that a character may
89 require more than a single byte to represent it, and so the two terms
90 are no longer equivalent. What matter are the characters as whole
91 entities, and not usually the bytes that comprise them. That's what
92 the term "Character Semantics" in the title of this section refers to.
93 Perl had to change internally to decouple "bytes" from "characters".
95 It is important that you too change your ideas, if you haven't already,
96 so that "byte" and "character" no longer mean the same thing in your
97 mind.
98 The basic building block of Perl strings has always been a "character".
100 The changes basically come down to that the implementation no longer
101 thinks that a character is always just a single byte.
102 There are various things to note:
104 · String handling functions, for the most part, continue to operate
106 in terms of characters. "length()", for example, returns the
107 number of characters in a string, just as before. But that number
108 no longer is necessarily the same as the number of bytes in the
109 string (there may be more bytes than characters). The other such
110 functions include "chop()", "chomp()", "substr()", "pos()",
111 "index()", "rindex()", "sort()", "sprintf()", and "write()".
112 The exceptions are:
114 · the bit-oriented "vec"
116 · the byte-oriented "pack"/"unpack" "C" format
120 However, the "W" specifier does operate on whole characters, as
122 does the "U" specifier.
123 · some operators that interact with the platform's operating
125 system
126 Operators dealing with filenames are examples.
128 · when the functions are called from within the scope of the
130 "use bytes" pragma
131 Likely, you should use this only for debugging anyway.
133 · Strings--including hash keys--and regular expression patterns may
135 contain characters that have ordinal values larger than 255.
136 If you use a Unicode editor to edit your program, Unicode
138 characters may occur directly within the literal strings in UTF-8
139 encoding, or UTF-16. (The former requires a "use utf8", the latter
140 may require a "BOM".)
141 "Creating Unicode" in perluniintro gives other ways to place non-
143 ASCII characters in your strings.
144 · The "chr()" and "ord()" functions work on whole characters.
146 · Regular expressions match whole characters. For example, "."
148 matches a whole character instead of only a single byte.
149 · The "tr///" operator translates whole characters. (Note that the
151 "tr///CU" functionality has been removed. For similar
152 functionality to that, see "pack('U0', ...)" and "pack('C0',
153 ...)").
154 · "scalar reverse()" reverses by character rather than by byte.
156 · The bit string operators, "& | ^ ~" and (starting in v5.22) "&. |.
158 ^. ~." can operate on bit strings encoded in UTF-8, but this can
159 give unexpected results if any of the strings contain code points
160 above 0xFF. Starting in v5.28, it is a fatal error to have such an
161 operand. Otherwise, the operation is performed on a non-UTF-8 copy
162 of the operand. If you're not sure about the encoding of a string,
163 downgrade it before using any of these operators; you can use
164 "utf8::utf8_downgrade()".
165 The bottom line is that Perl has always practiced "Character
167 Semantics", but with the advent of Unicode, that is now different than
168 "Byte Semantics".
169 ASCII Rules versus Unicode Rules
171 Before Unicode, when a character was a byte was a character, Perl knew
172 only about the 128 characters defined by ASCII, code points 0 through
173 127 (except for under "use locale"). That left the code points 128 to
174 255 as unassigned, and available for whatever use a program might want.
175 The only semantics they have is their ordinal numbers, and that they
176 are members of none of the non-negative character classes. None are
177 considered to match "\w" for example, but all match "\W".
178 Unicode, of course, assigns each of those code points a particular
180 meaning (along with ones above 255). To preserve backward
181 compatibility, Perl only uses the Unicode meanings when there is some
182 indication that Unicode is what is intended; otherwise the non-ASCII
183 code points remain treated as if they are unassigned.
184 Here are the ways that Perl knows that a string should be treated as
186 Unicode:
187 · Within the scope of "use utf8"
189 If the whole program is Unicode (signified by using 8-bit Unicode
191 Transformation Format), then all literal strings within it must be
192 Unicode.
193 · Within the scope of "use feature 'unicode_strings'"
195 This pragma was created so you can explicitly tell Perl that
197 operations executed within its scope are to use Unicode rules.
198 More operations are affected with newer perls. See "The "Unicode
199 Bug"".
200 · Within the scope of "use 5.012" or higher
202 This implicitly turns on "use feature 'unicode_strings'".
204 · Within the scope of "use locale 'not_characters'", or "use locale"
206 and the current locale is a UTF-8 locale.
207 The former is defined to imply Unicode handling; and the latter
209 indicates a Unicode locale, hence a Unicode interpretation of all
210 strings within it.
211 · When the string contains a Unicode-only code point
213 Perl has never accepted code points above 255 without them being
215 Unicode, so their use implies Unicode for the whole string.
216 · When the string contains a Unicode named code point "\N{...}"
218 The "\N{...}" construct explicitly refers to a Unicode code point,
220 even if it is one that is also in ASCII. Therefore the string
221 containing it must be Unicode.
222 · When the string has come from an external source marked as Unicode
224 The "-C" command line option can specify that certain inputs to the
226 program are Unicode, and the values of this can be read by your
227 Perl code, see "${^UNICODE}" in perlvar.
228 · When the string has been upgraded to UTF-8
230 The function "utf8::utf8_upgrade()" can be explicitly used to
232 permanently (unless a subsequent "utf8::utf8_downgrade()" is
233 called) cause a string to be treated as Unicode.
234 · There are additional methods for regular expression patterns
236 A pattern that is compiled with the "/u" or "/a" modifiers is
238 treated as Unicode (though there are some restrictions with "/a").
239 Under the "/d" and "/l" modifiers, there are several other
240 indications for Unicode; see "Character set modifiers" in perlre.
241 Note that all of the above are overridden within the scope of "use
243 bytes"; but you should be using this pragma only for debugging.
244 Note also that some interactions with the platform's operating system
246 never use Unicode rules.
247 When Unicode rules are in effect:
249 · Case translation operators use the Unicode case translation tables.
251 Note that "uc()", or "\U" in interpolated strings, translates to
253 uppercase, while "ucfirst", or "\u" in interpolated strings,
254 translates to titlecase in languages that make the distinction
255 (which is equivalent to uppercase in languages without the
256 distinction).
257 There is a CPAN module, "Unicode::Casing", which allows you to
259 define your own mappings to be used in "lc()", "lcfirst()", "uc()",
260 "ucfirst()", and "fc" (or their double-quoted string inlined
261 versions such as "\U"). (Prior to Perl 5.16, this functionality
262 was partially provided in the Perl core, but suffered from a number
263 of insurmountable drawbacks, so the CPAN module was written
264 instead.)
265 · Character classes in regular expressions match based on the
267 character properties specified in the Unicode properties database.
268 "\w" can be used to match a Japanese ideograph, for instance; and
270 "[[:digit:]]" a Bengali number.
271 · Named Unicode properties, scripts, and block ranges may be used
273 (like bracketed character classes) by using the "\p{}" "matches
274 property" construct and the "\P{}" negation, "doesn't match
275 property".
276 See "Unicode Character Properties" for more details.
278 You can define your own character properties and use them in the
280 regular expression with the "\p{}" or "\P{}" construct. See "User-
281 Defined Character Properties" for more details.
282 Extended Grapheme Clusters (Logical characters)
284 Consider a character, say "H". It could appear with various marks
285 around it, such as an acute accent, or a circumflex, or various hooks,
286 circles, arrows, etc., above, below, to one side or the other, etc.
287 There are many possibilities among the world's languages. The number
288 of combinations is astronomical, and if there were a character for each
289 combination, it would soon exhaust Unicode's more than a million
290 possible characters. So Unicode took a different approach: there is a
291 character for the base "H", and a character for each of the possible
292 marks, and these can be variously combined to get a final logical
293 character. So a logical character--what appears to be a single
294 character--can be a sequence of more than one individual characters.
295 The Unicode standard calls these "extended grapheme clusters" (which is
296 an improved version of the no-longer much used "grapheme cluster");
297 Perl furnishes the "\X" regular expression construct to match such
298 sequences in their entirety.
299 But Unicode's intent is to unify the existing character set standards
301 and practices, and several pre-existing standards have single
302 characters that mean the same thing as some of these combinations, like
303 ISO-8859-1, which has quite a few of them. For example, "LATIN CAPITAL
304 LETTER E WITH ACUTE" was already in this standard when Unicode came
305 along. Unicode therefore added it to its repertoire as that single
306 character. But this character is considered by Unicode to be
307 equivalent to the sequence consisting of the character "LATIN CAPITAL
308 LETTER E" followed by the character "COMBINING ACUTE ACCENT".
309 "LATIN CAPITAL LETTER E WITH ACUTE" is called a "pre-composed"
311 character, and its equivalence with the "E" and the "COMBINING ACCENT"
312 sequence is called canonical equivalence. All pre-composed characters
313 are said to have a decomposition (into the equivalent sequence), and
314 the decomposition type is also called canonical. A string may be
315 comprised as much as possible of precomposed characters, or it may be
316 comprised of entirely decomposed characters. Unicode calls these
317 respectively, "Normalization Form Composed" (NFC) and "Normalization
318 Form Decomposed". The "Unicode::Normalize" module contains functions
319 that convert between the two. A string may also have both composed
320 characters and decomposed characters; this module can be used to make
321 it all one or the other.
322 You may be presented with strings in any of these equivalent forms.
324 There is currently nothing in Perl 5 that ignores the differences. So
325 you'll have to specially handle it. The usual advice is to convert
326 your inputs to "NFD" before processing further.
327 For more detailed information, see <http://unicode.org/reports/tr15/>.
329 Unicode Character Properties
331 (The only time that Perl considers a sequence of individual code points
332 as a single logical character is in the "\X" construct, already
333 mentioned above. Therefore "character" in this discussion means a
334 single Unicode code point.)
335 Very nearly all Unicode character properties are accessible through
337 regular expressions by using the "\p{}" "matches property" construct
338 and the "\P{}" "doesn't match property" for its negation.
339 For instance, "\p{Uppercase}" matches any single character with the
341 Unicode "Uppercase" property, while "\p{L}" matches any character with
342 a "General_Category" of "L" (letter) property (see "General_Category"
343 below). Brackets are not required for single letter property names, so
344 "\p{L}" is equivalent to "\pL".
345 More formally, "\p{Uppercase}" matches any single character whose
347 Unicode "Uppercase" property value is "True", and "\P{Uppercase}"
348 matches any character whose "Uppercase" property value is "False", and
349 they could have been written as "\p{Uppercase=True}" and
350 "\p{Uppercase=False}", respectively.
351 This formality is needed when properties are not binary; that is, if
353 they can take on more values than just "True" and "False". For
354 example, the "Bidi_Class" property (see "Bidirectional Character Types"
355 below), can take on several different values, such as "Left", "Right",
356 "Whitespace", and others. To match these, one needs to specify both
357 the property name ("Bidi_Class"), AND the value being matched against
358 ("Left", "Right", etc.). This is done, as in the examples above, by
359 having the two components separated by an equal sign (or
360 interchangeably, a colon), like "\p{Bidi_Class: Left}".
361 All Unicode-defined character properties may be written in these
363 compound forms of "\p{property=value}" or "\p{property:value}", but
364 Perl provides some additional properties that are written only in the
365 single form, as well as single-form short-cuts for all binary
366 properties and certain others described below, in which you may omit
367 the property name and the equals or colon separator.
368 Most Unicode character properties have at least two synonyms (or
370 aliases if you prefer): a short one that is easier to type and a longer
371 one that is more descriptive and hence easier to understand. Thus the
372 "L" and "Letter" properties above are equivalent and can be used
373 interchangeably. Likewise, "Upper" is a synonym for "Uppercase", and
374 we could have written "\p{Uppercase}" equivalently as "\p{Upper}".
375 Also, there are typically various synonyms for the values the property
376 can be. For binary properties, "True" has 3 synonyms: "T", "Yes", and
377 "Y"; and "False" has correspondingly "F", "No", and "N". But be
378 careful. A short form of a value for one property may not mean the
379 same thing as the same short form for another. Thus, for the
380 "General_Category" property, "L" means "Letter", but for the
381 "Bidi_Class" property, "L" means "Left". A complete list of properties
382 and synonyms is in perluniprops.
383 Upper/lower case differences in property names and values are
385 irrelevant; thus "\p{Upper}" means the same thing as "\p{upper}" or
386 even "\p{UpPeR}". Similarly, you can add or subtract underscores
387 anywhere in the middle of a word, so that these are also equivalent to
388 "\p{U_p_p_e_r}". And white space is irrelevant adjacent to non-word
389 characters, such as the braces and the equals or colon separators, so
390 "\p{ Upper }" and "\p{ Upper_case : Y }" are equivalent to these as
391 well. In fact, white space and even hyphens can usually be added or
392 deleted anywhere. So even "\p{ Up-per case = Yes}" is equivalent. All
393 this is called "loose-matching" by Unicode. The few places where
394 stricter matching is used is in the middle of numbers, and in the Perl
395 extension properties that begin or end with an underscore. Stricter
396 matching cares about white space (except adjacent to non-word
397 characters), hyphens, and non-interior underscores.
398 You can also use negation in both "\p{}" and "\P{}" by introducing a
400 caret ("^") between the first brace and the property name: "\p{^Tamil}"
401 is equal to "\P{Tamil}".
402 Almost all properties are immune to case-insensitive matching. That
404 is, adding a "/i" regular expression modifier does not change what they
405 match. There are two sets that are affected. The first set is
406 "Uppercase_Letter", "Lowercase_Letter", and "Titlecase_Letter", all of
407 which match "Cased_Letter" under "/i" matching. And the second set is
408 "Uppercase", "Lowercase", and "Titlecase", all of which match "Cased"
409 under "/i" matching. This set also includes its subsets "PosixUpper"
410 and "PosixLower" both of which under "/i" match "PosixAlpha". (The
411 difference between these sets is that some things, such as Roman
412 numerals, come in both upper and lower case so they are "Cased", but
413 aren't considered letters, so they aren't "Cased_Letter"'s.)
414 See "Beyond Unicode code points" for special considerations when
416 matching Unicode properties against non-Unicode code points.
417 General_Category
419 Every Unicode character is assigned a general category, which is the
421 "most usual categorization of a character" (from
422 <http://www.unicode.org/reports/tr44>).
423 The compound way of writing these is like "\p{General_Category=Number}"
425 (short: "\p{gc:n}"). But Perl furnishes shortcuts in which everything
426 up through the equal or colon separator is omitted. So you can instead
427 just write "\pN".
428 Here are the short and long forms of the values the "General Category"
430 property can have:
431 Short Long
433 L Letter
435 LC, L& Cased_Letter (that is: [\p{Ll}\p{Lu}\p{Lt}])
436 Lu Uppercase_Letter
437 Ll Lowercase_Letter
438 Lt Titlecase_Letter
439 Lm Modifier_Letter
440 Lo Other_Letter
441 M Mark
443 Mn Nonspacing_Mark
444 Mc Spacing_Mark
445 Me Enclosing_Mark
446 N Number
448 Nd Decimal_Number (also Digit)
449 Nl Letter_Number
450 No Other_Number
451 P Punctuation (also Punct)
453 Pc Connector_Punctuation
454 Pd Dash_Punctuation
455 Ps Open_Punctuation
456 Pe Close_Punctuation
457 Pi Initial_Punctuation
458 (may behave like Ps or Pe depending on usage)
459 Pf Final_Punctuation
460 (may behave like Ps or Pe depending on usage)
461 Po Other_Punctuation
462 S Symbol
464 Sm Math_Symbol
465 Sc Currency_Symbol
466 Sk Modifier_Symbol
467 So Other_Symbol
468 Z Separator
470 Zs Space_Separator
471 Zl Line_Separator
472 Zp Paragraph_Separator
473 C Other
475 Cc Control (also Cntrl)
476 Cf Format
477 Cs Surrogate
478 Co Private_Use
479 Cn Unassigned
480 Single-letter properties match all characters in any of the two-letter
482 sub-properties starting with the same letter. "LC" and "L&" are
483 special: both are aliases for the set consisting of everything matched
484 by "Ll", "Lu", and "Lt".
485 Bidirectional Character Types
487 Because scripts differ in their directionality (Hebrew and Arabic are
489 written right to left, for example) Unicode supplies a "Bidi_Class"
490 property. Some of the values this property can have are:
491 Value Meaning
493 L Left-to-Right
495 LRE Left-to-Right Embedding
496 LRO Left-to-Right Override
497 R Right-to-Left
498 AL Arabic Letter
499 RLE Right-to-Left Embedding
500 RLO Right-to-Left Override
501 PDF Pop Directional Format
502 EN European Number
503 ES European Separator
504 ET European Terminator
505 AN Arabic Number
506 CS Common Separator
507 NSM Non-Spacing Mark
508 BN Boundary Neutral
509 B Paragraph Separator
510 S Segment Separator
511 WS Whitespace
512 ON Other Neutrals
513 This property is always written in the compound form. For example,
515 "\p{Bidi_Class:R}" matches characters that are normally written right
516 to left. Unlike the "General_Category" property, this property can
517 have more values added in a future Unicode release. Those listed above
518 comprised the complete set for many Unicode releases, but others were
519 added in Unicode 6.3; you can always find what the current ones are in
520 perluniprops. And <http://www.unicode.org/reports/tr9/> describes how
521 to use them.
522 Scripts
524 The world's languages are written in many different scripts. This
526 sentence (unless you're reading it in translation) is written in Latin,
527 while Russian is written in Cyrillic, and Greek is written in, well,
528 Greek; Japanese mainly in Hiragana or Katakana. There are many more.
529 The Unicode "Script" and "Script_Extensions" properties give what
531 script a given character is in. The "Script_Extensions" property is an
532 improved version of "Script", as demonstrated below. Either property
533 can be specified with the compound form like "\p{Script=Hebrew}"
534 (short: "\p{sc=hebr}"), or "\p{Script_Extensions=Javanese}" (short:
535 "\p{scx=java}"). In addition, Perl furnishes shortcuts for all
536 "Script_Extensions" property names. You can omit everything up through
537 the equals (or colon), and simply write "\p{Latin}" or "\P{Cyrillic}".
538 (This is not true for "Script", which is required to be written in the
539 compound form. Prior to Perl v5.26, the single form returned the plain
540 old "Script" version, but was changed because "Script_Extensions" gives
541 better results.)
542 The difference between these two properties involves characters that
544 are used in multiple scripts. For example the digits '0' through '9'
545 are used in many parts of the world. These are placed in a script
546 named "Common". Other characters are used in just a few scripts. For
547 example, the "KATAKANA-HIRAGANA DOUBLE HYPHEN" is used in both Japanese
548 scripts, Katakana and Hiragana, but nowhere else. The "Script"
549 property places all characters that are used in multiple scripts in the
550 "Common" script, while the "Script_Extensions" property places those
551 that are used in only a few scripts into each of those scripts; while
552 still using "Common" for those used in many scripts. Thus both these
553 match:
554 "0" =~ /\p{sc=Common}/ # Matches
556 "0" =~ /\p{scx=Common}/ # Matches
557 and only the first of these match:
559 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{sc=Common} # Matches
561 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{scx=Common} # No match
562 And only the last two of these match:
564 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{sc=Hiragana} # No match
566 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{sc=Katakana} # No match
567 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{scx=Hiragana} # Matches
568 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{scx=Katakana} # Matches
569 "Script_Extensions" is thus an improved "Script", in which there are
571 fewer characters in the "Common" script, and correspondingly more in
572 other scripts. It is new in Unicode version 6.0, and its data are
573 likely to change significantly in later releases, as things get sorted
574 out. New code should probably be using "Script_Extensions" and not
575 plain "Script". If you compile perl with a Unicode release that
576 doesn't have "Script_Extensions", the single form Perl extensions will
577 instead refer to the plain "Script" property. If you compile with a
578 version of Unicode that doesn't have the "Script" property, these
579 extensions will not be defined at all.
580 (Actually, besides "Common", the "Inherited" script, contains
582 characters that are used in multiple scripts. These are modifier
583 characters which inherit the script value of the controlling character.
584 Some of these are used in many scripts, and so go into "Inherited" in
585 both "Script" and "Script_Extensions". Others are used in just a few
586 scripts, so are in "Inherited" in "Script", but not in
587 "Script_Extensions".)
588 It is worth stressing that there are several different sets of digits
590 in Unicode that are equivalent to 0-9 and are matchable by "\d" in a
591 regular expression. If they are used in a single language only, they
592 are in that language's "Script" and "Script_Extensions". If they are
593 used in more than one script, they will be in "sc=Common", but only if
594 they are used in many scripts should they be in "scx=Common".
595 The explanation above has omitted some detail; refer to UAX#24 "Unicode
597 Script Property": <http://www.unicode.org/reports/tr24>.
598 A complete list of scripts and their shortcuts is in perluniprops.
600 Use of the "Is" Prefix
602 For backward compatibility (with ancient Perl 5.6), all properties
604 writable without using the compound form mentioned so far may have "Is"
605 or "Is_" prepended to their name, so "\P{Is_Lu}", for example, is equal
606 to "\P{Lu}", and "\p{IsScript:Arabic}" is equal to "\p{Arabic}".
607 Blocks
609 In addition to scripts, Unicode also defines blocks of characters. The
611 difference between scripts and blocks is that the concept of scripts is
612 closer to natural languages, while the concept of blocks is more of an
613 artificial grouping based on groups of Unicode characters with
614 consecutive ordinal values. For example, the "Basic Latin" block is all
615 the characters whose ordinals are between 0 and 127, inclusive; in
616 other words, the ASCII characters. The "Latin" script contains some
617 letters from this as well as several other blocks, like "Latin-1
618 Supplement", "Latin Extended-A", etc., but it does not contain all the
619 characters from those blocks. It does not, for example, contain the
620 digits 0-9, because those digits are shared across many scripts, and
621 hence are in the "Common" script.
622 For more about scripts versus blocks, see UAX#24 "Unicode Script
624 Property": <http://www.unicode.org/reports/tr24>
625 The "Script_Extensions" or "Script" properties are likely to be the
627 ones you want to use when processing natural language; the "Block"
628 property may occasionally be useful in working with the nuts and bolts
629 of Unicode.
630 Block names are matched in the compound form, like "\p{Block: Arrows}"
632 or "\p{Blk=Hebrew}". Unlike most other properties, only a few block
633 names have a Unicode-defined short name.
634 Perl also defines single form synonyms for the block property in cases
636 where these do not conflict with something else. But don't use any of
637 these, because they are unstable. Since these are Perl extensions,
638 they are subordinate to official Unicode property names; Unicode
639 doesn't know nor care about Perl's extensions. It may happen that a
640 name that currently means the Perl extension will later be changed
641 without warning to mean a different Unicode property in a future
642 version of the perl interpreter that uses a later Unicode release, and
643 your code would no longer work. The extensions are mentioned here for
644 completeness: Take the block name and prefix it with one of: "In" (for
645 example "\p{Blk=Arrows}" can currently be written as "\p{In_Arrows}");
646 or sometimes "Is" (like "\p{Is_Arrows}"); or sometimes no prefix at all
647 ("\p{Arrows}"). As of this writing (Unicode 9.0) there are no
648 conflicts with using the "In_" prefix, but there are plenty with the
649 other two forms. For example, "\p{Is_Hebrew}" and "\p{Hebrew}" mean
650 "\p{Script_Extensions=Hebrew}" which is NOT the same thing as
651 "\p{Blk=Hebrew}". Our advice used to be to use the "In_" prefix as a
652 single form way of specifying a block. But Unicode 8.0 added
653 properties whose names begin with "In", and it's now clear that it's
654 only luck that's so far prevented a conflict. Using "In" is only
655 marginally less typing than "Blk:", and the latter's meaning is clearer
656 anyway, and guaranteed to never conflict. So don't take chances. Use
657 "\p{Blk=foo}" for new code. And be sure that block is what you really
658 really want to do. In most cases scripts are what you want instead.
659 A complete list of blocks is in perluniprops.
661 Other Properties
663 There are many more properties than the very basic ones described here.
665 A complete list is in perluniprops.
666 Unicode defines all its properties in the compound form, so all single-
668 form properties are Perl extensions. Most of these are just synonyms
669 for the Unicode ones, but some are genuine extensions, including
670 several that are in the compound form. And quite a few of these are
671 actually recommended by Unicode (in
672 <http://www.unicode.org/reports/tr18>).
673 This section gives some details on all extensions that aren't just
675 synonyms for compound-form Unicode properties (for those properties,
676 you'll have to refer to the Unicode Standard
677 <http://www.unicode.org/reports/tr44>.
678 "\p{All}"
680 This matches every possible code point. It is equivalent to
681 "qr/./s". Unlike all the other non-user-defined "\p{}" property
682 matches, no warning is ever generated if this is property is
683 matched against a non-Unicode code point (see "Beyond Unicode code
684 points" below).
685 "\p{Alnum}"
687 This matches any "\p{Alphabetic}" or "\p{Decimal_Number}"
688 character.
689 "\p{Any}"
691 This matches any of the 1_114_112 Unicode code points. It is a
692 synonym for "\p{Unicode}".
693 "\p{ASCII}"
695 This matches any of the 128 characters in the US-ASCII character
696 set, which is a subset of Unicode.
697 "\p{Assigned}"
699 This matches any assigned code point; that is, any code point whose
700 general category is not "Unassigned" (or equivalently, not "Cn").
701 "\p{Blank}"
703 This is the same as "\h" and "\p{HorizSpace}": A character that
704 changes the spacing horizontally.
705 "\p{Decomposition_Type: Non_Canonical}" (Short: "\p{Dt=NonCanon}")
707 Matches a character that has a non-canonical decomposition.
708 The "Extended Grapheme Clusters (Logical characters)" section above
710 talked about canonical decompositions. However, many more
711 characters have a different type of decomposition, a "compatible"
712 or "non-canonical" decomposition. The sequences that form these
713 decompositions are not considered canonically equivalent to the
714 pre-composed character. An example is the "SUPERSCRIPT ONE". It
715 is somewhat like a regular digit 1, but not exactly; its
716 decomposition into the digit 1 is called a "compatible"
717 decomposition, specifically a "super" decomposition. There are
718 several such compatibility decompositions (see
719 <http://www.unicode.org/reports/tr44>), including one called
720 "compat", which means some miscellaneous type of decomposition that
721 doesn't fit into the other decomposition categories that Unicode
722 has chosen.
723 Note that most Unicode characters don't have a decomposition, so
725 their decomposition type is "None".
726 For your convenience, Perl has added the "Non_Canonical"
728 decomposition type to mean any of the several compatibility
729 decompositions.
730 "\p{Graph}"
732 Matches any character that is graphic. Theoretically, this means a
733 character that on a printer would cause ink to be used.
734 "\p{HorizSpace}"
736 This is the same as "\h" and "\p{Blank}": a character that changes
737 the spacing horizontally.
738 "\p{In=*}"
740 This is a synonym for "\p{Present_In=*}"
741 "\p{PerlSpace}"
743 This is the same as "\s", restricted to ASCII, namely "[ \f\n\r\t]"
744 and starting in Perl v5.18, a vertical tab.
745 Mnemonic: Perl's (original) space
747 "\p{PerlWord}"
749 This is the same as "\w", restricted to ASCII, namely
750 "[A-Za-z0-9_]"
751 Mnemonic: Perl's (original) word.
753 "\p{Posix...}"
755 There are several of these, which are equivalents, using the "\p{}"
756 notation, for Posix classes and are described in "POSIX Character
757 Classes" in perlrecharclass.
758 "\p{Present_In: *}" (Short: "\p{In=*}")
760 This property is used when you need to know in what Unicode
761 version(s) a character is.
762 The "*" above stands for some Unicode version number, such as 1.1
764 or 12.0; or the "*" can also be "Unassigned". This property will
765 match the code points whose final disposition has been settled as
766 of the Unicode release given by the version number; "\p{Present_In:
767 Unassigned}" will match those code points whose meaning has yet to
768 be assigned.
769 For example, "U+0041" "LATIN CAPITAL LETTER A" was present in the
771 very first Unicode release available, which is 1.1, so this
772 property is true for all valid "*" versions. On the other hand,
773 "U+1EFF" was not assigned until version 5.1 when it became "LATIN
774 SMALL LETTER Y WITH LOOP", so the only "*" that would match it are
775 5.1, 5.2, and later.
776 Unicode furnishes the "Age" property from which this is derived.
778 The problem with Age is that a strict interpretation of it (which
779 Perl takes) has it matching the precise release a code point's
780 meaning is introduced in. Thus "U+0041" would match only 1.1; and
781 "U+1EFF" only 5.1. This is not usually what you want.
782 Some non-Perl implementations of the Age property may change its
784 meaning to be the same as the Perl "Present_In" property; just be
785 aware of that.
786 Another confusion with both these properties is that the definition
788 is not that the code point has been assigned, but that the meaning
789 of the code point has been determined. This is because 66 code
790 points will always be unassigned, and so the "Age" for them is the
791 Unicode version in which the decision to make them so was made.
792 For example, "U+FDD0" is to be permanently unassigned to a
793 character, and the decision to do that was made in version 3.1, so
794 "\p{Age=3.1}" matches this character, as also does "\p{Present_In:
795 3.1}" and up.
796 "\p{Print}"
798 This matches any character that is graphical or blank, except
799 controls.
800 "\p{SpacePerl}"
802 This is the same as "\s", including beyond ASCII.
803 Mnemonic: Space, as modified by Perl. (It doesn't include the
805 vertical tab until v5.18, which both the Posix standard and Unicode
806 consider white space.)
807 "\p{Title}" and "\p{Titlecase}"
809 Under case-sensitive matching, these both match the same code
810 points as "\p{General Category=Titlecase_Letter}" ("\p{gc=lt}").
811 The difference is that under "/i" caseless matching, these match
812 the same as "\p{Cased}", whereas "\p{gc=lt}" matches
813 "\p{Cased_Letter").
814 "\p{Unicode}"
816 This matches any of the 1_114_112 Unicode code points. "\p{Any}".
817 "\p{VertSpace}"
819 This is the same as "\v": A character that changes the spacing
820 vertically.
821 "\p{Word}"
823 This is the same as "\w", including over 100_000 characters beyond
824 ASCII.
825 "\p{XPosix...}"
827 There are several of these, which are the standard Posix classes
828 extended to the full Unicode range. They are described in "POSIX
829 Character Classes" in perlrecharclass.
830 Wildcards in Property Values
832 Starting in Perl 5.30, it is possible to do do something like this:
833 qr!\p{numeric_value=/\A[0-5]\z/}!
835 or, by abbreviating and adding "/x",
837 qr! \p{nv= /(?x) \A [0-5] \z / }!
839 This matches all code points whose numeric value is one of 0, 1, 2, 3,
841 4, or 5. This particular example could instead have been written as
842 qr! \A [ \p{nv=0}\p{nv=1}\p{nv=2}\p{nv=3}\p{nv=4}\p{nv=5} ] \z !xx
844 in earlier perls, so in this case this feature just makes things easier
846 and shorter to write. If we hadn't included the "\A" and "\z", these
847 would have matched things like "1/2" because that contains a 1 (as well
848 as a 2). As written, it matches things like subscripts that have these
849 numeric values. If we only wanted the decimal digits with those
850 numeric values, we could say,
851 qr! (?[ \d & \p{nv=/[0-5]/ ]) }!x
853 The "\d" gets rid of needing to anchor the pattern, since it forces the
855 result to only match "[0-9]", and the "[0-5]" further restricts it.
856 The text in the above examples enclosed between the "/" characters can
858 be just about any regular expression. It is independent of the main
859 pattern, so doesn't share any capturing groups, etc. The delimiters
860 for it must be ASCII punctuation, but it may NOT be delimited by "{",
861 nor "}" nor contain a literal "}", as that delimits the end of the
862 enclosing "\p{}". Like any pattern, certain other delimiters are
863 terminated by their mirror images. These are "(", ""["", and "<". If
864 the delimiter is any of "-", "_", "+", or "\", or is the same delimiter
865 as is used for the enclosing pattern, it must be be preceded by a
866 backslash escape, both fore and aft.
867 Beware of using "$" to indicate to match the end of the string. It can
869 too easily be interpreted as being a punctuation variable, like $/.
870 No modifiers may follow the final delimiter. Instead, use
872 "(?adlupimnsx-imnsx)" in perlre and/or "(?adluimnsx-imnsx:pattern)" in
873 perlre to specify modifiers.
874 This feature is not available when the left-hand side is prefixed by
876 "Is_", nor for any form that is marked as "Discouraged" in
877 "Discouraged" in perluniprops.
878 Perl wraps your pattern with "(?iaa: ... )". This is because nothing
880 outside ASCII can match the Unicode property values available in this
881 release, and they should match caselessly. If your pattern has a
882 syntax error, this wrapping will be shown in the error message, even
883 though you didn't specify it yourself. This could be confusing if you
884 don't know about this.
885 This experimental feature has been added to begin to implement
887 <https://www.unicode.org/reports/tr18/#Wildcard_Properties>. Using it
888 will raise a (default-on) warning in the
889 "experimental::uniprop_wildcards" category. We reserve the right to
890 change its operation as we gain experience.
891 Your subpattern can be just about anything, but for it to have some
893 utility, it should match when called with either or both of a) the full
894 name of the property value with underscores (and/or spaces in the Block
895 property) and some things uppercase; or b) the property value in all
896 lowercase with spaces and underscores squeezed out. For example,
897 qr!\p{Blk=/Old I.*/}!
899 qr!\p{Blk=/oldi.*/}!
900 would match the same things.
902 A warning is issued if none of the legal values for a property are
904 matched by your pattern. It's likely that a future release will raise
905 a warning if your pattern ends up causing every possible code point to
906 match.
907 Another example that shows that within "\p{...}", "/x" isn't needed to
909 have spaces:
910 qr!\p{scx= /Hebrew|Greek/ }!
912 To be safe, we should have anchored the above example, to prevent
914 matches for something like "Hebrew_Braile", but there aren't any script
915 names like that.
916 There are certain properties that it doesn't currently work with.
918 These are:
919 Bidi Mirroring Glyph
921 Bidi Paired Bracket
922 Case Folding
923 Decomposition Mapping
924 Equivalent Unified Ideograph
925 Name
926 Name Alias
927 Lowercase Mapping
928 NFKC Case Fold
929 Titlecase Mapping
930 Uppercase Mapping
931 Nor is the "@unicode_property@" form implemented.
933 Here's a complete example of matching IPV4 internet protocol addresses
935 in any (single) script
936 no warnings 'experimental::script_run';
938 no warnings 'experimental::regex_sets';
939 no warnings 'experimental::uniprop_wildcards';
940 # Can match a substring, so this intermediate regex needs to have
942 # context or anchoring in its final use. Using nt=de yields decimal
943 # digits. When specifying a subset of these, we must include \d to
944 # prevent things like U+00B2 SUPERSCRIPT TWO from matching
945 my $zero_through_255 =
946 qr/ \b (*sr: # All from same sript
947 (?[ \p{nv=0} & \d ])* # Optional leading zeros
948 ( # Then one of:
949 \d{1,2} # 0 - 99
950 | (?[ \p{nv=1} & \d ]) \d{2} # 100 - 199
951 | (?[ \p{nv=2} & \d ])
952 ( (?[ \p{nv=:[0-4]:} & \d ]) \d # 200 - 249
953 | (?[ \p{nv=5} & \d ])
954 (?[ \p{nv=:[0-5]:} & \d ]) # 250 - 255
955 )
956 )
957 )
958 \b
959 /x;
960 my $ipv4 = qr/ \A (*sr: $zero_through_255
962 (?: [.] $zero_through_255 ) {3}
963 )
964 \z
965 /x;
966 User-Defined Character Properties
968 You can define your own binary character properties by defining
969 subroutines whose names begin with "In" or "Is". (The experimental
970 feature "(?[ ])" in perlre provides an alternative which allows more
971 complex definitions.) The subroutines can be defined in any package.
972 The user-defined properties can be used in the regular expression
973 "\p{}" and "\P{}" constructs; if you are using a user-defined property
974 from a package other than the one you are in, you must specify its
975 package in the "\p{}" or "\P{}" construct.
976 # assuming property Is_Foreign defined in Lang::
978 package main; # property package name required
979 if ($txt =~ /\p{Lang::IsForeign}+/) { ... }
980 package Lang; # property package name not required
982 if ($txt =~ /\p{IsForeign}+/) { ... }
983 Note that the effect is compile-time and immutable once defined.
985 However, the subroutines are passed a single parameter, which is 0 if
986 case-sensitive matching is in effect and non-zero if caseless matching
987 is in effect. The subroutine may return different values depending on
988 the value of the flag, and one set of values will immutably be in
989 effect for all case-sensitive matches, and the other set for all case-
990 insensitive matches.
991 Note that if the regular expression is tainted, then Perl will die
993 rather than calling the subroutine when the name of the subroutine is
994 determined by the tainted data.
995 The subroutines must return a specially-formatted string, with one or
997 more newline-separated lines. Each line must be one of the following:
998 · A single hexadecimal number denoting a code point to include.
1000 · Two hexadecimal numbers separated by horizontal whitespace (space
1002 or tabular characters) denoting a range of code points to include.
1003 The second number must not be smaller than the first.
1004 · Something to include, prefixed by "+": a built-in character
1006 property (prefixed by "utf8::") or a fully qualified (including
1007 package name) user-defined character property, to represent all the
1008 characters in that property; two hexadecimal code points for a
1009 range; or a single hexadecimal code point.
1010 · Something to exclude, prefixed by "-": an existing character
1012 property (prefixed by "utf8::") or a fully qualified (including
1013 package name) user-defined character property, to represent all the
1014 characters in that property; two hexadecimal code points for a
1015 range; or a single hexadecimal code point.
1016 · Something to negate, prefixed "!": an existing character property
1018 (prefixed by "utf8::") or a fully qualified (including package
1019 name) user-defined character property, to represent all the
1020 characters in that property; two hexadecimal code points for a
1021 range; or a single hexadecimal code point.
1022 · Something to intersect with, prefixed by "&": an existing character
1024 property (prefixed by "utf8::") or a fully qualified (including
1025 package name) user-defined character property, for all the
1026 characters except the characters in the property; two hexadecimal
1027 code points for a range; or a single hexadecimal code point.
1028 For example, to define a property that covers both the Japanese
1030 syllabaries (hiragana and katakana), you can define
1031 sub InKana {
1033 return <<END;
1034 3040\t309F
1035 30A0\t30FF
1036 END
1037 }
1038 Imagine that the here-doc end marker is at the beginning of the line.
1040 Now you can use "\p{InKana}" and "\P{InKana}".
1041 You could also have used the existing block property names:
1043 sub InKana {
1045 return <<'END';
1046 +utf8::InHiragana
1047 +utf8::InKatakana
1048 END
1049 }
1050 Suppose you wanted to match only the allocated characters, not the raw
1052 block ranges: in other words, you want to remove the unassigned
1053 characters:
1054 sub InKana {
1056 return <<'END';
1057 +utf8::InHiragana
1058 +utf8::InKatakana
1059 -utf8::IsCn
1060 END
1061 }
1062 The negation is useful for defining (surprise!) negated classes.
1064 sub InNotKana {
1066 return <<'END';
1067 !utf8::InHiragana
1068 -utf8::InKatakana
1069 +utf8::IsCn
1070 END
1071 }
1072 This will match all non-Unicode code points, since every one of them is
1074 not in Kana. You can use intersection to exclude these, if desired, as
1075 this modified example shows:
1076 sub InNotKana {
1078 return <<'END';
1079 !utf8::InHiragana
1080 -utf8::InKatakana
1081 +utf8::IsCn
1082 &utf8::Any
1083 END
1084 }
1085 &utf8::Any must be the last line in the definition.
1087 Intersection is used generally for getting the common characters
1089 matched by two (or more) classes. It's important to remember not to
1090 use "&" for the first set; that would be intersecting with nothing,
1091 resulting in an empty set. (Similarly using "-" for the first set does
1092 nothing).
1093 Unlike non-user-defined "\p{}" property matches, no warning is ever
1095 generated if these properties are matched against a non-Unicode code
1096 point (see "Beyond Unicode code points" below).
1097 User-Defined Case Mappings (for serious hackers only)
1099 This feature has been removed as of Perl 5.16. The CPAN module
1100 "Unicode::Casing" provides better functionality without the drawbacks
1101 that this feature had. If you are using a Perl earlier than 5.16, this
1102 feature was most fully documented in the 5.14 version of this pod:
1103 <http://perldoc.perl.org/5.14.0/perlunicode.html#User-Defined-Case-Mappings-%28for-serious-hackers-only%29>
1104 Character Encodings for Input and Output
1106 See Encode.
1107 Unicode Regular Expression Support Level
1109 The following list of Unicode supported features for regular
1110 expressions describes all features currently directly supported by core
1111 Perl. The references to "Level N" and the section numbers refer to
1112 UTS#18 "Unicode Regular Expressions"
1113 <http://www.unicode.org/reports/tr18>, version 18, October 2016.
1114 Level 1 - Basic Unicode Support
1116 RL1.1 Hex Notation - Done [1]
1118 RL1.2 Properties - Done [2]
1119 RL1.2a Compatibility Properties - Done [3]
1120 RL1.3 Subtraction and Intersection - Experimental [4]
1121 RL1.4 Simple Word Boundaries - Done [5]
1122 RL1.5 Simple Loose Matches - Done [6]
1123 RL1.6 Line Boundaries - Partial [7]
1124 RL1.7 Supplementary Code Points - Done [8]
1125 [1] "\N{U+...}" and "\x{...}"
1127 [2] "\p{...}" "\P{...}". This requirement is for a minimal list of
1128 properties. Perl supports these and all other Unicode character
1129 properties, as R2.7 asks (see "Unicode Character Properties" above).
1130 [3] Perl has "\d" "\D" "\s" "\S" "\w" "\W" "\X" "[:prop:]" "[:^prop:]",
1131 plus all the properties specified by
1132 <http://www.unicode.org/reports/tr18/#Compatibility_Properties>. These
1133 are described above in "Other Properties"
1134 [4] The experimental feature "(?[...])" starting in v5.18 accomplishes
1135 this.
1136 See "(?[ ])" in perlre. If you don't want to use an experimental
1138 feature, you can use one of the following:
1139 · Regular expression lookahead
1141 You can mimic class subtraction using lookahead. For example,
1143 what UTS#18 might write as
1144 [{Block=Greek}-[{UNASSIGNED}]]
1146 in Perl can be written as:
1148 (?!\p{Unassigned})\p{Block=Greek}
1150 (?=\p{Assigned})\p{Block=Greek}
1151 But in this particular example, you probably really want
1153 \p{Greek}
1155 which will match assigned characters known to be part of the
1157 Greek script.
1158 · CPAN module "Unicode::Regex::Set"
1160 It does implement the full UTS#18 grouping, intersection,
1162 union, and removal (subtraction) syntax.
1163 · "User-Defined Character Properties"
1165 "+" for union, "-" for removal (set-difference), "&" for
1167 intersection
1168 [5] "\b" "\B" meet most, but not all, the details of this requirement,
1170 but "\b{wb}" and "\B{wb}" do, as well as the stricter R2.3.
1171 [6] Note that Perl does Full case-folding in matching, not Simple:
1172 For example "U+1F88" is equivalent to "U+1F00 U+03B9", instead of
1174 just "U+1F80". This difference matters mainly for certain Greek
1175 capital letters with certain modifiers: the Full case-folding
1176 decomposes the letter, while the Simple case-folding would map it
1177 to a single character.
1178 [7] The reason this is considered to be only partially implemented is
1180 that Perl has "qr/\b{lb}/" and "Unicode::LineBreak" that are
1181 conformant with UAX#14 "Unicode Line Breaking Algorithm"
1182 <http://www.unicode.org/reports/tr14>. The regular expression
1183 construct provides default behavior, while the heavier-weight
1184 module provides customizable line breaking.
1185 But Perl treats "\n" as the start- and end-line delimiter, whereas
1187 Unicode specifies more characters that should be so-interpreted.
1188 These are:
1190 VT U+000B (\v in C)
1192 FF U+000C (\f)
1193 CR U+000D (\r)
1194 NEL U+0085
1195 LS U+2028
1196 PS U+2029
1197 "^" and "$" in regular expression patterns are supposed to match
1199 all these, but don't. These characters also don't, but should,
1200 affect "<>" $., and script line numbers.
1201 Also, lines should not be split within "CRLF" (i.e. there is no
1203 empty line between "\r" and "\n"). For "CRLF", try the ":crlf"
1204 layer (see PerlIO).
1205 [8] UTF-8/UTF-EBDDIC used in Perl allows not only "U+10000" to
1207 "U+10FFFF" but also beyond "U+10FFFF"
1208 Level 2 - Extended Unicode Support
1210 RL2.1 Canonical Equivalents - Retracted [9]
1212 by Unicode
1213 RL2.2 Extended Grapheme Clusters - Partial [10]
1214 RL2.3 Default Word Boundaries - Done [11]
1215 RL2.4 Default Case Conversion - Done
1216 RL2.5 Name Properties - Done
1217 RL2.6 Wildcards in Property Values - Partial [12]
1218 RL2.7 Full Properties - Done
1219 [9] Unicode has rewritten this portion of UTS#18 to say that getting
1221 canonical equivalence (see UAX#15 "Unicode Normalization Forms"
1222 <http://www.unicode.org/reports/tr15>) is basically to be done at the
1223 programmer level. Use NFD to write both your regular expressions and
1224 text to match them against (you can use Unicode::Normalize).
1225 [10] Perl has "\X" and "\b{gcb}" but we don't have a "Grapheme Cluster
1226 Mode".
1227 [11] see UAX#29 "Unicode Text Segmentation"
1228 <http://www.unicode.org/reports/tr29>,
1229 [12] see "Wildcards in Property Values" above.
1230 Level 3 - Tailored Support
1232 RL3.1 Tailored Punctuation - Missing
1234 RL3.2 Tailored Grapheme Clusters - Missing [13]
1235 RL3.3 Tailored Word Boundaries - Missing
1236 RL3.4 Tailored Loose Matches - Retracted by Unicode
1237 RL3.5 Tailored Ranges - Retracted by Unicode
1238 RL3.6 Context Matching - Partial [14]
1239 RL3.7 Incremental Matches - Missing
1240 RL3.8 Unicode Set Sharing - Retracted by Unicode
1241 RL3.9 Possible Match Sets - Missing
1242 RL3.10 Folded Matching - Retracted by Unicode
1243 RL3.11 Submatchers - Partial [15]
1244 [13] Perl has Unicode::Collate, but it isn't integrated with regular
1246 expressions. See UTS#10 "Unicode Collation Algorithms"
1247 <http://www.unicode.org/reports/tr10>.
1248 [14] Perl has "(?<=x)" and "(?=x)", but this requirement says that it
1249 should be possible to specify that matches may occur only in a
1250 substring with the lookaheads and lookbehinds able to see beyond that
1251 matchable portion.
1252 [15] Perl has user-defined properties ("User-Defined Character
1253 Properties") to look at single code points in ways beyond Unicode, and
1254 it might be possible, though probably not very clean, to use code
1255 blocks and things like "(?(DEFINE)...)" (see perlre) to do more
1256 specialized matching.
1257 Unicode Encodings
1259 Unicode characters are assigned to code points, which are abstract
1260 numbers. To use these numbers, various encodings are needed.
1261 · UTF-8
1263 UTF-8 is a variable-length (1 to 4 bytes), byte-order independent
1265 encoding. In most of Perl's documentation, including elsewhere in
1266 this document, the term "UTF-8" means also "UTF-EBCDIC". But in
1267 this section, "UTF-8" refers only to the encoding used on ASCII
1268 platforms. It is a superset of 7-bit US-ASCII, so anything encoded
1269 in ASCII has the identical representation when encoded in UTF-8.
1270 The following table is from Unicode 3.2.
1272 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
1274 U+0000..U+007F 00..7F
1276 U+0080..U+07FF * C2..DF 80..BF
1277 U+0800..U+0FFF E0 * A0..BF 80..BF
1278 U+1000..U+CFFF E1..EC 80..BF 80..BF
1279 U+D000..U+D7FF ED 80..9F 80..BF
1280 U+D800..U+DFFF +++++ utf16 surrogates, not legal utf8 +++++
1281 U+E000..U+FFFF EE..EF 80..BF 80..BF
1282 U+10000..U+3FFFF F0 * 90..BF 80..BF 80..BF
1283 U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
1284 U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
1285 Note the gaps marked by "*" before several of the byte entries
1287 above. These are caused by legal UTF-8 avoiding non-shortest
1288 encodings: it is technically possible to UTF-8-encode a single code
1289 point in different ways, but that is explicitly forbidden, and the
1290 shortest possible encoding should always be used (and that is what
1291 Perl does).
1292 Another way to look at it is via bits:
1294 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
1296 0aaaaaaa 0aaaaaaa
1298 00000bbbbbaaaaaa 110bbbbb 10aaaaaa
1299 ccccbbbbbbaaaaaa 1110cccc 10bbbbbb 10aaaaaa
1300 00000dddccccccbbbbbbaaaaaa 11110ddd 10cccccc 10bbbbbb 10aaaaaa
1301 As you can see, the continuation bytes all begin with "10", and the
1303 leading bits of the start byte tell how many bytes there are in the
1304 encoded character.
1305 The original UTF-8 specification allowed up to 6 bytes, to allow
1307 encoding of numbers up to "0x7FFF_FFFF". Perl continues to allow
1308 those, and has extended that up to 13 bytes to encode code points
1309 up to what can fit in a 64-bit word. However, Perl will warn if
1310 you output any of these as being non-portable; and under strict
1311 UTF-8 input protocols, they are forbidden. In addition, it is now
1312 illegal to use a code point larger than what a signed integer
1313 variable on your system can hold. On 32-bit ASCII systems, this
1314 means "0x7FFF_FFFF" is the legal maximum (much higher on 64-bit
1315 systems).
1316 · UTF-EBCDIC
1318 Like UTF-8, but EBCDIC-safe, in the way that UTF-8 is ASCII-safe.
1320 This means that all the basic characters (which includes all those
1321 that have ASCII equivalents (like "A", "0", "%", etc.) are the
1322 same in both EBCDIC and UTF-EBCDIC.)
1323 UTF-EBCDIC is used on EBCDIC platforms. It generally requires more
1325 bytes to represent a given code point than UTF-8 does; the largest
1326 Unicode code points take 5 bytes to represent (instead of 4 in
1327 UTF-8), and, extended for 64-bit words, it uses 14 bytes instead of
1328 13 bytes in UTF-8.
1329 · UTF-16, UTF-16BE, UTF-16LE, Surrogates, and "BOM"'s (Byte Order
1331 Marks)
1332 The followings items are mostly for reference and general Unicode
1334 knowledge, Perl doesn't use these constructs internally.
1335 Like UTF-8, UTF-16 is a variable-width encoding, but where UTF-8
1337 uses 8-bit code units, UTF-16 uses 16-bit code units. All code
1338 points occupy either 2 or 4 bytes in UTF-16: code points
1339 "U+0000..U+FFFF" are stored in a single 16-bit unit, and code
1340 points "U+10000..U+10FFFF" in two 16-bit units. The latter case is
1341 using surrogates, the first 16-bit unit being the high surrogate,
1342 and the second being the low surrogate.
1343 Surrogates are code points set aside to encode the
1345 "U+10000..U+10FFFF" range of Unicode code points in pairs of 16-bit
1346 units. The high surrogates are the range "U+D800..U+DBFF" and the
1347 low surrogates are the range "U+DC00..U+DFFF". The surrogate
1348 encoding is
1349 $hi = ($uni - 0x10000) / 0x400 + 0xD800;
1351 $lo = ($uni - 0x10000) % 0x400 + 0xDC00;
1352 and the decoding is
1354 $uni = 0x10000 + ($hi - 0xD800) * 0x400 + ($lo - 0xDC00);
1356 Because of the 16-bitness, UTF-16 is byte-order dependent. UTF-16
1358 itself can be used for in-memory computations, but if storage or
1359 transfer is required either UTF-16BE (big-endian) or UTF-16LE
1360 (little-endian) encodings must be chosen.
1361 This introduces another problem: what if you just know that your
1363 data is UTF-16, but you don't know which endianness? Byte Order
1364 Marks, or "BOM"'s, are a solution to this. A special character has
1365 been reserved in Unicode to function as a byte order marker: the
1366 character with the code point "U+FEFF" is the "BOM".
1367 The trick is that if you read a "BOM", you will know the byte
1369 order, since if it was written on a big-endian platform, you will
1370 read the bytes "0xFE 0xFF", but if it was written on a little-
1371 endian platform, you will read the bytes "0xFF 0xFE". (And if the
1372 originating platform was writing in ASCII platform UTF-8, you will
1373 read the bytes "0xEF 0xBB 0xBF".)
1374 The way this trick works is that the character with the code point
1376 "U+FFFE" is not supposed to be in input streams, so the sequence of
1377 bytes "0xFF 0xFE" is unambiguously ""BOM", represented in little-
1378 endian format" and cannot be "U+FFFE", represented in big-endian
1379 format".
1380 Surrogates have no meaning in Unicode outside their use in pairs to
1382 represent other code points. However, Perl allows them to be
1383 represented individually internally, for example by saying
1384 "chr(0xD801)", so that all code points, not just those valid for
1385 open interchange, are representable. Unicode does define semantics
1386 for them, such as their "General_Category" is "Cs". But because
1387 their use is somewhat dangerous, Perl will warn (using the warning
1388 category "surrogate", which is a sub-category of "utf8") if an
1389 attempt is made to do things like take the lower case of one, or
1390 match case-insensitively, or to output them. (But don't try this
1391 on Perls before 5.14.)
1392 · UTF-32, UTF-32BE, UTF-32LE
1394 The UTF-32 family is pretty much like the UTF-16 family, except
1396 that the units are 32-bit, and therefore the surrogate scheme is
1397 not needed. UTF-32 is a fixed-width encoding. The "BOM"
1398 signatures are "0x00 0x00 0xFE 0xFF" for BE and "0xFF 0xFE 0x00
1399 0x00" for LE.
1400 · UCS-2, UCS-4
1402 Legacy, fixed-width encodings defined by the ISO 10646 standard.
1404 UCS-2 is a 16-bit encoding. Unlike UTF-16, UCS-2 is not extensible
1405 beyond "U+FFFF", because it does not use surrogates. UCS-4 is a
1406 32-bit encoding, functionally identical to UTF-32 (the difference
1407 being that UCS-4 forbids neither surrogates nor code points larger
1408 than "0x10_FFFF").
1409 · UTF-7
1411 A seven-bit safe (non-eight-bit) encoding, which is useful if the
1413 transport or storage is not eight-bit safe. Defined by RFC 2152.
1414 Noncharacter code points
1416 66 code points are set aside in Unicode as "noncharacter code points".
1417 These all have the "Unassigned" ("Cn") "General_Category", and no
1418 character will ever be assigned to any of them. They are the 32 code
1419 points between "U+FDD0" and "U+FDEF" inclusive, and the 34 code points:
1420 U+FFFE U+FFFF
1422 U+1FFFE U+1FFFF
1423 U+2FFFE U+2FFFF
1424 ...
1425 U+EFFFE U+EFFFF
1426 U+FFFFE U+FFFFF
1427 U+10FFFE U+10FFFF
1428 Until Unicode 7.0, the noncharacters were "forbidden for use in open
1430 interchange of Unicode text data", so that code that processed those
1431 streams could use these code points as sentinels that could be mixed in
1432 with character data, and would always be distinguishable from that
1433 data. (Emphasis above and in the next paragraph are added in this
1434 document.)
1435 Unicode 7.0 changed the wording so that they are "not recommended for
1437 use in open interchange of Unicode text data". The 7.0 Standard goes
1438 on to say:
1439 "If a noncharacter is received in open interchange, an application
1441 is not required to interpret it in any way. It is good practice,
1442 however, to recognize it as a noncharacter and to take appropriate
1443 action, such as replacing it with "U+FFFD" replacement character,
1444 to indicate the problem in the text. It is not recommended to
1445 simply delete noncharacter code points from such text, because of
1446 the potential security issues caused by deleting uninterpreted
1447 characters. (See conformance clause C7 in Section 3.2, Conformance
1448 Requirements, and Unicode Technical Report #36, "Unicode Security
1449 Considerations"
1450 <http://www.unicode.org/reports/tr36/#Substituting_for_Ill_Formed_Subsequences>)."
1451 This change was made because it was found that various commercial tools
1453 like editors, or for things like source code control, had been written
1454 so that they would not handle program files that used these code
1455 points, effectively precluding their use almost entirely! And that was
1456 never the intent. They've always been meant to be usable within an
1457 application, or cooperating set of applications, at will.
1458 If you're writing code, such as an editor, that is supposed to be able
1460 to handle any Unicode text data, then you shouldn't be using these code
1461 points yourself, and instead allow them in the input. If you need
1462 sentinels, they should instead be something that isn't legal Unicode.
1463 For UTF-8 data, you can use the bytes 0xC1 and 0xC2 as sentinels, as
1464 they never appear in well-formed UTF-8. (There are equivalents for
1465 UTF-EBCDIC). You can also store your Unicode code points in integer
1466 variables and use negative values as sentinels.
1467 If you're not writing such a tool, then whether you accept
1469 noncharacters as input is up to you (though the Standard recommends
1470 that you not). If you do strict input stream checking with Perl, these
1471 code points continue to be forbidden. This is to maintain backward
1472 compatibility (otherwise potential security holes could open up, as an
1473 unsuspecting application that was written assuming the noncharacters
1474 would be filtered out before getting to it, could now, without warning,
1475 start getting them). To do strict checking, you can use the layer
1476 ":encoding('UTF-8')".
1477 Perl continues to warn (using the warning category "nonchar", which is
1479 a sub-category of "utf8") if an attempt is made to output
1480 noncharacters.
1481 Beyond Unicode code points
1483 The maximum Unicode code point is "U+10FFFF", and Unicode only defines
1484 operations on code points up through that. But Perl works on code
1485 points up to the maximum permissible signed number available on the
1486 platform. However, Perl will not accept these from input streams
1487 unless lax rules are being used, and will warn (using the warning
1488 category "non_unicode", which is a sub-category of "utf8") if any are
1489 output.
1490 Since Unicode rules are not defined on these code points, if a Unicode-
1492 defined operation is done on them, Perl uses what we believe are
1493 sensible rules, while generally warning, using the "non_unicode"
1494 category. For example, "uc("\x{11_0000}")" will generate such a
1495 warning, returning the input parameter as its result, since Perl
1496 defines the uppercase of every non-Unicode code point to be the code
1497 point itself. (All the case changing operations, not just uppercasing,
1498 work this way.)
1499 The situation with matching Unicode properties in regular expressions,
1501 the "\p{}" and "\P{}" constructs, against these code points is not as
1502 clear cut, and how these are handled has changed as we've gained
1503 experience.
1504 One possibility is to treat any match against these code points as
1506 undefined. But since Perl doesn't have the concept of a match being
1507 undefined, it converts this to failing or "FALSE". This is almost, but
1508 not quite, what Perl did from v5.14 (when use of these code points
1509 became generally reliable) through v5.18. The difference is that Perl
1510 treated all "\p{}" matches as failing, but all "\P{}" matches as
1511 succeeding.
1512 One problem with this is that it leads to unexpected, and confusing
1514 results in some cases:
1515 chr(0x110000) =~ \p{ASCII_Hex_Digit=True} # Failed on <= v5.18
1517 chr(0x110000) =~ \p{ASCII_Hex_Digit=False} # Failed! on <= v5.18
1518 That is, it treated both matches as undefined, and converted that to
1520 false (raising a warning on each). The first case is the expected
1521 result, but the second is likely counterintuitive: "How could both be
1522 false when they are complements?" Another problem was that the
1523 implementation optimized many Unicode property matches down to already
1524 existing simpler, faster operations, which don't raise the warning. We
1525 chose to not forgo those optimizations, which help the vast majority of
1526 matches, just to generate a warning for the unlikely event that an
1527 above-Unicode code point is being matched against.
1528 As a result of these problems, starting in v5.20, what Perl does is to
1530 treat non-Unicode code points as just typical unassigned Unicode
1531 characters, and matches accordingly. (Note: Unicode has atypical
1532 unassigned code points. For example, it has noncharacter code points,
1533 and ones that, when they do get assigned, are destined to be written
1534 Right-to-left, as Arabic and Hebrew are. Perl assumes that no non-
1535 Unicode code point has any atypical properties.)
1536 Perl, in most cases, will raise a warning when matching an above-
1538 Unicode code point against a Unicode property when the result is "TRUE"
1539 for "\p{}", and "FALSE" for "\P{}". For example:
1540 chr(0x110000) =~ \p{ASCII_Hex_Digit=True} # Fails, no warning
1542 chr(0x110000) =~ \p{ASCII_Hex_Digit=False} # Succeeds, with warning
1543 In both these examples, the character being matched is non-Unicode, so
1545 Unicode doesn't define how it should match. It clearly isn't an ASCII
1546 hex digit, so the first example clearly should fail, and so it does,
1547 with no warning. But it is arguable that the second example should
1548 have an undefined, hence "FALSE", result. So a warning is raised for
1549 it.
1550 Thus the warning is raised for many fewer cases than in earlier Perls,
1552 and only when what the result is could be arguable. It turns out that
1553 none of the optimizations made by Perl (or are ever likely to be made)
1554 cause the warning to be skipped, so it solves both problems of Perl's
1555 earlier approach. The most commonly used property that is affected by
1556 this change is "\p{Unassigned}" which is a short form for
1557 "\p{General_Category=Unassigned}". Starting in v5.20, all non-Unicode
1558 code points are considered "Unassigned". In earlier releases the
1559 matches failed because the result was considered undefined.
1560 The only place where the warning is not raised when it might ought to
1562 have been is if optimizations cause the whole pattern match to not even
1563 be attempted. For example, Perl may figure out that for a string to
1564 match a certain regular expression pattern, the string has to contain
1565 the substring "foobar". Before attempting the match, Perl may look for
1566 that substring, and if not found, immediately fail the match without
1567 actually trying it; so no warning gets generated even if the string
1568 contains an above-Unicode code point.
1569 This behavior is more "Do what I mean" than in earlier Perls for most
1571 applications. But it catches fewer issues for code that needs to be
1572 strictly Unicode compliant. Therefore there is an additional mode of
1573 operation available to accommodate such code. This mode is enabled if
1574 a regular expression pattern is compiled within the lexical scope where
1575 the "non_unicode" warning class has been made fatal, say by:
1576 use warnings FATAL => "non_unicode"
1578 (see warnings). In this mode of operation, Perl will raise the warning
1580 for all matches against a non-Unicode code point (not just the arguable
1581 ones), and it skips the optimizations that might cause the warning to
1582 not be output. (It currently still won't warn if the match isn't even
1583 attempted, like in the "foobar" example above.)
1584 In summary, Perl now normally treats non-Unicode code points as typical
1586 Unicode unassigned code points for regular expression matches, raising
1587 a warning only when it is arguable what the result should be. However,
1588 if this warning has been made fatal, it isn't skipped.
1589 There is one exception to all this. "\p{All}" looks like a Unicode
1591 property, but it is a Perl extension that is defined to be true for all
1592 possible code points, Unicode or not, so no warning is ever generated
1593 when matching this against a non-Unicode code point. (Prior to v5.20,
1594 it was an exact synonym for "\p{Any}", matching code points 0 through
1595 0x10FFFF.)
1596 Security Implications of Unicode
1598 First, read Unicode Security Considerations
1599 <http://www.unicode.org/reports/tr36>.
1600 Also, note the following:
1602 · Malformed UTF-8
1604 UTF-8 is very structured, so many combinations of bytes are
1606 invalid. In the past, Perl tried to soldier on and make some sense
1607 of invalid combinations, but this can lead to security holes, so
1608 now, if the Perl core needs to process an invalid combination, it
1609 will either raise a fatal error, or will replace those bytes by the
1610 sequence that forms the Unicode REPLACEMENT CHARACTER, for which
1611 purpose Unicode created it.
1612 Every code point can be represented by more than one possible
1614 syntactically valid UTF-8 sequence. Early on, both Unicode and
1615 Perl considered any of these to be valid, but now, all sequences
1616 longer than the shortest possible one are considered to be
1617 malformed.
1618 Unicode considers many code points to be illegal, or to be avoided.
1620 Perl generally accepts them, once they have passed through any
1621 input filters that may try to exclude them. These have been
1622 discussed above (see "Surrogates" under UTF-16 in "Unicode
1623 Encodings", "Noncharacter code points", and "Beyond Unicode code
1624 points").
1625 · Regular expression pattern matching may surprise you if you're not
1627 accustomed to Unicode. Starting in Perl 5.14, several pattern
1628 modifiers are available to control this, called the character set
1629 modifiers. Details are given in "Character set modifiers" in
1630 perlre.
1631 As discussed elsewhere, Perl has one foot (two hooves?) planted in each
1633 of two worlds: the old world of ASCII and single-byte locales, and the
1634 new world of Unicode, upgrading when necessary. If your legacy code
1635 does not explicitly use Unicode, no automatic switch-over to Unicode
1636 should happen.
1637 Unicode in Perl on EBCDIC
1639 Unicode is supported on EBCDIC platforms. See perlebcdic.
1640 Unless ASCII vs. EBCDIC issues are specifically being discussed,
1642 references to UTF-8 encoding in this document and elsewhere should be
1643 read as meaning UTF-EBCDIC on EBCDIC platforms. See "Unicode and UTF"
1644 in perlebcdic.
1645 Because UTF-EBCDIC is so similar to UTF-8, the differences are mostly
1647 hidden from you; "use utf8" (and NOT something like "use utfebcdic")
1648 declares the script is in the platform's "native" 8-bit encoding of
1649 Unicode. (Similarly for the ":utf8" layer.)
1650 Locales
1652 See "Unicode and UTF-8" in perllocale
1653 When Unicode Does Not Happen
1655 There are still many places where Unicode (in some encoding or another)
1656 could be given as arguments or received as results, or both in Perl,
1657 but it is not, in spite of Perl having extensive ways to input and
1658 output in Unicode, and a few other "entry points" like the @ARGV array
1659 (which can sometimes be interpreted as UTF-8).
1660 The following are such interfaces. Also, see "The "Unicode Bug"". For
1662 all of these interfaces Perl currently (as of v5.16.0) simply assumes
1663 byte strings both as arguments and results, or UTF-8 strings if the
1664 (deprecated) "encoding" pragma has been used.
1665 One reason that Perl does not attempt to resolve the role of Unicode in
1667 these situations is that the answers are highly dependent on the
1668 operating system and the file system(s). For example, whether
1669 filenames can be in Unicode and in exactly what kind of encoding, is
1670 not exactly a portable concept. Similarly for "qx" and "system": how
1671 well will the "command-line interface" (and which of them?) handle
1672 Unicode?
1673 · "chdir", "chmod", "chown", "chroot", "exec", "link", "lstat",
1675 "mkdir", "rename", "rmdir", "stat", "symlink", "truncate",
1676 "unlink", "utime", "-X"
1677 · %ENV
1679 · "glob" (aka the "<*>")
1681 · "open", "opendir", "sysopen"
1683 · "qx" (aka the backtick operator), "system"
1685 · "readdir", "readlink"
1687 The "Unicode Bug"
1689 The term, "Unicode bug" has been applied to an inconsistency with the
1690 code points in the "Latin-1 Supplement" block, that is, between 128 and
1691 255. Without a locale specified, unlike all other characters or code
1692 points, these characters can have very different semantics depending on
1693 the rules in effect. (Characters whose code points are above 255 force
1694 Unicode rules; whereas the rules for ASCII characters are the same
1695 under both ASCII and Unicode rules.)
1696 Under Unicode rules, these upper-Latin1 characters are interpreted as
1698 Unicode code points, which means they have the same semantics as
1699 Latin-1 (ISO-8859-1) and C1 controls.
1700 As explained in "ASCII Rules versus Unicode Rules", under ASCII rules,
1702 they are considered to be unassigned characters.
1703 This can lead to unexpected results. For example, a string's semantics
1705 can suddenly change if a code point above 255 is appended to it, which
1706 changes the rules from ASCII to Unicode. As an example, consider the
1707 following program and its output:
1708 $ perl -le'
1710 no feature "unicode_strings";
1711 $s1 = "\xC2";
1712 $s2 = "\x{2660}";
1713 for ($s1, $s2, $s1.$s2) {
1714 print /\w/ || 0;
1715 }
1716 '
1717 0
1718 0
1719 1
1720 If there's no "\w" in "s1" nor in "s2", why does their concatenation
1722 have one?
1723 This anomaly stems from Perl's attempt to not disturb older programs
1725 that didn't use Unicode, along with Perl's desire to add Unicode
1726 support seamlessly. But the result turned out to not be seamless. (By
1727 the way, you can choose to be warned when things like this happen. See
1728 "encoding::warnings".)
1729 "use feature 'unicode_strings'" was added, starting in Perl v5.12, to
1731 address this problem. It affects these things:
1732 · Changing the case of a scalar, that is, using "uc()", "ucfirst()",
1734 "lc()", and "lcfirst()", or "\L", "\U", "\u" and "\l" in double-
1735 quotish contexts, such as regular expression substitutions.
1736 Under "unicode_strings" starting in Perl 5.12.0, Unicode rules are
1738 generally used. See "lc" in perlfunc for details on how this works
1739 in combination with various other pragmas.
1740 · Using caseless ("/i") regular expression matching.
1742 Starting in Perl 5.14.0, regular expressions compiled within the
1744 scope of "unicode_strings" use Unicode rules even when executed or
1745 compiled into larger regular expressions outside the scope.
1746 · Matching any of several properties in regular expressions.
1748 These properties are "\b" (without braces), "\B" (without braces),
1750 "\s", "\S", "\w", "\W", and all the Posix character classes except
1751 "[[:ascii:]]".
1752 Starting in Perl 5.14.0, regular expressions compiled within the
1754 scope of "unicode_strings" use Unicode rules even when executed or
1755 compiled into larger regular expressions outside the scope.
1756 · In "quotemeta" or its inline equivalent "\Q".
1758 Starting in Perl 5.16.0, consistent quoting rules are used within
1760 the scope of "unicode_strings", as described in "quotemeta" in
1761 perlfunc. Prior to that, or outside its scope, no code points
1762 above 127 are quoted in UTF-8 encoded strings, but in byte encoded
1763 strings, code points between 128-255 are always quoted.
1764 · In the ".." or range operator.
1766 Starting in Perl 5.26.0, the range operator on strings treats their
1768 lengths consistently within the scope of "unicode_strings". Prior
1769 to that, or outside its scope, it could produce strings whose
1770 length in characters exceeded that of the right-hand side, where
1771 the right-hand side took up more bytes than the correct range
1772 endpoint.
1773 · In "split"'s special-case whitespace splitting.
1775 Starting in Perl 5.28.0, the "split" function with a pattern
1777 specified as a string containing a single space handles whitespace
1778 characters consistently within the scope of of "unicode_strings".
1779 Prior to that, or outside its scope, characters that are whitespace
1780 according to Unicode rules but not according to ASCII rules were
1781 treated as field contents rather than field separators when they
1782 appear in byte-encoded strings.
1783 You can see from the above that the effect of "unicode_strings"
1785 increased over several Perl releases. (And Perl's support for Unicode
1786 continues to improve; it's best to use the latest available release in
1787 order to get the most complete and accurate results possible.) Note
1788 that "unicode_strings" is automatically chosen if you "use 5.012" or
1789 higher.
1790 For Perls earlier than those described above, or when a string is
1792 passed to a function outside the scope of "unicode_strings", see the
1793 next section.
1794 Forcing Unicode in Perl (Or Unforcing Unicode in Perl)
1796 Sometimes (see "When Unicode Does Not Happen" or "The "Unicode Bug"")
1797 there are situations where you simply need to force a byte string into
1798 UTF-8, or vice versa. The standard module Encode can be used for this,
1799 or the low-level calls "utf8::upgrade($bytestring)" and
1800 "utf8::downgrade($utf8string[, FAIL_OK])".
1801 Note that "utf8::downgrade()" can fail if the string contains
1803 characters that don't fit into a byte.
1804 Calling either function on a string that already is in the desired
1806 state is a no-op.
1807 "ASCII Rules versus Unicode Rules" gives all the ways that a string is
1809 made to use Unicode rules.
1810 Using Unicode in XS
1812 See "Unicode Support" in perlguts for an introduction to Unicode at the
1813 XS level, and "Unicode Support" in perlapi for the API details.
1814 Hacking Perl to work on earlier Unicode versions (for very serious hackers
1816 only)
1817 Perl by default comes with the latest supported Unicode version built-
1818 in, but the goal is to allow you to change to use any earlier one. In
1819 Perls v5.20 and v5.22, however, the earliest usable version is Unicode
1820 5.1. Perl v5.18 and v5.24 are able to handle all earlier versions.
1821 Download the files in the desired version of Unicode from the Unicode
1823 web site <http://www.unicode.org>). These should replace the existing
1824 files in lib/unicore in the Perl source tree. Follow the instructions
1825 in README.perl in that directory to change some of their names, and
1826 then build perl (see INSTALL).
1827 Porting code from perl-5.6.X
1829 Perls starting in 5.8 have a different Unicode model from 5.6. In 5.6
1830 the programmer was required to use the "utf8" pragma to declare that a
1831 given scope expected to deal with Unicode data and had to make sure
1832 that only Unicode data were reaching that scope. If you have code that
1833 is working with 5.6, you will need some of the following adjustments to
1834 your code. The examples are written such that the code will continue to
1835 work under 5.6, so you should be safe to try them out.
1836 · A filehandle that should read or write UTF-8
1838 if ($] > 5.008) {
1840 binmode $fh, ":encoding(UTF-8)";
1841 }
1842 · A scalar that is going to be passed to some extension
1844 Be it "Compress::Zlib", "Apache::Request" or any extension that has
1846 no mention of Unicode in the manpage, you need to make sure that the
1847 UTF8 flag is stripped off. Note that at the time of this writing
1848 (January 2012) the mentioned modules are not UTF-8-aware. Please
1849 check the documentation to verify if this is still true.
1850 if ($] > 5.008) {
1852 require Encode;
1853 $val = Encode::encode("UTF-8", $val); # make octets
1854 }
1855 · A scalar we got back from an extension
1857 If you believe the scalar comes back as UTF-8, you will most likely
1859 want the UTF8 flag restored:
1860 if ($] > 5.008) {
1862 require Encode;
1863 $val = Encode::decode("UTF-8", $val);
1864 }
1865 · Same thing, if you are really sure it is UTF-8
1867 if ($] > 5.008) {
1869 require Encode;
1870 Encode::_utf8_on($val);
1871 }
1872 · A wrapper for DBI "fetchrow_array" and "fetchrow_hashref"
1874 When the database contains only UTF-8, a wrapper function or method
1876 is a convenient way to replace all your "fetchrow_array" and
1877 "fetchrow_hashref" calls. A wrapper function will also make it
1878 easier to adapt to future enhancements in your database driver. Note
1879 that at the time of this writing (January 2012), the DBI has no
1880 standardized way to deal with UTF-8 data. Please check the DBI
1881 documentation to verify if that is still true.
1882 sub fetchrow {
1884 # $what is one of fetchrow_{array,hashref}
1885 my($self, $sth, $what) = @_;
1886 if ($] < 5.008) {
1887 return $sth->$what;
1888 } else {
1889 require Encode;
1890 if (wantarray) {
1891 my @arr = $sth->$what;
1892 for (@arr) {
1893 defined && /[^\000-\177]/ && Encode::_utf8_on($_);
1894 }
1895 return @arr;
1896 } else {
1897 my $ret = $sth->$what;
1898 if (ref $ret) {
1899 for my $k (keys %$ret) {
1900 defined
1901 && /[^\000-\177]/
1902 && Encode::_utf8_on($_) for $ret->{$k};
1903 }
1904 return $ret;
1905 } else {
1906 defined && /[^\000-\177]/ && Encode::_utf8_on($_) for $ret;
1907 return $ret;
1908 }
1909 }
1910 }
1911 }
1912 · A large scalar that you know can only contain ASCII
1914 Scalars that contain only ASCII and are marked as UTF-8 are
1916 sometimes a drag to your program. If you recognize such a situation,
1917 just remove the UTF8 flag:
1918 utf8::downgrade($val) if $] > 5.008;
1920
1922 See also "The "Unicode Bug"" above.
1923 Interaction with Extensions
1925 When Perl exchanges data with an extension, the extension should be
1926 able to understand the UTF8 flag and act accordingly. If the extension
1927 doesn't recognize that flag, it's likely that the extension will return
1928 incorrectly-flagged data.
1929 So if you're working with Unicode data, consult the documentation of
1931 every module you're using if there are any issues with Unicode data
1932 exchange. If the documentation does not talk about Unicode at all,
1933 suspect the worst and probably look at the source to learn how the
1934 module is implemented. Modules written completely in Perl shouldn't
1935 cause problems. Modules that directly or indirectly access code written
1936 in other programming languages are at risk.
1937 For affected functions, the simple strategy to avoid data corruption is
1939 to always make the encoding of the exchanged data explicit. Choose an
1940 encoding that you know the extension can handle. Convert arguments
1941 passed to the extensions to that encoding and convert results back from
1942 that encoding. Write wrapper functions that do the conversions for you,
1943 so you can later change the functions when the extension catches up.
1944 To provide an example, let's say the popular "Foo::Bar::escape_html"
1946 function doesn't deal with Unicode data yet. The wrapper function would
1947 convert the argument to raw UTF-8 and convert the result back to Perl's
1948 internal representation like so:
1949 sub my_escape_html ($) {
1951 my($what) = shift;
1952 return unless defined $what;
1953 Encode::decode("UTF-8", Foo::Bar::escape_html(
1954 Encode::encode("UTF-8", $what)));
1955 }
1956 Sometimes, when the extension does not convert data but just stores and
1958 retrieves it, you will be able to use the otherwise dangerous
1959 "Encode::_utf8_on()" function. Let's say the popular "Foo::Bar"
1960 extension, written in C, provides a "param" method that lets you store
1961 and retrieve data according to these prototypes:
1962 $self->param($name, $value); # set a scalar
1964 $value = $self->param($name); # retrieve a scalar
1965 If it does not yet provide support for any encoding, one could write a
1967 derived class with such a "param" method:
1968 sub param {
1970 my($self,$name,$value) = @_;
1971 utf8::upgrade($name); # make sure it is UTF-8 encoded
1972 if (defined $value) {
1973 utf8::upgrade($value); # make sure it is UTF-8 encoded
1974 return $self->SUPER::param($name,$value);
1975 } else {
1976 my $ret = $self->SUPER::param($name);
1977 Encode::_utf8_on($ret); # we know, it is UTF-8 encoded
1978 return $ret;
1979 }
1980 }
1981 Some extensions provide filters on data entry/exit points, such as
1983 "DB_File::filter_store_key" and family. Look out for such filters in
1984 the documentation of your extensions; they can make the transition to
1985 Unicode data much easier.
1986 Speed
1988 Some functions are slower when working on UTF-8 encoded strings than on
1989 byte encoded strings. All functions that need to hop over characters
1990 such as "length()", "substr()" or "index()", or matching regular
1991 expressions can work much faster when the underlying data are byte-
1992 encoded.
1993 In Perl 5.8.0 the slowness was often quite spectacular; in Perl 5.8.1 a
1995 caching scheme was introduced which improved the situation. In
1996 general, operations with UTF-8 encoded strings are still slower. As an
1997 example, the Unicode properties (character classes) like "\p{Nd}" are
1998 known to be quite a bit slower (5-20 times) than their simpler
1999 counterparts like "[0-9]" (then again, there are hundreds of Unicode
2000 characters matching "Nd" compared with the 10 ASCII characters matching
2001 "[0-9]").
2002
2004 perlunitut, perluniintro, perluniprops, Encode, open, utf8, bytes,
2005 perlretut, "${^UNICODE}" in perlvar,
2006 <http://www.unicode.org/reports/tr44>).
2007perl v5.30.1 2019-11-29 PERLUNICODE(1)