3 perlunicode - Unicode support in Perl
7 If you haven't already, before reading this document, you should become
8 familiar with both L<perlunitut> and L<perluniintro>.
10 Unicode aims to B<UNI>-fy the en-B<CODE>-ings of all the world's
11 character sets into a single Standard. For quite a few of the various
12 coding standards that existed when Unicode was first created, converting
13 from each to Unicode essentially meant adding a constant to each code
14 point in the original standard, and converting back meant just
15 subtracting that same constant. For ASCII and ISO-8859-1, the constant
16 is 0. For ISO-8859-5, (Cyrillic) the constant is 864; for Hebrew
17 (ISO-8859-8), it's 1488; Thai (ISO-8859-11), 3424; and so forth. This
18 made it easy to do the conversions, and facilitated the adoption of
21 And it worked; nowadays, those legacy standards are rarely used. Most
22 everyone uses Unicode.
24 Unicode is a comprehensive standard. It specifies many things outside
25 the scope of Perl, such as how to display sequences of characters. For
26 a full discussion of all aspects of Unicode, see
27 L<https://www.unicode.org>.
29 =head2 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:
35 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.
39 Also, the use of Unicode may present security issues that aren't
40 obvious, see L</Security Implications of Unicode> below.
44 =item Safest if you C<use feature 'unicode_strings'>
46 In order to preserve backward compatibility, Perl does not turn
47 on full internal Unicode support unless the pragma
48 L<S<C<use feature 'unicode_strings'>>|feature/The 'unicode_strings' feature>
49 is specified. (This is automatically
50 selected if you S<C<use 5.012>> or higher.) Failure to do this can
51 trigger unexpected surprises. See L</The "Unicode Bug"> below.
53 This pragma doesn't affect I/O. Nor does it change the internal
54 representation of strings, only their interpretation. There are still
55 several places where Unicode isn't fully supported, such as in
58 =item Input and Output Layers
60 Use the C<:encoding(...)> layer to read from and write to
61 filehandles using the specified encoding. (See L<open>.)
63 =item You must convert your non-ASCII, non-UTF-8 Perl scripts to be
66 The L<encoding> module has been deprecated since perl 5.18 and the
67 perl internals it requires have been removed with perl 5.26.
69 =item C<use utf8> still needed to enable L<UTF-8|/Unicode Encodings> in scripts
71 If your Perl script is itself encoded in L<UTF-8|/Unicode Encodings>,
72 the S<C<use utf8>> pragma must be explicitly included to enable
73 recognition of that (in string or regular expression literals, or in
74 identifier names). B<This is the only time when an explicit S<C<use
75 utf8>> is needed.> (See L<utf8>).
77 If a Perl script begins with the bytes that form the UTF-8 encoding of
78 the Unicode BYTE ORDER MARK (C<BOM>, see L</Unicode Encodings>), those
79 bytes are completely ignored.
81 =item L<UTF-16|/Unicode Encodings> scripts autodetected
83 If a Perl script begins with the Unicode C<BOM> (UTF-16LE,
84 UTF16-BE), or if the script looks like non-C<BOM>-marked
85 UTF-16 of either endianness, Perl will correctly read in the script as
86 the appropriate Unicode encoding.
90 =head2 Byte and Character Semantics
92 Before Unicode, most encodings used 8 bits (a single byte) to encode
93 each character. Thus a character was a byte, and a byte was a
94 character, and there could be only 256 or fewer possible characters.
95 "Byte Semantics" in the title of this section refers to
96 this behavior. There was no need to distinguish between "Byte" and
99 Then along comes Unicode which has room for over a million characters
100 (and Perl allows for even more). This means that a character may
101 require more than a single byte to represent it, and so the two terms
102 are no longer equivalent. What matter are the characters as whole
103 entities, and not usually the bytes that comprise them. That's what the
104 term "Character Semantics" in the title of this section refers to.
106 Perl had to change internally to decouple "bytes" from "characters".
107 It is important that you too change your ideas, if you haven't already,
108 so that "byte" and "character" no longer mean the same thing in your
111 The basic building block of Perl strings has always been a "character".
112 The changes basically come down to that the implementation no longer
113 thinks that a character is always just a single byte.
115 There are various things to note:
121 String handling functions, for the most part, continue to operate in
122 terms of characters. C<length()>, for example, returns the number of
123 characters in a string, just as before. But that number no longer is
124 necessarily the same as the number of bytes in the string (there may be
125 more bytes than characters). The other such functions include
126 C<chop()>, C<chomp()>, C<substr()>, C<pos()>, C<index()>, C<rindex()>,
127 C<sort()>, C<sprintf()>, and C<write()>.
135 the bit-oriented C<vec>
141 the byte-oriented C<pack>/C<unpack> C<"C"> format
143 However, the C<W> specifier does operate on whole characters, as does the
148 some operators that interact with the platform's operating system
150 Operators dealing with filenames are examples.
154 when the functions are called from within the scope of the
155 S<C<L<use bytes|bytes>>> pragma
157 Likely, you should use this only for debugging anyway.
163 Strings--including hash keys--and regular expression patterns may
164 contain characters that have ordinal values larger than 255.
166 If you use a Unicode editor to edit your program, Unicode characters may
167 occur directly within the literal strings in UTF-8 encoding, or UTF-16.
168 (The former requires a C<use utf8>, the latter may require a C<BOM>.)
170 L<perluniintro/Creating Unicode> gives other ways to place non-ASCII
171 characters in your strings.
175 The C<chr()> and C<ord()> functions work on whole characters.
179 Regular expressions match whole characters. For example, C<"."> matches
180 a whole character instead of only a single byte.
184 The C<tr///> operator translates whole characters. (Note that the
185 C<tr///CU> functionality has been removed. For similar functionality to
186 that, see C<pack('U0', ...)> and C<pack('C0', ...)>).
190 C<scalar reverse()> reverses by character rather than by byte.
194 The bit string operators, C<& | ^ ~> and (starting in v5.22)
195 C<&. |. ^. ~.> can operate on bit strings encoded in UTF-8, but this
196 can give unexpected results if any of the strings contain code points
197 above 0xFF. Starting in v5.28, it is a fatal error to have such an
198 operand. Otherwise, the operation is performed on a non-UTF-8 copy of
199 the operand. If you're not sure about the encoding of a string,
200 downgrade it before using any of these operators; you can use
201 L<C<utf8::utf8_downgrade()>|utf8/Utility functions>.
205 The bottom line is that Perl has always practiced "Character Semantics",
206 but with the advent of Unicode, that is now different than "Byte
209 =head2 ASCII Rules versus Unicode Rules
211 Before Unicode, when a character was a byte was a character,
212 Perl knew only about the 128 characters defined by ASCII, code points 0
213 through 127 (except for under L<S<C<use locale>>|perllocale>). That
215 points 128 to 255 as unassigned, and available for whatever use a
216 program might want. The only semantics they have is their ordinal
217 numbers, and that they are members of none of the non-negative character
218 classes. None are considered to match C<\w> for example, but all match
221 Unicode, of course, assigns each of those code points a particular
222 meaning (along with ones above 255). To preserve backward
223 compatibility, Perl only uses the Unicode meanings when there is some
224 indication that Unicode is what is intended; otherwise the non-ASCII
225 code points remain treated as if they are unassigned.
227 Here are the ways that Perl knows that a string should be treated as
234 Within the scope of S<C<use utf8>>
236 If the whole program is Unicode (signified by using 8-bit B<U>nicode
237 B<T>ransformation B<F>ormat), then all literal strings within it must be
243 L<S<C<use feature 'unicode_strings'>>|feature/The 'unicode_strings' feature>
245 This pragma was created so you can explicitly tell Perl that operations
246 executed within its scope are to use Unicode rules. More operations are
247 affected with newer perls. See L</The "Unicode Bug">.
251 Within the scope of S<C<use 5.012>> or higher
253 This implicitly turns on S<C<use feature 'unicode_strings'>>.
258 L<S<C<use locale 'not_characters'>>|perllocale/Unicode and UTF-8>,
259 or L<S<C<use locale>>|perllocale> and the current
260 locale is a UTF-8 locale.
262 The former is defined to imply Unicode handling; and the latter
263 indicates a Unicode locale, hence a Unicode interpretation of all
268 When the string contains a Unicode-only code point
270 Perl has never accepted code points above 255 without them being
271 Unicode, so their use implies Unicode for the whole string.
275 When the string contains a Unicode named code point C<\N{...}>
277 The C<\N{...}> construct explicitly refers to a Unicode code point,
278 even if it is one that is also in ASCII. Therefore the string
279 containing it must be Unicode.
283 When the string has come from an external source marked as
286 The L<C<-C>|perlrun/-C [numberE<sol>list]> command line option can
287 specify that certain inputs to the program are Unicode, and the values
288 of this can be read by your Perl code, see L<perlvar/"${^UNICODE}">.
290 =item * When the string has been upgraded to UTF-8
292 The function L<C<utf8::utf8_upgrade()>|utf8/Utility functions>
293 can be explicitly used to permanently (unless a subsequent
294 C<utf8::utf8_downgrade()> is called) cause a string to be treated as
297 =item * There are additional methods for regular expression patterns
299 A pattern that is compiled with the C<< /u >> or C<< /a >> modifiers is
300 treated as Unicode (though there are some restrictions with C<< /a >>).
301 Under the C<< /d >> and C<< /l >> modifiers, there are several other
302 indications for Unicode; see L<perlre/Character set modifiers>.
306 Note that all of the above are overridden within the scope of
307 C<L<use bytes|bytes>>; but you should be using this pragma only for
310 Note also that some interactions with the platform's operating system
311 never use Unicode rules.
313 When Unicode rules are in effect:
319 Case translation operators use the Unicode case translation tables.
321 Note that C<uc()>, or C<\U> in interpolated strings, translates to
322 uppercase, while C<ucfirst>, or C<\u> in interpolated strings,
323 translates to titlecase in languages that make the distinction (which is
324 equivalent to uppercase in languages without the distinction).
326 There is a CPAN module, C<L<Unicode::Casing>>, which allows you to
327 define your own mappings to be used in C<lc()>, C<lcfirst()>, C<uc()>,
328 C<ucfirst()>, and C<fc> (or their double-quoted string inlined versions
329 such as C<\U>). (Prior to Perl 5.16, this functionality was partially
330 provided in the Perl core, but suffered from a number of insurmountable
331 drawbacks, so the CPAN module was written instead.)
335 Character classes in regular expressions match based on the character
336 properties specified in the Unicode properties database.
338 C<\w> can be used to match a Japanese ideograph, for instance; and
339 C<[[:digit:]]> a Bengali number.
343 Named Unicode properties, scripts, and block ranges may be used (like
344 bracketed character classes) by using the C<\p{}> "matches property"
345 construct and the C<\P{}> negation, "doesn't match property".
347 See L</"Unicode Character Properties"> for more details.
349 You can define your own character properties and use them
350 in the regular expression with the C<\p{}> or C<\P{}> construct.
351 See L</"User-Defined Character Properties"> for more details.
355 =head2 Extended Grapheme Clusters (Logical characters)
357 Consider a character, say C<H>. It could appear with various marks around it,
358 such as an acute accent, or a circumflex, or various hooks, circles, arrows,
359 I<etc.>, above, below, to one side or the other, I<etc>. There are many
360 possibilities among the world's languages. The number of combinations is
361 astronomical, and if there were a character for each combination, it would
362 soon exhaust Unicode's more than a million possible characters. So Unicode
363 took a different approach: there is a character for the base C<H>, and a
364 character for each of the possible marks, and these can be variously combined
365 to get a final logical character. So a logical character--what appears to be a
366 single character--can be a sequence of more than one individual characters.
367 The Unicode standard calls these "extended grapheme clusters" (which
368 is an improved version of the no-longer much used "grapheme cluster");
369 Perl furnishes the C<\X> regular expression construct to match such
370 sequences in their entirety.
372 But Unicode's intent is to unify the existing character set standards and
373 practices, and several pre-existing standards have single characters that
374 mean the same thing as some of these combinations, like ISO-8859-1,
375 which has quite a few of them. For example, C<"LATIN CAPITAL LETTER E
376 WITH ACUTE"> was already in this standard when Unicode came along.
377 Unicode therefore added it to its repertoire as that single character.
378 But this character is considered by Unicode to be equivalent to the
379 sequence consisting of the character C<"LATIN CAPITAL LETTER E">
380 followed by the character C<"COMBINING ACUTE ACCENT">.
382 C<"LATIN CAPITAL LETTER E WITH ACUTE"> is called a "pre-composed"
383 character, and its equivalence with the "E" and the "COMBINING ACCENT"
384 sequence is called canonical equivalence. All pre-composed characters
385 are said to have a decomposition (into the equivalent sequence), and the
386 decomposition type is also called canonical. A string may be comprised
387 as much as possible of precomposed characters, or it may be comprised of
388 entirely decomposed characters. Unicode calls these respectively,
389 "Normalization Form Composed" (NFC) and "Normalization Form Decomposed".
390 The C<L<Unicode::Normalize>> module contains functions that convert
391 between the two. A string may also have both composed characters and
392 decomposed characters; this module can be used to make it all one or the
395 You may be presented with strings in any of these equivalent forms.
396 There is currently nothing in Perl 5 that ignores the differences. So
397 you'll have to specially handle it. The usual advice is to convert your
398 inputs to C<NFD> before processing further.
400 For more detailed information, see L<http://unicode.org/reports/tr15/>.
402 =head2 Unicode Character Properties
404 (The only time that Perl considers a sequence of individual code
405 points as a single logical character is in the C<\X> construct, already
406 mentioned above. Therefore "character" in this discussion means a single
409 Very nearly all Unicode character properties are accessible through
410 regular expressions by using the C<\p{}> "matches property" construct
411 and the C<\P{}> "doesn't match property" for its negation.
413 For instance, C<\p{Uppercase}> matches any single character with the Unicode
414 C<"Uppercase"> property, while C<\p{L}> matches any character with a
415 C<General_Category> of C<"L"> (letter) property (see
416 L</General_Category> below). Brackets are not
417 required for single letter property names, so C<\p{L}> is equivalent to C<\pL>.
419 More formally, C<\p{Uppercase}> matches any single character whose Unicode
420 C<Uppercase> property value is C<True>, and C<\P{Uppercase}> matches any character
421 whose C<Uppercase> property value is C<False>, and they could have been written as
422 C<\p{Uppercase=True}> and C<\p{Uppercase=False}>, respectively.
424 This formality is needed when properties are not binary; that is, if they can
425 take on more values than just C<True> and C<False>. For example, the
426 C<Bidi_Class> property (see L</"Bidirectional Character Types"> below),
427 can take on several different
428 values, such as C<Left>, C<Right>, C<Whitespace>, and others. To match these, one needs
429 to specify both the property name (C<Bidi_Class>), AND the value being
431 (C<Left>, C<Right>, I<etc.>). This is done, as in the examples above, by having the
432 two components separated by an equal sign (or interchangeably, a colon), like
433 C<\p{Bidi_Class: Left}>.
435 All Unicode-defined character properties may be written in these compound forms
436 of C<\p{I<property>=I<value>}> or C<\p{I<property>:I<value>}>, but Perl provides some
437 additional properties that are written only in the single form, as well as
438 single-form short-cuts for all binary properties and certain others described
439 below, in which you may omit the property name and the equals or colon
442 Most Unicode character properties have at least two synonyms (or aliases if you
443 prefer): a short one that is easier to type and a longer one that is more
444 descriptive and hence easier to understand. Thus the C<"L"> and
445 C<"Letter"> properties above are equivalent and can be used
446 interchangeably. Likewise, C<"Upper"> is a synonym for C<"Uppercase">,
447 and we could have written C<\p{Uppercase}> equivalently as C<\p{Upper}>.
448 Also, there are typically various synonyms for the values the property
449 can be. For binary properties, C<"True"> has 3 synonyms: C<"T">,
450 C<"Yes">, and C<"Y">; and C<"False"> has correspondingly C<"F">,
451 C<"No">, and C<"N">. But be careful. A short form of a value for one
452 property may not mean the same thing as the short form spelled the same
454 Thus, for the C<L</General_Category>> property, C<"L"> means
455 C<"Letter">, but for the L<C<Bidi_Class>|/Bidirectional Character Types>
456 property, C<"L"> means C<"Left">. A complete list of properties and
457 synonyms is in L<perluniprops>.
459 Upper/lower case differences in property names and values are irrelevant;
460 thus C<\p{Upper}> means the same thing as C<\p{upper}> or even C<\p{UpPeR}>.
461 Similarly, you can add or subtract underscores anywhere in the middle of a
462 word, so that these are also equivalent to C<\p{U_p_p_e_r}>. And white space
463 is generally irrelevant adjacent to non-word characters, such as the
464 braces and the equals or colon separators, so C<\p{ Upper }> and
465 C<\p{ Upper_case : Y }> are equivalent to these as well. In fact, white
466 space and even hyphens can usually be added or deleted anywhere. So
467 even C<\p{ Up-per case = Yes}> is equivalent. All this is called
468 "loose-matching" by Unicode. The "name" property has some restrictions
469 on this due to a few outlier names. Full details are given in
470 L<https://www.unicode.org/reports/tr44/tr44-24.html#UAX44-LM2>.
472 The few places where stricter matching is
473 used is in the middle of numbers, the "name" property, and in the Perl
474 extension properties that begin or end with an underscore. Stricter
475 matching cares about white space (except adjacent to non-word
476 characters), hyphens, and non-interior underscores.
478 You can also use negation in both C<\p{}> and C<\P{}> by introducing a caret
479 (C<^>) between the first brace and the property name: C<\p{^Tamil}> is
480 equal to C<\P{Tamil}>.
482 Almost all properties are immune to case-insensitive matching. That is,
483 adding a C</i> regular expression modifier does not change what they
484 match. There are two sets that are affected.
488 and C<Titlecase_Letter>,
489 all of which match C<Cased_Letter> under C</i> matching.
490 And the second set is
494 all of which match C<Cased> under C</i> matching.
495 This set also includes its subsets C<PosixUpper> and C<PosixLower> both
496 of which under C</i> match C<PosixAlpha>.
497 (The difference between these sets is that some things, such as Roman
498 numerals, come in both upper and lower case so they are C<Cased>, but
499 aren't considered letters, so they aren't C<Cased_Letter>'s.)
501 See L</Beyond Unicode code points> for special considerations when
502 matching Unicode properties against non-Unicode code points.
504 =head3 B<General_Category>
506 Every Unicode character is assigned a general category, which is the "most
507 usual categorization of a character" (from
508 L<https://www.unicode.org/reports/tr44>).
510 The compound way of writing these is like C<\p{General_Category=Number}>
511 (short: C<\p{gc:n}>). But Perl furnishes shortcuts in which everything up
512 through the equal or colon separator is omitted. So you can instead just write
515 Here are the short and long forms of the values the C<General Category> property
521 LC, L& Cased_Letter (that is: [\p{Ll}\p{Lu}\p{Lt}])
534 Nd Decimal_Number (also Digit)
538 P Punctuation (also Punct)
539 Pc Connector_Punctuation
543 Pi Initial_Punctuation
544 (may behave like Ps or Pe depending on usage)
546 (may behave like Ps or Pe depending on usage)
558 Zp Paragraph_Separator
561 Cc Control (also Cntrl)
567 Single-letter properties match all characters in any of the
568 two-letter sub-properties starting with the same letter.
569 C<LC> and C<L&> are special: both are aliases for the set consisting of everything matched by C<Ll>, C<Lu>, and C<Lt>.
571 =head3 B<Bidirectional Character Types>
573 Because scripts differ in their directionality (Hebrew and Arabic are
574 written right to left, for example) Unicode supplies a C<Bidi_Class> property.
575 Some of the values this property can have are:
580 LRE Left-to-Right Embedding
581 LRO Left-to-Right Override
584 RLE Right-to-Left Embedding
585 RLO Right-to-Left Override
586 PDF Pop Directional Format
588 ES European Separator
589 ET European Terminator
594 B Paragraph Separator
599 This property is always written in the compound form.
600 For example, C<\p{Bidi_Class:R}> matches characters that are normally
601 written right to left. Unlike the
602 C<L</General_Category>> property, this
603 property can have more values added in a future Unicode release. Those
604 listed above comprised the complete set for many Unicode releases, but
605 others were added in Unicode 6.3; you can always find what the
606 current ones are in L<perluniprops>. And
607 L<https://www.unicode.org/reports/tr9/> describes how to use them.
611 The world's languages are written in many different scripts. This sentence
612 (unless you're reading it in translation) is written in Latin, while Russian is
613 written in Cyrillic, and Greek is written in, well, Greek; Japanese mainly in
614 Hiragana or Katakana. There are many more.
616 The Unicode C<Script> and C<Script_Extensions> properties give what
617 script a given character is in. The C<Script_Extensions> property is an
618 improved version of C<Script>, as demonstrated below. Either property
619 can be specified with the compound form like
620 C<\p{Script=Hebrew}> (short: C<\p{sc=hebr}>), or
621 C<\p{Script_Extensions=Javanese}> (short: C<\p{scx=java}>).
622 In addition, Perl furnishes shortcuts for all
623 C<Script_Extensions> property names. You can omit everything up through
624 the equals (or colon), and simply write C<\p{Latin}> or C<\P{Cyrillic}>.
625 (This is not true for C<Script>, which is required to be
626 written in the compound form. Prior to Perl v5.26, the single form
627 returned the plain old C<Script> version, but was changed because
628 C<Script_Extensions> gives better results.)
630 The difference between these two properties involves characters that are
631 used in multiple scripts. For example the digits '0' through '9' are
632 used in many parts of the world. These are placed in a script named
633 C<Common>. Other characters are used in just a few scripts. For
634 example, the C<"KATAKANA-HIRAGANA DOUBLE HYPHEN"> is used in both Japanese
635 scripts, Katakana and Hiragana, but nowhere else. The C<Script>
636 property places all characters that are used in multiple scripts in the
637 C<Common> script, while the C<Script_Extensions> property places those
638 that are used in only a few scripts into each of those scripts; while
639 still using C<Common> for those used in many scripts. Thus both these
642 "0" =~ /\p{sc=Common}/ # Matches
643 "0" =~ /\p{scx=Common}/ # Matches
645 and only the first of these match:
647 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{sc=Common} # Matches
648 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{scx=Common} # No match
650 And only the last two of these match:
652 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{sc=Hiragana} # No match
653 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{sc=Katakana} # No match
654 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{scx=Hiragana} # Matches
655 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{scx=Katakana} # Matches
657 C<Script_Extensions> is thus an improved C<Script>, in which there are
658 fewer characters in the C<Common> script, and correspondingly more in
659 other scripts. It is new in Unicode version 6.0, and its data are likely
660 to change significantly in later releases, as things get sorted out.
661 New code should probably be using C<Script_Extensions> and not plain
662 C<Script>. If you compile perl with a Unicode release that doesn't have
663 C<Script_Extensions>, the single form Perl extensions will instead refer
664 to the plain C<Script> property. If you compile with a version of
665 Unicode that doesn't have the C<Script> property, these extensions will
666 not be defined at all.
668 (Actually, besides C<Common>, the C<Inherited> script, contains
669 characters that are used in multiple scripts. These are modifier
670 characters which inherit the script value
671 of the controlling character. Some of these are used in many scripts,
672 and so go into C<Inherited> in both C<Script> and C<Script_Extensions>.
673 Others are used in just a few scripts, so are in C<Inherited> in
674 C<Script>, but not in C<Script_Extensions>.)
676 It is worth stressing that there are several different sets of digits in
677 Unicode that are equivalent to 0-9 and are matchable by C<\d> in a
678 regular expression. If they are used in a single language only, they
679 are in that language's C<Script> and C<Script_Extensions>. If they are
680 used in more than one script, they will be in C<sc=Common>, but only
681 if they are used in many scripts should they be in C<scx=Common>.
683 The explanation above has omitted some detail; refer to UAX#24 "Unicode
684 Script Property": L<https://www.unicode.org/reports/tr24>.
686 A complete list of scripts and their shortcuts is in L<perluniprops>.
688 =head3 B<Use of the C<"Is"> Prefix>
690 For backward compatibility (with ancient Perl 5.6), all properties writable
691 without using the compound form mentioned
692 so far may have C<Is> or C<Is_> prepended to their name, so C<\P{Is_Lu}>, for
693 example, is equal to C<\P{Lu}>, and C<\p{IsScript:Arabic}> is equal to
698 In addition to B<scripts>, Unicode also defines B<blocks> of
699 characters. The difference between scripts and blocks is that the
700 concept of scripts is closer to natural languages, while the concept
701 of blocks is more of an artificial grouping based on groups of Unicode
702 characters with consecutive ordinal values. For example, the C<"Basic Latin">
703 block is all the characters whose ordinals are between 0 and 127, inclusive; in
704 other words, the ASCII characters. The C<"Latin"> script contains some letters
705 from this as well as several other blocks, like C<"Latin-1 Supplement">,
706 C<"Latin Extended-A">, I<etc.>, but it does not contain all the characters from
707 those blocks. It does not, for example, contain the digits 0-9, because
708 those digits are shared across many scripts, and hence are in the
711 For more about scripts versus blocks, see UAX#24 "Unicode Script Property":
712 L<https://www.unicode.org/reports/tr24>
714 The C<Script_Extensions> or C<Script> properties are likely to be the
715 ones you want to use when processing
716 natural language; the C<Block> property may occasionally be useful in working
717 with the nuts and bolts of Unicode.
719 Block names are matched in the compound form, like C<\p{Block: Arrows}> or
720 C<\p{Blk=Hebrew}>. Unlike most other properties, only a few block names have a
721 Unicode-defined short name.
723 Perl also defines single form synonyms for the block property in cases
724 where these do not conflict with something else. But don't use any of
725 these, because they are unstable. Since these are Perl extensions, they
726 are subordinate to official Unicode property names; Unicode doesn't know
727 nor care about Perl's extensions. It may happen that a name that
728 currently means the Perl extension will later be changed without warning
729 to mean a different Unicode property in a future version of the perl
730 interpreter that uses a later Unicode release, and your code would no
731 longer work. The extensions are mentioned here for completeness: Take
732 the block name and prefix it with one of: C<In> (for example
733 C<\p{Blk=Arrows}> can currently be written as C<\p{In_Arrows}>); or
734 sometimes C<Is> (like C<\p{Is_Arrows}>); or sometimes no prefix at all
735 (C<\p{Arrows}>). As of this writing (Unicode 9.0) there are no
736 conflicts with using the C<In_> prefix, but there are plenty with the
737 other two forms. For example, C<\p{Is_Hebrew}> and C<\p{Hebrew}> mean
738 C<\p{Script_Extensions=Hebrew}> which is NOT the same thing as
739 C<\p{Blk=Hebrew}>. Our
740 advice used to be to use the C<In_> prefix as a single form way of
741 specifying a block. But Unicode 8.0 added properties whose names begin
742 with C<In>, and it's now clear that it's only luck that's so far
743 prevented a conflict. Using C<In> is only marginally less typing than
744 C<Blk:>, and the latter's meaning is clearer anyway, and guaranteed to
745 never conflict. So don't take chances. Use C<\p{Blk=foo}> for new
746 code. And be sure that block is what you really really want to do. In
747 most cases scripts are what you want instead.
749 A complete list of blocks is in L<perluniprops>.
751 =head3 B<Other Properties>
753 There are many more properties than the very basic ones described here.
754 A complete list is in L<perluniprops>.
756 Unicode defines all its properties in the compound form, so all single-form
757 properties are Perl extensions. Most of these are just synonyms for the
758 Unicode ones, but some are genuine extensions, including several that are in
759 the compound form. And quite a few of these are actually recommended by Unicode
760 (in L<https://www.unicode.org/reports/tr18>).
762 This section gives some details on all extensions that aren't just
763 synonyms for compound-form Unicode properties
764 (for those properties, you'll have to refer to the
765 L<Unicode Standard|https://www.unicode.org/reports/tr44>.
771 This matches every possible code point. It is equivalent to C<qr/./s>.
772 Unlike all the other non-user-defined C<\p{}> property matches, no
773 warning is ever generated if this is property is matched against a
774 non-Unicode code point (see L</Beyond Unicode code points> below).
776 =item B<C<\p{Alnum}>>
778 This matches any C<\p{Alphabetic}> or C<\p{Decimal_Number}> character.
782 This matches any of the 1_114_112 Unicode code points. It is a synonym
785 =item B<C<\p{ASCII}>>
787 This matches any of the 128 characters in the US-ASCII character set,
788 which is a subset of Unicode.
790 =item B<C<\p{Assigned}>>
792 This matches any assigned code point; that is, any code point whose L<general
793 category|/General_Category> is not C<Unassigned> (or equivalently, not C<Cn>).
795 =item B<C<\p{Blank}>>
797 This is the same as C<\h> and C<\p{HorizSpace}>: A character that changes the
798 spacing horizontally.
800 =item B<C<\p{Decomposition_Type: Non_Canonical}>> (Short: C<\p{Dt=NonCanon}>)
802 Matches a character that has any of the non-canonical decomposition
803 types. Canonical decompositions are introduced in the
804 L</Extended Grapheme Clusters (Logical characters)> section above.
805 However, many more characters have a different type of decomposition,
806 generically called "compatible" decompositions, or "non-canonical". The
807 sequences that form these decompositions are not considered canonically
808 equivalent to the pre-composed character. An example is the
809 C<"SUPERSCRIPT ONE">. It is somewhat like a regular digit 1, but not
810 exactly; its decomposition into the digit 1 is called a "compatible"
811 decomposition, specifically a "super" (for "superscript") decomposition.
812 There are several such compatibility decompositions (see
813 L<https://www.unicode.org/reports/tr44>). S<C<\p{Dt: Non_Canon}>> is a
814 Perl extension that uses just one name to refer to the union of all of
817 Most Unicode characters don't have a decomposition, so their
818 decomposition type is C<"None">. Hence, C<Non_Canonical> is equivalent
821 qr/(?[ \P{DT=Canonical} - \p{DT=None} ])/
823 (Note that one of the non-canonical decompositions is named "compat",
824 which could perhaps have been better named "miscellaneous". It includes
825 just the things that Unicode couldn't figure out a better generic name
828 =item B<C<\p{Graph}>>
830 Matches any character that is graphic. Theoretically, this means a character
831 that on a printer would cause ink to be used.
833 =item B<C<\p{HorizSpace}>>
835 This is the same as C<\h> and C<\p{Blank}>: a character that changes the
836 spacing horizontally.
840 This is a synonym for C<\p{Present_In=*}>
842 =item B<C<\p{PerlSpace}>>
844 This is the same as C<\s>, restricted to ASCII, namely C<S<[ \f\n\r\t]>>
845 and starting in Perl v5.18, a vertical tab.
847 Mnemonic: Perl's (original) space
849 =item B<C<\p{PerlWord}>>
851 This is the same as C<\w>, restricted to ASCII, namely C<[A-Za-z0-9_]>
853 Mnemonic: Perl's (original) word.
855 =item B<C<\p{Posix...}>>
857 There are several of these, which are equivalents, using the C<\p{}>
858 notation, for Posix classes and are described in
859 L<perlrecharclass/POSIX Character Classes>.
861 =item B<C<\p{Present_In: *}>> (Short: C<\p{In=*}>)
863 This property is used when you need to know in what Unicode version(s) a
866 The "*" above stands for some Unicode version number, such as
867 C<1.1> or C<12.0>; or the "*" can also be C<Unassigned>. This property will
868 match the code points whose final disposition has been settled as of the
869 Unicode release given by the version number; C<\p{Present_In: Unassigned}>
870 will match those code points whose meaning has yet to be assigned.
872 For example, C<U+0041> C<"LATIN CAPITAL LETTER A"> was present in the very first
873 Unicode release available, which is C<1.1>, so this property is true for all
874 valid "*" versions. On the other hand, C<U+1EFF> was not assigned until version
875 5.1 when it became C<"LATIN SMALL LETTER Y WITH LOOP">, so the only "*" that
876 would match it are 5.1, 5.2, and later.
878 Unicode furnishes the C<Age> property from which this is derived. The problem
879 with Age is that a strict interpretation of it (which Perl takes) has it
880 matching the precise release a code point's meaning is introduced in. Thus
881 C<U+0041> would match only 1.1; and C<U+1EFF> only 5.1. This is not usually what
884 Some non-Perl implementations of the Age property may change its meaning to be
885 the same as the Perl C<Present_In> property; just be aware of that.
887 Another confusion with both these properties is that the definition is not
888 that the code point has been I<assigned>, but that the meaning of the code point
889 has been I<determined>. This is because 66 code points will always be
890 unassigned, and so the C<Age> for them is the Unicode version in which the decision
891 to make them so was made. For example, C<U+FDD0> is to be permanently
892 unassigned to a character, and the decision to do that was made in version 3.1,
893 so C<\p{Age=3.1}> matches this character, as also does C<\p{Present_In: 3.1}> and up.
895 =item B<C<\p{Print}>>
897 This matches any character that is graphical or blank, except controls.
899 =item B<C<\p{SpacePerl}>>
901 This is the same as C<\s>, including beyond ASCII.
903 Mnemonic: Space, as modified by Perl. (It doesn't include the vertical tab
904 until v5.18, which both the Posix standard and Unicode consider white space.)
906 =item B<C<\p{Title}>> and B<C<\p{Titlecase}>>
908 Under case-sensitive matching, these both match the same code points as
909 C<\p{General Category=Titlecase_Letter}> (C<\p{gc=lt}>). The difference
910 is that under C</i> caseless matching, these match the same as
911 C<\p{Cased}>, whereas C<\p{gc=lt}> matches C<\p{Cased_Letter>).
913 =item B<C<\p{Unicode}>>
915 This matches any of the 1_114_112 Unicode code points.
918 =item B<C<\p{VertSpace}>>
920 This is the same as C<\v>: A character that changes the spacing vertically.
924 This is the same as C<\w>, including over 100_000 characters beyond ASCII.
926 =item B<C<\p{XPosix...}>>
928 There are several of these, which are the standard Posix classes
929 extended to the full Unicode range. They are described in
930 L<perlrecharclass/POSIX Character Classes>.
934 =head2 Comparison of C<\N{...}> and C<\p{name=...}>
936 Starting in Perl 5.32, you can specify a character by its name in
937 regular expression patterns using C<\p{name=...}>. This is in addition
938 to the longstanding method of using C<\N{...}>. The following
939 summarizes the differences between these two:
942 can interpolate only with eval yes [1]
943 custom names yes no [2]
944 name aliases yes yes [3]
945 named sequences yes yes [4]
946 name value parsing exact Unicode loose [5]
952 The ability to interpolate means you can do something like
954 qr/\p{na=latin capital letter $which}/
956 and specify C<$which> elsewhere.
960 You can create your own names for characters, and override official
961 ones when using C<\N{...}>. See L<charnames/CUSTOM ALIASES>.
965 Some characters have multiple names (synonyms).
969 Some particular sequences of characters are given a single name, in
970 addition to their individual ones.
974 Exact name value matching means you have to specify case, hyphens,
975 underscores, and spaces precisely in the name you want. Loose matching
976 follows the Unicode rules
977 L<https://www.unicode.org/reports/tr44/tr44-24.html#UAX44-LM2>,
978 where these are mostly irrelevant. Except for a few outlier character
979 names, these are the same rules as are already used for any other
984 =head2 Wildcards in Property Values
986 Starting in Perl 5.30, it is possible to do something like this:
988 qr!\p{numeric_value=/\A[0-5]\z/}!
990 or, by abbreviating and adding C</x>,
992 qr! \p{nv= /(?x) \A [0-5] \z / }!
994 This matches all code points whose numeric value is one of 0, 1, 2, 3,
995 4, or 5. This particular example could instead have been written as
997 qr! \A [ \p{nv=0}\p{nv=1}\p{nv=2}\p{nv=3}\p{nv=4}\p{nv=5} ] \z !xx
999 in earlier perls, so in this case this feature just makes things easier
1000 and shorter to write. If we hadn't included the C<\A> and C<\z>, these
1001 would have matched things like C<1E<sol>2> because that contains a 1 (as
1002 well as a 2). As written, it matches things like subscripts that have
1003 these numeric values. If we only wanted the decimal digits with those
1004 numeric values, we could say,
1006 qr! (?[ \d & \p{nv=/[0-5]/ ]) }!x
1008 The C<\d> gets rid of needing to anchor the pattern, since it forces the
1009 result to only match C<[0-9]>, and the C<[0-5]> further restricts it.
1011 The text in the above examples enclosed between the C<"E<sol>">
1012 characters can be just about any regular expression. It is independent
1013 of the main pattern, so doesn't share any capturing groups, I<etc>. The
1014 delimiters for it must be ASCII punctuation, but it may NOT be
1015 delimited by C<"{">, nor C<"}"> nor contain a literal C<"}">, as that
1016 delimits the end of the enclosing C<\p{}>. Like any pattern, certain
1017 other delimiters are terminated by their mirror images. These are
1018 C<"(">, C<"[>", and C<"E<lt>">. If the delimiter is any of C<"-">,
1019 C<"_">, C<"+">, or C<"\">, or is the same delimiter as is used for the
1020 enclosing pattern, it must be preceded by a backslash escape, both
1023 Beware of using C<"$"> to indicate to match the end of the string. It
1024 can too easily be interpreted as being a punctuation variable, like
1027 No modifiers may follow the final delimiter. Instead, use
1028 L<perlre/(?adlupimnsx-imnsx)> and/or
1029 L<perlre/(?adluimnsx-imnsx:pattern)> to specify modifiers.
1030 However, certain modifiers are illegal in your wildcard subpattern.
1031 The only character set modifier specifiable is C</aa>;
1032 any other character set, and C<-m>, and C<p>, and C<s> are all illegal.
1033 Specifying modifiers like C<qr/.../gc> that aren't legal in the
1034 C<(?...)> notation normally raise a warning, but with wildcard
1035 subpatterns, their use is an error. The C<m> modifier is ineffective;
1036 everything that matches will be a single line.
1038 By default, your pattern is matched case-insensitively, as if C</i> had
1039 been specified. You can change this by saying C<(?-i)> in your pattern.
1041 There are also certain operations that are illegal. You can't nest
1042 C<\p{...}> and C<\P{...}> calls within a wildcard subpattern, and C<\G>
1043 doesn't make sense, so is also prohibited.
1045 And the C<*> quantifier (or its equivalent C<(0,}>) is illegal.
1047 This feature is not available when the left-hand side is prefixed by
1048 C<Is_>, nor for any form that is marked as "Discouraged" in
1049 L<perluniprops/Discouraged>.
1051 This experimental feature has been added to begin to implement
1052 L<https://www.unicode.org/reports/tr18/#Wildcard_Properties>. Using it
1053 will raise a (default-on) warning in the
1054 C<experimental::uniprop_wildcards> category. We reserve the right to
1055 change its operation as we gain experience.
1057 Your subpattern can be just about anything, but for it to have some
1058 utility, it should match when called with either or both of
1059 a) the full name of the property value with underscores (and/or spaces
1060 in the Block property) and some things uppercase; or b) the property
1061 value in all lowercase with spaces and underscores squeezed out. For
1064 qr!\p{Blk=/Old I.*/}!
1065 qr!\p{Blk=/oldi.*/}!
1067 would match the same things.
1069 Another example that shows that within C<\p{...}>, C</x> isn't needed to
1072 qr!\p{scx= /Hebrew|Greek/ }!
1074 To be safe, we should have anchored the above example, to prevent
1075 matches for something like C<Hebrew_Braille>, but there aren't
1076 any script names like that, so far.
1077 A warning is issued if none of the legal values for a property are
1078 matched by your pattern. It's likely that a future release will raise a
1079 warning if your pattern ends up causing every possible code point to
1082 Starting in 5.32, the Name, Name Aliases, and Named Sequences properties
1083 are allowed to be matched. They are considered to be a single
1084 combination property, just as has long been the case for C<\N{}>. Loose
1085 matching doesn't work in exactly the same way for these as it does for
1086 the values of other properties. The rules are given in
1087 L<https://www.unicode.org/reports/tr44/tr44-24.html#UAX44-LM2>. As a
1088 result, Perl doesn't try loose matching for you, like it does in other
1089 properties. All letters in names are uppercase, but you can add C<(?i)>
1090 to your subpattern to ignore case. If you're uncertain where a blank
1091 is, you can use C< ?> in your subpattern. No character name contains an
1092 underscore, so don't bother trying to match one. The use of hyphens is
1093 particularly problematic; refer to the above link. But note that, as of
1094 Unicode 13.0, the only script in modern usage which has weirdnesses with
1095 these is Tibetan; also the two Korean characters U+116C HANGUL JUNGSEONG
1096 OE and U+1180 HANGUL JUNGSEONG O-E. Unicode makes no promises to not
1097 add hyphen-problematic names in the future.
1099 Using wildcards on these is resource intensive, given the hundreds of
1100 thousands of legal names that must be checked against.
1102 An example of using Name property wildcards is
1104 qr!\p{name=/(SMILING|GRINNING) FACE/}!
1108 qr/(?[ \p{name=\/CJK\/} - \p{ideographic} ])/
1110 which is the 200-ish (as of Unicode 13.0) CJK characters that aren't
1113 There are certain properties that wildcard subpatterns don't currently
1114 work with. These are:
1116 Bidi Mirroring Glyph
1119 Decomposition Mapping
1120 Equivalent Unified Ideograph
1126 Nor is the C<@I<unicode_property>@> form implemented.
1128 Here's a complete example of matching IPV4 internet protocol addresses
1129 in any (single) script
1131 no warnings 'experimental::uniprop_wildcards';
1133 # Can match a substring, so this intermediate regex needs to have
1134 # context or anchoring in its final use. Using nt=de yields decimal
1135 # digits. When specifying a subset of these, we must include \d to
1136 # prevent things like U+00B2 SUPERSCRIPT TWO from matching
1137 my $zero_through_255 =
1138 qr/ \b (*sr: # All from same sript
1139 (?[ \p{nv=0} & \d ])* # Optional leading zeros
1142 | (?[ \p{nv=1} & \d ]) \d{2} # 100 - 199
1143 | (?[ \p{nv=2} & \d ])
1144 ( (?[ \p{nv=:[0-4]:} & \d ]) \d # 200 - 249
1145 | (?[ \p{nv=5} & \d ])
1146 (?[ \p{nv=:[0-5]:} & \d ]) # 250 - 255
1153 my $ipv4 = qr/ \A (*sr: $zero_through_255
1154 (?: [.] $zero_through_255 ) {3}
1159 =head2 User-Defined Character Properties
1161 You can define your own binary character properties by defining subroutines
1162 whose names begin with C<"In"> or C<"Is">. (The regex sets feature
1163 L<perlre/(?[ ])> provides an alternative which allows more complex
1164 definitions.) The subroutines can be defined in any
1165 package. They override any Unicode properties expressed as the same
1166 names. The user-defined properties can be used in the regular
1168 C<\p{}> and C<\P{}> constructs; if you are using a user-defined property from a
1169 package other than the one you are in, you must specify its package in the
1170 C<\p{}> or C<\P{}> construct.
1172 # assuming property IsForeign defined in Lang::
1173 package main; # property package name required
1174 if ($txt =~ /\p{Lang::IsForeign}+/) { ... }
1176 package Lang; # property package name not required
1177 if ($txt =~ /\p{IsForeign}+/) { ... }
1180 Note that the effect is compile-time and immutable once defined.
1181 However, the subroutines are passed a single parameter, which is 0 if
1182 case-sensitive matching is in effect and non-zero if caseless matching
1183 is in effect. The subroutine may return different values depending on
1184 the value of the flag, and one set of values will immutably be in effect
1185 for all case-sensitive matches, and the other set for all case-insensitive
1188 Note that if the regular expression is tainted, then Perl will die rather
1189 than calling the subroutine when the name of the subroutine is
1190 determined by the tainted data.
1192 The subroutines must return a specially-formatted string, with one
1193 or more newline-separated lines. Each line must be one of the following:
1199 A single hexadecimal number denoting a code point to include.
1203 Two hexadecimal numbers separated by horizontal whitespace (space or
1204 tabular characters) denoting a range of code points to include. The
1205 second number must not be smaller than the first.
1209 Something to include, prefixed by C<"+">: a built-in character
1210 property (prefixed by C<"utf8::">) or a fully qualified (including package
1211 name) user-defined character property,
1212 to represent all the characters in that property; two hexadecimal code
1213 points for a range; or a single hexadecimal code point.
1217 Something to exclude, prefixed by C<"-">: an existing character
1218 property (prefixed by C<"utf8::">) or a fully qualified (including package
1219 name) user-defined character property,
1220 to represent all the characters in that property; two hexadecimal code
1221 points for a range; or a single hexadecimal code point.
1225 Something to negate, prefixed C<"!">: an existing character
1226 property (prefixed by C<"utf8::">) or a fully qualified (including package
1227 name) user-defined character property,
1228 to represent all the characters in that property; two hexadecimal code
1229 points for a range; or a single hexadecimal code point.
1233 Something to intersect with, prefixed by C<"&">: an existing character
1234 property (prefixed by C<"utf8::">) or a fully qualified (including package
1235 name) user-defined character property,
1236 for all the characters except the characters in the property; two
1237 hexadecimal code points for a range; or a single hexadecimal code point.
1241 For example, to define a property that covers both the Japanese
1242 syllabaries (hiragana and katakana), you can define
1251 Imagine that the here-doc end marker is at the beginning of the line.
1252 Now you can use C<\p{InKana}> and C<\P{InKana}>.
1254 You could also have used the existing block property names:
1263 Suppose you wanted to match only the allocated characters,
1264 not the raw block ranges: in other words, you want to remove
1265 the unassigned characters:
1275 The negation is useful for defining (surprise!) negated classes.
1285 This will match all non-Unicode code points, since every one of them is
1286 not in Kana. You can use intersection to exclude these, if desired, as
1287 this modified example shows:
1298 C<&utf8::Any> must be the last line in the definition.
1300 Intersection is used generally for getting the common characters matched
1301 by two (or more) classes. It's important to remember not to use C<"&"> for
1302 the first set; that would be intersecting with nothing, resulting in an
1303 empty set. (Similarly using C<"-"> for the first set does nothing).
1305 Unlike non-user-defined C<\p{}> property matches, no warning is ever
1306 generated if these properties are matched against a non-Unicode code
1307 point (see L</Beyond Unicode code points> below).
1309 =head2 User-Defined Case Mappings (for serious hackers only)
1311 B<This feature has been removed as of Perl 5.16.>
1312 The CPAN module C<L<Unicode::Casing>> provides better functionality without
1313 the drawbacks that this feature had. If you are using a Perl earlier
1314 than 5.16, this feature was most fully documented in the 5.14 version of
1316 L<http://perldoc.perl.org/5.14.0/perlunicode.html#User-Defined-Case-Mappings-%28for-serious-hackers-only%29>
1318 =head2 Character Encodings for Input and Output
1322 =head2 Unicode Regular Expression Support Level
1324 The following list of Unicode supported features for regular expressions describes
1325 all features currently directly supported by core Perl. The references
1326 to "Level I<N>" and the section numbers refer to
1327 L<UTS#18 "Unicode Regular Expressions"|https://www.unicode.org/reports/tr18>,
1328 version 18, October 2016.
1330 =head3 Level 1 - Basic Unicode Support
1332 RL1.1 Hex Notation - Done [1]
1333 RL1.2 Properties - Done [2]
1334 RL1.2a Compatibility Properties - Done [3]
1335 RL1.3 Subtraction and Intersection - Done [4]
1336 RL1.4 Simple Word Boundaries - Done [5]
1337 RL1.5 Simple Loose Matches - Done [6]
1338 RL1.6 Line Boundaries - Partial [7]
1339 RL1.7 Supplementary Code Points - Done [8]
1343 =item [1] C<\N{U+...}> and C<\x{...}>
1346 C<\p{...}> C<\P{...}>. This requirement is for a minimal list of
1347 properties. Perl supports these. See R2.7 for other properties.
1351 Perl has C<\d> C<\D> C<\s> C<\S> C<\w> C<\W> C<\X> C<[:I<prop>:]>
1352 C<[:^I<prop>:]>, plus all the properties specified by
1353 L<https://www.unicode.org/reports/tr18/#Compatibility_Properties>. These
1354 are described above in L</Other Properties>
1358 The regex sets feature C<"(?[...])"> starting in v5.18 accomplishes
1359 this. See L<perlre/(?[ ])>.
1362 C<\b> C<\B> meet most, but not all, the details of this requirement, but
1363 C<\b{wb}> and C<\B{wb}> do, as well as the stricter R2.3.
1367 Note that Perl does Full case-folding in matching, not Simple:
1369 For example C<U+1F88> is equivalent to C<U+1F00 U+03B9>, instead of just
1370 C<U+1F80>. This difference matters mainly for certain Greek capital
1371 letters with certain modifiers: the Full case-folding decomposes the
1372 letter, while the Simple case-folding would map it to a single
1377 The reason this is considered to be only partially implemented is that
1378 Perl has L<C<qrE<sol>\b{lb}E<sol>>|perlrebackslash/\b{lb}> and
1379 C<L<Unicode::LineBreak>> that are conformant with
1380 L<UAX#14 "Unicode Line Breaking Algorithm"|https://www.unicode.org/reports/tr14>.
1381 The regular expression construct provides default behavior, while the
1382 heavier-weight module provides customizable line breaking.
1384 But Perl treats C<\n> as the start- and end-line
1385 delimiter, whereas Unicode specifies more characters that should be
1397 C<^> and C<$> in regular expression patterns are supposed to match all
1399 These characters also don't, but should, affect C<< <> >> C<$.>, and
1400 script line numbers.
1402 Also, lines should not be split within C<CRLF> (i.e. there is no
1403 empty line between C<\r> and C<\n>). For C<CRLF>, try the C<:crlf>
1404 layer (see L<PerlIO>).
1407 UTF-8/UTF-EBDDIC used in Perl allows not only C<U+10000> to
1408 C<U+10FFFF> but also beyond C<U+10FFFF>
1412 =head3 Level 2 - Extended Unicode Support
1414 RL2.1 Canonical Equivalents - Retracted [9]
1416 RL2.2 Extended Grapheme Clusters and - Partial [10]
1417 Character Classes with Strings
1418 RL2.3 Default Word Boundaries - Done [11]
1419 RL2.4 Default Case Conversion - Done
1420 RL2.5 Name Properties - Done
1421 RL2.6 Wildcards in Property Values - Partial [12]
1422 RL2.7 Full Properties - Partial [13]
1423 RL2.8 Optional Properties - Partial [14]
1428 Unicode has rewritten this portion of UTS#18 to say that getting
1429 canonical equivalence (see UAX#15
1430 L<"Unicode Normalization Forms"|https://www.unicode.org/reports/tr15>)
1431 is basically to be done at the programmer level. Use NFD to write
1432 both your regular expressions and text to match them against (you
1433 can use L<Unicode::Normalize>).
1436 Perl has C<\X> and C<\b{gcb}>. Unicode has retracted their "Grapheme
1437 Cluster Mode", and recently added string properties, which Perl does not
1441 L<UAX#29 "Unicode Text Segmentation"|https://www.unicode.org/reports/tr29>,
1444 L</Wildcards in Property Values> above.
1447 Perl supports all the properties in the Unicode Character Database
1448 (UCD). It does not yet support the listed properties that come from
1449 other Unicode sources.
1452 The only optional property that Perl supports is Named Sequence. None
1453 of these properties are in the UCD.
1457 =head3 Level 3 - Tailored Support
1459 This has been retracted by Unicode.
1461 =head2 Unicode Encodings
1463 Unicode characters are assigned to I<code points>, which are abstract
1464 numbers. To use these numbers, various encodings are needed.
1472 UTF-8 is a variable-length (1 to 4 bytes), byte-order independent
1473 encoding. In most of Perl's documentation, including elsewhere in this
1474 document, the term "UTF-8" means also "UTF-EBCDIC". But in this section,
1475 "UTF-8" refers only to the encoding used on ASCII platforms. It is a
1476 superset of 7-bit US-ASCII, so anything encoded in ASCII has the
1477 identical representation when encoded in UTF-8.
1479 The following table is from Unicode 3.2.
1481 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
1483 U+0000..U+007F 00..7F
1484 U+0080..U+07FF * C2..DF 80..BF
1485 U+0800..U+0FFF E0 * A0..BF 80..BF
1486 U+1000..U+CFFF E1..EC 80..BF 80..BF
1487 U+D000..U+D7FF ED 80..9F 80..BF
1488 U+D800..U+DFFF +++++ utf16 surrogates, not legal utf8 +++++
1489 U+E000..U+FFFF EE..EF 80..BF 80..BF
1490 U+10000..U+3FFFF F0 * 90..BF 80..BF 80..BF
1491 U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
1492 U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
1494 Note the gaps marked by "*" before several of the byte entries above. These are
1495 caused by legal UTF-8 avoiding non-shortest encodings: it is technically
1496 possible to UTF-8-encode a single code point in different ways, but that is
1497 explicitly forbidden, and the shortest possible encoding should always be used
1498 (and that is what Perl does).
1500 Another way to look at it is via bits:
1502 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
1505 00000bbbbbaaaaaa 110bbbbb 10aaaaaa
1506 ccccbbbbbbaaaaaa 1110cccc 10bbbbbb 10aaaaaa
1507 00000dddccccccbbbbbbaaaaaa 11110ddd 10cccccc 10bbbbbb 10aaaaaa
1509 As you can see, the continuation bytes all begin with C<"10">, and the
1510 leading bits of the start byte tell how many bytes there are in the
1513 The original UTF-8 specification allowed up to 6 bytes, to allow
1514 encoding of numbers up to C<0x7FFF_FFFF>. Perl continues to allow those,
1515 and has extended that up to 13 bytes to encode code points up to what
1516 can fit in a 64-bit word. However, Perl will warn if you output any of
1517 these as being non-portable; and under strict UTF-8 input protocols,
1518 they are forbidden. In addition, it is now illegal to use a code point
1519 larger than what a signed integer variable on your system can hold. On
1520 32-bit ASCII systems, this means C<0x7FFF_FFFF> is the legal maximum
1521 (much higher on 64-bit systems).
1527 Like UTF-8, but EBCDIC-safe, in the way that UTF-8 is ASCII-safe.
1528 This means that all the basic characters (which includes all
1529 those that have ASCII equivalents (like C<"A">, C<"0">, C<"%">, I<etc.>)
1530 are the same in both EBCDIC and UTF-EBCDIC.)
1532 UTF-EBCDIC is used on EBCDIC platforms. It generally requires more
1533 bytes to represent a given code point than UTF-8 does; the largest
1534 Unicode code points take 5 bytes to represent (instead of 4 in UTF-8),
1535 and, extended for 64-bit words, it uses 14 bytes instead of 13 bytes in
1540 UTF-16, UTF-16BE, UTF-16LE, Surrogates, and C<BOM>'s (Byte Order Marks)
1542 The followings items are mostly for reference and general Unicode
1543 knowledge, Perl doesn't use these constructs internally.
1545 Like UTF-8, UTF-16 is a variable-width encoding, but where
1546 UTF-8 uses 8-bit code units, UTF-16 uses 16-bit code units.
1547 All code points occupy either 2 or 4 bytes in UTF-16: code points
1548 C<U+0000..U+FFFF> are stored in a single 16-bit unit, and code
1549 points C<U+10000..U+10FFFF> in two 16-bit units. The latter case is
1550 using I<surrogates>, the first 16-bit unit being the I<high
1551 surrogate>, and the second being the I<low surrogate>.
1553 Surrogates are code points set aside to encode the C<U+10000..U+10FFFF>
1554 range of Unicode code points in pairs of 16-bit units. The I<high
1555 surrogates> are the range C<U+D800..U+DBFF> and the I<low surrogates>
1556 are the range C<U+DC00..U+DFFF>. The surrogate encoding is
1558 $hi = ($uni - 0x10000) / 0x400 + 0xD800;
1559 $lo = ($uni - 0x10000) % 0x400 + 0xDC00;
1563 $uni = 0x10000 + ($hi - 0xD800) * 0x400 + ($lo - 0xDC00);
1565 Because of the 16-bitness, UTF-16 is byte-order dependent. UTF-16
1566 itself can be used for in-memory computations, but if storage or
1567 transfer is required either UTF-16BE (big-endian) or UTF-16LE
1568 (little-endian) encodings must be chosen.
1570 This introduces another problem: what if you just know that your data
1571 is UTF-16, but you don't know which endianness? Byte Order Marks, or
1572 C<BOM>'s, are a solution to this. A special character has been reserved
1573 in Unicode to function as a byte order marker: the character with the
1574 code point C<U+FEFF> is the C<BOM>.
1576 The trick is that if you read a C<BOM>, you will know the byte order,
1577 since if it was written on a big-endian platform, you will read the
1578 bytes C<0xFE 0xFF>, but if it was written on a little-endian platform,
1579 you will read the bytes C<0xFF 0xFE>. (And if the originating platform
1580 was writing in ASCII platform UTF-8, you will read the bytes
1583 The way this trick works is that the character with the code point
1584 C<U+FFFE> is not supposed to be in input streams, so the
1585 sequence of bytes C<0xFF 0xFE> is unambiguously "C<BOM>, represented in
1586 little-endian format" and cannot be C<U+FFFE>, represented in big-endian
1589 Surrogates have no meaning in Unicode outside their use in pairs to
1590 represent other code points. However, Perl allows them to be
1591 represented individually internally, for example by saying
1592 C<chr(0xD801)>, so that all code points, not just those valid for open
1594 representable. Unicode does define semantics for them, such as their
1595 C<L</General_Category>> is C<"Cs">. But because their use is somewhat dangerous,
1596 Perl will warn (using the warning category C<"surrogate">, which is a
1597 sub-category of C<"utf8">) if an attempt is made
1598 to do things like take the lower case of one, or match
1599 case-insensitively, or to output them. (But don't try this on Perls
1604 UTF-32, UTF-32BE, UTF-32LE
1606 The UTF-32 family is pretty much like the UTF-16 family, except that
1607 the units are 32-bit, and therefore the surrogate scheme is not
1608 needed. UTF-32 is a fixed-width encoding. The C<BOM> signatures are
1609 C<0x00 0x00 0xFE 0xFF> for BE and C<0xFF 0xFE 0x00 0x00> for LE.
1615 Legacy, fixed-width encodings defined by the ISO 10646 standard. UCS-2 is a 16-bit
1616 encoding. Unlike UTF-16, UCS-2 is not extensible beyond C<U+FFFF>,
1617 because it does not use surrogates. UCS-4 is a 32-bit encoding,
1618 functionally identical to UTF-32 (the difference being that
1619 UCS-4 forbids neither surrogates nor code points larger than C<0x10_FFFF>).
1625 A seven-bit safe (non-eight-bit) encoding, which is useful if the
1626 transport or storage is not eight-bit safe. Defined by RFC 2152.
1630 =head2 Noncharacter code points
1632 66 code points are set aside in Unicode as "noncharacter code points".
1633 These all have the C<Unassigned> (C<Cn>) C<L</General_Category>>, and
1634 no character will ever be assigned to any of them. They are the 32 code
1635 points between C<U+FDD0> and C<U+FDEF> inclusive, and the 34 code
1646 Until Unicode 7.0, the noncharacters were "B<forbidden> for use in open
1647 interchange of Unicode text data", so that code that processed those
1648 streams could use these code points as sentinels that could be mixed in
1649 with character data, and would always be distinguishable from that data.
1650 (Emphasis above and in the next paragraph are added in this document.)
1652 Unicode 7.0 changed the wording so that they are "B<not recommended> for
1653 use in open interchange of Unicode text data". The 7.0 Standard goes on
1658 "If a noncharacter is received in open interchange, an application is
1659 not required to interpret it in any way. It is good practice, however,
1660 to recognize it as a noncharacter and to take appropriate action, such
1661 as replacing it with C<U+FFFD> replacement character, to indicate the
1662 problem in the text. It is not recommended to simply delete
1663 noncharacter code points from such text, because of the potential
1664 security issues caused by deleting uninterpreted characters. (See
1665 conformance clause C7 in Section 3.2, Conformance Requirements, and
1666 L<Unicode Technical Report #36, "Unicode Security
1667 Considerations"|https://www.unicode.org/reports/tr36/#Substituting_for_Ill_Formed_Subsequences>)."
1671 This change was made because it was found that various commercial tools
1672 like editors, or for things like source code control, had been written
1673 so that they would not handle program files that used these code points,
1674 effectively precluding their use almost entirely! And that was never
1675 the intent. They've always been meant to be usable within an
1676 application, or cooperating set of applications, at will.
1678 If you're writing code, such as an editor, that is supposed to be able
1679 to handle any Unicode text data, then you shouldn't be using these code
1680 points yourself, and instead allow them in the input. If you need
1681 sentinels, they should instead be something that isn't legal Unicode.
1682 For UTF-8 data, you can use the bytes 0xC1 and 0xC2 as sentinels, as
1683 they never appear in well-formed UTF-8. (There are equivalents for
1684 UTF-EBCDIC). You can also store your Unicode code points in integer
1685 variables and use negative values as sentinels.
1687 If you're not writing such a tool, then whether you accept noncharacters
1688 as input is up to you (though the Standard recommends that you not). If
1689 you do strict input stream checking with Perl, these code points
1690 continue to be forbidden. This is to maintain backward compatibility
1691 (otherwise potential security holes could open up, as an unsuspecting
1692 application that was written assuming the noncharacters would be
1693 filtered out before getting to it, could now, without warning, start
1694 getting them). To do strict checking, you can use the layer
1695 C<:encoding('UTF-8')>.
1697 Perl continues to warn (using the warning category C<"nonchar">, which
1698 is a sub-category of C<"utf8">) if an attempt is made to output
1701 =head2 Beyond Unicode code points
1703 The maximum Unicode code point is C<U+10FFFF>, and Unicode only defines
1704 operations on code points up through that. But Perl works on code
1705 points up to the maximum permissible signed number available on the
1706 platform. However, Perl will not accept these from input streams unless
1707 lax rules are being used, and will warn (using the warning category
1708 C<"non_unicode">, which is a sub-category of C<"utf8">) if any are output.
1710 Since Unicode rules are not defined on these code points, if a
1711 Unicode-defined operation is done on them, Perl uses what we believe are
1712 sensible rules, while generally warning, using the C<"non_unicode">
1713 category. For example, C<uc("\x{11_0000}")> will generate such a
1714 warning, returning the input parameter as its result, since Perl defines
1715 the uppercase of every non-Unicode code point to be the code point
1716 itself. (All the case changing operations, not just uppercasing, work
1719 The situation with matching Unicode properties in regular expressions,
1720 the C<\p{}> and C<\P{}> constructs, against these code points is not as
1721 clear cut, and how these are handled has changed as we've gained
1724 One possibility is to treat any match against these code points as
1725 undefined. But since Perl doesn't have the concept of a match being
1726 undefined, it converts this to failing or C<FALSE>. This is almost, but
1727 not quite, what Perl did from v5.14 (when use of these code points
1728 became generally reliable) through v5.18. The difference is that Perl
1729 treated all C<\p{}> matches as failing, but all C<\P{}> matches as
1732 One problem with this is that it leads to unexpected, and confusing
1733 results in some cases:
1735 chr(0x110000) =~ \p{ASCII_Hex_Digit=True} # Failed on <= v5.18
1736 chr(0x110000) =~ \p{ASCII_Hex_Digit=False} # Failed! on <= v5.18
1738 That is, it treated both matches as undefined, and converted that to
1739 false (raising a warning on each). The first case is the expected
1740 result, but the second is likely counterintuitive: "How could both be
1741 false when they are complements?" Another problem was that the
1742 implementation optimized many Unicode property matches down to already
1743 existing simpler, faster operations, which don't raise the warning. We
1744 chose to not forgo those optimizations, which help the vast majority of
1745 matches, just to generate a warning for the unlikely event that an
1746 above-Unicode code point is being matched against.
1748 As a result of these problems, starting in v5.20, what Perl does is
1749 to treat non-Unicode code points as just typical unassigned Unicode
1750 characters, and matches accordingly. (Note: Unicode has atypical
1751 unassigned code points. For example, it has noncharacter code points,
1752 and ones that, when they do get assigned, are destined to be written
1753 Right-to-left, as Arabic and Hebrew are. Perl assumes that no
1754 non-Unicode code point has any atypical properties.)
1756 Perl, in most cases, will raise a warning when matching an above-Unicode
1757 code point against a Unicode property when the result is C<TRUE> for
1758 C<\p{}>, and C<FALSE> for C<\P{}>. For example:
1760 chr(0x110000) =~ \p{ASCII_Hex_Digit=True} # Fails, no warning
1761 chr(0x110000) =~ \p{ASCII_Hex_Digit=False} # Succeeds, with warning
1763 In both these examples, the character being matched is non-Unicode, so
1764 Unicode doesn't define how it should match. It clearly isn't an ASCII
1765 hex digit, so the first example clearly should fail, and so it does,
1766 with no warning. But it is arguable that the second example should have
1767 an undefined, hence C<FALSE>, result. So a warning is raised for it.
1769 Thus the warning is raised for many fewer cases than in earlier Perls,
1770 and only when what the result is could be arguable. It turns out that
1771 none of the optimizations made by Perl (or are ever likely to be made)
1772 cause the warning to be skipped, so it solves both problems of Perl's
1773 earlier approach. The most commonly used property that is affected by
1774 this change is C<\p{Unassigned}> which is a short form for
1775 C<\p{General_Category=Unassigned}>. Starting in v5.20, all non-Unicode
1776 code points are considered C<Unassigned>. In earlier releases the
1777 matches failed because the result was considered undefined.
1779 The only place where the warning is not raised when it might ought to
1780 have been is if optimizations cause the whole pattern match to not even
1781 be attempted. For example, Perl may figure out that for a string to
1782 match a certain regular expression pattern, the string has to contain
1783 the substring C<"foobar">. Before attempting the match, Perl may look
1784 for that substring, and if not found, immediately fail the match without
1785 actually trying it; so no warning gets generated even if the string
1786 contains an above-Unicode code point.
1788 This behavior is more "Do what I mean" than in earlier Perls for most
1789 applications. But it catches fewer issues for code that needs to be
1790 strictly Unicode compliant. Therefore there is an additional mode of
1791 operation available to accommodate such code. This mode is enabled if a
1792 regular expression pattern is compiled within the lexical scope where
1793 the C<"non_unicode"> warning class has been made fatal, say by:
1795 use warnings FATAL => "non_unicode"
1797 (see L<warnings>). In this mode of operation, Perl will raise the
1798 warning for all matches against a non-Unicode code point (not just the
1799 arguable ones), and it skips the optimizations that might cause the
1800 warning to not be output. (It currently still won't warn if the match
1801 isn't even attempted, like in the C<"foobar"> example above.)
1803 In summary, Perl now normally treats non-Unicode code points as typical
1804 Unicode unassigned code points for regular expression matches, raising a
1805 warning only when it is arguable what the result should be. However, if
1806 this warning has been made fatal, it isn't skipped.
1808 There is one exception to all this. C<\p{All}> looks like a Unicode
1809 property, but it is a Perl extension that is defined to be true for all
1810 possible code points, Unicode or not, so no warning is ever generated
1811 when matching this against a non-Unicode code point. (Prior to v5.20,
1812 it was an exact synonym for C<\p{Any}>, matching code points C<0>
1813 through C<0x10FFFF>.)
1815 =head2 Security Implications of Unicode
1818 L<Unicode Security Considerations|https://www.unicode.org/reports/tr36>.
1820 Also, note the following:
1828 UTF-8 is very structured, so many combinations of bytes are invalid. In
1829 the past, Perl tried to soldier on and make some sense of invalid
1830 combinations, but this can lead to security holes, so now, if the Perl
1831 core needs to process an invalid combination, it will either raise a
1832 fatal error, or will replace those bytes by the sequence that forms the
1833 Unicode REPLACEMENT CHARACTER, for which purpose Unicode created it.
1835 Every code point can be represented by more than one possible
1836 syntactically valid UTF-8 sequence. Early on, both Unicode and Perl
1837 considered any of these to be valid, but now, all sequences longer
1838 than the shortest possible one are considered to be malformed.
1840 Unicode considers many code points to be illegal, or to be avoided.
1841 Perl generally accepts them, once they have passed through any input
1842 filters that may try to exclude them. These have been discussed above
1843 (see "Surrogates" under UTF-16 in L</Unicode Encodings>,
1844 L</Noncharacter code points>, and L</Beyond Unicode code points>).
1848 Regular expression pattern matching may surprise you if you're not
1849 accustomed to Unicode. Starting in Perl 5.14, several pattern
1850 modifiers are available to control this, called the character set
1851 modifiers. Details are given in L<perlre/Character set modifiers>.
1855 As discussed elsewhere, Perl has one foot (two hooves?) planted in
1856 each of two worlds: the old world of ASCII and single-byte locales, and
1857 the new world of Unicode, upgrading when necessary.
1858 If your legacy code does not explicitly use Unicode, no automatic
1859 switch-over to Unicode should happen.
1861 =head2 Unicode in Perl on EBCDIC
1863 Unicode is supported on EBCDIC platforms. See L<perlebcdic>.
1865 Unless ASCII vs. EBCDIC issues are specifically being discussed,
1866 references to UTF-8 encoding in this document and elsewhere should be
1867 read as meaning UTF-EBCDIC on EBCDIC platforms.
1868 See L<perlebcdic/Unicode and UTF>.
1870 Because UTF-EBCDIC is so similar to UTF-8, the differences are mostly
1871 hidden from you; S<C<use utf8>> (and NOT something like
1872 S<C<use utfebcdic>>) declares the script is in the platform's
1873 "native" 8-bit encoding of Unicode. (Similarly for the C<":utf8">
1878 See L<perllocale/Unicode and UTF-8>
1880 =head2 When Unicode Does Not Happen
1882 There are still many places where Unicode (in some encoding or
1883 another) could be given as arguments or received as results, or both in
1884 Perl, but it is not, in spite of Perl having extensive ways to input and
1885 output in Unicode, and a few other "entry points" like the C<@ARGV>
1886 array (which can sometimes be interpreted as UTF-8).
1888 The following are such interfaces. Also, see L</The "Unicode Bug">.
1889 For all of these interfaces Perl
1890 currently (as of v5.16.0) simply assumes byte strings both as arguments
1891 and results, or UTF-8 strings if the (deprecated) C<encoding> pragma has been used.
1893 One reason that Perl does not attempt to resolve the role of Unicode in
1894 these situations is that the answers are highly dependent on the operating
1895 system and the file system(s). For example, whether filenames can be
1896 in Unicode and in exactly what kind of encoding, is not exactly a
1897 portable concept. Similarly for C<qx> and C<system>: how well will the
1898 "command-line interface" (and which of them?) handle Unicode?
1904 C<chdir>, C<chmod>, C<chown>, C<chroot>, C<exec>, C<link>, C<lstat>, C<mkdir>,
1905 C<rename>, C<rmdir>, C<stat>, C<symlink>, C<truncate>, C<unlink>, C<utime>, C<-X>
1913 C<glob> (aka the C<E<lt>*E<gt>>)
1917 C<open>, C<opendir>, C<sysopen>
1921 C<qx> (aka the backtick operator), C<system>
1925 C<readdir>, C<readlink>
1929 =head2 The "Unicode Bug"
1931 The term, "Unicode bug" has been applied to an inconsistency with the
1932 code points in the C<Latin-1 Supplement> block, that is, between
1933 128 and 255. Without a locale specified, unlike all other characters or
1934 code points, these characters can have very different semantics
1935 depending on the rules in effect. (Characters whose code points are
1936 above 255 force Unicode rules; whereas the rules for ASCII characters
1937 are the same under both ASCII and Unicode rules.)
1939 Under Unicode rules, these upper-Latin1 characters are interpreted as
1940 Unicode code points, which means they have the same semantics as Latin-1
1941 (ISO-8859-1) and C1 controls.
1943 As explained in L</ASCII Rules versus Unicode Rules>, under ASCII rules,
1944 they are considered to be unassigned characters.
1946 This can lead to unexpected results. For example, a string's
1947 semantics can suddenly change if a code point above 255 is appended to
1948 it, which changes the rules from ASCII to Unicode. As an
1949 example, consider the following program and its output:
1952 no feature "unicode_strings";
1955 for ($s1, $s2, $s1.$s2) {
1963 If there's no C<\w> in C<s1> nor in C<s2>, why does their concatenation
1966 This anomaly stems from Perl's attempt to not disturb older programs that
1967 didn't use Unicode, along with Perl's desire to add Unicode support
1968 seamlessly. But the result turned out to not be seamless. (By the way,
1969 you can choose to be warned when things like this happen. See
1970 C<L<encoding::warnings>>.)
1972 L<S<C<use feature 'unicode_strings'>>|feature/The 'unicode_strings' feature>
1973 was added, starting in Perl v5.12, to address this problem. It affects
1980 Changing the case of a scalar, that is, using C<uc()>, C<ucfirst()>, C<lc()>,
1981 and C<lcfirst()>, or C<\L>, C<\U>, C<\u> and C<\l> in double-quotish
1982 contexts, such as regular expression substitutions.
1984 Under C<unicode_strings> starting in Perl 5.12.0, Unicode rules are
1985 generally used. See L<perlfunc/lc> for details on how this works
1986 in combination with various other pragmas.
1990 Using caseless (C</i>) regular expression matching.
1992 Starting in Perl 5.14.0, regular expressions compiled within
1993 the scope of C<unicode_strings> use Unicode rules
1994 even when executed or compiled into larger
1995 regular expressions outside the scope.
1999 Matching any of several properties in regular expressions.
2001 These properties are C<\b> (without braces), C<\B> (without braces),
2002 C<\s>, C<\S>, C<\w>, C<\W>, and all the Posix character classes
2003 I<except> C<[[:ascii:]]>.
2005 Starting in Perl 5.14.0, regular expressions compiled within
2006 the scope of C<unicode_strings> use Unicode rules
2007 even when executed or compiled into larger
2008 regular expressions outside the scope.
2012 In C<quotemeta> or its inline equivalent C<\Q>.
2014 Starting in Perl 5.16.0, consistent quoting rules are used within the
2015 scope of C<unicode_strings>, as described in L<perlfunc/quotemeta>.
2016 Prior to that, or outside its scope, no code points above 127 are quoted
2017 in UTF-8 encoded strings, but in byte encoded strings, code points
2018 between 128-255 are always quoted.
2022 In the C<..> or L<range|perlop/Range Operators> operator.
2024 Starting in Perl 5.26.0, the range operator on strings treats their lengths
2025 consistently within the scope of C<unicode_strings>. Prior to that, or
2026 outside its scope, it could produce strings whose length in characters
2027 exceeded that of the right-hand side, where the right-hand side took up more
2028 bytes than the correct range endpoint.
2032 In L<< C<split>'s special-case whitespace splitting|perlfunc/split >>.
2034 Starting in Perl 5.28.0, the C<split> function with a pattern specified as
2035 a string containing a single space handles whitespace characters consistently
2036 within the scope of C<unicode_strings>. Prior to that, or outside its scope,
2037 characters that are whitespace according to Unicode rules but not according to
2038 ASCII rules were treated as field contents rather than field separators when
2039 they appear in byte-encoded strings.
2043 You can see from the above that the effect of C<unicode_strings>
2044 increased over several Perl releases. (And Perl's support for Unicode
2045 continues to improve; it's best to use the latest available release in
2046 order to get the most complete and accurate results possible.) Note that
2047 C<unicode_strings> is automatically chosen if you S<C<use 5.012>> or
2050 For Perls earlier than those described above, or when a string is passed
2051 to a function outside the scope of C<unicode_strings>, see the next section.
2053 =head2 Forcing Unicode in Perl (Or Unforcing Unicode in Perl)
2055 Sometimes (see L</"When Unicode Does Not Happen"> or L</The "Unicode Bug">)
2056 there are situations where you simply need to force a byte
2057 string into UTF-8, or vice versa. The standard module L<Encode> can be
2058 used for this, or the low-level calls
2059 L<C<utf8::upgrade($bytestring)>|utf8/Utility functions> and
2060 L<C<utf8::downgrade($utf8string[, FAIL_OK])>|utf8/Utility functions>.
2062 Note that C<utf8::downgrade()> can fail if the string contains characters
2063 that don't fit into a byte.
2065 Calling either function on a string that already is in the desired state is a
2068 L</ASCII Rules versus Unicode Rules> gives all the ways that a string is
2069 made to use Unicode rules.
2071 =head2 Using Unicode in XS
2073 See L<perlguts/"Unicode Support"> for an introduction to Unicode at
2074 the XS level, and L<perlapi/Unicode Support> for the API details.
2076 =head2 Hacking Perl to work on earlier Unicode versions (for very serious hackers only)
2078 Perl by default comes with the latest supported Unicode version built-in, but
2079 the goal is to allow you to change to use any earlier one. In Perls
2080 v5.20 and v5.22, however, the earliest usable version is Unicode 5.1.
2081 Perl v5.18 and v5.24 are able to handle all earlier versions.
2083 Download the files in the desired version of Unicode from the Unicode web
2084 site L<https://www.unicode.org>). These should replace the existing files in
2085 F<lib/unicore> in the Perl source tree. Follow the instructions in
2086 F<README.perl> in that directory to change some of their names, and then build
2087 perl (see L<INSTALL>).
2089 =head2 Porting code from perl-5.6.X
2091 Perls starting in 5.8 have a different Unicode model from 5.6. In 5.6 the
2092 programmer was required to use the C<utf8> pragma to declare that a
2093 given scope expected to deal with Unicode data and had to make sure that
2094 only Unicode data were reaching that scope. If you have code that is
2095 working with 5.6, you will need some of the following adjustments to
2096 your code. The examples are written such that the code will continue to
2097 work under 5.6, so you should be safe to try them out.
2103 A filehandle that should read or write UTF-8
2106 binmode $fh, ":encoding(UTF-8)";
2111 A scalar that is going to be passed to some extension
2113 Be it C<Compress::Zlib>, C<Apache::Request> or any extension that has no
2114 mention of Unicode in the manpage, you need to make sure that the
2115 UTF8 flag is stripped off. Note that at the time of this writing
2116 (January 2012) the mentioned modules are not UTF-8-aware. Please
2117 check the documentation to verify if this is still true.
2121 $val = Encode::encode("UTF-8", $val); # make octets
2126 A scalar we got back from an extension
2128 If you believe the scalar comes back as UTF-8, you will most likely
2129 want the UTF8 flag restored:
2133 $val = Encode::decode("UTF-8", $val);
2138 Same thing, if you are really sure it is UTF-8
2142 Encode::_utf8_on($val);
2147 A wrapper for L<DBI> C<fetchrow_array> and C<fetchrow_hashref>
2149 When the database contains only UTF-8, a wrapper function or method is
2150 a convenient way to replace all your C<fetchrow_array> and
2151 C<fetchrow_hashref> calls. A wrapper function will also make it easier to
2152 adapt to future enhancements in your database driver. Note that at the
2153 time of this writing (January 2012), the DBI has no standardized way
2154 to deal with UTF-8 data. Please check the L<DBI documentation|DBI> to verify if
2158 # $what is one of fetchrow_{array,hashref}
2159 my($self, $sth, $what) = @_;
2165 my @arr = $sth->$what;
2167 defined && /[^\000-\177]/ && Encode::_utf8_on($_);
2171 my $ret = $sth->$what;
2173 for my $k (keys %$ret) {
2176 && Encode::_utf8_on($_) for $ret->{$k};
2180 defined && /[^\000-\177]/ && Encode::_utf8_on($_) for $ret;
2190 A large scalar that you know can only contain ASCII
2192 Scalars that contain only ASCII and are marked as UTF-8 are sometimes
2193 a drag to your program. If you recognize such a situation, just remove
2196 utf8::downgrade($val) if $] > 5.008;
2202 See also L</The "Unicode Bug"> above.
2204 =head2 Interaction with Extensions
2206 When Perl exchanges data with an extension, the extension should be
2207 able to understand the UTF8 flag and act accordingly. If the
2208 extension doesn't recognize that flag, it's likely that the extension
2209 will return incorrectly-flagged data.
2211 So if you're working with Unicode data, consult the documentation of
2212 every module you're using if there are any issues with Unicode data
2213 exchange. If the documentation does not talk about Unicode at all,
2214 suspect the worst and probably look at the source to learn how the
2215 module is implemented. Modules written completely in Perl shouldn't
2216 cause problems. Modules that directly or indirectly access code written
2217 in other programming languages are at risk.
2219 For affected functions, the simple strategy to avoid data corruption is
2220 to always make the encoding of the exchanged data explicit. Choose an
2221 encoding that you know the extension can handle. Convert arguments passed
2222 to the extensions to that encoding and convert results back from that
2223 encoding. Write wrapper functions that do the conversions for you, so
2224 you can later change the functions when the extension catches up.
2226 To provide an example, let's say the popular C<Foo::Bar::escape_html>
2227 function doesn't deal with Unicode data yet. The wrapper function
2228 would convert the argument to raw UTF-8 and convert the result back to
2229 Perl's internal representation like so:
2231 sub my_escape_html ($) {
2233 return unless defined $what;
2234 Encode::decode("UTF-8", Foo::Bar::escape_html(
2235 Encode::encode("UTF-8", $what)));
2238 Sometimes, when the extension does not convert data but just stores
2239 and retrieves it, you will be able to use the otherwise
2240 dangerous L<C<Encode::_utf8_on()>|Encode/_utf8_on> function. Let's say
2241 the popular C<Foo::Bar> extension, written in C, provides a C<param>
2242 method that lets you store and retrieve data according to these prototypes:
2244 $self->param($name, $value); # set a scalar
2245 $value = $self->param($name); # retrieve a scalar
2247 If it does not yet provide support for any encoding, one could write a
2248 derived class with such a C<param> method:
2251 my($self,$name,$value) = @_;
2252 utf8::upgrade($name); # make sure it is UTF-8 encoded
2253 if (defined $value) {
2254 utf8::upgrade($value); # make sure it is UTF-8 encoded
2255 return $self->SUPER::param($name,$value);
2257 my $ret = $self->SUPER::param($name);
2258 Encode::_utf8_on($ret); # we know, it is UTF-8 encoded
2263 Some extensions provide filters on data entry/exit points, such as
2264 C<DB_File::filter_store_key> and family. Look out for such filters in
2265 the documentation of your extensions; they can make the transition to
2266 Unicode data much easier.
2270 Some functions are slower when working on UTF-8 encoded strings than
2271 on byte encoded strings. All functions that need to hop over
2272 characters such as C<length()>, C<substr()> or C<index()>, or matching
2273 regular expressions can work B<much> faster when the underlying data are
2276 In Perl 5.8.0 the slowness was often quite spectacular; in Perl 5.8.1
2277 a caching scheme was introduced which improved the situation. In general,
2278 operations with UTF-8 encoded strings are still slower. As an example,
2279 the Unicode properties (character classes) like C<\p{Nd}> are known to
2280 be quite a bit slower (5-20 times) than their simpler counterparts
2281 like C<[0-9]> (then again, there are hundreds of Unicode characters matching
2282 C<Nd> compared with the 10 ASCII characters matching C<[0-9]>).
2286 L<perlunitut>, L<perluniintro>, L<perluniprops>, L<Encode>, L<open>, L<utf8>, L<bytes>,
2287 L<perlretut>, L<perlvar/"${^UNICODE}">,
2288 L<https://www.unicode.org/reports/tr44>).