3 perlunicode - Unicode support in Perl
7 =head2 Important Caveats
9 Unicode support is an extensive requirement. While Perl does not
10 implement the Unicode standard or the accompanying technical reports
11 from cover to cover, Perl does support many Unicode features.
13 People who want to learn to use Unicode in Perl, should probably read
14 the L<Perl Unicode tutorial, perlunitut|perlunitut> and
15 L<perluniintro>, before reading
16 this reference document.
18 Also, the use of Unicode may present security issues that aren't obvious.
19 Read L<Unicode Security Considerations|http://www.unicode.org/reports/tr36>.
23 =item Safest if you "use feature 'unicode_strings'"
25 In order to preserve backward compatibility, Perl does not turn
26 on full internal Unicode support unless the pragma
27 C<use feature 'unicode_strings'> is specified. (This is automatically
28 selected if you use C<use 5.012> or higher.) Failure to do this can
29 trigger unexpected surprises. See L</The "Unicode Bug"> below.
31 This pragma doesn't affect I/O, and there are still several places
32 where Unicode isn't fully supported, such as in filenames.
34 =item Input and Output Layers
36 Perl knows when a filehandle uses Perl's internal Unicode encodings
37 (UTF-8, or UTF-EBCDIC if in EBCDIC) if the filehandle is opened with
38 the ":encoding(utf8)" layer. Other encodings can be converted to Perl's
39 encoding on input or from Perl's encoding on output by use of the
40 ":encoding(...)" layer. See L<open>.
42 To indicate that Perl source itself is in UTF-8, use C<use utf8;>.
44 =item C<use utf8> still needed to enable UTF-8/UTF-EBCDIC in scripts
46 As a compatibility measure, the C<use utf8> pragma must be explicitly
47 included to enable recognition of UTF-8 in the Perl scripts themselves
48 (in string or regular expression literals, or in identifier names) on
49 ASCII-based machines or to recognize UTF-EBCDIC on EBCDIC-based
50 machines. B<These are the only times when an explicit C<use utf8>
51 is needed.> See L<utf8>.
53 =item BOM-marked scripts and UTF-16 scripts autodetected
55 If a Perl script begins marked with the Unicode BOM (UTF-16LE, UTF16-BE,
56 or UTF-8), or if the script looks like non-BOM-marked UTF-16 of either
57 endianness, Perl will correctly read in the script as Unicode.
58 (BOMless UTF-8 cannot be effectively recognized or differentiated from
59 ISO 8859-1 or other eight-bit encodings.)
61 =item C<use encoding> needed to upgrade non-Latin-1 byte strings
63 By default, there is a fundamental asymmetry in Perl's Unicode model:
64 implicit upgrading from byte strings to Unicode strings assumes that
65 they were encoded in I<ISO 8859-1 (Latin-1)>, but Unicode strings are
66 downgraded with UTF-8 encoding. This happens because the first 256
67 codepoints in Unicode happens to agree with Latin-1.
69 See L</"Byte and Character Semantics"> for more details.
73 =head2 Byte and Character Semantics
75 Beginning with version 5.6, Perl uses logically-wide characters to
76 represent strings internally.
78 Starting in Perl 5.14, Perl-level operations work with
79 characters rather than bytes within the scope of a
80 C<L<use feature 'unicode_strings'|feature>> (or equivalently
81 C<use 5.012> or higher). (This is not true if bytes have been
82 explicitly requested by C<L<use bytes|bytes>>, nor necessarily true
83 for interactions with the platform's operating system.)
85 For earlier Perls, and when C<unicode_strings> is not in effect, Perl
86 provides a fairly safe environment that can handle both types of
87 semantics in programs. For operations where Perl can unambiguously
88 decide that the input data are characters, Perl switches to character
89 semantics. For operations where this determination cannot be made
90 without additional information from the user, Perl decides in favor of
91 compatibility and chooses to use byte semantics.
93 When C<use locale> is in effect (which overrides
94 C<use feature 'unicode_strings'> in the same scope), Perl uses the
96 with the current locale. Otherwise, Perl uses the platform's native
97 byte semantics for characters whose code points are less than 256, and
98 Unicode semantics for those greater than 255. On EBCDIC platforms, this
99 is almost seamless, as the EBCDIC code pages that Perl handles are
100 equivalent to Unicode's first 256 code points. (The exception is that
101 EBCDIC regular expression case-insensitive matching rules are not as
102 as robust as Unicode's.) But on ASCII platforms, Perl uses US-ASCII
103 (or Basic Latin in Unicode terminology) byte semantics, meaning that characters
104 whose ordinal numbers are in the range 128 - 255 are undefined except for their
105 ordinal numbers. This means that none have case (upper and lower), nor are any
106 a member of character classes, like C<[:alpha:]> or C<\w>. (But all do belong
107 to the C<\W> class or the Perl regular expression extension C<[:^alpha:]>.)
109 This behavior preserves compatibility with earlier versions of Perl,
110 which allowed byte semantics in Perl operations only if
111 none of the program's inputs were marked as being a source of Unicode
112 character data. Such data may come from filehandles, from calls to
113 external programs, from information provided by the system (such as %ENV),
114 or from literals and constants in the source text.
116 The C<utf8> pragma is primarily a compatibility device that enables
117 recognition of UTF-(8|EBCDIC) in literals encountered by the parser.
118 Note that this pragma is only required while Perl defaults to byte
119 semantics; when character semantics become the default, this pragma
120 may become a no-op. See L<utf8>.
122 If strings operating under byte semantics and strings with Unicode
123 character data are concatenated, the new string will have
124 character semantics. This can cause surprises: See L</BUGS>, below.
125 You can choose to be warned when this happens. See L<encoding::warnings>.
127 Under character semantics, many operations that formerly operated on
128 bytes now operate on characters. A character in Perl is
129 logically just a number ranging from 0 to 2**31 or so. Larger
130 characters may encode into longer sequences of bytes internally, but
131 this internal detail is mostly hidden for Perl code.
132 See L<perluniintro> for more.
134 =head2 Effects of Character Semantics
136 Character semantics have the following effects:
142 Strings--including hash keys--and regular expression patterns may
143 contain characters that have an ordinal value larger than 255.
145 If you use a Unicode editor to edit your program, Unicode characters may
146 occur directly within the literal strings in UTF-8 encoding, or UTF-16.
147 (The former requires a BOM or C<use utf8>, the latter requires a BOM.)
149 Unicode characters can also be added to a string by using the C<\N{U+...}>
150 notation. The Unicode code for the desired character, in hexadecimal,
151 should be placed in the braces, after the C<U>. For instance, a smiley face is
154 Alternatively, you can use the C<\x{...}> notation for characters 0x100 and
155 above. For characters below 0x100 you may get byte semantics instead of
156 character semantics; see L</The "Unicode Bug">. On EBCDIC machines there is
157 the additional problem that the value for such characters gives the EBCDIC
158 character rather than the Unicode one.
162 use charnames ':full';
164 you can use the C<\N{...}> notation and put the official Unicode
165 character name within the braces, such as C<\N{WHITE SMILING FACE}>.
170 If an appropriate L<encoding> is specified, identifiers within the
171 Perl script may contain Unicode alphanumeric characters, including
172 ideographs. Perl does not currently attempt to canonicalize variable
177 Regular expressions match characters instead of bytes. "." matches
178 a character instead of a byte.
182 Bracketed character classes in regular expressions match characters instead of
183 bytes and match against the character properties specified in the
184 Unicode properties database. C<\w> can be used to match a Japanese
185 ideograph, for instance.
189 Named Unicode properties, scripts, and block ranges may be used (like bracketed
190 character classes) by using the C<\p{}> "matches property" construct and
191 the C<\P{}> negation, "doesn't match property".
192 See L</"Unicode Character Properties"> for more details.
194 You can define your own character properties and use them
195 in the regular expression with the C<\p{}> or C<\P{}> construct.
196 See L</"User-Defined Character Properties"> for more details.
200 The special pattern C<\X> matches a logical character, an "extended grapheme
201 cluster" in Standardese. In Unicode what appears to the user to be a single
202 character, for example an accented C<G>, may in fact be composed of a sequence
203 of characters, in this case a C<G> followed by an accent character. C<\X>
204 will match the entire sequence.
208 The C<tr///> operator translates characters instead of bytes. Note
209 that the C<tr///CU> functionality has been removed. For similar
210 functionality see pack('U0', ...) and pack('C0', ...).
214 Case translation operators use the Unicode case translation tables
215 when character input is provided. Note that C<uc()>, or C<\U> in
216 interpolated strings, translates to uppercase, while C<ucfirst>,
217 or C<\u> in interpolated strings, translates to titlecase in languages
218 that make the distinction (which is equivalent to uppercase in languages
219 without the distinction).
223 Most operators that deal with positions or lengths in a string will
224 automatically switch to using character positions, including
225 C<chop()>, C<chomp()>, C<substr()>, C<pos()>, C<index()>, C<rindex()>,
226 C<sprintf()>, C<write()>, and C<length()>. An operator that
227 specifically does not switch is C<vec()>. Operators that really don't
228 care include operators that treat strings as a bucket of bits such as
229 C<sort()>, and operators dealing with filenames.
233 The C<pack()>/C<unpack()> letter C<C> does I<not> change, since it is often
234 used for byte-oriented formats. Again, think C<char> in the C language.
236 There is a new C<U> specifier that converts between Unicode characters
237 and code points. There is also a C<W> specifier that is the equivalent of
238 C<chr>/C<ord> and properly handles character values even if they are above 255.
242 The C<chr()> and C<ord()> functions work on characters, similar to
243 C<pack("W")> and C<unpack("W")>, I<not> C<pack("C")> and
244 C<unpack("C")>. C<pack("C")> and C<unpack("C")> are methods for
245 emulating byte-oriented C<chr()> and C<ord()> on Unicode strings.
246 While these methods reveal the internal encoding of Unicode strings,
247 that is not something one normally needs to care about at all.
251 The bit string operators, C<& | ^ ~>, can operate on character data.
252 However, for backward compatibility, such as when using bit string
253 operations when characters are all less than 256 in ordinal value, one
254 should not use C<~> (the bit complement) with characters of both
255 values less than 256 and values greater than 256. Most importantly,
256 DeMorgan's laws (C<~($x|$y) eq ~$x&~$y> and C<~($x&$y) eq ~$x|~$y>)
257 will not hold. The reason for this mathematical I<faux pas> is that
258 the complement cannot return B<both> the 8-bit (byte-wide) bit
259 complement B<and> the full character-wide bit complement.
263 There is a CPAN module, L<Unicode::Casing>, which allows you to define
264 your own mappings to be used in C<lc()>, C<lcfirst()>, C<uc()>, and
265 C<ucfirst()> (or their double-quoted string inlined versions such as
266 C<\U>). (Prior to Perl 5.16, this functionality was partially provided
267 in the Perl core, but suffered from a number of insurmountable
268 drawbacks, so the CPAN module was written instead.)
276 And finally, C<scalar reverse()> reverses by character rather than by byte.
280 =head2 Unicode Character Properties
282 (The only time that Perl considers a sequence of individual code
283 points as a single logical character is in the C<\X> construct, already
284 mentioned above. Therefore "character" in this discussion means a single
287 Very nearly all Unicode character properties are accessible through
288 regular expressions by using the C<\p{}> "matches property" construct
289 and the C<\P{}> "doesn't match property" for its negation.
291 For instance, C<\p{Uppercase}> matches any single character with the Unicode
292 "Uppercase" property, while C<\p{L}> matches any character with a
293 General_Category of "L" (letter) property. Brackets are not
294 required for single letter property names, so C<\p{L}> is equivalent to C<\pL>.
296 More formally, C<\p{Uppercase}> matches any single character whose Unicode
297 Uppercase property value is True, and C<\P{Uppercase}> matches any character
298 whose Uppercase property value is False, and they could have been written as
299 C<\p{Uppercase=True}> and C<\p{Uppercase=False}>, respectively.
301 This formality is needed when properties are not binary; that is, if they can
302 take on more values than just True and False. For example, the Bidi_Class (see
303 L</"Bidirectional Character Types"> below), can take on several different
304 values, such as Left, Right, Whitespace, and others. To match these, one needs
305 to specify both the property name (Bidi_Class), AND the value being
307 (Left, Right, etc.). This is done, as in the examples above, by having the
308 two components separated by an equal sign (or interchangeably, a colon), like
309 C<\p{Bidi_Class: Left}>.
311 All Unicode-defined character properties may be written in these compound forms
312 of C<\p{property=value}> or C<\p{property:value}>, but Perl provides some
313 additional properties that are written only in the single form, as well as
314 single-form short-cuts for all binary properties and certain others described
315 below, in which you may omit the property name and the equals or colon
318 Most Unicode character properties have at least two synonyms (or aliases if you
319 prefer): a short one that is easier to type and a longer one that is more
320 descriptive and hence easier to understand. Thus the "L" and "Letter" properties
321 above are equivalent and can be used interchangeably. Likewise,
322 "Upper" is a synonym for "Uppercase", and we could have written
323 C<\p{Uppercase}> equivalently as C<\p{Upper}>. Also, there are typically
324 various synonyms for the values the property can be. For binary properties,
325 "True" has 3 synonyms: "T", "Yes", and "Y"; and "False has correspondingly "F",
326 "No", and "N". But be careful. A short form of a value for one property may
327 not mean the same thing as the same short form for another. Thus, for the
328 General_Category property, "L" means "Letter", but for the Bidi_Class property,
329 "L" means "Left". A complete list of properties and synonyms is in
332 Upper/lower case differences in property names and values are irrelevant;
333 thus C<\p{Upper}> means the same thing as C<\p{upper}> or even C<\p{UpPeR}>.
334 Similarly, you can add or subtract underscores anywhere in the middle of a
335 word, so that these are also equivalent to C<\p{U_p_p_e_r}>. And white space
336 is irrelevant adjacent to non-word characters, such as the braces and the equals
337 or colon separators, so C<\p{ Upper }> and C<\p{ Upper_case : Y }> are
338 equivalent to these as well. In fact, white space and even
339 hyphens can usually be added or deleted anywhere. So even C<\p{ Up-per case = Yes}> is
340 equivalent. All this is called "loose-matching" by Unicode. The few places
341 where stricter matching is used is in the middle of numbers, and in the Perl
342 extension properties that begin or end with an underscore. Stricter matching
343 cares about white space (except adjacent to non-word characters),
344 hyphens, and non-interior underscores.
346 You can also use negation in both C<\p{}> and C<\P{}> by introducing a caret
347 (^) between the first brace and the property name: C<\p{^Tamil}> is
348 equal to C<\P{Tamil}>.
350 Almost all properties are immune to case-insensitive matching. That is,
351 adding a C</i> regular expression modifier does not change what they
352 match. There are two sets that are affected.
356 and C<Titlecase_Letter>,
357 all of which match C<Cased_Letter> under C</i> matching.
358 And the second set is
362 all of which match C<Cased> under C</i> matching.
363 This set also includes its subsets C<PosixUpper> and C<PosixLower> both
364 of which under C</i> matching match C<PosixAlpha>.
365 (The difference between these sets is that some things, such as Roman
366 numerals, come in both upper and lower case so they are C<Cased>, but aren't considered
367 letters, so they aren't C<Cased_Letter>s.)
369 =head3 B<General_Category>
371 Every Unicode character is assigned a general category, which is the "most
372 usual categorization of a character" (from
373 L<http://www.unicode.org/reports/tr44>).
375 The compound way of writing these is like C<\p{General_Category=Number}>
376 (short, C<\p{gc:n}>). But Perl furnishes shortcuts in which everything up
377 through the equal or colon separator is omitted. So you can instead just write
380 Here are the short and long forms of the General Category properties:
385 LC, L& Cased_Letter (that is: [\p{Ll}\p{Lu}\p{Lt}])
398 Nd Decimal_Number (also Digit)
402 P Punctuation (also Punct)
403 Pc Connector_Punctuation
407 Pi Initial_Punctuation
408 (may behave like Ps or Pe depending on usage)
410 (may behave like Ps or Pe depending on usage)
422 Zp Paragraph_Separator
425 Cc Control (also Cntrl)
431 Single-letter properties match all characters in any of the
432 two-letter sub-properties starting with the same letter.
433 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>.
435 =head3 B<Bidirectional Character Types>
437 Because scripts differ in their directionality (Hebrew and Arabic are
438 written right to left, for example) Unicode supplies these properties in
439 the Bidi_Class class:
444 LRE Left-to-Right Embedding
445 LRO Left-to-Right Override
448 RLE Right-to-Left Embedding
449 RLO Right-to-Left Override
450 PDF Pop Directional Format
452 ES European Separator
453 ET European Terminator
458 B Paragraph Separator
463 This property is always written in the compound form.
464 For example, C<\p{Bidi_Class:R}> matches characters that are normally
465 written right to left.
469 The world's languages are written in many different scripts. This sentence
470 (unless you're reading it in translation) is written in Latin, while Russian is
471 written in Cyrillic, and Greek is written in, well, Greek; Japanese mainly in
472 Hiragana or Katakana. There are many more.
474 The Unicode Script and Script_Extensions properties give what script a
475 given character is in. Either property can be specified with the
477 C<\p{Script=Hebrew}> (short: C<\p{sc=hebr}>), or
478 C<\p{Script_Extensions=Javanese}> (short: C<\p{scx=java}>).
479 In addition, Perl furnishes shortcuts for all
480 C<Script> property names. You can omit everything up through the equals
481 (or colon), and simply write C<\p{Latin}> or C<\P{Cyrillic}>.
482 (This is not true for C<Script_Extensions>, which is required to be
483 written in the compound form.)
485 The difference between these two properties involves characters that are
486 used in multiple scripts. For example the digits '0' through '9' are
487 used in many parts of the world. These are placed in a script named
488 C<Common>. Other characters are used in just a few scripts. For
489 example, the "KATAKANA-HIRAGANA DOUBLE HYPHEN" is used in both Japanese
490 scripts, Katakana and Hiragana, but nowhere else. The C<Script>
491 property places all characters that are used in multiple scripts in the
492 C<Common> script, while the C<Script_Extensions> property places those
493 that are used in only a few scripts into each of those scripts; while
494 still using C<Common> for those used in many scripts. Thus both these
497 "0" =~ /\p{sc=Common}/ # Matches
498 "0" =~ /\p{scx=Common}/ # Matches
500 and only the first of these match:
502 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{sc=Common} # Matches
503 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{scx=Common} # No match
505 And only the last two of these match:
507 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{sc=Hiragana} # No match
508 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{sc=Katakana} # No match
509 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{scx=Hiragana} # Matches
510 "\N{KATAKANA-HIRAGANA DOUBLE HYPHEN}" =~ /\p{scx=Katakana} # Matches
512 C<Script_Extensions> is thus an improved C<Script>, in which there are
513 fewer characters in the C<Common> script, and correspondingly more in
514 other scripts. It is new in Unicode version 6.0, and its data are likely
515 to change significantly in later releases, as things get sorted out.
517 (Actually, besides C<Common>, the C<Inherited> script, contains
518 characters that are used in multiple scripts. These are modifier
519 characters which modify other characters, and inherit the script value
520 of the controlling character. Some of these are used in many scripts,
521 and so go into C<Inherited> in both C<Script> and C<Script_Extensions>.
522 Others are used in just a few scripts, so are in C<Inherited> in
523 C<Script>, but not in C<Script_Extensions>.)
525 It is worth stressing that there are several different sets of digits in
526 Unicode that are equivalent to 0-9 and are matchable by C<\d> in a
527 regular expression. If they are used in a single language only, they
528 are in that language's C<Script> and C<Script_Extension>. If they are
529 used in more than one script, they will be in C<sc=Common>, but only
530 if they are used in many scripts should they be in C<scx=Common>.
532 A complete list of scripts and their shortcuts is in L<perluniprops>.
534 =head3 B<Use of "Is" Prefix>
536 For backward compatibility (with Perl 5.6), all properties mentioned
537 so far may have C<Is> or C<Is_> prepended to their name, so C<\P{Is_Lu}>, for
538 example, is equal to C<\P{Lu}>, and C<\p{IsScript:Arabic}> is equal to
543 In addition to B<scripts>, Unicode also defines B<blocks> of
544 characters. The difference between scripts and blocks is that the
545 concept of scripts is closer to natural languages, while the concept
546 of blocks is more of an artificial grouping based on groups of Unicode
547 characters with consecutive ordinal values. For example, the "Basic Latin"
548 block is all characters whose ordinals are between 0 and 127, inclusive; in
549 other words, the ASCII characters. The "Latin" script contains some letters
550 from this as well as several other blocks, like "Latin-1 Supplement",
551 "Latin Extended-A", etc., but it does not contain all the characters from
552 those blocks. It does not, for example, contain the digits 0-9, because
553 those digits are shared across many scripts, and hence are in the
556 For more about scripts versus blocks, see UAX#24 "Unicode Script Property":
557 L<http://www.unicode.org/reports/tr24>
559 The C<Script> or C<Script_Extensions> properties are likely to be the
560 ones you want to use when processing
561 natural language; the Block property may occasionally be useful in working
562 with the nuts and bolts of Unicode.
564 Block names are matched in the compound form, like C<\p{Block: Arrows}> or
565 C<\p{Blk=Hebrew}>. Unlike most other properties, only a few block names have a
566 Unicode-defined short name. But Perl does provide a (slight) shortcut: You
567 can say, for example C<\p{In_Arrows}> or C<\p{In_Hebrew}>. For backwards
568 compatibility, the C<In> prefix may be omitted if there is no naming conflict
569 with a script or any other property, and you can even use an C<Is> prefix
570 instead in those cases. But it is not a good idea to do this, for a couple
577 It is confusing. There are many naming conflicts, and you may forget some.
578 For example, C<\p{Hebrew}> means the I<script> Hebrew, and NOT the I<block>
579 Hebrew. But would you remember that 6 months from now?
583 It is unstable. A new version of Unicode may pre-empt the current meaning by
584 creating a property with the same name. There was a time in very early Unicode
585 releases when C<\p{Hebrew}> would have matched the I<block> Hebrew; now it
590 Some people prefer to always use C<\p{Block: foo}> and C<\p{Script: bar}>
591 instead of the shortcuts, whether for clarity, because they can't remember the
592 difference between 'In' and 'Is' anyway, or they aren't confident that those who
593 eventually will read their code will know that difference.
595 A complete list of blocks and their shortcuts is in L<perluniprops>.
597 =head3 B<Other Properties>
599 There are many more properties than the very basic ones described here.
600 A complete list is in L<perluniprops>.
602 Unicode defines all its properties in the compound form, so all single-form
603 properties are Perl extensions. Most of these are just synonyms for the
604 Unicode ones, but some are genuine extensions, including several that are in
605 the compound form. And quite a few of these are actually recommended by Unicode
606 (in L<http://www.unicode.org/reports/tr18>).
608 This section gives some details on all extensions that aren't just
609 synonyms for compound-form Unicode properties
610 (for those properties, you'll have to refer to the
611 L<Unicode Standard|http://www.unicode.org/reports/tr44>.
617 This matches any of the 1_114_112 Unicode code points. It is a synonym for
620 =item B<C<\p{Alnum}>>
622 This matches any C<\p{Alphabetic}> or C<\p{Decimal_Number}> character.
626 This matches any of the 1_114_112 Unicode code points. It is a synonym for
629 =item B<C<\p{ASCII}>>
631 This matches any of the 128 characters in the US-ASCII character set,
632 which is a subset of Unicode.
634 =item B<C<\p{Assigned}>>
636 This matches any assigned code point; that is, any code point whose general
637 category is not Unassigned (or equivalently, not Cn).
639 =item B<C<\p{Blank}>>
641 This is the same as C<\h> and C<\p{HorizSpace}>: A character that changes the
642 spacing horizontally.
644 =item B<C<\p{Decomposition_Type: Non_Canonical}>> (Short: C<\p{Dt=NonCanon}>)
646 Matches a character that has a non-canonical decomposition.
648 To understand the use of this rarely used property=value combination, it is
649 necessary to know some basics about decomposition.
650 Consider a character, say H. It could appear with various marks around it,
651 such as an acute accent, or a circumflex, or various hooks, circles, arrows,
652 I<etc.>, above, below, to one side or the other, etc. There are many
653 possibilities among the world's languages. The number of combinations is
654 astronomical, and if there were a character for each combination, it would
655 soon exhaust Unicode's more than a million possible characters. So Unicode
656 took a different approach: there is a character for the base H, and a
657 character for each of the possible marks, and these can be variously combined
658 to get a final logical character. So a logical character--what appears to be a
659 single character--can be a sequence of more than one individual characters.
660 This is called an "extended grapheme cluster"; Perl furnishes the C<\X>
661 regular expression construct to match such sequences.
663 But Unicode's intent is to unify the existing character set standards and
664 practices, and several pre-existing standards have single characters that
665 mean the same thing as some of these combinations. An example is ISO-8859-1,
666 which has quite a few of these in the Latin-1 range, an example being "LATIN
667 CAPITAL LETTER E WITH ACUTE". Because this character was in this pre-existing
668 standard, Unicode added it to its repertoire. But this character is considered
669 by Unicode to be equivalent to the sequence consisting of the character
670 "LATIN CAPITAL LETTER E" followed by the character "COMBINING ACUTE ACCENT".
672 "LATIN CAPITAL LETTER E WITH ACUTE" is called a "pre-composed" character, and
673 its equivalence with the sequence is called canonical equivalence. All
674 pre-composed characters are said to have a decomposition (into the equivalent
675 sequence), and the decomposition type is also called canonical.
677 However, many more characters have a different type of decomposition, a
678 "compatible" or "non-canonical" decomposition. The sequences that form these
679 decompositions are not considered canonically equivalent to the pre-composed
680 character. An example, again in the Latin-1 range, is the "SUPERSCRIPT ONE".
681 It is somewhat like a regular digit 1, but not exactly; its decomposition
682 into the digit 1 is called a "compatible" decomposition, specifically a
683 "super" decomposition. There are several such compatibility
684 decompositions (see L<http://www.unicode.org/reports/tr44>), including one
685 called "compat", which means some miscellaneous type of decomposition
686 that doesn't fit into the decomposition categories that Unicode has chosen.
688 Note that most Unicode characters don't have a decomposition, so their
689 decomposition type is "None".
691 For your convenience, Perl has added the C<Non_Canonical> decomposition
692 type to mean any of the several compatibility decompositions.
694 =item B<C<\p{Graph}>>
696 Matches any character that is graphic. Theoretically, this means a character
697 that on a printer would cause ink to be used.
699 =item B<C<\p{HorizSpace}>>
701 This is the same as C<\h> and C<\p{Blank}>: a character that changes the
702 spacing horizontally.
706 This is a synonym for C<\p{Present_In=*}>
708 =item B<C<\p{PerlSpace}>>
710 This is the same as C<\s>, restricted to ASCII, namely C<S<[ \f\n\r\t]>>.
712 Mnemonic: Perl's (original) space
714 =item B<C<\p{PerlWord}>>
716 This is the same as C<\w>, restricted to ASCII, namely C<[A-Za-z0-9_]>
718 Mnemonic: Perl's (original) word.
720 =item B<C<\p{Posix...}>>
722 There are several of these, which are equivalents using the C<\p>
723 notation for Posix classes and are described in
724 L<perlrecharclass/POSIX Character Classes>.
726 =item B<C<\p{Present_In: *}>> (Short: C<\p{In=*}>)
728 This property is used when you need to know in what Unicode version(s) a
731 The "*" above stands for some two digit Unicode version number, such as
732 C<1.1> or C<4.0>; or the "*" can also be C<Unassigned>. This property will
733 match the code points whose final disposition has been settled as of the
734 Unicode release given by the version number; C<\p{Present_In: Unassigned}>
735 will match those code points whose meaning has yet to be assigned.
737 For example, C<U+0041> "LATIN CAPITAL LETTER A" was present in the very first
738 Unicode release available, which is C<1.1>, so this property is true for all
739 valid "*" versions. On the other hand, C<U+1EFF> was not assigned until version
740 5.1 when it became "LATIN SMALL LETTER Y WITH LOOP", so the only "*" that
741 would match it are 5.1, 5.2, and later.
743 Unicode furnishes the C<Age> property from which this is derived. The problem
744 with Age is that a strict interpretation of it (which Perl takes) has it
745 matching the precise release a code point's meaning is introduced in. Thus
746 C<U+0041> would match only 1.1; and C<U+1EFF> only 5.1. This is not usually what
749 Some non-Perl implementations of the Age property may change its meaning to be
750 the same as the Perl Present_In property; just be aware of that.
752 Another confusion with both these properties is that the definition is not
753 that the code point has been I<assigned>, but that the meaning of the code point
754 has been I<determined>. This is because 66 code points will always be
755 unassigned, and so the Age for them is the Unicode version in which the decision
756 to make them so was made. For example, C<U+FDD0> is to be permanently
757 unassigned to a character, and the decision to do that was made in version 3.1,
758 so C<\p{Age=3.1}> matches this character, as also does C<\p{Present_In: 3.1}> and up.
760 =item B<C<\p{Print}>>
762 This matches any character that is graphical or blank, except controls.
764 =item B<C<\p{SpacePerl}>>
766 This is the same as C<\s>, including beyond ASCII.
768 Mnemonic: Space, as modified by Perl. (It doesn't include the vertical tab
769 which both the Posix standard and Unicode consider white space.)
771 =item B<C<\p{VertSpace}>>
773 This is the same as C<\v>: A character that changes the spacing vertically.
777 This is the same as C<\w>, including over 100_000 characters beyond ASCII.
779 =item B<C<\p{XPosix...}>>
781 There are several of these, which are the standard Posix classes
782 extended to the full Unicode range. They are described in
783 L<perlrecharclass/POSIX Character Classes>.
787 =head2 User-Defined Character Properties
789 You can define your own binary character properties by defining subroutines
790 whose names begin with "In" or "Is". The subroutines can be defined in any
791 package. The user-defined properties can be used in the regular expression
792 C<\p> and C<\P> constructs; if you are using a user-defined property from a
793 package other than the one you are in, you must specify its package in the
794 C<\p> or C<\P> construct.
796 # assuming property Is_Foreign defined in Lang::
797 package main; # property package name required
798 if ($txt =~ /\p{Lang::IsForeign}+/) { ... }
800 package Lang; # property package name not required
801 if ($txt =~ /\p{IsForeign}+/) { ... }
804 Note that the effect is compile-time and immutable once defined.
805 However, the subroutines are passed a single parameter, which is 0 if
806 case-sensitive matching is in effect and non-zero if caseless matching
807 is in effect. The subroutine may return different values depending on
808 the value of the flag, and one set of values will immutably be in effect
809 for all case-sensitive matches, and the other set for all case-insensitive
812 Note that if the regular expression is tainted, then Perl will die rather
813 than calling the subroutine, where the name of the subroutine is
814 determined by the tainted data.
816 The subroutines must return a specially-formatted string, with one
817 or more newline-separated lines. Each line must be one of the following:
823 A single hexadecimal number denoting a Unicode code point to include.
827 Two hexadecimal numbers separated by horizontal whitespace (space or
828 tabular characters) denoting a range of Unicode code points to include.
832 Something to include, prefixed by "+": a built-in character
833 property (prefixed by "utf8::") or a user-defined character property,
834 to represent all the characters in that property; two hexadecimal code
835 points for a range; or a single hexadecimal code point.
839 Something to exclude, prefixed by "-": an existing character
840 property (prefixed by "utf8::") or a user-defined character property,
841 to represent all the characters in that property; two hexadecimal code
842 points for a range; or a single hexadecimal code point.
846 Something to negate, prefixed "!": an existing character
847 property (prefixed by "utf8::") or a user-defined character property,
848 to represent all the characters in that property; two hexadecimal code
849 points for a range; or a single hexadecimal code point.
853 Something to intersect with, prefixed by "&": an existing character
854 property (prefixed by "utf8::") or a user-defined character property,
855 for all the characters except the characters in the property; two
856 hexadecimal code points for a range; or a single hexadecimal code point.
860 For example, to define a property that covers both the Japanese
861 syllabaries (hiragana and katakana), you can define
870 Imagine that the here-doc end marker is at the beginning of the line.
871 Now you can use C<\p{InKana}> and C<\P{InKana}>.
873 You could also have used the existing block property names:
882 Suppose you wanted to match only the allocated characters,
883 not the raw block ranges: in other words, you want to remove
894 The negation is useful for defining (surprise!) negated classes.
904 Intersection is useful for getting the common characters matched by
905 two (or more) classes.
914 It's important to remember not to use "&" for the first set; that
915 would be intersecting with nothing, resulting in an empty set.
917 =head2 User-Defined Case Mappings (for serious hackers only)
919 B<This feature has been removed as of Perl 5.16.>
920 The CPAN module L<Unicode::Casing> provides better functionality without
921 the drawbacks that this feature had. If you are using a Perl earlier
922 than 5.16, this feature was most fully documented in the 5.14 version of
924 L<http://perldoc.perl.org/5.14.0/perlunicode.html#User-Defined-Case-Mappings-%28for-serious-hackers-only%29>
926 =head2 Character Encodings for Input and Output
930 =head2 Unicode Regular Expression Support Level
932 The following list of Unicode supported features for regular expressions describes
933 all features currently directly supported by core Perl. The references to "Level N"
934 and the section numbers refer to the Unicode Technical Standard #18,
935 "Unicode Regular Expressions", version 13, from August 2008.
941 Level 1 - Basic Unicode Support
943 RL1.1 Hex Notation - done [1]
944 RL1.2 Properties - done [2][3]
945 RL1.2a Compatibility Properties - done [4]
946 RL1.3 Subtraction and Intersection - MISSING [5]
947 RL1.4 Simple Word Boundaries - done [6]
948 RL1.5 Simple Loose Matches - done [7]
949 RL1.6 Line Boundaries - MISSING [8][9]
950 RL1.7 Supplementary Code Points - done [10]
954 [3] supports not only minimal list, but all Unicode character
955 properties (see Unicode Character Properties above)
956 [4] \d \D \s \S \w \W \X [:prop:] [:^prop:]
957 [5] can use regular expression look-ahead [a] or
958 user-defined character properties [b] to emulate set
961 [7] note that Perl does Full case-folding in matching (but with
962 bugs), not Simple: for example U+1F88 is equivalent to
963 U+1F00 U+03B9, instead of just U+1F80. This difference
964 matters mainly for certain Greek capital letters with certain
965 modifiers: the Full case-folding decomposes the letter,
966 while the Simple case-folding would map it to a single
968 [8] should do ^ and $ also on U+000B (\v in C), FF (\f), CR
969 (\r), CRLF (\r\n), NEL (U+0085), LS (U+2028), and PS
970 (U+2029); should also affect <>, $., and script line
971 numbers; should not split lines within CRLF [c] (i.e. there
972 is no empty line between \r and \n)
973 [9] Linebreaking conformant with UAX#14 "Unicode Line Breaking
974 Algorithm" is available through the Unicode::LineBreaking
976 [10] UTF-8/UTF-EBDDIC used in Perl allows not only U+10000 to
977 U+10FFFF but also beyond U+10FFFF
979 [a] You can mimic class subtraction using lookahead.
980 For example, what UTS#18 might write as
982 [{Greek}-[{UNASSIGNED}]]
984 in Perl can be written as:
986 (?!\p{Unassigned})\p{InGreekAndCoptic}
987 (?=\p{Assigned})\p{InGreekAndCoptic}
989 But in this particular example, you probably really want
993 which will match assigned characters known to be part of the Greek script.
995 Also see the L<Unicode::Regex::Set> module, it does implement the full
996 UTS#18 grouping, intersection, union, and removal (subtraction) syntax.
998 [b] '+' for union, '-' for removal (set-difference), '&' for intersection
999 (see L</"User-Defined Character Properties">)
1001 [c] Try the C<:crlf> layer (see L<PerlIO>).
1005 Level 2 - Extended Unicode Support
1007 RL2.1 Canonical Equivalents - MISSING [10][11]
1008 RL2.2 Default Grapheme Clusters - MISSING [12]
1009 RL2.3 Default Word Boundaries - MISSING [14]
1010 RL2.4 Default Loose Matches - MISSING [15]
1011 RL2.5 Name Properties - DONE
1012 RL2.6 Wildcard Properties - MISSING
1014 [10] see UAX#15 "Unicode Normalization Forms"
1015 [11] have Unicode::Normalize but not integrated to regexes
1016 [12] have \X but we don't have a "Grapheme Cluster Mode"
1017 [14] see UAX#29, Word Boundaries
1018 [15] see UAX#21 "Case Mappings"
1022 Level 3 - Tailored Support
1024 RL3.1 Tailored Punctuation - MISSING
1025 RL3.2 Tailored Grapheme Clusters - MISSING [17][18]
1026 RL3.3 Tailored Word Boundaries - MISSING
1027 RL3.4 Tailored Loose Matches - MISSING
1028 RL3.5 Tailored Ranges - MISSING
1029 RL3.6 Context Matching - MISSING [19]
1030 RL3.7 Incremental Matches - MISSING
1031 ( RL3.8 Unicode Set Sharing )
1032 RL3.9 Possible Match Sets - MISSING
1033 RL3.10 Folded Matching - MISSING [20]
1034 RL3.11 Submatchers - MISSING
1036 [17] see UAX#10 "Unicode Collation Algorithms"
1037 [18] have Unicode::Collate but not integrated to regexes
1038 [19] have (?<=x) and (?=x), but look-aheads or look-behinds
1039 should see outside of the target substring
1040 [20] need insensitive matching for linguistic features other
1041 than case; for example, hiragana to katakana, wide and
1042 narrow, simplified Han to traditional Han (see UTR#30
1043 "Character Foldings")
1047 =head2 Unicode Encodings
1049 Unicode characters are assigned to I<code points>, which are abstract
1050 numbers. To use these numbers, various encodings are needed.
1058 UTF-8 is a variable-length (1 to 4 bytes), byte-order independent
1059 encoding. For ASCII (and we really do mean 7-bit ASCII, not another
1060 8-bit encoding), UTF-8 is transparent.
1062 The following table is from Unicode 3.2.
1064 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
1066 U+0000..U+007F 00..7F
1067 U+0080..U+07FF * C2..DF 80..BF
1068 U+0800..U+0FFF E0 * A0..BF 80..BF
1069 U+1000..U+CFFF E1..EC 80..BF 80..BF
1070 U+D000..U+D7FF ED 80..9F 80..BF
1071 U+D800..U+DFFF +++++ utf16 surrogates, not legal utf8 +++++
1072 U+E000..U+FFFF EE..EF 80..BF 80..BF
1073 U+10000..U+3FFFF F0 * 90..BF 80..BF 80..BF
1074 U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
1075 U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
1077 Note the gaps marked by "*" before several of the byte entries above. These are
1078 caused by legal UTF-8 avoiding non-shortest encodings: it is technically
1079 possible to UTF-8-encode a single code point in different ways, but that is
1080 explicitly forbidden, and the shortest possible encoding should always be used
1081 (and that is what Perl does).
1083 Another way to look at it is via bits:
1085 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
1088 00000bbbbbaaaaaa 110bbbbb 10aaaaaa
1089 ccccbbbbbbaaaaaa 1110cccc 10bbbbbb 10aaaaaa
1090 00000dddccccccbbbbbbaaaaaa 11110ddd 10cccccc 10bbbbbb 10aaaaaa
1092 As you can see, the continuation bytes all begin with "10", and the
1093 leading bits of the start byte tell how many bytes there are in the
1096 The original UTF-8 specification allowed up to 6 bytes, to allow
1097 encoding of numbers up to 0x7FFF_FFFF. Perl continues to allow those,
1098 and has extended that up to 13 bytes to encode code points up to what
1099 can fit in a 64-bit word. However, Perl will warn if you output any of
1100 these as being non-portable; and under strict UTF-8 input protocols,
1103 The Unicode non-character code points are also disallowed in UTF-8 in
1104 "open interchange". See L</Non-character code points>.
1110 Like UTF-8 but EBCDIC-safe, in the way that UTF-8 is ASCII-safe.
1114 UTF-16, UTF-16BE, UTF-16LE, Surrogates, and BOMs (Byte Order Marks)
1116 The followings items are mostly for reference and general Unicode
1117 knowledge, Perl doesn't use these constructs internally.
1119 Like UTF-8, UTF-16 is a variable-width encoding, but where
1120 UTF-8 uses 8-bit code units, UTF-16 uses 16-bit code units.
1121 All code points occupy either 2 or 4 bytes in UTF-16: code points
1122 C<U+0000..U+FFFF> are stored in a single 16-bit unit, and code
1123 points C<U+10000..U+10FFFF> in two 16-bit units. The latter case is
1124 using I<surrogates>, the first 16-bit unit being the I<high
1125 surrogate>, and the second being the I<low surrogate>.
1127 Surrogates are code points set aside to encode the C<U+10000..U+10FFFF>
1128 range of Unicode code points in pairs of 16-bit units. The I<high
1129 surrogates> are the range C<U+D800..U+DBFF> and the I<low surrogates>
1130 are the range C<U+DC00..U+DFFF>. The surrogate encoding is
1132 $hi = ($uni - 0x10000) / 0x400 + 0xD800;
1133 $lo = ($uni - 0x10000) % 0x400 + 0xDC00;
1137 $uni = 0x10000 + ($hi - 0xD800) * 0x400 + ($lo - 0xDC00);
1139 Because of the 16-bitness, UTF-16 is byte-order dependent. UTF-16
1140 itself can be used for in-memory computations, but if storage or
1141 transfer is required either UTF-16BE (big-endian) or UTF-16LE
1142 (little-endian) encodings must be chosen.
1144 This introduces another problem: what if you just know that your data
1145 is UTF-16, but you don't know which endianness? Byte Order Marks, or
1146 BOMs, are a solution to this. A special character has been reserved
1147 in Unicode to function as a byte order marker: the character with the
1148 code point C<U+FEFF> is the BOM.
1150 The trick is that if you read a BOM, you will know the byte order,
1151 since if it was written on a big-endian platform, you will read the
1152 bytes C<0xFE 0xFF>, but if it was written on a little-endian platform,
1153 you will read the bytes C<0xFF 0xFE>. (And if the originating platform
1154 was writing in UTF-8, you will read the bytes C<0xEF 0xBB 0xBF>.)
1156 The way this trick works is that the character with the code point
1157 C<U+FFFE> is not supposed to be in input streams, so the
1158 sequence of bytes C<0xFF 0xFE> is unambiguously "BOM, represented in
1159 little-endian format" and cannot be C<U+FFFE>, represented in big-endian
1162 Surrogates have no meaning in Unicode outside their use in pairs to
1163 represent other code points. However, Perl allows them to be
1164 represented individually internally, for example by saying
1165 C<chr(0xD801)>, so that all code points, not just those valid for open
1167 representable. Unicode does define semantics for them, such as their
1168 General Category is "Cs". But because their use is somewhat dangerous,
1169 Perl will warn (using the warning category "surrogate", which is a
1170 sub-category of "utf8") if an attempt is made
1171 to do things like take the lower case of one, or match
1172 case-insensitively, or to output them. (But don't try this on Perls
1177 UTF-32, UTF-32BE, UTF-32LE
1179 The UTF-32 family is pretty much like the UTF-16 family, expect that
1180 the units are 32-bit, and therefore the surrogate scheme is not
1181 needed. UTF-32 is a fixed-width encoding. The BOM signatures are
1182 C<0x00 0x00 0xFE 0xFF> for BE and C<0xFF 0xFE 0x00 0x00> for LE.
1188 Legacy, fixed-width encodings defined by the ISO 10646 standard. UCS-2 is a 16-bit
1189 encoding. Unlike UTF-16, UCS-2 is not extensible beyond C<U+FFFF>,
1190 because it does not use surrogates. UCS-4 is a 32-bit encoding,
1191 functionally identical to UTF-32 (the difference being that
1192 UCS-4 forbids neither surrogates nor code points larger than 0x10_FFFF).
1198 A seven-bit safe (non-eight-bit) encoding, which is useful if the
1199 transport or storage is not eight-bit safe. Defined by RFC 2152.
1203 =head2 Non-character code points
1205 66 code points are set aside in Unicode as "non-character code points".
1206 These all have the Unassigned (Cn) General Category, and they never will
1207 be assigned. These are never supposed to be in legal Unicode input
1208 streams, so that code can use them as sentinels that can be mixed in
1209 with character data, and they always will be distinguishable from that data.
1210 To keep them out of Perl input streams, strict UTF-8 should be
1211 specified, such as by using the layer C<:encoding('UTF-8')>. The
1212 non-character code points are the 32 between U+FDD0 and U+FDEF, and the
1213 34 code points U+FFFE, U+FFFF, U+1FFFE, U+1FFFF, ... U+10FFFE, U+10FFFF.
1214 Some people are under the mistaken impression that these are "illegal",
1215 but that is not true. An application or cooperating set of applications
1216 can legally use them at will internally; but these code points are
1217 "illegal for open interchange". Therefore, Perl will not accept these
1218 from input streams unless lax rules are being used, and will warn
1219 (using the warning category "nonchar", which is a sub-category of "utf8") if
1220 an attempt is made to output them.
1222 =head2 Beyond Unicode code points
1224 The maximum Unicode code point is U+10FFFF. But Perl accepts code
1225 points up to the maximum permissible unsigned number available on the
1226 platform. However, Perl will not accept these from input streams unless
1227 lax rules are being used, and will warn (using the warning category
1228 "non_unicode", which is a sub-category of "utf8") if an attempt is made to
1229 operate on or output them. For example, C<uc(0x11_0000)> will generate
1230 this warning, returning the input parameter as its result, as the upper
1231 case of every non-Unicode code point is the code point itself.
1233 =head2 Security Implications of Unicode
1235 Read L<Unicode Security Considerations|http://www.unicode.org/reports/tr36>.
1236 Also, note the following:
1244 Unfortunately, the original specification of UTF-8 leaves some room for
1245 interpretation of how many bytes of encoded output one should generate
1246 from one input Unicode character. Strictly speaking, the shortest
1247 possible sequence of UTF-8 bytes should be generated,
1248 because otherwise there is potential for an input buffer overflow at
1249 the receiving end of a UTF-8 connection. Perl always generates the
1250 shortest length UTF-8, and with warnings on, Perl will warn about
1251 non-shortest length UTF-8 along with other malformations, such as the
1252 surrogates, which are not Unicode code points valid for interchange.
1256 Regular expression pattern matching may surprise you if you're not
1257 accustomed to Unicode. Starting in Perl 5.14, several pattern
1258 modifiers are available to control this, called the character set
1259 modifiers. Details are given in L<perlre/Character set modifiers>.
1263 As discussed elsewhere, Perl has one foot (two hooves?) planted in
1264 each of two worlds: the old world of bytes and the new world of
1265 characters, upgrading from bytes to characters when necessary.
1266 If your legacy code does not explicitly use Unicode, no automatic
1267 switch-over to characters should happen. Characters shouldn't get
1268 downgraded to bytes, either. It is possible to accidentally mix bytes
1269 and characters, however (see L<perluniintro>), in which case C<\w> in
1270 regular expressions might start behaving differently (unless the C</a>
1271 modifier is in effect). Review your code. Use warnings and the C<strict> pragma.
1273 =head2 Unicode in Perl on EBCDIC
1275 The way Unicode is handled on EBCDIC platforms is still
1276 experimental. On such platforms, references to UTF-8 encoding in this
1277 document and elsewhere should be read as meaning the UTF-EBCDIC
1278 specified in Unicode Technical Report 16, unless ASCII vs. EBCDIC issues
1279 are specifically discussed. There is no C<utfebcdic> pragma or
1280 ":utfebcdic" layer; rather, "utf8" and ":utf8" are reused to mean
1281 the platform's "natural" 8-bit encoding of Unicode. See L<perlebcdic>
1282 for more discussion of the issues.
1286 See L<perllocale/Unicode and UTF-8>
1288 =head2 When Unicode Does Not Happen
1290 While Perl does have extensive ways to input and output in Unicode,
1291 and a few other "entry points" like the @ARGV array (which can sometimes be
1292 interpreted as UTF-8), there are still many places where Unicode
1293 (in some encoding or another) could be given as arguments or received as
1294 results, or both, but it is not.
1296 The following are such interfaces. Also, see L</The "Unicode Bug">.
1297 For all of these interfaces Perl
1298 currently (as of 5.8.3) simply assumes byte strings both as arguments
1299 and results, or UTF-8 strings if the (problematic) C<encoding> pragma has been used.
1301 One reason that Perl does not attempt to resolve the role of Unicode in
1302 these situations is that the answers are highly dependent on the operating
1303 system and the file system(s). For example, whether filenames can be
1304 in Unicode and in exactly what kind of encoding, is not exactly a
1305 portable concept. Similarly for C<qx> and C<system>: how well will the
1306 "command-line interface" (and which of them?) handle Unicode?
1312 chdir, chmod, chown, chroot, exec, link, lstat, mkdir,
1313 rename, rmdir, stat, symlink, truncate, unlink, utime, -X
1325 open, opendir, sysopen
1329 qx (aka the backtick operator), system
1337 =head2 The "Unicode Bug"
1339 The term, the "Unicode bug" has been applied to an inconsistency
1340 on ASCII platforms with the
1341 Unicode code points in the Latin-1 Supplement block, that
1342 is, between 128 and 255. Without a locale specified, unlike all other
1343 characters or code points, these characters have very different semantics in
1344 byte semantics versus character semantics, unless
1345 C<use feature 'unicode_strings'> is specified.
1346 (The lesson here is to specify C<unicode_strings> to avoid the
1349 In character semantics they are interpreted as Unicode code points, which means
1350 they have the same semantics as Latin-1 (ISO-8859-1).
1352 In byte semantics, they are considered to be unassigned characters, meaning
1353 that the only semantics they have is their ordinal numbers, and that they are
1354 not members of various character classes. None are considered to match C<\w>
1355 for example, but all match C<\W>.
1357 The behavior is known to have effects on these areas:
1363 Changing the case of a scalar, that is, using C<uc()>, C<ucfirst()>, C<lc()>,
1364 and C<lcfirst()>, or C<\L>, C<\U>, C<\u> and C<\l> in regular expression
1369 Using caseless (C</i>) regular expression matching
1373 Matching any of several properties in regular expressions, namely C<\b>,
1374 C<\B>, C<\s>, C<\S>, C<\w>, C<\W>, and all the Posix character classes
1375 I<except> C<[[:ascii:]]>.
1379 In C<quotemeta> or its inline equivalent C<\Q>, no characters
1380 code points above 127 are quoted in UTF-8 encoded strings, but in
1381 byte encoded strings, code points between 128-255 are always quoted.
1385 This behavior can lead to unexpected results in which a string's semantics
1386 suddenly change if a code point above 255 is appended to or removed from it,
1387 which changes the string's semantics from byte to character or vice versa. As
1388 an example, consider the following program and its output:
1391 no feature 'unicode_strings';
1394 for ($s1, $s2, $s1.$s2) {
1402 If there's no C<\w> in C<s1> or in C<s2>, why does their concatenation have one?
1404 This anomaly stems from Perl's attempt to not disturb older programs that
1405 didn't use Unicode, and hence had no semantics for characters outside of the
1406 ASCII range (except in a locale), along with Perl's desire to add Unicode
1407 support seamlessly. The result wasn't seamless: these characters were
1410 Starting in Perl 5.14, C<use feature 'unicode_strings'> can be used to
1411 cause Perl to use Unicode semantics on all string operations within the
1412 scope of the feature subpragma. Regular expressions compiled in its
1413 scope retain that behavior even when executed or compiled into larger
1414 regular expressions outside the scope. (The pragma does not, however,
1415 affect the C<quotemeta> behavior. Nor does it affect the deprecated
1416 user-defined case changing operations--these still require a UTF-8
1417 encoded string to operate.)
1419 In Perl 5.12, the subpragma affected casing changes, but not regular
1420 expressions. See L<perlfunc/lc> for details on how this pragma works in
1421 combination with various others for casing.
1423 For earlier Perls, or when a string is passed to a function outside the
1424 subpragma's scope, a workaround is to always call C<utf8::upgrade($string)>,
1425 or to use the standard module L<Encode>. Also, a scalar that has any characters
1426 whose ordinal is above 0x100, or which were specified using either of the
1427 C<\N{...}> notations, will automatically have character semantics.
1429 =head2 Forcing Unicode in Perl (Or Unforcing Unicode in Perl)
1431 Sometimes (see L</"When Unicode Does Not Happen"> or L</The "Unicode Bug">)
1432 there are situations where you simply need to force a byte
1433 string into UTF-8, or vice versa. The low-level calls
1434 utf8::upgrade($bytestring) and utf8::downgrade($utf8string[, FAIL_OK]) are
1437 Note that utf8::downgrade() can fail if the string contains characters
1438 that don't fit into a byte.
1440 Calling either function on a string that already is in the desired state is a
1443 =head2 Using Unicode in XS
1445 If you want to handle Perl Unicode in XS extensions, you may find the
1446 following C APIs useful. See also L<perlguts/"Unicode Support"> for an
1447 explanation about Unicode at the XS level, and L<perlapi> for the API
1454 C<DO_UTF8(sv)> returns true if the C<UTF8> flag is on and the bytes
1455 pragma is not in effect. C<SvUTF8(sv)> returns true if the C<UTF8>
1456 flag is on; the bytes pragma is ignored. The C<UTF8> flag being on
1457 does B<not> mean that there are any characters of code points greater
1458 than 255 (or 127) in the scalar or that there are even any characters
1459 in the scalar. What the C<UTF8> flag means is that the sequence of
1460 octets in the representation of the scalar is the sequence of UTF-8
1461 encoded code points of the characters of a string. The C<UTF8> flag
1462 being off means that each octet in this representation encodes a
1463 single character with code point 0..255 within the string. Perl's
1464 Unicode model is not to use UTF-8 until it is absolutely necessary.
1468 C<uvchr_to_utf8(buf, chr)> writes a Unicode character code point into
1469 a buffer encoding the code point as UTF-8, and returns a pointer
1470 pointing after the UTF-8 bytes. It works appropriately on EBCDIC machines.
1474 C<utf8_to_uvchr(buf, lenp)> reads UTF-8 encoded bytes from a buffer and
1475 returns the Unicode character code point and, optionally, the length of
1476 the UTF-8 byte sequence. It works appropriately on EBCDIC machines.
1480 C<utf8_length(start, end)> returns the length of the UTF-8 encoded buffer
1481 in characters. C<sv_len_utf8(sv)> returns the length of the UTF-8 encoded
1486 C<sv_utf8_upgrade(sv)> converts the string of the scalar to its UTF-8
1487 encoded form. C<sv_utf8_downgrade(sv)> does the opposite, if
1488 possible. C<sv_utf8_encode(sv)> is like sv_utf8_upgrade except that
1489 it does not set the C<UTF8> flag. C<sv_utf8_decode()> does the
1490 opposite of C<sv_utf8_encode()>. Note that none of these are to be
1491 used as general-purpose encoding or decoding interfaces: C<use Encode>
1492 for that. C<sv_utf8_upgrade()> is affected by the encoding pragma
1493 but C<sv_utf8_downgrade()> is not (since the encoding pragma is
1494 designed to be a one-way street).
1498 C<is_utf8_char(s)> returns true if the pointer points to a valid UTF-8
1503 C<is_utf8_string(buf, len)> returns true if C<len> bytes of the buffer
1508 C<UTF8SKIP(buf)> will return the number of bytes in the UTF-8 encoded
1509 character in the buffer. C<UNISKIP(chr)> will return the number of bytes
1510 required to UTF-8-encode the Unicode character code point. C<UTF8SKIP()>
1511 is useful for example for iterating over the characters of a UTF-8
1512 encoded buffer; C<UNISKIP()> is useful, for example, in computing
1513 the size required for a UTF-8 encoded buffer.
1517 C<utf8_distance(a, b)> will tell the distance in characters between the
1518 two pointers pointing to the same UTF-8 encoded buffer.
1522 C<utf8_hop(s, off)> will return a pointer to a UTF-8 encoded buffer
1523 that is C<off> (positive or negative) Unicode characters displaced
1524 from the UTF-8 buffer C<s>. Be careful not to overstep the buffer:
1525 C<utf8_hop()> will merrily run off the end or the beginning of the
1526 buffer if told to do so.
1530 C<pv_uni_display(dsv, spv, len, pvlim, flags)> and
1531 C<sv_uni_display(dsv, ssv, pvlim, flags)> are useful for debugging the
1532 output of Unicode strings and scalars. By default they are useful
1533 only for debugging--they display B<all> characters as hexadecimal code
1534 points--but with the flags C<UNI_DISPLAY_ISPRINT>,
1535 C<UNI_DISPLAY_BACKSLASH>, and C<UNI_DISPLAY_QQ> you can make the
1536 output more readable.
1540 C<foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2)> can be used to
1541 compare two strings case-insensitively in Unicode. For case-sensitive
1542 comparisons you can just use C<memEQ()> and C<memNE()> as usual, except
1543 if one string is in utf8 and the other isn't.
1547 For more information, see L<perlapi>, and F<utf8.c> and F<utf8.h>
1548 in the Perl source code distribution.
1550 =head2 Hacking Perl to work on earlier Unicode versions (for very serious hackers only)
1552 Perl by default comes with the latest supported Unicode version built in, but
1553 you can change to use any earlier one.
1555 Download the files in the desired version of Unicode from the Unicode web
1556 site L<http://www.unicode.org>). These should replace the existing files in
1557 F<lib/unicore> in the Perl source tree. Follow the instructions in
1558 F<README.perl> in that directory to change some of their names, and then build
1559 perl (see L<INSTALL>).
1561 It is even possible to copy the built files to a different directory, and then
1562 change F<utf8_heavy.pl> in the directory C<$Config{privlib}> to point to the
1563 new directory, or maybe make a copy of that directory before making the change,
1564 and using C<@INC> or the C<-I> run-time flag to switch between versions at will
1565 (but because of caching, not in the middle of a process), but all this is
1566 beyond the scope of these instructions.
1570 =head2 Interaction with Locales
1572 See L<perllocale/Unicode and UTF-8>
1574 =head2 Problems with characters in the Latin-1 Supplement range
1576 See L</The "Unicode Bug">
1578 =head2 Interaction with Extensions
1580 When Perl exchanges data with an extension, the extension should be
1581 able to understand the UTF8 flag and act accordingly. If the
1582 extension doesn't recognize that flag, it's likely that the extension
1583 will return incorrectly-flagged data.
1585 So if you're working with Unicode data, consult the documentation of
1586 every module you're using if there are any issues with Unicode data
1587 exchange. If the documentation does not talk about Unicode at all,
1588 suspect the worst and probably look at the source to learn how the
1589 module is implemented. Modules written completely in Perl shouldn't
1590 cause problems. Modules that directly or indirectly access code written
1591 in other programming languages are at risk.
1593 For affected functions, the simple strategy to avoid data corruption is
1594 to always make the encoding of the exchanged data explicit. Choose an
1595 encoding that you know the extension can handle. Convert arguments passed
1596 to the extensions to that encoding and convert results back from that
1597 encoding. Write wrapper functions that do the conversions for you, so
1598 you can later change the functions when the extension catches up.
1600 To provide an example, let's say the popular Foo::Bar::escape_html
1601 function doesn't deal with Unicode data yet. The wrapper function
1602 would convert the argument to raw UTF-8 and convert the result back to
1603 Perl's internal representation like so:
1605 sub my_escape_html ($) {
1607 return unless defined $what;
1608 Encode::decode_utf8(Foo::Bar::escape_html(
1609 Encode::encode_utf8($what)));
1612 Sometimes, when the extension does not convert data but just stores
1613 and retrieves them, you will be able to use the otherwise
1614 dangerous Encode::_utf8_on() function. Let's say the popular
1615 C<Foo::Bar> extension, written in C, provides a C<param> method that
1616 lets you store and retrieve data according to these prototypes:
1618 $self->param($name, $value); # set a scalar
1619 $value = $self->param($name); # retrieve a scalar
1621 If it does not yet provide support for any encoding, one could write a
1622 derived class with such a C<param> method:
1625 my($self,$name,$value) = @_;
1626 utf8::upgrade($name); # make sure it is UTF-8 encoded
1627 if (defined $value) {
1628 utf8::upgrade($value); # make sure it is UTF-8 encoded
1629 return $self->SUPER::param($name,$value);
1631 my $ret = $self->SUPER::param($name);
1632 Encode::_utf8_on($ret); # we know, it is UTF-8 encoded
1637 Some extensions provide filters on data entry/exit points, such as
1638 DB_File::filter_store_key and family. Look out for such filters in
1639 the documentation of your extensions, they can make the transition to
1640 Unicode data much easier.
1644 Some functions are slower when working on UTF-8 encoded strings than
1645 on byte encoded strings. All functions that need to hop over
1646 characters such as length(), substr() or index(), or matching regular
1647 expressions can work B<much> faster when the underlying data are
1650 In Perl 5.8.0 the slowness was often quite spectacular; in Perl 5.8.1
1651 a caching scheme was introduced which will hopefully make the slowness
1652 somewhat less spectacular, at least for some operations. In general,
1653 operations with UTF-8 encoded strings are still slower. As an example,
1654 the Unicode properties (character classes) like C<\p{Nd}> are known to
1655 be quite a bit slower (5-20 times) than their simpler counterparts
1656 like C<\d> (then again, there are hundreds of Unicode characters matching C<Nd>
1657 compared with the 10 ASCII characters matching C<d>).
1659 =head2 Problems on EBCDIC platforms
1661 There are several known problems with Perl on EBCDIC platforms. If you
1662 want to use Perl there, send email to perlbug@perl.org.
1664 In earlier versions, when byte and character data were concatenated,
1665 the new string was sometimes created by
1666 decoding the byte strings as I<ISO 8859-1 (Latin-1)>, even if the
1667 old Unicode string used EBCDIC.
1669 If you find any of these, please report them as bugs.
1671 =head2 Porting code from perl-5.6.X
1673 Perl 5.8 has a different Unicode model from 5.6. In 5.6 the programmer
1674 was required to use the C<utf8> pragma to declare that a given scope
1675 expected to deal with Unicode data and had to make sure that only
1676 Unicode data were reaching that scope. If you have code that is
1677 working with 5.6, you will need some of the following adjustments to
1678 your code. The examples are written such that the code will continue
1679 to work under 5.6, so you should be safe to try them out.
1685 A filehandle that should read or write UTF-8
1688 binmode $fh, ":encoding(utf8)";
1693 A scalar that is going to be passed to some extension
1695 Be it Compress::Zlib, Apache::Request or any extension that has no
1696 mention of Unicode in the manpage, you need to make sure that the
1697 UTF8 flag is stripped off. Note that at the time of this writing
1698 (October 2002) the mentioned modules are not UTF-8-aware. Please
1699 check the documentation to verify if this is still true.
1703 $val = Encode::encode_utf8($val); # make octets
1708 A scalar we got back from an extension
1710 If you believe the scalar comes back as UTF-8, you will most likely
1711 want the UTF8 flag restored:
1715 $val = Encode::decode_utf8($val);
1720 Same thing, if you are really sure it is UTF-8
1724 Encode::_utf8_on($val);
1729 A wrapper for fetchrow_array and fetchrow_hashref
1731 When the database contains only UTF-8, a wrapper function or method is
1732 a convenient way to replace all your fetchrow_array and
1733 fetchrow_hashref calls. A wrapper function will also make it easier to
1734 adapt to future enhancements in your database driver. Note that at the
1735 time of this writing (October 2002), the DBI has no standardized way
1736 to deal with UTF-8 data. Please check the documentation to verify if
1740 # $what is one of fetchrow_{array,hashref}
1741 my($self, $sth, $what) = @_;
1747 my @arr = $sth->$what;
1749 defined && /[^\000-\177]/ && Encode::_utf8_on($_);
1753 my $ret = $sth->$what;
1755 for my $k (keys %$ret) {
1758 && Encode::_utf8_on($_) for $ret->{$k};
1762 defined && /[^\000-\177]/ && Encode::_utf8_on($_) for $ret;
1772 A large scalar that you know can only contain ASCII
1774 Scalars that contain only ASCII and are marked as UTF-8 are sometimes
1775 a drag to your program. If you recognize such a situation, just remove
1778 utf8::downgrade($val) if $] > 5.007;
1784 L<perlunitut>, L<perluniintro>, L<perluniprops>, L<Encode>, L<open>, L<utf8>, L<bytes>,
1785 L<perlretut>, L<perlvar/"${^UNICODE}">
1786 L<http://www.unicode.org/reports/tr44>).