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.
15 =item Input and Output Layers
17 Perl knows when a filehandle uses Perl's internal Unicode encodings
18 (UTF-8, or UTF-EBCDIC if in EBCDIC) if the filehandle is opened with
19 the ":utf8" layer. Other encodings can be converted to Perl's
20 encoding on input or from Perl's encoding on output by use of the
21 ":encoding(...)" layer. See L<open>.
23 To indicate that Perl source itself is using a particular encoding,
26 =item Regular Expressions
28 The regular expression compiler produces polymorphic opcodes. That is,
29 the pattern adapts to the data and automatically switches to the Unicode
30 character scheme when presented with Unicode data--or instead uses
31 a traditional byte scheme when presented with byte data.
33 =item C<use utf8> still needed to enable UTF-8/UTF-EBCDIC in scripts
35 As a compatibility measure, the C<use utf8> pragma must be explicitly
36 included to enable recognition of UTF-8 in the Perl scripts themselves
37 (in string or regular expression literals, or in identifier names) on
38 ASCII-based machines or to recognize UTF-EBCDIC on EBCDIC-based
39 machines. B<These are the only times when an explicit C<use utf8>
40 is needed.> See L<utf8>.
42 You can also use the C<encoding> pragma to change the default encoding
43 of the data in your script; see L<encoding>.
45 =item C<use encoding> needed to upgrade non-Latin-1 byte strings
47 By default, there is a fundamental asymmetry in Perl's unicode model:
48 implicit upgrading from byte strings to Unicode strings assumes that
49 they were encoded in I<ISO 8859-1 (Latin-1)>, but Unicode strings are
50 downgraded with UTF-8 encoding. This happens because the first 256
51 codepoints in Unicode happens to agree with Latin-1.
53 If you wish to interpret byte strings as UTF-8 instead, use the
58 See L</"Byte and Character Semantics"> for more details.
62 =head2 Byte and Character Semantics
64 Beginning with version 5.6, Perl uses logically-wide characters to
65 represent strings internally.
67 In future, Perl-level operations will be expected to work with
68 characters rather than bytes.
70 However, as an interim compatibility measure, Perl aims to
71 provide a safe migration path from byte semantics to character
72 semantics for programs. For operations where Perl can unambiguously
73 decide that the input data are characters, Perl switches to
74 character semantics. For operations where this determination cannot
75 be made without additional information from the user, Perl decides in
76 favor of compatibility and chooses to use byte semantics.
78 This behavior preserves compatibility with earlier versions of Perl,
79 which allowed byte semantics in Perl operations only if
80 none of the program's inputs were marked as being as source of Unicode
81 character data. Such data may come from filehandles, from calls to
82 external programs, from information provided by the system (such as %ENV),
83 or from literals and constants in the source text.
85 The C<bytes> pragma will always, regardless of platform, force byte
86 semantics in a particular lexical scope. See L<bytes>.
88 The C<utf8> pragma is primarily a compatibility device that enables
89 recognition of UTF-(8|EBCDIC) in literals encountered by the parser.
90 Note that this pragma is only required while Perl defaults to byte
91 semantics; when character semantics become the default, this pragma
92 may become a no-op. See L<utf8>.
94 Unless explicitly stated, Perl operators use character semantics
95 for Unicode data and byte semantics for non-Unicode data.
96 The decision to use character semantics is made transparently. If
97 input data comes from a Unicode source--for example, if a character
98 encoding layer is added to a filehandle or a literal Unicode
99 string constant appears in a program--character semantics apply.
100 Otherwise, byte semantics are in effect. The C<bytes> pragma should
101 be used to force byte semantics on Unicode data.
103 If strings operating under byte semantics and strings with Unicode
104 character data are concatenated, the new string will be created by
105 decoding the byte strings as I<ISO 8859-1 (Latin-1)>, even if the
106 old Unicode string used EBCDIC. This translation is done without
107 regard to the system's native 8-bit encoding. To change this for
108 systems with non-Latin-1 and non-EBCDIC native encodings, use the
109 C<encoding> pragma. See L<encoding>.
111 Under character semantics, many operations that formerly operated on
112 bytes now operate on characters. A character in Perl is
113 logically just a number ranging from 0 to 2**31 or so. Larger
114 characters may encode into longer sequences of bytes internally, but
115 this internal detail is mostly hidden for Perl code.
116 See L<perluniintro> for more.
118 =head2 Effects of Character Semantics
120 Character semantics have the following effects:
126 Strings--including hash keys--and regular expression patterns may
127 contain characters that have an ordinal value larger than 255.
129 If you use a Unicode editor to edit your program, Unicode characters
130 may occur directly within the literal strings in one of the various
131 Unicode encodings (UTF-8, UTF-EBCDIC, UCS-2, etc.), but will be recognized
132 as such and converted to Perl's internal representation only if the
133 appropriate L<encoding> is specified.
135 Unicode characters can also be added to a string by using the
136 C<\x{...}> notation. The Unicode code for the desired character, in
137 hexadecimal, should be placed in the braces. For instance, a smiley
138 face is C<\x{263A}>. This encoding scheme only works for characters
139 with a code of 0x100 or above.
143 use charnames ':full';
145 you can use the C<\N{...}> notation and put the official Unicode
146 character name within the braces, such as C<\N{WHITE SMILING FACE}>.
151 If an appropriate L<encoding> is specified, identifiers within the
152 Perl script may contain Unicode alphanumeric characters, including
153 ideographs. Perl does not currently attempt to canonicalize variable
158 Regular expressions match characters instead of bytes. "." matches
159 a character instead of a byte. The C<\C> pattern is provided to force
160 a match a single byte--a C<char> in C, hence C<\C>.
164 Character classes in regular expressions match characters instead of
165 bytes and match against the character properties specified in the
166 Unicode properties database. C<\w> can be used to match a Japanese
167 ideograph, for instance.
169 (However, and as a limitation of the current implementation, using
170 C<\w> or C<\W> I<inside> a C<[...]> character class will still match
171 with byte semantics.)
175 Named Unicode properties, scripts, and block ranges may be used like
176 character classes via the C<\p{}> "matches property" construct and
177 the C<\P{}> negation, "doesn't match property".
179 For instance, C<\p{Lu}> matches any character with the Unicode "Lu"
180 (Letter, uppercase) property, while C<\p{M}> matches any character
181 with an "M" (mark--accents and such) property. Brackets are not
182 required for single letter properties, so C<\p{M}> is equivalent to
183 C<\pM>. Many predefined properties are available, such as
184 C<\p{Mirrored}> and C<\p{Tibetan}>.
186 The official Unicode script and block names have spaces and dashes as
187 separators, but for convenience you can use dashes, spaces, or
188 underbars, and case is unimportant. It is recommended, however, that
189 for consistency you use the following naming: the official Unicode
190 script, property, or block name (see below for the additional rules
191 that apply to block names) with whitespace and dashes removed, and the
192 words "uppercase-first-lowercase-rest". C<Latin-1 Supplement> thus
193 becomes C<Latin1Supplement>.
195 You can also use negation in both C<\p{}> and C<\P{}> by introducing a caret
196 (^) between the first brace and the property name: C<\p{^Tamil}> is
197 equal to C<\P{Tamil}>.
199 B<NOTE: the properties, scripts, and blocks listed here are as of
200 Unicode 3.2.0, March 2002, or Perl 5.8.0, July 2002. Unicode 4.0.0
201 came out in April 2003, and Perl 5.8.1 in September 2003.>
203 Here are the basic Unicode General Category properties, followed by their
204 long form. You can use either; C<\p{Lu}> and C<\p{UppercaseLetter}>,
205 for instance, are identical.
228 Pc ConnectorPunctuation
232 Pi InitialPunctuation
233 (may behave like Ps or Pe depending on usage)
235 (may behave like Ps or Pe depending on usage)
247 Zp ParagraphSeparator
252 Cs Surrogate (not usable)
256 Single-letter properties match all characters in any of the
257 two-letter sub-properties starting with the same letter.
258 C<LC> and C<L&> are special cases, which are aliases for the set of
259 C<Ll>, C<Lu>, and C<Lt>.
261 Because Perl hides the need for the user to understand the internal
262 representation of Unicode characters, there is no need to implement
263 the somewhat messy concept of surrogates. C<Cs> is therefore not
266 Because scripts differ in their directionality--Hebrew is
267 written right to left, for example--Unicode supplies these properties in
273 LRE Left-to-Right Embedding
274 LRO Left-to-Right Override
276 AL Right-to-Left Arabic
277 RLE Right-to-Left Embedding
278 RLO Right-to-Left Override
279 PDF Pop Directional Format
281 ES European Number Separator
282 ET European Number Terminator
284 CS Common Number Separator
287 B Paragraph Separator
292 For example, C<\p{BidiClass:R}> matches characters that are normally
293 written right to left.
299 The script names which can be used by C<\p{...}> and C<\P{...}>,
300 such as in C<\p{Latin}> or C<\p{Cyrillic}>, are as follows:
347 Extended property classes can supplement the basic
348 properties, defined by the F<PropList> Unicode database:
363 LogicalOrderException
364 NoncharacterCodePoint
366 OtherDefaultIgnorableCodePoint
378 and there are further derived properties:
380 Alphabetic Lu + Ll + Lt + Lm + Lo + OtherAlphabetic
381 Lowercase Ll + OtherLowercase
382 Uppercase Lu + OtherUppercase
385 ID_Start Lu + Ll + Lt + Lm + Lo + Nl
386 ID_Continue ID_Start + Mn + Mc + Nd + Pc
389 Assigned Any non-Cn character (i.e. synonym for \P{Cn})
390 Unassigned Synonym for \p{Cn}
391 Common Any character (or unassigned code point)
392 not explicitly assigned to a script
394 For backward compatibility (with Perl 5.6), all properties mentioned
395 so far may have C<Is> prepended to their name, so C<\P{IsLu}>, for
396 example, is equal to C<\P{Lu}>.
400 In addition to B<scripts>, Unicode also defines B<blocks> of
401 characters. The difference between scripts and blocks is that the
402 concept of scripts is closer to natural languages, while the concept
403 of blocks is more of an artificial grouping based on groups of 256
404 Unicode characters. For example, the C<Latin> script contains letters
405 from many blocks but does not contain all the characters from those
406 blocks. It does not, for example, contain digits, because digits are
407 shared across many scripts. Digits and similar groups, like
408 punctuation, are in a category called C<Common>.
410 For more about scripts, see the UTR #24:
412 http://www.unicode.org/unicode/reports/tr24/
414 For more about blocks, see:
416 http://www.unicode.org/Public/UNIDATA/Blocks.txt
418 Block names are given with the C<In> prefix. For example, the
419 Katakana block is referenced via C<\p{InKatakana}>. The C<In>
420 prefix may be omitted if there is no naming conflict with a script
421 or any other property, but it is recommended that C<In> always be used
422 for block tests to avoid confusion.
424 These block names are supported:
426 InAlphabeticPresentationForms
428 InArabicPresentationFormsA
429 InArabicPresentationFormsB
440 InByzantineMusicalSymbols
442 InCJKCompatibilityForms
443 InCJKCompatibilityIdeographs
444 InCJKCompatibilityIdeographsSupplement
445 InCJKRadicalsSupplement
446 InCJKSymbolsAndPunctuation
447 InCJKUnifiedIdeographs
448 InCJKUnifiedIdeographsExtensionA
449 InCJKUnifiedIdeographsExtensionB
451 InCombiningDiacriticalMarks
452 InCombiningDiacriticalMarksforSymbols
457 InCyrillicSupplementary
461 InEnclosedAlphanumerics
462 InEnclosedCJKLettersAndMonths
472 InHalfwidthAndFullwidthForms
473 InHangulCompatibilityJamo
478 InHighPrivateUseSurrogates
482 InIdeographicDescriptionCharacters
487 InKatakanaPhoneticExtensions
492 InLatinExtendedAdditional
497 InMathematicalAlphanumericSymbols
498 InMathematicalOperators
499 InMiscellaneousMathematicalSymbolsA
500 InMiscellaneousMathematicalSymbolsB
501 InMiscellaneousSymbols
502 InMiscellaneousTechnical
509 InOpticalCharacterRecognition
515 InSpacingModifierLetters
517 InSuperscriptsAndSubscripts
518 InSupplementalArrowsA
519 InSupplementalArrowsB
520 InSupplementalMathematicalOperators
521 InSupplementaryPrivateUseAreaA
522 InSupplementaryPrivateUseAreaB
532 InUnifiedCanadianAboriginalSyllabics
541 The special pattern C<\X> matches any extended Unicode
542 sequence--"a combining character sequence" in Standardese--where the
543 first character is a base character and subsequent characters are mark
544 characters that apply to the base character. C<\X> is equivalent to
549 The C<tr///> operator translates characters instead of bytes. Note
550 that the C<tr///CU> functionality has been removed. For similar
551 functionality see pack('U0', ...) and pack('C0', ...).
555 Case translation operators use the Unicode case translation tables
556 when character input is provided. Note that C<uc()>, or C<\U> in
557 interpolated strings, translates to uppercase, while C<ucfirst>,
558 or C<\u> in interpolated strings, translates to titlecase in languages
559 that make the distinction.
563 Most operators that deal with positions or lengths in a string will
564 automatically switch to using character positions, including
565 C<chop()>, C<chomp()>, C<substr()>, C<pos()>, C<index()>, C<rindex()>,
566 C<sprintf()>, C<write()>, and C<length()>. Operators that
567 specifically do not switch include C<vec()>, C<pack()>, and
568 C<unpack()>. Operators that really don't care include
569 operators that treats strings as a bucket of bits such as C<sort()>,
570 and operators dealing with filenames.
574 The C<pack()>/C<unpack()> letters C<c> and C<C> do I<not> change,
575 since they are often used for byte-oriented formats. Again, think
576 C<char> in the C language.
578 There is a new C<U> specifier that converts between Unicode characters
583 The C<chr()> and C<ord()> functions work on characters, similar to
584 C<pack("U")> and C<unpack("U")>, I<not> C<pack("C")> and
585 C<unpack("C")>. C<pack("C")> and C<unpack("C")> are methods for
586 emulating byte-oriented C<chr()> and C<ord()> on Unicode strings.
587 While these methods reveal the internal encoding of Unicode strings,
588 that is not something one normally needs to care about at all.
592 The bit string operators, C<& | ^ ~>, can operate on character data.
593 However, for backward compatibility, such as when using bit string
594 operations when characters are all less than 256 in ordinal value, one
595 should not use C<~> (the bit complement) with characters of both
596 values less than 256 and values greater than 256. Most importantly,
597 DeMorgan's laws (C<~($x|$y) eq ~$x&~$y> and C<~($x&$y) eq ~$x|~$y>)
598 will not hold. The reason for this mathematical I<faux pas> is that
599 the complement cannot return B<both> the 8-bit (byte-wide) bit
600 complement B<and> the full character-wide bit complement.
604 lc(), uc(), lcfirst(), and ucfirst() work for the following cases:
610 the case mapping is from a single Unicode character to another
611 single Unicode character, or
615 the case mapping is from a single Unicode character to more
616 than one Unicode character.
620 Things to do with locales (Lithuanian, Turkish, Azeri) do B<not> work
621 since Perl does not understand the concept of Unicode locales.
623 See the Unicode Technical Report #21, Case Mappings, for more details.
631 And finally, C<scalar reverse()> reverses by character rather than by byte.
635 =head2 User-Defined Character Properties
637 You can define your own character properties by defining subroutines
638 whose names begin with "In" or "Is". The subroutines can be defined in
639 any package. The user-defined properties can be used in the regular
640 expression C<\p> and C<\P> constructs; if you are using a user-defined
641 property from a package other than the one you are in, you must specify
642 its package in the C<\p> or C<\P> construct.
644 # assuming property IsForeign defined in Lang::
645 package main; # property package name required
646 if ($txt =~ /\p{Lang::IsForeign}+/) { ... }
648 package Lang; # property package name not required
649 if ($txt =~ /\p{IsForeign}+/) { ... }
652 Note that the effect is compile-time and immutable once defined.
654 The subroutines must return a specially-formatted string, with one
655 or more newline-separated lines. Each line must be one of the following:
661 Two hexadecimal numbers separated by horizontal whitespace (space or
662 tabular characters) denoting a range of Unicode code points to include.
666 Something to include, prefixed by "+": a built-in character
667 property (prefixed by "utf8::") or a user-defined character property,
668 to represent all the characters in that property; two hexadecimal code
669 points for a range; or a single hexadecimal code point.
673 Something to exclude, prefixed by "-": an existing character
674 property (prefixed by "utf8::") or a user-defined character property,
675 to represent all the characters in that property; two hexadecimal code
676 points for a range; or a single hexadecimal code point.
680 Something to negate, prefixed "!": an existing character
681 property (prefixed by "utf8::") or a user-defined character property,
682 to represent all the characters in that property; two hexadecimal code
683 points for a range; or a single hexadecimal code point.
687 Something to intersect with, prefixed by "&": an existing character
688 property (prefixed by "utf8::") or a user-defined character property,
689 for all the characters except the characters in the property; two
690 hexadecimal code points for a range; or a single hexadecimal code point.
694 For example, to define a property that covers both the Japanese
695 syllabaries (hiragana and katakana), you can define
704 Imagine that the here-doc end marker is at the beginning of the line.
705 Now you can use C<\p{InKana}> and C<\P{InKana}>.
707 You could also have used the existing block property names:
716 Suppose you wanted to match only the allocated characters,
717 not the raw block ranges: in other words, you want to remove
728 The negation is useful for defining (surprise!) negated classes.
738 Intersection is useful for getting the common characters matched by
739 two (or more) classes.
748 It's important to remember not to use "&" for the first set -- that
749 would be intersecting with nothing (resulting in an empty set).
751 You can also define your own mappings to be used in the lc(),
752 lcfirst(), uc(), and ucfirst() (or their string-inlined versions).
753 The principle is the same: define subroutines in the C<main> package
754 with names like C<ToLower> (for lc() and lcfirst()), C<ToTitle> (for
755 the first character in ucfirst()), and C<ToUpper> (for uc(), and the
756 rest of the characters in ucfirst()).
758 The string returned by the subroutines needs now to be three
759 hexadecimal numbers separated by tabulators: start of the source
760 range, end of the source range, and start of the destination range.
769 defines an uc() mapping that causes only the characters "a", "b", and
770 "c" to be mapped to "A", "B", "C", all other characters will remain
773 If there is no source range to speak of, that is, the mapping is from
774 a single character to another single character, leave the end of the
775 source range empty, but the two tabulator characters are still needed.
784 defines a lc() mapping that causes only "A" to be mapped to "a", all
785 other characters will remain unchanged.
787 (For serious hackers only) If you want to introspect the default
788 mappings, you can find the data in the directory
789 C<$Config{privlib}>/F<unicore/To/>. The mapping data is returned as
790 the here-document, and the C<utf8::ToSpecFoo> are special exception
791 mappings derived from <$Config{privlib}>/F<unicore/SpecialCasing.txt>.
792 The C<Digit> and C<Fold> mappings that one can see in the directory
793 are not directly user-accessible, one can use either the
794 C<Unicode::UCD> module, or just match case-insensitively (that's when
795 the C<Fold> mapping is used).
797 A final note on the user-defined property tests and mappings: they
798 will be used only if the scalar has been marked as having Unicode
799 characters. Old byte-style strings will not be affected.
801 =head2 Character Encodings for Input and Output
805 =head2 Unicode Regular Expression Support Level
807 The following list of Unicode support for regular expressions describes
808 all the features currently supported. The references to "Level N"
809 and the section numbers refer to the Unicode Technical Report 18,
810 "Unicode Regular Expression Guidelines", version 6 (Unicode 3.2.0,
817 Level 1 - Basic Unicode Support
819 2.1 Hex Notation - done [1]
820 Named Notation - done [2]
821 2.2 Categories - done [3][4]
822 2.3 Subtraction - MISSING [5][6]
823 2.4 Simple Word Boundaries - done [7]
824 2.5 Simple Loose Matches - done [8]
825 2.6 End of Line - MISSING [9][10]
829 [ 3] . \p{...} \P{...}
830 [ 4] support for scripts (see UTR#24 Script Names), blocks,
831 binary properties, enumerated non-binary properties, and
832 numeric properties (as listed in UTR#18 Other Properties)
834 [ 6] can use regular expression look-ahead [a]
835 or user-defined character properties [b] to emulate subtraction
836 [ 7] include Letters in word characters
837 [ 8] note that Perl does Full case-folding in matching, not Simple:
838 for example U+1F88 is equivalent with U+1F00 U+03B9,
839 not with 1F80. This difference matters for certain Greek
840 capital letters with certain modifiers: the Full case-folding
841 decomposes the letter, while the Simple case-folding would map
842 it to a single character.
843 [ 9] see UTR #13 Unicode Newline Guidelines
844 [10] should do ^ and $ also on \x{85}, \x{2028} and \x{2029}
845 (should also affect <>, $., and script line numbers)
846 (the \x{85}, \x{2028} and \x{2029} do match \s)
848 [a] You can mimic class subtraction using lookahead.
849 For example, what UTR #18 might write as
851 [{Greek}-[{UNASSIGNED}]]
853 in Perl can be written as:
855 (?!\p{Unassigned})\p{InGreekAndCoptic}
856 (?=\p{Assigned})\p{InGreekAndCoptic}
858 But in this particular example, you probably really want
862 which will match assigned characters known to be part of the Greek script.
864 Also see the Unicode::Regex::Set module, it does implement the full
865 UTR #18 grouping, intersection, union, and removal (subtraction) syntax.
867 [b] See L</"User-Defined Character Properties">.
871 Level 2 - Extended Unicode Support
873 3.1 Surrogates - MISSING [11]
874 3.2 Canonical Equivalents - MISSING [12][13]
875 3.3 Locale-Independent Graphemes - MISSING [14]
876 3.4 Locale-Independent Words - MISSING [15]
877 3.5 Locale-Independent Loose Matches - MISSING [16]
879 [11] Surrogates are solely a UTF-16 concept and Perl's internal
880 representation is UTF-8. The Encode module does UTF-16, though.
881 [12] see UTR#15 Unicode Normalization
882 [13] have Unicode::Normalize but not integrated to regexes
883 [14] have \X but at this level . should equal that
884 [15] need three classes, not just \w and \W
885 [16] see UTR#21 Case Mappings
889 Level 3 - Locale-Sensitive Support
891 4.1 Locale-Dependent Categories - MISSING
892 4.2 Locale-Dependent Graphemes - MISSING [16][17]
893 4.3 Locale-Dependent Words - MISSING
894 4.4 Locale-Dependent Loose Matches - MISSING
895 4.5 Locale-Dependent Ranges - MISSING
897 [16] see UTR#10 Unicode Collation Algorithms
898 [17] have Unicode::Collate but not integrated to regexes
902 =head2 Unicode Encodings
904 Unicode characters are assigned to I<code points>, which are abstract
905 numbers. To use these numbers, various encodings are needed.
913 UTF-8 is a variable-length (1 to 6 bytes, current character allocations
914 require 4 bytes), byte-order independent encoding. For ASCII (and we
915 really do mean 7-bit ASCII, not another 8-bit encoding), UTF-8 is
918 The following table is from Unicode 3.2.
920 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
922 U+0000..U+007F 00..7F
923 U+0080..U+07FF C2..DF 80..BF
924 U+0800..U+0FFF E0 A0..BF 80..BF
925 U+1000..U+CFFF E1..EC 80..BF 80..BF
926 U+D000..U+D7FF ED 80..9F 80..BF
927 U+D800..U+DFFF ******* ill-formed *******
928 U+E000..U+FFFF EE..EF 80..BF 80..BF
929 U+10000..U+3FFFF F0 90..BF 80..BF 80..BF
930 U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
931 U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
933 Note the C<A0..BF> in C<U+0800..U+0FFF>, the C<80..9F> in
934 C<U+D000...U+D7FF>, the C<90..B>F in C<U+10000..U+3FFFF>, and the
935 C<80...8F> in C<U+100000..U+10FFFF>. The "gaps" are caused by legal
936 UTF-8 avoiding non-shortest encodings: it is technically possible to
937 UTF-8-encode a single code point in different ways, but that is
938 explicitly forbidden, and the shortest possible encoding should always
939 be used. So that's what Perl does.
941 Another way to look at it is via bits:
943 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
946 00000bbbbbaaaaaa 110bbbbb 10aaaaaa
947 ccccbbbbbbaaaaaa 1110cccc 10bbbbbb 10aaaaaa
948 00000dddccccccbbbbbbaaaaaa 11110ddd 10cccccc 10bbbbbb 10aaaaaa
950 As you can see, the continuation bytes all begin with C<10>, and the
951 leading bits of the start byte tell how many bytes the are in the
958 Like UTF-8 but EBCDIC-safe, in the way that UTF-8 is ASCII-safe.
962 UTF-16, UTF-16BE, UTF-16LE, Surrogates, and BOMs (Byte Order Marks)
964 The followings items are mostly for reference and general Unicode
965 knowledge, Perl doesn't use these constructs internally.
967 UTF-16 is a 2 or 4 byte encoding. The Unicode code points
968 C<U+0000..U+FFFF> are stored in a single 16-bit unit, and the code
969 points C<U+10000..U+10FFFF> in two 16-bit units. The latter case is
970 using I<surrogates>, the first 16-bit unit being the I<high
971 surrogate>, and the second being the I<low surrogate>.
973 Surrogates are code points set aside to encode the C<U+10000..U+10FFFF>
974 range of Unicode code points in pairs of 16-bit units. The I<high
975 surrogates> are the range C<U+D800..U+DBFF>, and the I<low surrogates>
976 are the range C<U+DC00..U+DFFF>. The surrogate encoding is
978 $hi = ($uni - 0x10000) / 0x400 + 0xD800;
979 $lo = ($uni - 0x10000) % 0x400 + 0xDC00;
983 $uni = 0x10000 + ($hi - 0xD800) * 0x400 + ($lo - 0xDC00);
985 If you try to generate surrogates (for example by using chr()), you
986 will get a warning if warnings are turned on, because those code
987 points are not valid for a Unicode character.
989 Because of the 16-bitness, UTF-16 is byte-order dependent. UTF-16
990 itself can be used for in-memory computations, but if storage or
991 transfer is required either UTF-16BE (big-endian) or UTF-16LE
992 (little-endian) encodings must be chosen.
994 This introduces another problem: what if you just know that your data
995 is UTF-16, but you don't know which endianness? Byte Order Marks, or
996 BOMs, are a solution to this. A special character has been reserved
997 in Unicode to function as a byte order marker: the character with the
998 code point C<U+FEFF> is the BOM.
1000 The trick is that if you read a BOM, you will know the byte order,
1001 since if it was written on a big-endian platform, you will read the
1002 bytes C<0xFE 0xFF>, but if it was written on a little-endian platform,
1003 you will read the bytes C<0xFF 0xFE>. (And if the originating platform
1004 was writing in UTF-8, you will read the bytes C<0xEF 0xBB 0xBF>.)
1006 The way this trick works is that the character with the code point
1007 C<U+FFFE> is guaranteed not to be a valid Unicode character, so the
1008 sequence of bytes C<0xFF 0xFE> is unambiguously "BOM, represented in
1009 little-endian format" and cannot be C<U+FFFE>, represented in big-endian
1014 UTF-32, UTF-32BE, UTF-32LE
1016 The UTF-32 family is pretty much like the UTF-16 family, expect that
1017 the units are 32-bit, and therefore the surrogate scheme is not
1018 needed. The BOM signatures will be C<0x00 0x00 0xFE 0xFF> for BE and
1019 C<0xFF 0xFE 0x00 0x00> for LE.
1025 Encodings defined by the ISO 10646 standard. UCS-2 is a 16-bit
1026 encoding. Unlike UTF-16, UCS-2 is not extensible beyond C<U+FFFF>,
1027 because it does not use surrogates. UCS-4 is a 32-bit encoding,
1028 functionally identical to UTF-32.
1034 A seven-bit safe (non-eight-bit) encoding, which is useful if the
1035 transport or storage is not eight-bit safe. Defined by RFC 2152.
1039 =head2 Security Implications of Unicode
1047 Unfortunately, the specification of UTF-8 leaves some room for
1048 interpretation of how many bytes of encoded output one should generate
1049 from one input Unicode character. Strictly speaking, the shortest
1050 possible sequence of UTF-8 bytes should be generated,
1051 because otherwise there is potential for an input buffer overflow at
1052 the receiving end of a UTF-8 connection. Perl always generates the
1053 shortest length UTF-8, and with warnings on Perl will warn about
1054 non-shortest length UTF-8 along with other malformations, such as the
1055 surrogates, which are not real Unicode code points.
1059 Regular expressions behave slightly differently between byte data and
1060 character (Unicode) data. For example, the "word character" character
1061 class C<\w> will work differently depending on if data is eight-bit bytes
1064 In the first case, the set of C<\w> characters is either small--the
1065 default set of alphabetic characters, digits, and the "_"--or, if you
1066 are using a locale (see L<perllocale>), the C<\w> might contain a few
1067 more letters according to your language and country.
1069 In the second case, the C<\w> set of characters is much, much larger.
1070 Most importantly, even in the set of the first 256 characters, it will
1071 probably match different characters: unlike most locales, which are
1072 specific to a language and country pair, Unicode classifies all the
1073 characters that are letters I<somewhere> as C<\w>. For example, your
1074 locale might not think that LATIN SMALL LETTER ETH is a letter (unless
1075 you happen to speak Icelandic), but Unicode does.
1077 As discussed elsewhere, Perl has one foot (two hooves?) planted in
1078 each of two worlds: the old world of bytes and the new world of
1079 characters, upgrading from bytes to characters when necessary.
1080 If your legacy code does not explicitly use Unicode, no automatic
1081 switch-over to characters should happen. Characters shouldn't get
1082 downgraded to bytes, either. It is possible to accidentally mix bytes
1083 and characters, however (see L<perluniintro>), in which case C<\w> in
1084 regular expressions might start behaving differently. Review your
1085 code. Use warnings and the C<strict> pragma.
1089 =head2 Unicode in Perl on EBCDIC
1091 The way Unicode is handled on EBCDIC platforms is still
1092 experimental. On such platforms, references to UTF-8 encoding in this
1093 document and elsewhere should be read as meaning the UTF-EBCDIC
1094 specified in Unicode Technical Report 16, unless ASCII vs. EBCDIC issues
1095 are specifically discussed. There is no C<utfebcdic> pragma or
1096 ":utfebcdic" layer; rather, "utf8" and ":utf8" are reused to mean
1097 the platform's "natural" 8-bit encoding of Unicode. See L<perlebcdic>
1098 for more discussion of the issues.
1102 Usually locale settings and Unicode do not affect each other, but
1103 there are a couple of exceptions:
1109 You can enable automatic UTF-8-ification of your standard file
1110 handles, default C<open()> layer, and C<@ARGV> by using either
1111 the C<-C> command line switch or the C<PERL_UNICODE> environment
1112 variable, see L<perlrun> for the documentation of the C<-C> switch.
1116 Perl tries really hard to work both with Unicode and the old
1117 byte-oriented world. Most often this is nice, but sometimes Perl's
1118 straddling of the proverbial fence causes problems.
1122 =head2 When Unicode Does Not Happen
1124 While Perl does have extensive ways to input and output in Unicode,
1125 and few other 'entry points' like the @ARGV which can be interpreted
1126 as Unicode (UTF-8), there still are many places where Unicode (in some
1127 encoding or another) could be given as arguments or received as
1128 results, or both, but it is not.
1130 The following are such interfaces. For all of these interfaces Perl
1131 currently (as of 5.8.3) simply assumes byte strings both as arguments
1132 and results, or UTF-8 strings if the C<encoding> pragma has been used.
1134 One reason why Perl does not attempt to resolve the role of Unicode in
1135 this cases is that the answers are highly dependent on the operating
1136 system and the file system(s). For example, whether filenames can be
1137 in Unicode, and in exactly what kind of encoding, is not exactly a
1138 portable concept. Similarly for the qx and system: how well will the
1139 'command line interface' (and which of them?) handle Unicode?
1145 chmod, chmod, chown, chroot, exec, link, lstat, mkdir,
1146 rename, rmdir, stat, symlink, truncate, unlink, utime, -X
1158 open, opendir, sysopen
1162 qx (aka the backtick operator), system
1170 =head2 Forcing Unicode in Perl (Or Unforcing Unicode in Perl)
1172 Sometimes (see L</"When Unicode Does Not Happen">) there are
1173 situations where you simply need to force Perl to believe that a byte
1174 string is UTF-8, or vice versa. The low-level calls
1175 utf8::upgrade($bytestring) and utf8::downgrade($utf8string) are
1178 Do not use them without careful thought, though: Perl may easily get
1179 very confused, angry, or even crash, if you suddenly change the 'nature'
1180 of scalar like that. Especially careful you have to be if you use the
1181 utf8::upgrade(): any random byte string is not valid UTF-8.
1183 =head2 Using Unicode in XS
1185 If you want to handle Perl Unicode in XS extensions, you may find the
1186 following C APIs useful. See also L<perlguts/"Unicode Support"> for an
1187 explanation about Unicode at the XS level, and L<perlapi> for the API
1194 C<DO_UTF8(sv)> returns true if the C<UTF8> flag is on and the bytes
1195 pragma is not in effect. C<SvUTF8(sv)> returns true is the C<UTF8>
1196 flag is on; the bytes pragma is ignored. The C<UTF8> flag being on
1197 does B<not> mean that there are any characters of code points greater
1198 than 255 (or 127) in the scalar or that there are even any characters
1199 in the scalar. What the C<UTF8> flag means is that the sequence of
1200 octets in the representation of the scalar is the sequence of UTF-8
1201 encoded code points of the characters of a string. The C<UTF8> flag
1202 being off means that each octet in this representation encodes a
1203 single character with code point 0..255 within the string. Perl's
1204 Unicode model is not to use UTF-8 until it is absolutely necessary.
1208 C<uvuni_to_utf8(buf, chr)> writes a Unicode character code point into
1209 a buffer encoding the code point as UTF-8, and returns a pointer
1210 pointing after the UTF-8 bytes.
1214 C<utf8_to_uvuni(buf, lenp)> reads UTF-8 encoded bytes from a buffer and
1215 returns the Unicode character code point and, optionally, the length of
1216 the UTF-8 byte sequence.
1220 C<utf8_length(start, end)> returns the length of the UTF-8 encoded buffer
1221 in characters. C<sv_len_utf8(sv)> returns the length of the UTF-8 encoded
1226 C<sv_utf8_upgrade(sv)> converts the string of the scalar to its UTF-8
1227 encoded form. C<sv_utf8_downgrade(sv)> does the opposite, if
1228 possible. C<sv_utf8_encode(sv)> is like sv_utf8_upgrade except that
1229 it does not set the C<UTF8> flag. C<sv_utf8_decode()> does the
1230 opposite of C<sv_utf8_encode()>. Note that none of these are to be
1231 used as general-purpose encoding or decoding interfaces: C<use Encode>
1232 for that. C<sv_utf8_upgrade()> is affected by the encoding pragma
1233 but C<sv_utf8_downgrade()> is not (since the encoding pragma is
1234 designed to be a one-way street).
1238 C<is_utf8_char(s)> returns true if the pointer points to a valid UTF-8
1243 C<is_utf8_string(buf, len)> returns true if C<len> bytes of the buffer
1248 C<UTF8SKIP(buf)> will return the number of bytes in the UTF-8 encoded
1249 character in the buffer. C<UNISKIP(chr)> will return the number of bytes
1250 required to UTF-8-encode the Unicode character code point. C<UTF8SKIP()>
1251 is useful for example for iterating over the characters of a UTF-8
1252 encoded buffer; C<UNISKIP()> is useful, for example, in computing
1253 the size required for a UTF-8 encoded buffer.
1257 C<utf8_distance(a, b)> will tell the distance in characters between the
1258 two pointers pointing to the same UTF-8 encoded buffer.
1262 C<utf8_hop(s, off)> will return a pointer to an UTF-8 encoded buffer
1263 that is C<off> (positive or negative) Unicode characters displaced
1264 from the UTF-8 buffer C<s>. Be careful not to overstep the buffer:
1265 C<utf8_hop()> will merrily run off the end or the beginning of the
1266 buffer if told to do so.
1270 C<pv_uni_display(dsv, spv, len, pvlim, flags)> and
1271 C<sv_uni_display(dsv, ssv, pvlim, flags)> are useful for debugging the
1272 output of Unicode strings and scalars. By default they are useful
1273 only for debugging--they display B<all> characters as hexadecimal code
1274 points--but with the flags C<UNI_DISPLAY_ISPRINT>,
1275 C<UNI_DISPLAY_BACKSLASH>, and C<UNI_DISPLAY_QQ> you can make the
1276 output more readable.
1280 C<ibcmp_utf8(s1, pe1, u1, l1, u1, s2, pe2, l2, u2)> can be used to
1281 compare two strings case-insensitively in Unicode. For case-sensitive
1282 comparisons you can just use C<memEQ()> and C<memNE()> as usual.
1286 For more information, see L<perlapi>, and F<utf8.c> and F<utf8.h>
1287 in the Perl source code distribution.
1291 =head2 Interaction with Locales
1293 Use of locales with Unicode data may lead to odd results. Currently,
1294 Perl attempts to attach 8-bit locale info to characters in the range
1295 0..255, but this technique is demonstrably incorrect for locales that
1296 use characters above that range when mapped into Unicode. Perl's
1297 Unicode support will also tend to run slower. Use of locales with
1298 Unicode is discouraged.
1300 =head2 Interaction with Extensions
1302 When Perl exchanges data with an extension, the extension should be
1303 able to understand the UTF-8 flag and act accordingly. If the
1304 extension doesn't know about the flag, it's likely that the extension
1305 will return incorrectly-flagged data.
1307 So if you're working with Unicode data, consult the documentation of
1308 every module you're using if there are any issues with Unicode data
1309 exchange. If the documentation does not talk about Unicode at all,
1310 suspect the worst and probably look at the source to learn how the
1311 module is implemented. Modules written completely in Perl shouldn't
1312 cause problems. Modules that directly or indirectly access code written
1313 in other programming languages are at risk.
1315 For affected functions, the simple strategy to avoid data corruption is
1316 to always make the encoding of the exchanged data explicit. Choose an
1317 encoding that you know the extension can handle. Convert arguments passed
1318 to the extensions to that encoding and convert results back from that
1319 encoding. Write wrapper functions that do the conversions for you, so
1320 you can later change the functions when the extension catches up.
1322 To provide an example, let's say the popular Foo::Bar::escape_html
1323 function doesn't deal with Unicode data yet. The wrapper function
1324 would convert the argument to raw UTF-8 and convert the result back to
1325 Perl's internal representation like so:
1327 sub my_escape_html ($) {
1329 return unless defined $what;
1330 Encode::decode_utf8(Foo::Bar::escape_html(Encode::encode_utf8($what)));
1333 Sometimes, when the extension does not convert data but just stores
1334 and retrieves them, you will be in a position to use the otherwise
1335 dangerous Encode::_utf8_on() function. Let's say the popular
1336 C<Foo::Bar> extension, written in C, provides a C<param> method that
1337 lets you store and retrieve data according to these prototypes:
1339 $self->param($name, $value); # set a scalar
1340 $value = $self->param($name); # retrieve a scalar
1342 If it does not yet provide support for any encoding, one could write a
1343 derived class with such a C<param> method:
1346 my($self,$name,$value) = @_;
1347 utf8::upgrade($name); # make sure it is UTF-8 encoded
1349 utf8::upgrade($value); # make sure it is UTF-8 encoded
1350 return $self->SUPER::param($name,$value);
1352 my $ret = $self->SUPER::param($name);
1353 Encode::_utf8_on($ret); # we know, it is UTF-8 encoded
1358 Some extensions provide filters on data entry/exit points, such as
1359 DB_File::filter_store_key and family. Look out for such filters in
1360 the documentation of your extensions, they can make the transition to
1361 Unicode data much easier.
1365 Some functions are slower when working on UTF-8 encoded strings than
1366 on byte encoded strings. All functions that need to hop over
1367 characters such as length(), substr() or index(), or matching regular
1368 expressions can work B<much> faster when the underlying data are
1371 In Perl 5.8.0 the slowness was often quite spectacular; in Perl 5.8.1
1372 a caching scheme was introduced which will hopefully make the slowness
1373 somewhat less spectacular, at least for some operations. In general,
1374 operations with UTF-8 encoded strings are still slower. As an example,
1375 the Unicode properties (character classes) like C<\p{Nd}> are known to
1376 be quite a bit slower (5-20 times) than their simpler counterparts
1377 like C<\d> (then again, there 268 Unicode characters matching C<Nd>
1378 compared with the 10 ASCII characters matching C<d>).
1380 =head2 Porting code from perl-5.6.X
1382 Perl 5.8 has a different Unicode model from 5.6. In 5.6 the programmer
1383 was required to use the C<utf8> pragma to declare that a given scope
1384 expected to deal with Unicode data and had to make sure that only
1385 Unicode data were reaching that scope. If you have code that is
1386 working with 5.6, you will need some of the following adjustments to
1387 your code. The examples are written such that the code will continue
1388 to work under 5.6, so you should be safe to try them out.
1394 A filehandle that should read or write UTF-8
1397 binmode $fh, ":utf8";
1402 A scalar that is going to be passed to some extension
1404 Be it Compress::Zlib, Apache::Request or any extension that has no
1405 mention of Unicode in the manpage, you need to make sure that the
1406 UTF-8 flag is stripped off. Note that at the time of this writing
1407 (October 2002) the mentioned modules are not UTF-8-aware. Please
1408 check the documentation to verify if this is still true.
1412 $val = Encode::encode_utf8($val); # make octets
1417 A scalar we got back from an extension
1419 If you believe the scalar comes back as UTF-8, you will most likely
1420 want the UTF-8 flag restored:
1424 $val = Encode::decode_utf8($val);
1429 Same thing, if you are really sure it is UTF-8
1433 Encode::_utf8_on($val);
1438 A wrapper for fetchrow_array and fetchrow_hashref
1440 When the database contains only UTF-8, a wrapper function or method is
1441 a convenient way to replace all your fetchrow_array and
1442 fetchrow_hashref calls. A wrapper function will also make it easier to
1443 adapt to future enhancements in your database driver. Note that at the
1444 time of this writing (October 2002), the DBI has no standardized way
1445 to deal with UTF-8 data. Please check the documentation to verify if
1449 my($self, $sth, $what) = @_; # $what is one of fetchrow_{array,hashref}
1455 my @arr = $sth->$what;
1457 defined && /[^\000-\177]/ && Encode::_utf8_on($_);
1461 my $ret = $sth->$what;
1463 for my $k (keys %$ret) {
1464 defined && /[^\000-\177]/ && Encode::_utf8_on($_) for $ret->{$k};
1468 defined && /[^\000-\177]/ && Encode::_utf8_on($_) for $ret;
1478 A large scalar that you know can only contain ASCII
1480 Scalars that contain only ASCII and are marked as UTF-8 are sometimes
1481 a drag to your program. If you recognize such a situation, just remove
1484 utf8::downgrade($val) if $] > 5.007;
1490 L<perluniintro>, L<encoding>, L<Encode>, L<open>, L<utf8>, L<bytes>,
1491 L<perlretut>, L<perlvar/"${^UNICODE}">
projects / p5sagit/p5-mst-13.2.git / blob