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>.
47 =head2 Byte and Character Semantics
49 Beginning with version 5.6, Perl uses logically-wide characters to
50 represent strings internally.
52 In future, Perl-level operations will be expected to work with
53 characters rather than bytes.
55 However, as an interim compatibility measure, Perl aims to
56 provide a safe migration path from byte semantics to character
57 semantics for programs. For operations where Perl can unambiguously
58 decide that the input data are characters, Perl switches to
59 character semantics. For operations where this determination cannot
60 be made without additional information from the user, Perl decides in
61 favor of compatibility and chooses to use byte semantics.
63 This behavior preserves compatibility with earlier versions of Perl,
64 which allowed byte semantics in Perl operations only if
65 none of the program's inputs were marked as being as source of Unicode
66 character data. Such data may come from filehandles, from calls to
67 external programs, from information provided by the system (such as %ENV),
68 or from literals and constants in the source text.
70 The C<bytes> pragma will always, regardless of platform, force byte
71 semantics in a particular lexical scope. See L<bytes>.
73 The C<utf8> pragma is primarily a compatibility device that enables
74 recognition of UTF-(8|EBCDIC) in literals encountered by the parser.
75 Note that this pragma is only required while Perl defaults to byte
76 semantics; when character semantics become the default, this pragma
77 may become a no-op. See L<utf8>.
79 Unless explicitly stated, Perl operators use character semantics
80 for Unicode data and byte semantics for non-Unicode data.
81 The decision to use character semantics is made transparently. If
82 input data comes from a Unicode source--for example, if a character
83 encoding layer is added to a filehandle or a literal Unicode
84 string constant appears in a program--character semantics apply.
85 Otherwise, byte semantics are in effect. The C<bytes> pragma should
86 be used to force byte semantics on Unicode data.
88 If strings operating under byte semantics and strings with Unicode
89 character data are concatenated, the new string will be upgraded to
90 I<ISO 8859-1 (Latin-1)>, even if the old Unicode string used EBCDIC.
91 This translation is done without regard to the system's native 8-bit
92 encoding, so to change this for systems with non-Latin-1 and
93 non-EBCDIC native encodings use the C<encoding> pragma. See
96 Under character semantics, many operations that formerly operated on
97 bytes now operate on characters. A character in Perl is
98 logically just a number ranging from 0 to 2**31 or so. Larger
99 characters may encode into longer sequences of bytes internally, but
100 this internal detail is mostly hidden for Perl code.
101 See L<perluniintro> for more.
103 =head2 Effects of Character Semantics
105 Character semantics have the following effects:
111 Strings--including hash keys--and regular expression patterns may
112 contain characters that have an ordinal value larger than 255.
114 If you use a Unicode editor to edit your program, Unicode characters
115 may occur directly within the literal strings in one of the various
116 Unicode encodings (UTF-8, UTF-EBCDIC, UCS-2, etc.), but will be recognized
117 as such and converted to Perl's internal representation only if the
118 appropriate L<encoding> is specified.
120 Unicode characters can also be added to a string by using the
121 C<\x{...}> notation. The Unicode code for the desired character, in
122 hexadecimal, should be placed in the braces. For instance, a smiley
123 face is C<\x{263A}>. This encoding scheme only works for characters
124 with a code of 0x100 or above.
128 use charnames ':full';
130 you can use the C<\N{...}> notation and put the official Unicode
131 character name within the braces, such as C<\N{WHITE SMILING FACE}>.
136 If an appropriate L<encoding> is specified, identifiers within the
137 Perl script may contain Unicode alphanumeric characters, including
138 ideographs. Perl does not currently attempt to canonicalize variable
143 Regular expressions match characters instead of bytes. "." matches
144 a character instead of a byte. The C<\C> pattern is provided to force
145 a match a single byte--a C<char> in C, hence C<\C>.
149 Character classes in regular expressions match characters instead of
150 bytes and match against the character properties specified in the
151 Unicode properties database. C<\w> can be used to match a Japanese
152 ideograph, for instance.
156 Named Unicode properties, scripts, and block ranges may be used like
157 character classes via the C<\p{}> "matches property" construct and
158 the C<\P{}> negation, "doesn't match property".
160 For instance, C<\p{Lu}> matches any character with the Unicode "Lu"
161 (Letter, uppercase) property, while C<\p{M}> matches any character
162 with an "M" (mark--accents and such) property. Brackets are not
163 required for single letter properties, so C<\p{M}> is equivalent to
164 C<\pM>. Many predefined properties are available, such as
165 C<\p{Mirrored}> and C<\p{Tibetan}>.
167 The official Unicode script and block names have spaces and dashes as
168 separators, but for convenience you can use dashes, spaces, or
169 underbars, and case is unimportant. It is recommended, however, that
170 for consistency you use the following naming: the official Unicode
171 script, property, or block name (see below for the additional rules
172 that apply to block names) with whitespace and dashes removed, and the
173 words "uppercase-first-lowercase-rest". C<Latin-1 Supplement> thus
174 becomes C<Latin1Supplement>.
176 You can also use negation in both C<\p{}> and C<\P{}> by introducing a caret
177 (^) between the first brace and the property name: C<\p{^Tamil}> is
178 equal to C<\P{Tamil}>.
180 B<NOTE: the properties, scripts, and blocks listed here are as of
181 Unicode 3.2.0, March 2002, or Perl 5.8.0, July 2002. Unicode 4.0.0
182 came out in April 2003, and Perl 5.8.1 in September 2003.>
184 Here are the basic Unicode General Category properties, followed by their
185 long form. You can use either; C<\p{Lu}> and C<\p{UppercaseLetter}>,
186 for instance, are identical.
208 Pc ConnectorPunctuation
212 Pi InitialPunctuation
213 (may behave like Ps or Pe depending on usage)
215 (may behave like Ps or Pe depending on usage)
227 Zp ParagraphSeparator
232 Cs Surrogate (not usable)
236 Single-letter properties match all characters in any of the
237 two-letter sub-properties starting with the same letter.
238 C<L&> is a special case, which is an alias for C<Ll>, C<Lu>, and C<Lt>.
240 Because Perl hides the need for the user to understand the internal
241 representation of Unicode characters, there is no need to implement
242 the somewhat messy concept of surrogates. C<Cs> is therefore not
245 Because scripts differ in their directionality--Hebrew is
246 written right to left, for example--Unicode supplies these properties:
251 BidiLRE Left-to-Right Embedding
252 BidiLRO Left-to-Right Override
254 BidiAL Right-to-Left Arabic
255 BidiRLE Right-to-Left Embedding
256 BidiRLO Right-to-Left Override
257 BidiPDF Pop Directional Format
258 BidiEN European Number
259 BidiES European Number Separator
260 BidiET European Number Terminator
262 BidiCS Common Number Separator
263 BidiNSM Non-Spacing Mark
264 BidiBN Boundary Neutral
265 BidiB Paragraph Separator
266 BidiS Segment Separator
268 BidiON Other Neutrals
270 For example, C<\p{BidiR}> matches characters that are normally
271 written right to left.
277 The script names which can be used by C<\p{...}> and C<\P{...}>,
278 such as in C<\p{Latin}> or C<\p{Cyrillic}>, are as follows:
325 Extended property classes can supplement the basic
326 properties, defined by the F<PropList> Unicode database:
341 LogicalOrderException
342 NoncharacterCodePoint
344 OtherDefaultIgnorableCodePoint
356 and there are further derived properties:
358 Alphabetic Lu + Ll + Lt + Lm + Lo + OtherAlphabetic
359 Lowercase Ll + OtherLowercase
360 Uppercase Lu + OtherUppercase
363 ID_Start Lu + Ll + Lt + Lm + Lo + Nl
364 ID_Continue ID_Start + Mn + Mc + Nd + Pc
367 Assigned Any non-Cn character (i.e. synonym for \P{Cn})
368 Unassigned Synonym for \p{Cn}
369 Common Any character (or unassigned code point)
370 not explicitly assigned to a script
372 For backward compatibility (with Perl 5.6), all properties mentioned
373 so far may have C<Is> prepended to their name, so C<\P{IsLu}>, for
374 example, is equal to C<\P{Lu}>.
378 In addition to B<scripts>, Unicode also defines B<blocks> of
379 characters. The difference between scripts and blocks is that the
380 concept of scripts is closer to natural languages, while the concept
381 of blocks is more of an artificial grouping based on groups of 256
382 Unicode characters. For example, the C<Latin> script contains letters
383 from many blocks but does not contain all the characters from those
384 blocks. It does not, for example, contain digits, because digits are
385 shared across many scripts. Digits and similar groups, like
386 punctuation, are in a category called C<Common>.
388 For more about scripts, see the UTR #24:
390 http://www.unicode.org/unicode/reports/tr24/
392 For more about blocks, see:
394 http://www.unicode.org/Public/UNIDATA/Blocks.txt
396 Block names are given with the C<In> prefix. For example, the
397 Katakana block is referenced via C<\p{InKatakana}>. The C<In>
398 prefix may be omitted if there is no naming conflict with a script
399 or any other property, but it is recommended that C<In> always be used
400 for block tests to avoid confusion.
402 These block names are supported:
404 InAlphabeticPresentationForms
406 InArabicPresentationFormsA
407 InArabicPresentationFormsB
418 InByzantineMusicalSymbols
420 InCJKCompatibilityForms
421 InCJKCompatibilityIdeographs
422 InCJKCompatibilityIdeographsSupplement
423 InCJKRadicalsSupplement
424 InCJKSymbolsAndPunctuation
425 InCJKUnifiedIdeographs
426 InCJKUnifiedIdeographsExtensionA
427 InCJKUnifiedIdeographsExtensionB
429 InCombiningDiacriticalMarks
430 InCombiningDiacriticalMarksforSymbols
435 InCyrillicSupplementary
439 InEnclosedAlphanumerics
440 InEnclosedCJKLettersAndMonths
450 InHalfwidthAndFullwidthForms
451 InHangulCompatibilityJamo
456 InHighPrivateUseSurrogates
460 InIdeographicDescriptionCharacters
465 InKatakanaPhoneticExtensions
470 InLatinExtendedAdditional
475 InMathematicalAlphanumericSymbols
476 InMathematicalOperators
477 InMiscellaneousMathematicalSymbolsA
478 InMiscellaneousMathematicalSymbolsB
479 InMiscellaneousSymbols
480 InMiscellaneousTechnical
487 InOpticalCharacterRecognition
493 InSpacingModifierLetters
495 InSuperscriptsAndSubscripts
496 InSupplementalArrowsA
497 InSupplementalArrowsB
498 InSupplementalMathematicalOperators
499 InSupplementaryPrivateUseAreaA
500 InSupplementaryPrivateUseAreaB
510 InUnifiedCanadianAboriginalSyllabics
519 The special pattern C<\X> matches any extended Unicode
520 sequence--"a combining character sequence" in Standardese--where the
521 first character is a base character and subsequent characters are mark
522 characters that apply to the base character. C<\X> is equivalent to
527 The C<tr///> operator translates characters instead of bytes. Note
528 that the C<tr///CU> functionality has been removed. For similar
529 functionality see pack('U0', ...) and pack('C0', ...).
533 Case translation operators use the Unicode case translation tables
534 when character input is provided. Note that C<uc()>, or C<\U> in
535 interpolated strings, translates to uppercase, while C<ucfirst>,
536 or C<\u> in interpolated strings, translates to titlecase in languages
537 that make the distinction.
541 Most operators that deal with positions or lengths in a string will
542 automatically switch to using character positions, including
543 C<chop()>, C<substr()>, C<pos()>, C<index()>, C<rindex()>,
544 C<sprintf()>, C<write()>, and C<length()>. Operators that
545 specifically do not switch include C<vec()>, C<pack()>, and
546 C<unpack()>. Operators that really don't care include C<chomp()>,
547 operators that treats strings as a bucket of bits such as C<sort()>,
548 and operators dealing with filenames.
552 The C<pack()>/C<unpack()> letters C<c> and C<C> do I<not> change,
553 since they are often used for byte-oriented formats. Again, think
554 C<char> in the C language.
556 There is a new C<U> specifier that converts between Unicode characters
561 The C<chr()> and C<ord()> functions work on characters, similar to
562 C<pack("U")> and C<unpack("U")>, I<not> C<pack("C")> and
563 C<unpack("C")>. C<pack("C")> and C<unpack("C")> are methods for
564 emulating byte-oriented C<chr()> and C<ord()> on Unicode strings.
565 While these methods reveal the internal encoding of Unicode strings,
566 that is not something one normally needs to care about at all.
570 The bit string operators, C<& | ^ ~>, can operate on character data.
571 However, for backward compatibility, such as when using bit string
572 operations when characters are all less than 256 in ordinal value, one
573 should not use C<~> (the bit complement) with characters of both
574 values less than 256 and values greater than 256. Most importantly,
575 DeMorgan's laws (C<~($x|$y) eq ~$x&~$y> and C<~($x&$y) eq ~$x|~$y>)
576 will not hold. The reason for this mathematical I<faux pas> is that
577 the complement cannot return B<both> the 8-bit (byte-wide) bit
578 complement B<and> the full character-wide bit complement.
582 lc(), uc(), lcfirst(), and ucfirst() work for the following cases:
588 the case mapping is from a single Unicode character to another
589 single Unicode character, or
593 the case mapping is from a single Unicode character to more
594 than one Unicode character.
598 Things to do with locales (Lithuanian, Turkish, Azeri) do B<not> work
599 since Perl does not understand the concept of Unicode locales.
601 See the Unicode Technical Report #21, Case Mappings, for more details.
609 And finally, C<scalar reverse()> reverses by character rather than by byte.
613 =head2 User-Defined Character Properties
615 You can define your own character properties by defining subroutines
616 whose names begin with "In" or "Is". The subroutines must be defined
617 in the C<main> package. The user-defined properties can be used in the
618 regular expression C<\p> and C<\P> constructs. Note that the effect
619 is compile-time and immutable once defined.
621 The subroutines must return a specially-formatted string, with one
622 or more newline-separated lines. Each line must be one of the following:
628 Two hexadecimal numbers separated by horizontal whitespace (space or
629 tabular characters) denoting a range of Unicode code points to include.
633 Something to include, prefixed by "+": a built-in character
634 property (prefixed by "utf8::"), to represent all the characters in that
635 property; two hexadecimal code points for a range; or a single
636 hexadecimal code point.
640 Something to exclude, prefixed by "-": an existing character
641 property (prefixed by "utf8::"), for all the characters in that
642 property; two hexadecimal code points for a range; or a single
643 hexadecimal code point.
647 Something to negate, prefixed "!": an existing character
648 property (prefixed by "utf8::") for all the characters except the
649 characters in the property; two hexadecimal code points for a range;
650 or a single hexadecimal code point.
654 For example, to define a property that covers both the Japanese
655 syllabaries (hiragana and katakana), you can define
664 Imagine that the here-doc end marker is at the beginning of the line.
665 Now you can use C<\p{InKana}> and C<\P{InKana}>.
667 You could also have used the existing block property names:
676 Suppose you wanted to match only the allocated characters,
677 not the raw block ranges: in other words, you want to remove
688 The negation is useful for defining (surprise!) negated classes.
698 You can also define your own mappings to be used in the lc(),
699 lcfirst(), uc(), and ucfirst() (or their string-inlined versions).
700 The principle is the same: define subroutines in the C<main> package
701 with names like C<ToLower> (for lc() and lcfirst()), C<ToTitle> (for
702 the first character in ucfirst()), and C<ToUpper> (for uc(), and the
703 rest of the characters in ucfirst()).
705 The string returned by the subroutines needs now to be three
706 hexadecimal numbers separated by tabulators: start of the source
707 range, end of the source range, and start of the destination range.
716 defines an uc() mapping that causes only the characters "a", "b", and
717 "c" to be mapped to "A", "B", "C", all other characters will remain
720 If there is no source range to speak of, that is, the mapping is from
721 a single character to another single character, leave the end of the
722 source range empty, but the two tabulator characters are still needed.
731 defines a lc() mapping that causes only "A" to be mapped to "a", all
732 other characters will remain unchanged.
734 (For serious hackers only) If you want to introspect the default
735 mappings, you can find the data in the directory
736 C<$Config{privlib}>/F<unicore/To/>. The mapping data is returned as
737 the here-document, and the C<utf8::ToSpecFoo> are special exception
738 mappings derived from <$Config{privlib}>/F<unicore/SpecialCasing.txt>.
739 The C<Digit> and C<Fold> mappings that one can see in the directory
740 are not directly user-accessible, one can use either the
741 C<Unicode::UCD> module, or just match case-insensitively (that's when
742 the C<Fold> mapping is used).
744 A final note on the user-defined property tests and mappings: they
745 will be used only if the scalar has been marked as having Unicode
746 characters. Old byte-style strings will not be affected.
748 =head2 Character Encodings for Input and Output
752 =head2 Unicode Regular Expression Support Level
754 The following list of Unicode support for regular expressions describes
755 all the features currently supported. The references to "Level N"
756 and the section numbers refer to the Unicode Technical Report 18,
757 "Unicode Regular Expression Guidelines", version 6 (Unicode 3.2.0,
764 Level 1 - Basic Unicode Support
766 2.1 Hex Notation - done [1]
767 Named Notation - done [2]
768 2.2 Categories - done [3][4]
769 2.3 Subtraction - MISSING [5][6]
770 2.4 Simple Word Boundaries - done [7]
771 2.5 Simple Loose Matches - done [8]
772 2.6 End of Line - MISSING [9][10]
776 [ 3] . \p{...} \P{...}
777 [ 4] now scripts (see UTR#24 Script Names) in addition to blocks
779 [ 6] can use regular expression look-ahead [a]
780 or user-defined character properties [b] to emulate subtraction
781 [ 7] include Letters in word characters
782 [ 8] note that Perl does Full case-folding in matching, not Simple:
783 for example U+1F88 is equivalent with U+1F00 U+03B9,
784 not with 1F80. This difference matters for certain Greek
785 capital letters with certain modifiers: the Full case-folding
786 decomposes the letter, while the Simple case-folding would map
787 it to a single character.
788 [ 9] see UTR #13 Unicode Newline Guidelines
789 [10] should do ^ and $ also on \x{85}, \x{2028} and \x{2029}
790 (should also affect <>, $., and script line numbers)
791 (the \x{85}, \x{2028} and \x{2029} do match \s)
793 [a] You can mimic class subtraction using lookahead.
794 For example, what UTR #18 might write as
796 [{Greek}-[{UNASSIGNED}]]
798 in Perl can be written as:
800 (?!\p{Unassigned})\p{InGreekAndCoptic}
801 (?=\p{Assigned})\p{InGreekAndCoptic}
803 But in this particular example, you probably really want
807 which will match assigned characters known to be part of the Greek script.
809 Also see the Unicode::Regex::Set module, it does implement the full
810 UTR #18 grouping, intersection, union, and removal (subtraction) syntax.
812 [b] See L</"User-Defined Character Properties">.
816 Level 2 - Extended Unicode Support
818 3.1 Surrogates - MISSING [11]
819 3.2 Canonical Equivalents - MISSING [12][13]
820 3.3 Locale-Independent Graphemes - MISSING [14]
821 3.4 Locale-Independent Words - MISSING [15]
822 3.5 Locale-Independent Loose Matches - MISSING [16]
824 [11] Surrogates are solely a UTF-16 concept and Perl's internal
825 representation is UTF-8. The Encode module does UTF-16, though.
826 [12] see UTR#15 Unicode Normalization
827 [13] have Unicode::Normalize but not integrated to regexes
828 [14] have \X but at this level . should equal that
829 [15] need three classes, not just \w and \W
830 [16] see UTR#21 Case Mappings
834 Level 3 - Locale-Sensitive Support
836 4.1 Locale-Dependent Categories - MISSING
837 4.2 Locale-Dependent Graphemes - MISSING [16][17]
838 4.3 Locale-Dependent Words - MISSING
839 4.4 Locale-Dependent Loose Matches - MISSING
840 4.5 Locale-Dependent Ranges - MISSING
842 [16] see UTR#10 Unicode Collation Algorithms
843 [17] have Unicode::Collate but not integrated to regexes
847 =head2 Unicode Encodings
849 Unicode characters are assigned to I<code points>, which are abstract
850 numbers. To use these numbers, various encodings are needed.
858 UTF-8 is a variable-length (1 to 6 bytes, current character allocations
859 require 4 bytes), byte-order independent encoding. For ASCII (and we
860 really do mean 7-bit ASCII, not another 8-bit encoding), UTF-8 is
863 The following table is from Unicode 3.2.
865 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
867 U+0000..U+007F 00..7F
868 U+0080..U+07FF C2..DF 80..BF
869 U+0800..U+0FFF E0 A0..BF 80..BF
870 U+1000..U+CFFF E1..EC 80..BF 80..BF
871 U+D000..U+D7FF ED 80..9F 80..BF
872 U+D800..U+DFFF ******* ill-formed *******
873 U+E000..U+FFFF EE..EF 80..BF 80..BF
874 U+10000..U+3FFFF F0 90..BF 80..BF 80..BF
875 U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
876 U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
878 Note the C<A0..BF> in C<U+0800..U+0FFF>, the C<80..9F> in
879 C<U+D000...U+D7FF>, the C<90..B>F in C<U+10000..U+3FFFF>, and the
880 C<80...8F> in C<U+100000..U+10FFFF>. The "gaps" are caused by legal
881 UTF-8 avoiding non-shortest encodings: it is technically possible to
882 UTF-8-encode a single code point in different ways, but that is
883 explicitly forbidden, and the shortest possible encoding should always
884 be used. So that's what Perl does.
886 Another way to look at it is via bits:
888 Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte
891 00000bbbbbaaaaaa 110bbbbb 10aaaaaa
892 ccccbbbbbbaaaaaa 1110cccc 10bbbbbb 10aaaaaa
893 00000dddccccccbbbbbbaaaaaa 11110ddd 10cccccc 10bbbbbb 10aaaaaa
895 As you can see, the continuation bytes all begin with C<10>, and the
896 leading bits of the start byte tell how many bytes the are in the
903 Like UTF-8 but EBCDIC-safe, in the way that UTF-8 is ASCII-safe.
907 UTF-16, UTF-16BE, UTF16-LE, Surrogates, and BOMs (Byte Order Marks)
909 The followings items are mostly for reference and general Unicode
910 knowledge, Perl doesn't use these constructs internally.
912 UTF-16 is a 2 or 4 byte encoding. The Unicode code points
913 C<U+0000..U+FFFF> are stored in a single 16-bit unit, and the code
914 points C<U+10000..U+10FFFF> in two 16-bit units. The latter case is
915 using I<surrogates>, the first 16-bit unit being the I<high
916 surrogate>, and the second being the I<low surrogate>.
918 Surrogates are code points set aside to encode the C<U+10000..U+10FFFF>
919 range of Unicode code points in pairs of 16-bit units. The I<high
920 surrogates> are the range C<U+D800..U+DBFF>, and the I<low surrogates>
921 are the range C<U+DC00..U+DFFF>. The surrogate encoding is
923 $hi = ($uni - 0x10000) / 0x400 + 0xD800;
924 $lo = ($uni - 0x10000) % 0x400 + 0xDC00;
928 $uni = 0x10000 + ($hi - 0xD800) * 0x400 + ($lo - 0xDC00);
930 If you try to generate surrogates (for example by using chr()), you
931 will get a warning if warnings are turned on, because those code
932 points are not valid for a Unicode character.
934 Because of the 16-bitness, UTF-16 is byte-order dependent. UTF-16
935 itself can be used for in-memory computations, but if storage or
936 transfer is required either UTF-16BE (big-endian) or UTF-16LE
937 (little-endian) encodings must be chosen.
939 This introduces another problem: what if you just know that your data
940 is UTF-16, but you don't know which endianness? Byte Order Marks, or
941 BOMs, are a solution to this. A special character has been reserved
942 in Unicode to function as a byte order marker: the character with the
943 code point C<U+FEFF> is the BOM.
945 The trick is that if you read a BOM, you will know the byte order,
946 since if it was written on a big-endian platform, you will read the
947 bytes C<0xFE 0xFF>, but if it was written on a little-endian platform,
948 you will read the bytes C<0xFF 0xFE>. (And if the originating platform
949 was writing in UTF-8, you will read the bytes C<0xEF 0xBB 0xBF>.)
951 The way this trick works is that the character with the code point
952 C<U+FFFE> is guaranteed not to be a valid Unicode character, so the
953 sequence of bytes C<0xFF 0xFE> is unambiguously "BOM, represented in
954 little-endian format" and cannot be C<U+FFFE>, represented in big-endian
959 UTF-32, UTF-32BE, UTF32-LE
961 The UTF-32 family is pretty much like the UTF-16 family, expect that
962 the units are 32-bit, and therefore the surrogate scheme is not
963 needed. The BOM signatures will be C<0x00 0x00 0xFE 0xFF> for BE and
964 C<0xFF 0xFE 0x00 0x00> for LE.
970 Encodings defined by the ISO 10646 standard. UCS-2 is a 16-bit
971 encoding. Unlike UTF-16, UCS-2 is not extensible beyond C<U+FFFF>,
972 because it does not use surrogates. UCS-4 is a 32-bit encoding,
973 functionally identical to UTF-32.
979 A seven-bit safe (non-eight-bit) encoding, which is useful if the
980 transport or storage is not eight-bit safe. Defined by RFC 2152.
984 =head2 Security Implications of Unicode
992 Unfortunately, the specification of UTF-8 leaves some room for
993 interpretation of how many bytes of encoded output one should generate
994 from one input Unicode character. Strictly speaking, the shortest
995 possible sequence of UTF-8 bytes should be generated,
996 because otherwise there is potential for an input buffer overflow at
997 the receiving end of a UTF-8 connection. Perl always generates the
998 shortest length UTF-8, and with warnings on Perl will warn about
999 non-shortest length UTF-8 along with other malformations, such as the
1000 surrogates, which are not real Unicode code points.
1004 Regular expressions behave slightly differently between byte data and
1005 character (Unicode) data. For example, the "word character" character
1006 class C<\w> will work differently depending on if data is eight-bit bytes
1009 In the first case, the set of C<\w> characters is either small--the
1010 default set of alphabetic characters, digits, and the "_"--or, if you
1011 are using a locale (see L<perllocale>), the C<\w> might contain a few
1012 more letters according to your language and country.
1014 In the second case, the C<\w> set of characters is much, much larger.
1015 Most importantly, even in the set of the first 256 characters, it will
1016 probably match different characters: unlike most locales, which are
1017 specific to a language and country pair, Unicode classifies all the
1018 characters that are letters I<somewhere> as C<\w>. For example, your
1019 locale might not think that LATIN SMALL LETTER ETH is a letter (unless
1020 you happen to speak Icelandic), but Unicode does.
1022 As discussed elsewhere, Perl has one foot (two hooves?) planted in
1023 each of two worlds: the old world of bytes and the new world of
1024 characters, upgrading from bytes to characters when necessary.
1025 If your legacy code does not explicitly use Unicode, no automatic
1026 switch-over to characters should happen. Characters shouldn't get
1027 downgraded to bytes, either. It is possible to accidentally mix bytes
1028 and characters, however (see L<perluniintro>), in which case C<\w> in
1029 regular expressions might start behaving differently. Review your
1030 code. Use warnings and the C<strict> pragma.
1034 =head2 Unicode in Perl on EBCDIC
1036 The way Unicode is handled on EBCDIC platforms is still
1037 experimental. On such platforms, references to UTF-8 encoding in this
1038 document and elsewhere should be read as meaning the UTF-EBCDIC
1039 specified in Unicode Technical Report 16, unless ASCII vs. EBCDIC issues
1040 are specifically discussed. There is no C<utfebcdic> pragma or
1041 ":utfebcdic" layer; rather, "utf8" and ":utf8" are reused to mean
1042 the platform's "natural" 8-bit encoding of Unicode. See L<perlebcdic>
1043 for more discussion of the issues.
1047 Usually locale settings and Unicode do not affect each other, but
1048 there are a couple of exceptions:
1054 You can enable automatic UTF-8-ification of your standard file
1055 handles, default C<open()> layer, and C<@ARGV> by using either
1056 the C<-C> command line switch or the C<PERL_UNICODE> environment
1057 variable, see L<perlrun> for the documentation of the C<-C> switch.
1061 Perl tries really hard to work both with Unicode and the old
1062 byte-oriented world. Most often this is nice, but sometimes Perl's
1063 straddling of the proverbial fence causes problems.
1067 =head2 When Unicode Does Not Happen
1069 While Perl does have extensive ways to input and output in Unicode,
1070 and few other 'entry points' like the @ARGV which can be interpreted
1071 as Unicode (UTF-8), there still are many places where Unicode (in some
1072 encoding or another) could be given as arguments or received as
1073 results, or both, but it is not.
1075 The following are such interfaces. For all of these Perl currently
1076 (as of 5.8.1) simply assumes byte strings both as arguments and results.
1078 One reason why Perl does not attempt to resolve the role of Unicode in
1079 this cases is that the answers are highly dependent on the operating
1080 system and the file system(s). For example, whether filenames can be
1081 in Unicode, and in exactly what kind of encoding, is not exactly a
1082 portable concept. Similarly for the qx and system: how well will the
1083 'command line interface' (and which of them?) handle Unicode?
1089 chmod, chmod, chown, chroot, exec, link, mkdir
1090 rename, rmdir stat, symlink, truncate, unlink, utime
1102 open, opendir, sysopen
1106 qx (aka the backtick operator), system
1114 =head2 Forcing Unicode in Perl (Or Unforcing Unicode in Perl)
1116 Sometimes (see L</"When Unicode Does Not Happen">) there are
1117 situations where you simply need to force Perl to believe that a byte
1118 string is UTF-8, or vice versa. The low-level calls
1119 utf8::upgrade($bytestring) and utf8::downgrade($utf8string) are
1122 Do not use them without careful thought, though: Perl may easily get
1123 very confused, angry, or even crash, if you suddenly change the 'nature'
1124 of scalar like that. Especially careful you have to be if you use the
1125 utf8::upgrade(): any random byte string is not valid UTF-8.
1127 =head2 Using Unicode in XS
1129 If you want to handle Perl Unicode in XS extensions, you may find the
1130 following C APIs useful. See also L<perlguts/"Unicode Support"> for an
1131 explanation about Unicode at the XS level, and L<perlapi> for the API
1138 C<DO_UTF8(sv)> returns true if the C<UTF8> flag is on and the bytes
1139 pragma is not in effect. C<SvUTF8(sv)> returns true is the C<UTF8>
1140 flag is on; the bytes pragma is ignored. The C<UTF8> flag being on
1141 does B<not> mean that there are any characters of code points greater
1142 than 255 (or 127) in the scalar or that there are even any characters
1143 in the scalar. What the C<UTF8> flag means is that the sequence of
1144 octets in the representation of the scalar is the sequence of UTF-8
1145 encoded code points of the characters of a string. The C<UTF8> flag
1146 being off means that each octet in this representation encodes a
1147 single character with code point 0..255 within the string. Perl's
1148 Unicode model is not to use UTF-8 until it is absolutely necessary.
1152 C<uvuni_to_utf8(buf, chr>) writes a Unicode character code point into
1153 a buffer encoding the code point as UTF-8, and returns a pointer
1154 pointing after the UTF-8 bytes.
1158 C<utf8_to_uvuni(buf, lenp)> reads UTF-8 encoded bytes from a buffer and
1159 returns the Unicode character code point and, optionally, the length of
1160 the UTF-8 byte sequence.
1164 C<utf8_length(start, end)> returns the length of the UTF-8 encoded buffer
1165 in characters. C<sv_len_utf8(sv)> returns the length of the UTF-8 encoded
1170 C<sv_utf8_upgrade(sv)> converts the string of the scalar to its UTF-8
1171 encoded form. C<sv_utf8_downgrade(sv)> does the opposite, if
1172 possible. C<sv_utf8_encode(sv)> is like sv_utf8_upgrade except that
1173 it does not set the C<UTF8> flag. C<sv_utf8_decode()> does the
1174 opposite of C<sv_utf8_encode()>. Note that none of these are to be
1175 used as general-purpose encoding or decoding interfaces: C<use Encode>
1176 for that. C<sv_utf8_upgrade()> is affected by the encoding pragma
1177 but C<sv_utf8_downgrade()> is not (since the encoding pragma is
1178 designed to be a one-way street).
1182 C<is_utf8_char(s)> returns true if the pointer points to a valid UTF-8
1187 C<is_utf8_string(buf, len)> returns true if C<len> bytes of the buffer
1192 C<UTF8SKIP(buf)> will return the number of bytes in the UTF-8 encoded
1193 character in the buffer. C<UNISKIP(chr)> will return the number of bytes
1194 required to UTF-8-encode the Unicode character code point. C<UTF8SKIP()>
1195 is useful for example for iterating over the characters of a UTF-8
1196 encoded buffer; C<UNISKIP()> is useful, for example, in computing
1197 the size required for a UTF-8 encoded buffer.
1201 C<utf8_distance(a, b)> will tell the distance in characters between the
1202 two pointers pointing to the same UTF-8 encoded buffer.
1206 C<utf8_hop(s, off)> will return a pointer to an UTF-8 encoded buffer
1207 that is C<off> (positive or negative) Unicode characters displaced
1208 from the UTF-8 buffer C<s>. Be careful not to overstep the buffer:
1209 C<utf8_hop()> will merrily run off the end or the beginning of the
1210 buffer if told to do so.
1214 C<pv_uni_display(dsv, spv, len, pvlim, flags)> and
1215 C<sv_uni_display(dsv, ssv, pvlim, flags)> are useful for debugging the
1216 output of Unicode strings and scalars. By default they are useful
1217 only for debugging--they display B<all> characters as hexadecimal code
1218 points--but with the flags C<UNI_DISPLAY_ISPRINT>,
1219 C<UNI_DISPLAY_BACKSLASH>, and C<UNI_DISPLAY_QQ> you can make the
1220 output more readable.
1224 C<ibcmp_utf8(s1, pe1, u1, l1, u1, s2, pe2, l2, u2)> can be used to
1225 compare two strings case-insensitively in Unicode. For case-sensitive
1226 comparisons you can just use C<memEQ()> and C<memNE()> as usual.
1230 For more information, see L<perlapi>, and F<utf8.c> and F<utf8.h>
1231 in the Perl source code distribution.
1235 =head2 Interaction with Locales
1237 Use of locales with Unicode data may lead to odd results. Currently,
1238 Perl attempts to attach 8-bit locale info to characters in the range
1239 0..255, but this technique is demonstrably incorrect for locales that
1240 use characters above that range when mapped into Unicode. Perl's
1241 Unicode support will also tend to run slower. Use of locales with
1242 Unicode is discouraged.
1244 =head2 Interaction with Extensions
1246 When Perl exchanges data with an extension, the extension should be
1247 able to understand the UTF-8 flag and act accordingly. If the
1248 extension doesn't know about the flag, it's likely that the extension
1249 will return incorrectly-flagged data.
1251 So if you're working with Unicode data, consult the documentation of
1252 every module you're using if there are any issues with Unicode data
1253 exchange. If the documentation does not talk about Unicode at all,
1254 suspect the worst and probably look at the source to learn how the
1255 module is implemented. Modules written completely in Perl shouldn't
1256 cause problems. Modules that directly or indirectly access code written
1257 in other programming languages are at risk.
1259 For affected functions, the simple strategy to avoid data corruption is
1260 to always make the encoding of the exchanged data explicit. Choose an
1261 encoding that you know the extension can handle. Convert arguments passed
1262 to the extensions to that encoding and convert results back from that
1263 encoding. Write wrapper functions that do the conversions for you, so
1264 you can later change the functions when the extension catches up.
1266 To provide an example, let's say the popular Foo::Bar::escape_html
1267 function doesn't deal with Unicode data yet. The wrapper function
1268 would convert the argument to raw UTF-8 and convert the result back to
1269 Perl's internal representation like so:
1271 sub my_escape_html ($) {
1273 return unless defined $what;
1274 Encode::decode_utf8(Foo::Bar::escape_html(Encode::encode_utf8($what)));
1277 Sometimes, when the extension does not convert data but just stores
1278 and retrieves them, you will be in a position to use the otherwise
1279 dangerous Encode::_utf8_on() function. Let's say the popular
1280 C<Foo::Bar> extension, written in C, provides a C<param> method that
1281 lets you store and retrieve data according to these prototypes:
1283 $self->param($name, $value); # set a scalar
1284 $value = $self->param($name); # retrieve a scalar
1286 If it does not yet provide support for any encoding, one could write a
1287 derived class with such a C<param> method:
1290 my($self,$name,$value) = @_;
1291 utf8::upgrade($name); # make sure it is UTF-8 encoded
1293 utf8::upgrade($value); # make sure it is UTF-8 encoded
1294 return $self->SUPER::param($name,$value);
1296 my $ret = $self->SUPER::param($name);
1297 Encode::_utf8_on($ret); # we know, it is UTF-8 encoded
1302 Some extensions provide filters on data entry/exit points, such as
1303 DB_File::filter_store_key and family. Look out for such filters in
1304 the documentation of your extensions, they can make the transition to
1305 Unicode data much easier.
1309 Some functions are slower when working on UTF-8 encoded strings than
1310 on byte encoded strings. All functions that need to hop over
1311 characters such as length(), substr() or index(), or matching regular
1312 expressions can work B<much> faster when the underlying data are
1315 In Perl 5.8.0 the slowness was often quite spectacular; in Perl 5.8.1
1316 a caching scheme was introduced which will hopefully make the slowness
1317 somewhat less spectacular. Operations with UTF-8 encoded strings are
1318 still slower, though.
1320 =head2 Porting code from perl-5.6.X
1322 Perl 5.8 has a different Unicode model from 5.6. In 5.6 the programmer
1323 was required to use the C<utf8> pragma to declare that a given scope
1324 expected to deal with Unicode data and had to make sure that only
1325 Unicode data were reaching that scope. If you have code that is
1326 working with 5.6, you will need some of the following adjustments to
1327 your code. The examples are written such that the code will continue
1328 to work under 5.6, so you should be safe to try them out.
1334 A filehandle that should read or write UTF-8
1337 binmode $fh, ":utf8";
1342 A scalar that is going to be passed to some extension
1344 Be it Compress::Zlib, Apache::Request or any extension that has no
1345 mention of Unicode in the manpage, you need to make sure that the
1346 UTF-8 flag is stripped off. Note that at the time of this writing
1347 (October 2002) the mentioned modules are not UTF-8-aware. Please
1348 check the documentation to verify if this is still true.
1352 $val = Encode::encode_utf8($val); # make octets
1357 A scalar we got back from an extension
1359 If you believe the scalar comes back as UTF-8, you will most likely
1360 want the UTF-8 flag restored:
1364 $val = Encode::decode_utf8($val);
1369 Same thing, if you are really sure it is UTF-8
1373 Encode::_utf8_on($val);
1378 A wrapper for fetchrow_array and fetchrow_hashref
1380 When the database contains only UTF-8, a wrapper function or method is
1381 a convenient way to replace all your fetchrow_array and
1382 fetchrow_hashref calls. A wrapper function will also make it easier to
1383 adapt to future enhancements in your database driver. Note that at the
1384 time of this writing (October 2002), the DBI has no standardized way
1385 to deal with UTF-8 data. Please check the documentation to verify if
1389 my($self, $sth, $what) = @_; # $what is one of fetchrow_{array,hashref}
1395 my @arr = $sth->$what;
1397 defined && /[^\000-\177]/ && Encode::_utf8_on($_);
1401 my $ret = $sth->$what;
1403 for my $k (keys %$ret) {
1404 defined && /[^\000-\177]/ && Encode::_utf8_on($_) for $ret->{$k};
1408 defined && /[^\000-\177]/ && Encode::_utf8_on($_) for $ret;
1418 A large scalar that you know can only contain ASCII
1420 Scalars that contain only ASCII and are marked as UTF-8 are sometimes
1421 a drag to your program. If you recognize such a situation, just remove
1424 utf8::downgrade($val) if $] > 5.007;
1430 L<perluniintro>, L<encoding>, L<Encode>, L<open>, L<utf8>, L<bytes>,
1431 L<perlretut>, L<perlvar/"${^UNICODE}">