Commit | Line | Data |
393fec97 |
1 | =head1 NAME |
2 | |
3 | perlunicode - Unicode support in Perl |
4 | |
5 | =head1 DESCRIPTION |
6 | |
0a1f2d14 |
7 | =head2 Important Caveats |
21bad921 |
8 | |
c349b1b9 |
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. |
21bad921 |
12 | |
13a2d996 |
13 | =over 4 |
21bad921 |
14 | |
15 | =item Input and Output Disciplines |
16 | |
75daf61c |
17 | A filehandle can be marked as containing perl's internal Unicode |
18 | encoding (UTF-8 or UTF-EBCDIC) by opening it with the ":utf8" layer. |
0a1f2d14 |
19 | Other encodings can be converted to perl's encoding on input, or from |
c349b1b9 |
20 | perl's encoding on output by use of the ":encoding(...)" layer. |
21 | See L<open>. |
22 | |
d1be9408 |
23 | To mark the Perl source itself as being in a particular encoding, |
c349b1b9 |
24 | see L<encoding>. |
21bad921 |
25 | |
26 | =item Regular Expressions |
27 | |
c349b1b9 |
28 | The regular expression compiler produces polymorphic opcodes. That is, |
29 | the pattern adapts to the data and automatically switch to the Unicode |
30 | character scheme when presented with Unicode data, or a traditional |
31 | byte scheme when presented with byte data. |
21bad921 |
32 | |
ad0029c4 |
33 | =item C<use utf8> still needed to enable UTF-8/UTF-EBCDIC in scripts |
21bad921 |
34 | |
75daf61c |
35 | The C<utf8> pragma implements the tables used for Unicode support. |
c349b1b9 |
36 | However, these tables are automatically loaded on demand, so the |
37 | C<utf8> pragma should not normally be used. |
21bad921 |
38 | |
c349b1b9 |
39 | As a compatibility measure, this pragma must be explicitly used to |
40 | enable recognition of UTF-8 in the Perl scripts themselves on ASCII |
41 | based machines or recognize UTF-EBCDIC on EBCDIC based machines. |
42 | B<NOTE: this should be the only place where an explicit C<use utf8> |
43 | is needed>. |
21bad921 |
44 | |
1768d7eb |
45 | You can also use the C<encoding> pragma to change the default encoding |
6ec9efec |
46 | of the data in your script; see L<encoding>. |
1768d7eb |
47 | |
21bad921 |
48 | =back |
49 | |
50 | =head2 Byte and Character semantics |
393fec97 |
51 | |
52 | Beginning with version 5.6, Perl uses logically wide characters to |
53 | represent strings internally. This internal representation of strings |
b3419ed8 |
54 | uses either the UTF-8 or the UTF-EBCDIC encoding. |
393fec97 |
55 | |
75daf61c |
56 | In future, Perl-level operations can be expected to work with |
57 | characters rather than bytes, in general. |
393fec97 |
58 | |
75daf61c |
59 | However, as strictly an interim compatibility measure, Perl aims to |
60 | provide a safe migration path from byte semantics to character |
61 | semantics for programs. For operations where Perl can unambiguously |
62 | decide that the input data is characters, Perl now switches to |
63 | character semantics. For operations where this determination cannot |
64 | be made without additional information from the user, Perl decides in |
65 | favor of compatibility, and chooses to use byte semantics. |
8cbd9a7a |
66 | |
67 | This behavior preserves compatibility with earlier versions of Perl, |
68 | which allowed byte semantics in Perl operations, but only as long as |
69 | none of the program's inputs are marked as being as source of Unicode |
70 | character data. Such data may come from filehandles, from calls to |
71 | external programs, from information provided by the system (such as %ENV), |
21bad921 |
72 | or from literals and constants in the source text. |
8cbd9a7a |
73 | |
c349b1b9 |
74 | On Windows platforms, if the C<-C> command line switch is used, (or the |
75daf61c |
75 | ${^WIDE_SYSTEM_CALLS} global flag is set to C<1>), all system calls |
76 | will use the corresponding wide character APIs. Note that this is |
c349b1b9 |
77 | currently only implemented on Windows since other platforms lack an |
78 | API standard on this area. |
8cbd9a7a |
79 | |
75daf61c |
80 | Regardless of the above, the C<bytes> pragma can always be used to |
81 | force byte semantics in a particular lexical scope. See L<bytes>. |
8cbd9a7a |
82 | |
83 | The C<utf8> pragma is primarily a compatibility device that enables |
75daf61c |
84 | recognition of UTF-(8|EBCDIC) in literals encountered by the parser. |
7dedd01f |
85 | Note that this pragma is only required until a future version of Perl |
86 | in which character semantics will become the default. This pragma may |
87 | then become a no-op. See L<utf8>. |
8cbd9a7a |
88 | |
89 | Unless mentioned otherwise, Perl operators will use character semantics |
90 | when they are dealing with Unicode data, and byte semantics otherwise. |
91 | Thus, character semantics for these operations apply transparently; if |
92 | the input data came from a Unicode source (for example, by adding a |
93 | character encoding discipline to the filehandle whence it came, or a |
94 | literal UTF-8 string constant in the program), character semantics |
95 | apply; otherwise, byte semantics are in effect. To force byte semantics |
8058d7ab |
96 | on Unicode data, the C<bytes> pragma should be used. |
393fec97 |
97 | |
0a378802 |
98 | Notice that if you concatenate strings with byte semantics and strings |
99 | with Unicode character data, the bytes will by default be upgraded |
100 | I<as if they were ISO 8859-1 (Latin-1)> (or if in EBCDIC, after a |
101 | translation to ISO 8859-1). To change this, use the C<encoding> |
102 | pragma, see L<encoding>. |
7dedd01f |
103 | |
393fec97 |
104 | Under character semantics, many operations that formerly operated on |
75daf61c |
105 | bytes change to operating on characters. For ASCII data this makes no |
106 | difference, because UTF-8 stores ASCII in single bytes, but for any |
107 | character greater than C<chr(127)>, the character B<may> be stored in |
393fec97 |
108 | a sequence of two or more bytes, all of which have the high bit set. |
2796c109 |
109 | |
110 | For C1 controls or Latin 1 characters on an EBCDIC platform the |
111 | character may be stored in a UTF-EBCDIC multi byte sequence. But by |
112 | and large, the user need not worry about this, because Perl hides it |
113 | from the user. A character in Perl is logically just a number ranging |
114 | from 0 to 2**32 or so. Larger characters encode to longer sequences |
115 | of bytes internally, but again, this is just an internal detail which |
116 | is hidden at the Perl level. |
393fec97 |
117 | |
8cbd9a7a |
118 | =head2 Effects of character semantics |
393fec97 |
119 | |
120 | Character semantics have the following effects: |
121 | |
122 | =over 4 |
123 | |
124 | =item * |
125 | |
126 | Strings and patterns may contain characters that have an ordinal value |
21bad921 |
127 | larger than 255. |
393fec97 |
128 | |
75daf61c |
129 | Presuming you use a Unicode editor to edit your program, such |
130 | characters will typically occur directly within the literal strings as |
131 | UTF-8 (or UTF-EBCDIC on EBCDIC platforms) characters, but you can also |
132 | specify a particular character with an extension of the C<\x> |
133 | notation. UTF-X characters are specified by putting the hexadecimal |
134 | code within curlies after the C<\x>. For instance, a Unicode smiley |
135 | face is C<\x{263A}>. |
393fec97 |
136 | |
137 | =item * |
138 | |
139 | Identifiers within the Perl script may contain Unicode alphanumeric |
140 | characters, including ideographs. (You are currently on your own when |
75daf61c |
141 | it comes to using the canonical forms of characters--Perl doesn't |
142 | (yet) attempt to canonicalize variable names for you.) |
393fec97 |
143 | |
393fec97 |
144 | =item * |
145 | |
146 | Regular expressions match characters instead of bytes. For instance, |
147 | "." matches a character instead of a byte. (However, the C<\C> pattern |
75daf61c |
148 | is provided to force a match a single byte ("C<char>" in C, hence C<\C>).) |
393fec97 |
149 | |
393fec97 |
150 | =item * |
151 | |
152 | Character classes in regular expressions match characters instead of |
153 | bytes, and match against the character properties specified in the |
75daf61c |
154 | Unicode properties database. So C<\w> can be used to match an |
155 | ideograph, for instance. |
393fec97 |
156 | |
393fec97 |
157 | =item * |
158 | |
cfc01aea |
159 | Named Unicode properties and block ranges may be used as character |
393fec97 |
160 | classes via the new C<\p{}> (matches property) and C<\P{}> (doesn't |
161 | match property) constructs. For instance, C<\p{Lu}> matches any |
162 | character with the Unicode uppercase property, while C<\p{M}> matches |
9fdf68be |
163 | any mark character. Single letter properties may omit the brackets, |
164 | so that can be written C<\pM> also. Many predefined character classes |
a1cc1cb1 |
165 | are available, such as C<\p{IsMirrored}> and C<\p{InTibetan}>. |
4193bef7 |
166 | |
167 | The C<\p{Is...}> test for "general properties" such as "letter", |
168 | "digit", while the C<\p{In...}> test for Unicode scripts and blocks. |
169 | |
cfc01aea |
170 | The official Unicode script and block names have spaces and dashes as |
e150c829 |
171 | separators, but for convenience you can have dashes, spaces, and |
172 | underbars at every word division, and you need not care about correct |
173 | casing. It is recommended, however, that for consistency you use the |
174 | following naming: the official Unicode script, block, or property name |
175 | (see below for the additional rules that apply to block names), |
176 | with whitespace and dashes replaced with underbar, and the words |
177 | "uppercase-first-lowercase-rest". That is, "Latin-1 Supplement" |
178 | becomes "Latin_1_Supplement". |
4193bef7 |
179 | |
a1cc1cb1 |
180 | You can also negate both C<\p{}> and C<\P{}> by introducing a caret |
e150c829 |
181 | (^) between the first curly and the property name: C<\p{^In_Tamil}> is |
182 | equal to C<\P{In_Tamil}>. |
4193bef7 |
183 | |
61247495 |
184 | The C<In> and C<Is> can be left out: C<\p{Greek}> is equal to |
e150c829 |
185 | C<\p{In_Greek}>, C<\P{Pd}> is equal to C<\P{Pd}>. |
393fec97 |
186 | |
d73e5302 |
187 | Short Long |
188 | |
189 | L Letter |
e150c829 |
190 | Lu Uppercase_Letter |
191 | Ll Lowercase_Letter |
192 | Lt Titlecase_Letter |
193 | Lm Modifier_Letter |
194 | Lo Other_Letter |
d73e5302 |
195 | |
196 | M Mark |
e150c829 |
197 | Mn Nonspacing_Mark |
198 | Mc Spacing_Mark |
199 | Me Enclosing_Mark |
d73e5302 |
200 | |
201 | N Number |
e150c829 |
202 | Nd Decimal_Number |
203 | Nl Letter_Number |
204 | No Other_Number |
d73e5302 |
205 | |
206 | P Punctuation |
e150c829 |
207 | Pc Connector_Punctuation |
208 | Pd Dash_Punctuation |
209 | Ps Open_Punctuation |
210 | Pe Close_Punctuation |
211 | Pi Initial_Punctuation |
d73e5302 |
212 | (may behave like Ps or Pe depending on usage) |
e150c829 |
213 | Pf Final_Punctuation |
d73e5302 |
214 | (may behave like Ps or Pe depending on usage) |
e150c829 |
215 | Po Other_Punctuation |
d73e5302 |
216 | |
217 | S Symbol |
e150c829 |
218 | Sm Math_Symbol |
219 | Sc Currency_Symbol |
220 | Sk Modifier_Symbol |
221 | So Other_Symbol |
d73e5302 |
222 | |
223 | Z Separator |
e150c829 |
224 | Zs Space_Separator |
225 | Zl Line_Separator |
226 | Zp Paragraph_Separator |
d73e5302 |
227 | |
228 | C Other |
e150c829 |
229 | Cc Control |
230 | Cf Format |
231 | Cs Surrogate |
232 | Co Private_Use |
233 | Cn Unassigned |
1ac13f9a |
234 | |
235 | There's also C<L&> which is an alias for C<Ll>, C<Lu>, and C<Lt>. |
32293815 |
236 | |
d73e5302 |
237 | The following reserved ranges have C<In> tests: |
238 | |
e150c829 |
239 | CJK_Ideograph_Extension_A |
240 | CJK_Ideograph |
241 | Hangul_Syllable |
242 | Non_Private_Use_High_Surrogate |
243 | Private_Use_High_Surrogate |
244 | Low_Surrogate |
245 | Private_Surrogate |
246 | CJK_Ideograph_Extension_B |
247 | Plane_15_Private_Use |
248 | Plane_16_Private_Use |
d73e5302 |
249 | |
250 | For example C<"\x{AC00}" =~ \p{HangulSyllable}> will test true. |
e9ad1727 |
251 | (Handling of surrogates is not implemented yet, because Perl |
0675fa52 |
252 | uses UTF-8 and not UTF-16 internally to represent Unicode. |
253 | So you really can't use the "Cs" category.) |
d73e5302 |
254 | |
32293815 |
255 | Additionally, because scripts differ in their directionality |
256 | (for example Hebrew is written right to left), all characters |
257 | have their directionality defined: |
258 | |
d73e5302 |
259 | BidiL Left-to-Right |
260 | BidiLRE Left-to-Right Embedding |
261 | BidiLRO Left-to-Right Override |
262 | BidiR Right-to-Left |
263 | BidiAL Right-to-Left Arabic |
264 | BidiRLE Right-to-Left Embedding |
265 | BidiRLO Right-to-Left Override |
266 | BidiPDF Pop Directional Format |
267 | BidiEN European Number |
268 | BidiES European Number Separator |
269 | BidiET European Number Terminator |
270 | BidiAN Arabic Number |
271 | BidiCS Common Number Separator |
272 | BidiNSM Non-Spacing Mark |
273 | BidiBN Boundary Neutral |
274 | BidiB Paragraph Separator |
275 | BidiS Segment Separator |
276 | BidiWS Whitespace |
277 | BidiON Other Neutrals |
32293815 |
278 | |
210b36aa |
279 | =back |
280 | |
2796c109 |
281 | =head2 Scripts |
282 | |
75daf61c |
283 | The scripts available for C<\p{In...}> and C<\P{In...}>, for example |
cfc01aea |
284 | C<\p{InLatin}> or \p{InCyrillic>, are as follows: |
2796c109 |
285 | |
1ac13f9a |
286 | Arabic |
e9ad1727 |
287 | Armenian |
1ac13f9a |
288 | Bengali |
e9ad1727 |
289 | Bopomofo |
290 | Canadian-Aboriginal |
291 | Cherokee |
292 | Cyrillic |
293 | Deseret |
294 | Devanagari |
295 | Ethiopic |
296 | Georgian |
297 | Gothic |
298 | Greek |
1ac13f9a |
299 | Gujarati |
e9ad1727 |
300 | Gurmukhi |
301 | Han |
302 | Hangul |
303 | Hebrew |
304 | Hiragana |
305 | Inherited |
1ac13f9a |
306 | Kannada |
e9ad1727 |
307 | Katakana |
308 | Khmer |
1ac13f9a |
309 | Lao |
e9ad1727 |
310 | Latin |
311 | Malayalam |
312 | Mongolian |
1ac13f9a |
313 | Myanmar |
1ac13f9a |
314 | Ogham |
e9ad1727 |
315 | Old-Italic |
316 | Oriya |
1ac13f9a |
317 | Runic |
e9ad1727 |
318 | Sinhala |
319 | Syriac |
320 | Tamil |
321 | Telugu |
322 | Thaana |
323 | Thai |
324 | Tibetan |
1ac13f9a |
325 | Yi |
1ac13f9a |
326 | |
327 | There are also extended property classes that supplement the basic |
328 | properties, defined by the F<PropList> Unicode database: |
329 | |
e9ad1727 |
330 | ASCII_Hex_Digit |
1ac13f9a |
331 | Bidi_Control |
1ac13f9a |
332 | Dash |
1ac13f9a |
333 | Diacritic |
334 | Extender |
e9ad1727 |
335 | Hex_Digit |
336 | Hyphen |
337 | Ideographic |
338 | Join_Control |
339 | Noncharacter_Code_Point |
340 | Other_Alphabetic |
1ac13f9a |
341 | Other_Lowercase |
e9ad1727 |
342 | Other_Math |
1ac13f9a |
343 | Other_Uppercase |
e9ad1727 |
344 | Quotation_Mark |
e150c829 |
345 | White_Space |
1ac13f9a |
346 | |
347 | and further derived properties: |
348 | |
349 | Alphabetic Lu + Ll + Lt + Lm + Lo + Other_Alphabetic |
350 | Lowercase Ll + Other_Lowercase |
351 | Uppercase Lu + Other_Uppercase |
352 | Math Sm + Other_Math |
353 | |
354 | ID_Start Lu + Ll + Lt + Lm + Lo + Nl |
355 | ID_Continue ID_Start + Mn + Mc + Nd + Pc |
356 | |
357 | Any Any character |
358 | Assigned Any non-Cn character |
359 | Common Any character (or unassigned code point) |
e150c829 |
360 | not explicitly assigned to a script |
2796c109 |
361 | |
362 | =head2 Blocks |
363 | |
364 | In addition to B<scripts>, Unicode also defines B<blocks> of |
365 | characters. The difference between scripts and blocks is that the |
e9ad1727 |
366 | scripts concept is closer to natural languages, while the blocks |
2796c109 |
367 | concept is more an artificial grouping based on groups of 256 Unicode |
368 | characters. For example, the C<Latin> script contains letters from |
e9ad1727 |
369 | many blocks. On the other hand, the C<Latin> script does not contain |
cfc01aea |
370 | all the characters from those blocks. It does not, for example, contain |
e9ad1727 |
371 | digits because digits are shared across many scripts. Digits and |
372 | other similar groups, like punctuation, are in a category called |
373 | C<Common>. |
2796c109 |
374 | |
cfc01aea |
375 | For more about scripts, see the UTR #24: |
376 | |
377 | http://www.unicode.org/unicode/reports/tr24/ |
378 | |
379 | For more about blocks, see: |
380 | |
381 | http://www.unicode.org/Public/UNIDATA/Blocks.txt |
2796c109 |
382 | |
383 | Because there are overlaps in naming (there are, for example, both |
384 | a script called C<Katakana> and a block called C<Katakana>, the block |
385 | version has C<Block> appended to its name, C<\p{InKatakanaBlock}>. |
386 | |
387 | Notice that this definition was introduced in Perl 5.8.0: in Perl |
e150c829 |
388 | 5.6 only the blocks were used; in Perl 5.8.0 scripts became the |
61247495 |
389 | preferential Unicode character class definition; this meant that |
390 | the definitions of some character classes changed (the ones in the |
2796c109 |
391 | below list that have the C<Block> appended). |
392 | |
e9ad1727 |
393 | Alphabetic Presentation Forms |
394 | Arabic Block |
395 | Arabic Presentation Forms-A |
396 | Arabic Presentation Forms-B |
397 | Armenian Block |
398 | Arrows |
71d929cb |
399 | Basic Latin |
e9ad1727 |
400 | Bengali Block |
401 | Block Elements |
402 | Bopomofo Block |
403 | Bopomofo Extended |
404 | Box Drawing |
405 | Braille Patterns |
406 | Byzantine Musical Symbols |
407 | CJK Compatibility |
408 | CJK Compatibility Forms |
409 | CJK Compatibility Ideographs |
410 | CJK Compatibility Ideographs Supplement |
411 | CJK Radicals Supplement |
412 | CJK Symbols and Punctuation |
413 | CJK Unified Ideographs |
414 | CJK Unified Ideographs Extension A |
415 | CJK Unified Ideographs Extension B |
416 | Cherokee Block |
71d929cb |
417 | Combining Diacritical Marks |
e9ad1727 |
418 | Combining Half Marks |
419 | Combining Marks for Symbols |
420 | Control Pictures |
421 | Currency Symbols |
71d929cb |
422 | Cyrillic Block |
e9ad1727 |
423 | Deseret Block |
71d929cb |
424 | Devanagari Block |
e9ad1727 |
425 | Dingbats |
426 | Enclosed Alphanumerics |
427 | Enclosed CJK Letters and Months |
428 | Ethiopic Block |
429 | General Punctuation |
430 | Geometric Shapes |
71d929cb |
431 | Georgian Block |
e9ad1727 |
432 | Gothic Block |
433 | Greek Block |
434 | Greek Extended |
435 | Gujarati Block |
436 | Gurmukhi Block |
437 | Halfwidth and Fullwidth Forms |
438 | Hangul Compatibility Jamo |
71d929cb |
439 | Hangul Jamo |
e9ad1727 |
440 | Hangul Syllables |
441 | Hebrew Block |
442 | High Private Use Surrogates |
443 | High Surrogates |
444 | Hiragana Block |
445 | IPA Extensions |
446 | Ideographic Description Characters |
447 | Kanbun |
448 | Kangxi Radicals |
449 | Kannada Block |
450 | Katakana Block |
71d929cb |
451 | Khmer Block |
e9ad1727 |
452 | Lao Block |
453 | Latin 1 Supplement |
71d929cb |
454 | Latin Extended Additional |
e9ad1727 |
455 | Latin Extended-A |
456 | Latin Extended-B |
71d929cb |
457 | Letterlike Symbols |
e9ad1727 |
458 | Low Surrogates |
459 | Malayalam Block |
460 | Mathematical Alphanumeric Symbols |
71d929cb |
461 | Mathematical Operators |
e9ad1727 |
462 | Miscellaneous Symbols |
71d929cb |
463 | Miscellaneous Technical |
e9ad1727 |
464 | Mongolian Block |
465 | Musical Symbols |
466 | Myanmar Block |
467 | Number Forms |
468 | Ogham Block |
469 | Old Italic Block |
71d929cb |
470 | Optical Character Recognition |
e9ad1727 |
471 | Oriya Block |
71d929cb |
472 | Private Use |
e9ad1727 |
473 | Runic Block |
474 | Sinhala Block |
71d929cb |
475 | Small Form Variants |
e9ad1727 |
476 | Spacing Modifier Letters |
2796c109 |
477 | Specials |
e9ad1727 |
478 | Superscripts and Subscripts |
479 | Syriac Block |
2796c109 |
480 | Tags |
e9ad1727 |
481 | Tamil Block |
482 | Telugu Block |
483 | Thaana Block |
484 | Thai Block |
485 | Tibetan Block |
486 | Unified Canadian Aboriginal Syllabics |
487 | Yi Radicals |
488 | Yi Syllables |
32293815 |
489 | |
210b36aa |
490 | =over 4 |
491 | |
393fec97 |
492 | =item * |
493 | |
494 | The special pattern C<\X> match matches any extended Unicode sequence |
495 | (a "combining character sequence" in Standardese), where the first |
496 | character is a base character and subsequent characters are mark |
497 | characters that apply to the base character. It is equivalent to |
498 | C<(?:\PM\pM*)>. |
499 | |
393fec97 |
500 | =item * |
501 | |
383e7cdd |
502 | The C<tr///> operator translates characters instead of bytes. Note |
503 | that the C<tr///CU> functionality has been removed, as the interface |
504 | was a mistake. For similar functionality see pack('U0', ...) and |
505 | pack('C0', ...). |
393fec97 |
506 | |
393fec97 |
507 | =item * |
508 | |
509 | Case translation operators use the Unicode case translation tables |
44bc797b |
510 | when provided character input. Note that C<uc()> (also known as C<\U> |
511 | in doublequoted strings) translates to uppercase, while C<ucfirst> |
512 | (also known as C<\u> in doublequoted strings) translates to titlecase |
513 | (for languages that make the distinction). Naturally the |
514 | corresponding backslash sequences have the same semantics. |
393fec97 |
515 | |
516 | =item * |
517 | |
518 | Most operators that deal with positions or lengths in the string will |
75daf61c |
519 | automatically switch to using character positions, including |
520 | C<chop()>, C<substr()>, C<pos()>, C<index()>, C<rindex()>, |
521 | C<sprintf()>, C<write()>, and C<length()>. Operators that |
522 | specifically don't switch include C<vec()>, C<pack()>, and |
523 | C<unpack()>. Operators that really don't care include C<chomp()>, as |
524 | well as any other operator that treats a string as a bucket of bits, |
525 | such as C<sort()>, and the operators dealing with filenames. |
393fec97 |
526 | |
527 | =item * |
528 | |
529 | The C<pack()>/C<unpack()> letters "C<c>" and "C<C>" do I<not> change, |
530 | since they're often used for byte-oriented formats. (Again, think |
531 | "C<char>" in the C language.) However, there is a new "C<U>" specifier |
532 | that will convert between UTF-8 characters and integers. (It works |
533 | outside of the utf8 pragma too.) |
534 | |
535 | =item * |
536 | |
537 | The C<chr()> and C<ord()> functions work on characters. This is like |
538 | C<pack("U")> and C<unpack("U")>, not like C<pack("C")> and |
539 | C<unpack("C")>. In fact, the latter are how you now emulate |
35bcd338 |
540 | byte-oriented C<chr()> and C<ord()> for Unicode strings. |
541 | (Note that this reveals the internal UTF-8 encoding of strings and |
542 | you are not supposed to do that unless you know what you are doing.) |
393fec97 |
543 | |
544 | =item * |
545 | |
a1ca4561 |
546 | The bit string operators C<& | ^ ~> can operate on character data. |
547 | However, for backward compatibility reasons (bit string operations |
75daf61c |
548 | when the characters all are less than 256 in ordinal value) one should |
549 | not mix C<~> (the bit complement) and characters both less than 256 and |
a1ca4561 |
550 | equal or greater than 256. Most importantly, the DeMorgan's laws |
551 | (C<~($x|$y) eq ~$x&~$y>, C<~($x&$y) eq ~$x|~$y>) won't hold. |
552 | Another way to look at this is that the complement cannot return |
75daf61c |
553 | B<both> the 8-bit (byte) wide bit complement B<and> the full character |
a1ca4561 |
554 | wide bit complement. |
555 | |
556 | =item * |
557 | |
983ffd37 |
558 | lc(), uc(), lcfirst(), and ucfirst() work for the following cases: |
559 | |
560 | =over 8 |
561 | |
562 | =item * |
563 | |
564 | the case mapping is from a single Unicode character to another |
565 | single Unicode character |
566 | |
567 | =item * |
568 | |
569 | the case mapping is from a single Unicode character to more |
570 | than one Unicode character |
571 | |
572 | =back |
573 | |
210b36aa |
574 | What doesn't yet work are the following cases: |
983ffd37 |
575 | |
576 | =over 8 |
577 | |
578 | =item * |
579 | |
580 | the "final sigma" (Greek) |
581 | |
582 | =item * |
583 | |
584 | anything to with locales (Lithuanian, Turkish, Azeri) |
585 | |
586 | =back |
587 | |
588 | See the Unicode Technical Report #21, Case Mappings, for more details. |
ac1256e8 |
589 | |
590 | =item * |
591 | |
393fec97 |
592 | And finally, C<scalar reverse()> reverses by character rather than by byte. |
593 | |
594 | =back |
595 | |
8cbd9a7a |
596 | =head2 Character encodings for input and output |
597 | |
7221edc9 |
598 | See L<Encode>. |
8cbd9a7a |
599 | |
393fec97 |
600 | =head1 CAVEATS |
601 | |
602 | As of yet, there is no method for automatically coercing input and |
b3419ed8 |
603 | output to some encoding other than UTF-8 or UTF-EBCDIC. This is planned |
604 | in the near future, however. |
393fec97 |
605 | |
8cbd9a7a |
606 | Whether an arbitrary piece of data will be treated as "characters" or |
607 | "bytes" by internal operations cannot be divined at the current time. |
393fec97 |
608 | |
609 | Use of locales with utf8 may lead to odd results. Currently there is |
610 | some attempt to apply 8-bit locale info to characters in the range |
611 | 0..255, but this is demonstrably incorrect for locales that use |
612 | characters above that range (when mapped into Unicode). It will also |
613 | tend to run slower. Avoidance of locales is strongly encouraged. |
614 | |
776f8809 |
615 | =head1 UNICODE REGULAR EXPRESSION SUPPORT LEVEL |
616 | |
617 | The following list of Unicode regular expression support describes |
618 | feature by feature the Unicode support implemented in Perl as of Perl |
619 | 5.8.0. The "Level N" and the section numbers refer to the Unicode |
620 | Technical Report 18, "Unicode Regular Expression Guidelines". |
621 | |
622 | =over 4 |
623 | |
624 | =item * |
625 | |
626 | Level 1 - Basic Unicode Support |
627 | |
628 | 2.1 Hex Notation - done [1] |
629 | Named Notation - done [2] |
630 | 2.2 Categories - done [3][4] |
631 | 2.3 Subtraction - MISSING [5][6] |
632 | 2.4 Simple Word Boundaries - done [7] |
90a59240 |
633 | 2.5 Simple Loose Matches - done [8] |
776f8809 |
634 | 2.6 End of Line - MISSING [9][10] |
635 | |
636 | [ 1] \x{...} |
637 | [ 2] \N{...} |
638 | [ 3] . \p{Is...} \P{Is...} |
29bdacb8 |
639 | [ 4] now scripts (see UTR#24 Script Names) in addition to blocks |
776f8809 |
640 | [ 5] have negation |
29bdacb8 |
641 | [ 6] can use look-ahead to emulate subtraction (*) |
776f8809 |
642 | [ 7] include Letters in word characters |
90a59240 |
643 | [ 8] see UTR#21 Case Mappings: Perl implements 1:1 mappings |
776f8809 |
644 | [ 9] see UTR#13 Unicode Newline Guidelines |
ec83e909 |
645 | [10] should do ^ and $ also on \x{85}, \x{2028} and \x{2029}) |
646 | (should also affect <>, $., and script line numbers) |
647 | |
dbe420b4 |
648 | (*) You can mimic class subtraction using lookahead. |
649 | For example, what TR18 might write as |
29bdacb8 |
650 | |
dbe420b4 |
651 | [{Greek}-[{UNASSIGNED}]] |
652 | |
653 | in Perl can be written as: |
654 | |
655 | (?!\p{UNASSIGNED})\p{GreekBlock} |
656 | (?=\p{ASSIGNED})\p{GreekBlock} |
657 | |
658 | But in this particular example, you probably really want |
659 | |
660 | \p{Greek} |
661 | |
662 | which will match assigned characters known to be part of the Greek script. |
29bdacb8 |
663 | |
664 | In other words: the matched character must not be a non-assigned |
665 | character, but it must be in the block of modern Greek characters. |
666 | |
776f8809 |
667 | =item * |
668 | |
669 | Level 2 - Extended Unicode Support |
670 | |
671 | 3.1 Surrogates - MISSING |
672 | 3.2 Canonical Equivalents - MISSING [11][12] |
673 | 3.3 Locale-Independent Graphemes - MISSING [13] |
674 | 3.4 Locale-Independent Words - MISSING [14] |
675 | 3.5 Locale-Independent Loose Matches - MISSING [15] |
676 | |
677 | [11] see UTR#15 Unicode Normalization |
678 | [12] have Unicode::Normalize but not integrated to regexes |
679 | [13] have \X but at this level . should equal that |
680 | [14] need three classes, not just \w and \W |
681 | [15] see UTR#21 Case Mappings |
682 | |
683 | =item * |
684 | |
685 | Level 3 - Locale-Sensitive Support |
686 | |
687 | 4.1 Locale-Dependent Categories - MISSING |
688 | 4.2 Locale-Dependent Graphemes - MISSING [16][17] |
689 | 4.3 Locale-Dependent Words - MISSING |
690 | 4.4 Locale-Dependent Loose Matches - MISSING |
691 | 4.5 Locale-Dependent Ranges - MISSING |
692 | |
693 | [16] see UTR#10 Unicode Collation Algorithms |
694 | [17] have Unicode::Collate but not integrated to regexes |
695 | |
696 | =back |
697 | |
c349b1b9 |
698 | =head2 Unicode Encodings |
699 | |
700 | Unicode characters are assigned to I<code points> which are abstract |
86bbd6d1 |
701 | numbers. To use these numbers various encodings are needed. |
c349b1b9 |
702 | |
703 | =over 4 |
704 | |
705 | =item UTF-8 |
706 | |
86bbd6d1 |
707 | UTF-8 is the encoding used internally by Perl. UTF-8 is a variable |
c349b1b9 |
708 | length (1 to 6 bytes, current character allocations require 4 bytes), |
86bbd6d1 |
709 | byteorder independent encoding. For ASCII, UTF-8 is transparent |
710 | (and we really do mean 7-bit ASCII, not any 8-bit encoding). |
c349b1b9 |
711 | |
05632f9a |
712 | The following table is from Unicode 3.1. |
713 | |
714 | Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte |
715 | |
716 | U+0000..U+007F 00..7FÂ Â Â |
717 | U+0080..U+07FF C2..DF 80..BFÂ Â Â |
718 | U+0800..U+0FFF E0 A0..BF 80..BFÂ Â |
719 | U+1000..U+FFFF E1..EF 80..BF 80..BFÂ Â |
720 | U+10000..U+3FFFF F0 90..BF 80..BF 80..BF |
721 | U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF |
722 | U+100000..U+10FFFF F4 80..8F 80..BF 80..BF |
723 | |
724 | Or, another way to look at it, as bits: |
725 | |
726 | Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte |
727 | |
728 | 0aaaaaaa 0aaaaaaa |
729 | 00000bbbbbaaaaaa 110bbbbb 10aaaaaa |
730 | ccccbbbbbbaaaaaa 1110cccc 10bbbbbb 10aaaaaa |
731 | 00000dddccccccbbbbbbaaaaaa 11110ddd 10cccccc 10bbbbbb 10aaaaaa |
732 | |
733 | As you can see, the continuation bytes all begin with C<10>, and the |
734 | leading bits of the start byte tells how many bytes the are in the |
735 | encoded character. |
736 | |
dbe420b4 |
737 | =item UTF-EBDIC |
738 | |
739 | Like UTF-8, but EBDCIC-safe, as UTF-8 is ASCII-safe. |
740 | |
c349b1b9 |
741 | =item UTF-16, UTF-16BE, UTF16-LE, Surrogates, and BOMs (Byte Order Marks) |
742 | |
dbe420b4 |
743 | (The followings items are mostly for reference, Perl doesn't |
744 | use them internally.) |
745 | |
c349b1b9 |
746 | UTF-16 is a 2 or 4 byte encoding. The Unicode code points |
747 | 0x0000..0xFFFF are stored in two 16-bit units, and the code points |
dbe420b4 |
748 | 0x010000..0x10FFFF in two 16-bit units. The latter case is |
c349b1b9 |
749 | using I<surrogates>, the first 16-bit unit being the I<high |
750 | surrogate>, and the second being the I<low surrogate>. |
751 | |
752 | Surrogates are code points set aside to encode the 0x01000..0x10FFFF |
753 | range of Unicode code points in pairs of 16-bit units. The I<high |
754 | surrogates> are the range 0xD800..0xDBFF, and the I<low surrogates> |
755 | are the range 0xDC00..0xDFFFF. The surrogate encoding is |
756 | |
757 | $hi = ($uni - 0x10000) / 0x400 + 0xD800; |
758 | $lo = ($uni - 0x10000) % 0x400 + 0xDC00; |
759 | |
760 | and the decoding is |
761 | |
762 | $uni = 0x10000 + ($hi - 0xD8000) * 0x400 + ($lo - 0xDC00); |
763 | |
9466bab6 |
764 | If you try to generate surrogates (for example by using chr()), you |
dbe420b4 |
765 | will get an error because firstly a surrogate on its own is meaningless, |
766 | and secondly because Perl encodes its Unicode characters in UTF-8 |
767 | (not 16-bit numbers), which makes the encoded character doubly illegal. |
9466bab6 |
768 | |
86bbd6d1 |
769 | Because of the 16-bitness, UTF-16 is byteorder dependent. UTF-16 |
c349b1b9 |
770 | itself can be used for in-memory computations, but if storage or |
86bbd6d1 |
771 | transfer is required, either UTF-16BE (Big Endian) or UTF-16LE |
c349b1b9 |
772 | (Little Endian) must be chosen. |
773 | |
774 | This introduces another problem: what if you just know that your data |
775 | is UTF-16, but you don't know which endianness? Byte Order Marks |
776 | (BOMs) are a solution to this. A special character has been reserved |
86bbd6d1 |
777 | in Unicode to function as a byte order marker: the character with the |
778 | code point 0xFEFF is the BOM. |
042da322 |
779 | |
c349b1b9 |
780 | The trick is that if you read a BOM, you will know the byte order, |
781 | since if it was written on a big endian platform, you will read the |
86bbd6d1 |
782 | bytes 0xFE 0xFF, but if it was written on a little endian platform, |
783 | you will read the bytes 0xFF 0xFE. (And if the originating platform |
784 | was writing in UTF-8, you will read the bytes 0xEF 0xBB 0xBF.) |
042da322 |
785 | |
86bbd6d1 |
786 | The way this trick works is that the character with the code point |
787 | 0xFFFE is guaranteed not to be a valid Unicode character, so the |
788 | sequence of bytes 0xFF 0xFE is unambiguously "BOM, represented in |
042da322 |
789 | little-endian format" and cannot be "0xFFFE, represented in big-endian |
790 | format". |
c349b1b9 |
791 | |
792 | =item UTF-32, UTF-32BE, UTF32-LE |
793 | |
794 | The UTF-32 family is pretty much like the UTF-16 family, expect that |
042da322 |
795 | the units are 32-bit, and therefore the surrogate scheme is not |
796 | needed. The BOM signatures will be 0x00 0x00 0xFE 0xFF for BE and |
797 | 0xFF 0xFE 0x00 0x00 for LE. |
c349b1b9 |
798 | |
799 | =item UCS-2, UCS-4 |
800 | |
86bbd6d1 |
801 | Encodings defined by the ISO 10646 standard. UCS-2 is a 16-bit |
802 | encoding, UCS-4 is a 32-bit encoding. Unlike UTF-16, UCS-2 |
803 | is not extensible beyond 0xFFFF, because it does not use surrogates. |
c349b1b9 |
804 | |
805 | =item UTF-7 |
806 | |
807 | A seven-bit safe (non-eight-bit) encoding, useful if the |
808 | transport/storage is not eight-bit safe. Defined by RFC 2152. |
809 | |
95a1a48b |
810 | =back |
811 | |
bf0fa0b2 |
812 | =head2 Security Implications of Malformed UTF-8 |
813 | |
814 | Unfortunately, the specification of UTF-8 leaves some room for |
815 | interpretation of how many bytes of encoded output one should generate |
816 | from one input Unicode character. Strictly speaking, one is supposed |
817 | to always generate the shortest possible sequence of UTF-8 bytes, |
818 | because otherwise there is potential for input buffer overflow at the |
819 | receiving end of a UTF-8 connection. Perl always generates the shortest |
820 | length UTF-8, and with warnings on (C<-w> or C<use warnings;>) Perl will |
821 | warn about non-shortest length UTF-8 (and other malformations, too, |
822 | such as the surrogates, which are not real character code points.) |
823 | |
c349b1b9 |
824 | =head2 Unicode in Perl on EBCDIC |
825 | |
826 | The way Unicode is handled on EBCDIC platforms is still rather |
86bbd6d1 |
827 | experimental. On such a platform, references to UTF-8 encoding in this |
c349b1b9 |
828 | document and elsewhere should be read as meaning UTF-EBCDIC as |
829 | specified in Unicode Technical Report 16 unless ASCII vs EBCDIC issues |
830 | are specifically discussed. There is no C<utfebcdic> pragma or |
86bbd6d1 |
831 | ":utfebcdic" layer, rather, "utf8" and ":utf8" are re-used to mean |
832 | the platform's "natural" 8-bit encoding of Unicode. See L<perlebcdic> |
833 | for more discussion of the issues. |
c349b1b9 |
834 | |
95a1a48b |
835 | =head2 Using Unicode in XS |
836 | |
837 | If you want to handle Perl Unicode in XS extensions, you may find |
838 | the following C APIs useful: |
839 | |
840 | =over 4 |
841 | |
842 | =item * |
843 | |
844 | DO_UTF8(sv) returns true if the UTF8 flag is on and the bytes |
845 | pragma is not in effect. SvUTF8(sv) returns true is the UTF8 |
846 | flag is on, the bytes pragma is ignored. Remember that UTF8 |
847 | flag being on does not mean that there would be any characters |
848 | of code points greater than 255 or 127 in the scalar, or that |
849 | there even are any characters in the scalar. The UTF8 flag |
850 | means that any characters added to the string will be encoded |
851 | in UTF8 if the code points of the characters are greater than |
852 | 255. Not "if greater than 127", since Perl's Unicode model |
853 | is not to use UTF-8 until it's really necessary. |
854 | |
855 | =item * |
856 | |
857 | uvuni_to_utf8(buf, chr) writes a Unicode character code point into a |
cfc01aea |
858 | buffer encoding the code point as UTF-8, and returns a pointer |
95a1a48b |
859 | pointing after the UTF-8 bytes. |
860 | |
861 | =item * |
862 | |
863 | utf8_to_uvuni(buf, lenp) reads UTF-8 encoded bytes from a buffer and |
864 | returns the Unicode character code point (and optionally the length of |
865 | the UTF-8 byte sequence). |
866 | |
867 | =item * |
868 | |
869 | utf8_length(s, len) returns the length of the UTF-8 encoded buffer in |
870 | characters. sv_len_utf8(sv) returns the length of the UTF-8 encoded |
871 | scalar. |
872 | |
873 | =item * |
874 | |
875 | sv_utf8_upgrade(sv) converts the string of the scalar to its UTF-8 |
876 | encoded form. sv_utf8_downgrade(sv) does the opposite (if possible). |
877 | sv_utf8_encode(sv) is like sv_utf8_upgrade but the UTF8 flag does not |
878 | get turned on. sv_utf8_decode() does the opposite of sv_utf8_encode(). |
879 | |
880 | =item * |
881 | |
882 | is_utf8_char(buf) returns true if the buffer points to valid UTF-8. |
883 | |
884 | =item * |
885 | |
886 | is_utf8_string(buf, len) returns true if the len bytes of the buffer |
887 | are valid UTF-8. |
888 | |
889 | =item * |
890 | |
891 | UTF8SKIP(buf) will return the number of bytes in the UTF-8 encoded |
892 | character in the buffer. UNISKIP(chr) will return the number of bytes |
893 | required to UTF-8-encode the Unicode character code point. |
894 | |
895 | =item * |
896 | |
897 | utf8_distance(a, b) will tell the distance in characters between the |
898 | two pointers pointing to the same UTF-8 encoded buffer. |
899 | |
900 | =item * |
901 | |
902 | utf8_hop(s, off) will return a pointer to an UTF-8 encoded buffer that |
903 | is C<off> (positive or negative) Unicode characters displaced from the |
904 | UTF-8 buffer C<s>. |
905 | |
d2cc3551 |
906 | =item * |
907 | |
908 | pv_uni_display(dsv, spv, len, pvlim, flags) and sv_uni_display(dsv, |
909 | ssv, pvlim, flags) are useful for debug output of Unicode strings and |
910 | scalars (only for debug: they display B<all> characters as hexadecimal |
911 | code points). |
912 | |
913 | =item * |
914 | |
332ddc25 |
915 | ibcmp_utf8(s1, u1, len1, s2, u2, len2) can be used to compare two |
916 | strings case-insensitively in Unicode. (For case-sensitive |
917 | comparisons you can just use memEQ() and memNE() as usual.) |
d2cc3551 |
918 | |
c349b1b9 |
919 | =back |
920 | |
95a1a48b |
921 | For more information, see L<perlapi>, and F<utf8.c> and F<utf8.h> |
922 | in the Perl source code distribution. |
923 | |
393fec97 |
924 | =head1 SEE ALSO |
925 | |
72ff2908 |
926 | L<perluniintro>, L<encoding>, L<Encode>, L<open>, L<utf8>, L<bytes>, |
927 | L<perlretut>, L<perlvar/"${^WIDE_SYSTEM_CALLS}"> |
393fec97 |
928 | |
929 | =cut |