Commit | Line | Data |
393fec97 |
1 | =head1 NAME |
2 | |
3 | perlunicode - Unicode support in Perl |
4 | |
5 | =head1 DESCRIPTION |
6 | |
21bad921 |
7 | =head2 Important Caveat |
8 | |
393fec97 |
9 | WARNING: The implementation of Unicode support in Perl is incomplete. |
21bad921 |
10 | |
11 | The following areas need further work. |
12 | |
13 | =over |
14 | |
15 | =item Input and Output Disciplines |
16 | |
17 | There is currently no easy way to mark data read from a file or other |
18 | external source as being utf8. This will be one of the major areas of |
19 | focus in the near future. |
20 | |
21 | =item Regular Expressions |
22 | |
23 | The existing regular expression compiler does not produce polymorphic |
24 | opcodes. This means that the determination on whether to match Unicode |
25 | characters is made when the pattern is compiled, based on whether the |
26 | pattern contains Unicode characters, and not when the matching happens |
27 | at run time. This needs to be changed to adaptively match Unicode if |
28 | the string to be matched is Unicode. |
29 | |
30 | =item C<use utf8> still needed to enable a few features |
31 | |
32 | The C<utf8> pragma implements the tables used for Unicode support. These |
33 | tables are automatically loaded on demand, so the C<utf8> pragma need not |
34 | normally be used. |
35 | |
36 | However, as a compatibility measure, this pragma must be explicitly used |
37 | to enable recognition of UTF-8 encoded literals and identifiers in the |
38 | source text. |
39 | |
40 | =back |
41 | |
42 | =head2 Byte and Character semantics |
393fec97 |
43 | |
44 | Beginning with version 5.6, Perl uses logically wide characters to |
45 | represent strings internally. This internal representation of strings |
46 | uses the UTF-8 encoding. |
47 | |
21bad921 |
48 | In future, Perl-level operations can be expected to work with characters |
393fec97 |
49 | rather than bytes, in general. |
50 | |
8cbd9a7a |
51 | However, as strictly an interim compatibility measure, Perl v5.6 aims to |
52 | provide a safe migration path from byte semantics to character semantics |
53 | for programs. For operations where Perl can unambiguously decide that the |
54 | input data is characters, Perl now switches to character semantics. |
55 | For operations where this determination cannot be made without additional |
56 | information from the user, Perl decides in favor of compatibility, and |
57 | chooses to use byte semantics. |
58 | |
59 | This behavior preserves compatibility with earlier versions of Perl, |
60 | which allowed byte semantics in Perl operations, but only as long as |
61 | none of the program's inputs are marked as being as source of Unicode |
62 | character data. Such data may come from filehandles, from calls to |
63 | external programs, from information provided by the system (such as %ENV), |
21bad921 |
64 | or from literals and constants in the source text. |
8cbd9a7a |
65 | |
46487f74 |
66 | If the C<-C> command line switch is used, (or the ${^WIDE_SYSTEM_CALLS} |
67 | global flag is set to C<1>), all system calls will use the |
3969a896 |
68 | corresponding wide character APIs. This is currently only implemented |
46487f74 |
69 | on Windows. |
8cbd9a7a |
70 | |
8058d7ab |
71 | Regardless of the above, the C<bytes> pragma can always be used to force |
72 | byte semantics in a particular lexical scope. See L<bytes>. |
8cbd9a7a |
73 | |
74 | The C<utf8> pragma is primarily a compatibility device that enables |
21bad921 |
75 | recognition of UTF-8 in literals encountered by the parser. It may also |
76 | be used for enabling some of the more experimental Unicode support features. |
8cbd9a7a |
77 | Note that this pragma is only required until a future version of Perl |
78 | in which character semantics will become the default. This pragma may |
79 | then become a no-op. See L<utf8>. |
80 | |
81 | Unless mentioned otherwise, Perl operators will use character semantics |
82 | when they are dealing with Unicode data, and byte semantics otherwise. |
83 | Thus, character semantics for these operations apply transparently; if |
84 | the input data came from a Unicode source (for example, by adding a |
85 | character encoding discipline to the filehandle whence it came, or a |
86 | literal UTF-8 string constant in the program), character semantics |
87 | apply; otherwise, byte semantics are in effect. To force byte semantics |
8058d7ab |
88 | on Unicode data, the C<bytes> pragma should be used. |
393fec97 |
89 | |
90 | Under character semantics, many operations that formerly operated on |
91 | bytes change to operating on characters. For ASCII data this makes |
92 | no difference, because UTF-8 stores ASCII in single bytes, but for |
21bad921 |
93 | any character greater than C<chr(127)>, the character may be stored in |
393fec97 |
94 | a sequence of two or more bytes, all of which have the high bit set. |
95 | But by and large, the user need not worry about this, because Perl |
96 | hides it from the user. A character in Perl is logically just a number |
97 | ranging from 0 to 2**32 or so. Larger characters encode to longer |
98 | sequences of bytes internally, but again, this is just an internal |
99 | detail which is hidden at the Perl level. |
100 | |
8cbd9a7a |
101 | =head2 Effects of character semantics |
393fec97 |
102 | |
103 | Character semantics have the following effects: |
104 | |
105 | =over 4 |
106 | |
107 | =item * |
108 | |
109 | Strings and patterns may contain characters that have an ordinal value |
21bad921 |
110 | larger than 255. |
393fec97 |
111 | |
112 | Presuming you use a Unicode editor to edit your program, such characters |
113 | will typically occur directly within the literal strings as UTF-8 |
114 | characters, but you can also specify a particular character with an |
115 | extension of the C<\x> notation. UTF-8 characters are specified by |
116 | putting the hexadecimal code within curlies after the C<\x>. For instance, |
4375e838 |
117 | a Unicode smiley face is C<\x{263A}>. |
393fec97 |
118 | |
119 | =item * |
120 | |
121 | Identifiers within the Perl script may contain Unicode alphanumeric |
122 | characters, including ideographs. (You are currently on your own when |
123 | it comes to using the canonical forms of characters--Perl doesn't (yet) |
124 | attempt to canonicalize variable names for you.) |
125 | |
393fec97 |
126 | =item * |
127 | |
128 | Regular expressions match characters instead of bytes. For instance, |
129 | "." matches a character instead of a byte. (However, the C<\C> pattern |
130 | is provided to force a match a single byte ("C<char>" in C, hence |
131 | C<\C>).) |
132 | |
393fec97 |
133 | =item * |
134 | |
135 | Character classes in regular expressions match characters instead of |
136 | bytes, and match against the character properties specified in the |
137 | Unicode properties database. So C<\w> can be used to match an ideograph, |
138 | for instance. |
139 | |
393fec97 |
140 | =item * |
141 | |
142 | Named Unicode properties and block ranges make be used as character |
143 | classes via the new C<\p{}> (matches property) and C<\P{}> (doesn't |
144 | match property) constructs. For instance, C<\p{Lu}> matches any |
145 | character with the Unicode uppercase property, while C<\p{M}> matches |
146 | any mark character. Single letter properties may omit the brackets, so |
147 | that can be written C<\pM> also. Many predefined character classes are |
148 | available, such as C<\p{IsMirrored}> and C<\p{InTibetan}>. |
149 | |
393fec97 |
150 | =item * |
151 | |
152 | The special pattern C<\X> match matches any extended Unicode sequence |
153 | (a "combining character sequence" in Standardese), where the first |
154 | character is a base character and subsequent characters are mark |
155 | characters that apply to the base character. It is equivalent to |
156 | C<(?:\PM\pM*)>. |
157 | |
393fec97 |
158 | =item * |
159 | |
160 | The C<tr///> operator translates characters instead of bytes. It can also |
21bad921 |
161 | be forced to translate between 8-bit codes and UTF-8. For instance, if you |
162 | know your input in Latin-1, you can say: |
393fec97 |
163 | |
393fec97 |
164 | while (<>) { |
165 | tr/\0-\xff//CU; # latin1 char to utf8 |
166 | ... |
167 | } |
168 | |
169 | Similarly you could translate your output with |
170 | |
171 | tr/\0-\x{ff}//UC; # utf8 to latin1 char |
172 | |
173 | No, C<s///> doesn't take /U or /C (yet?). |
174 | |
393fec97 |
175 | =item * |
176 | |
177 | Case translation operators use the Unicode case translation tables |
178 | when provided character input. Note that C<uc()> translates to |
179 | uppercase, while C<ucfirst> translates to titlecase (for languages |
180 | that make the distinction). Naturally the corresponding backslash |
181 | sequences have the same semantics. |
182 | |
183 | =item * |
184 | |
185 | Most operators that deal with positions or lengths in the string will |
186 | automatically switch to using character positions, including C<chop()>, |
187 | C<substr()>, C<pos()>, C<index()>, C<rindex()>, C<sprintf()>, |
188 | C<write()>, and C<length()>. Operators that specifically don't switch |
189 | include C<vec()>, C<pack()>, and C<unpack()>. Operators that really |
190 | don't care include C<chomp()>, as well as any other operator that |
191 | treats a string as a bucket of bits, such as C<sort()>, and the |
192 | operators dealing with filenames. |
193 | |
194 | =item * |
195 | |
196 | The C<pack()>/C<unpack()> letters "C<c>" and "C<C>" do I<not> change, |
197 | since they're often used for byte-oriented formats. (Again, think |
198 | "C<char>" in the C language.) However, there is a new "C<U>" specifier |
199 | that will convert between UTF-8 characters and integers. (It works |
200 | outside of the utf8 pragma too.) |
201 | |
202 | =item * |
203 | |
204 | The C<chr()> and C<ord()> functions work on characters. This is like |
205 | C<pack("U")> and C<unpack("U")>, not like C<pack("C")> and |
206 | C<unpack("C")>. In fact, the latter are how you now emulate |
207 | byte-oriented C<chr()> and C<ord()> under utf8. |
208 | |
209 | =item * |
210 | |
211 | And finally, C<scalar reverse()> reverses by character rather than by byte. |
212 | |
213 | =back |
214 | |
8cbd9a7a |
215 | =head2 Character encodings for input and output |
216 | |
217 | [XXX: This feature is not yet implemented.] |
218 | |
393fec97 |
219 | =head1 CAVEATS |
220 | |
221 | As of yet, there is no method for automatically coercing input and |
222 | output to some encoding other than UTF-8. This is planned in the near |
223 | future, however. |
224 | |
8cbd9a7a |
225 | Whether an arbitrary piece of data will be treated as "characters" or |
226 | "bytes" by internal operations cannot be divined at the current time. |
393fec97 |
227 | |
228 | Use of locales with utf8 may lead to odd results. Currently there is |
229 | some attempt to apply 8-bit locale info to characters in the range |
230 | 0..255, but this is demonstrably incorrect for locales that use |
231 | characters above that range (when mapped into Unicode). It will also |
232 | tend to run slower. Avoidance of locales is strongly encouraged. |
233 | |
234 | =head1 SEE ALSO |
235 | |
8058d7ab |
236 | L<bytes>, L<utf8>, L<perlvar/"${^WIDE_SYSTEM_CALLS}"> |
393fec97 |
237 | |
238 | =cut |