.\" Automatically generated by Pod::Man 2.25 (Pod::Simple 3.16) .\" .\" Standard preamble: .\" ======================================================================== .de Sp \" Vertical space (when we can't use .PP) .if t .sp .5v .if n .sp .. .de Vb \" Begin verbatim text .ft CW .nf .ne \\$1 .. .de Ve \" End verbatim text .ft R .fi .. .\" Set up some character translations and predefined strings. \*(-- will .\" give an unbreakable dash, \*(PI will give pi, \*(L" will give a left .\" double quote, and \*(R" will give a right double quote. \*(C+ will .\" give a nicer C++. Capital omega is used to do unbreakable dashes and .\" therefore won't be available. \*(C` and \*(C' expand to `' in nroff, .\" nothing in troff, for use with C<>. .tr \(*W- .ds C+ C\v'-.1v'\h'-1p'\s-2+\h'-1p'+\s0\v'.1v'\h'-1p' .ie n \{\ . ds -- \(*W- . ds PI pi . if (\n(.H=4u)&(1m=24u) .ds -- \(*W\h'-12u'\(*W\h'-12u'-\" diablo 10 pitch . if (\n(.H=4u)&(1m=20u) .ds -- \(*W\h'-12u'\(*W\h'-8u'-\" diablo 12 pitch . ds L" "" . ds R" "" . ds C` "" . ds C' "" 'br\} .el\{\ . ds -- \|\(em\| . ds PI \(*p . ds L" `` . ds R" '' 'br\} .\" .\" Escape single quotes in literal strings from groff's Unicode transform. .ie \n(.g .ds Aq \(aq .el .ds Aq ' .\" .\" If the F register is turned on, we'll generate index entries on stderr for .\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index .\" entries marked with X<> in POD. Of course, you'll have to process the .\" output yourself in some meaningful fashion. .ie \nF \{\ . de IX . tm Index:\\$1\t\\n%\t"\\$2" .. . nr % 0 . rr F .\} .el \{\ . de IX .. .\} .\" .\" Accent mark definitions (@(#)ms.acc 1.5 88/02/08 SMI; from UCB 4.2). .\" Fear. Run. Save yourself. No user-serviceable parts. . \" fudge factors for nroff and troff .if n \{\ . ds #H 0 . ds #V .8m . ds #F .3m . ds #[ \f1 . ds #] \fP .\} .if t \{\ . ds #H ((1u-(\\\\n(.fu%2u))*.13m) . ds #V .6m . ds #F 0 . ds #[ \& . ds #] \& .\} . \" simple accents for nroff and troff .if n \{\ . ds ' \& . ds ` \& . ds ^ \& . ds , \& . ds ~ ~ . ds / .\} .if t \{\ . ds ' \\k:\h'-(\\n(.wu*8/10-\*(#H)'\'\h"|\\n:u" . ds ` \\k:\h'-(\\n(.wu*8/10-\*(#H)'\`\h'|\\n:u' . ds ^ \\k:\h'-(\\n(.wu*10/11-\*(#H)'^\h'|\\n:u' . ds , \\k:\h'-(\\n(.wu*8/10)',\h'|\\n:u' . ds ~ \\k:\h'-(\\n(.wu-\*(#H-.1m)'~\h'|\\n:u' . ds / \\k:\h'-(\\n(.wu*8/10-\*(#H)'\z\(sl\h'|\\n:u' .\} . \" troff and (daisy-wheel) nroff accents .ds : \\k:\h'-(\\n(.wu*8/10-\*(#H+.1m+\*(#F)'\v'-\*(#V'\z.\h'.2m+\*(#F'.\h'|\\n:u'\v'\*(#V' .ds 8 \h'\*(#H'\(*b\h'-\*(#H' .ds o \\k:\h'-(\\n(.wu+\w'\(de'u-\*(#H)/2u'\v'-.3n'\*(#[\z\(de\v'.3n'\h'|\\n:u'\*(#] .ds d- \h'\*(#H'\(pd\h'-\w'~'u'\v'-.25m'\f2\(hy\fP\v'.25m'\h'-\*(#H' .ds D- D\\k:\h'-\w'D'u'\v'-.11m'\z\(hy\v'.11m'\h'|\\n:u' .ds th \*(#[\v'.3m'\s+1I\s-1\v'-.3m'\h'-(\w'I'u*2/3)'\s-1o\s+1\*(#] .ds Th \*(#[\s+2I\s-2\h'-\w'I'u*3/5'\v'-.3m'o\v'.3m'\*(#] .ds ae a\h'-(\w'a'u*4/10)'e .ds Ae A\h'-(\w'A'u*4/10)'E . \" corrections for vroff .if v .ds ~ \\k:\h'-(\\n(.wu*9/10-\*(#H)'\s-2\u~\d\s+2\h'|\\n:u' .if v .ds ^ \\k:\h'-(\\n(.wu*10/11-\*(#H)'\v'-.4m'^\v'.4m'\h'|\\n:u' . \" for low resolution devices (crt and lpr) .if \n(.H>23 .if \n(.V>19 \ \{\ . ds : e . ds 8 ss . ds o a . ds d- d\h'-1'\(ga . ds D- D\h'-1'\(hy . ds th \o'bp' . ds Th \o'LP' . ds ae ae . ds Ae AE .\} .rm #[ #] #H #V #F C .\" ======================================================================== .\" .IX Title "PERLUNICODE 1" .TH PERLUNICODE 1 "2011-09-26" "perl v5.14.2" "Perl Programmers Reference Guide" .\" For nroff, turn off justification. Always turn off hyphenation; it makes .\" way too many mistakes in technical documents. .if n .ad l .nh .SH "NAME" perlunicode \- Unicode support in Perl .SH "DESCRIPTION" .IX Header "DESCRIPTION" .SS "Important Caveats" .IX Subsection "Important Caveats" Unicode support is an extensive requirement. While Perl does not implement the Unicode standard or the accompanying technical reports from cover to cover, Perl does support many Unicode features. .PP People who want to learn to use Unicode in Perl, should probably read the Perl Unicode tutorial, perlunitut and perluniintro, before reading this reference document. .PP Also, the use of Unicode may present security issues that aren't obvious. Read Unicode Security Considerations . .ie n .IP "Safest if you ""use feature 'unicode_strings'""" 4 .el .IP "Safest if you ``use feature 'unicode_strings'''" 4 .IX Item "Safest if you use feature 'unicode_strings'" In order to preserve backward compatibility, Perl does not turn on full internal Unicode support unless the pragma \&\f(CW\*(C`use feature \*(Aqunicode_strings\*(Aq\*(C'\fR is specified. (This is automatically selected if you use \f(CW\*(C`use 5.012\*(C'\fR or higher.) Failure to do this can trigger unexpected surprises. See \*(L"The \*(R"Unicode Bug"" below. .Sp This pragma doesn't affect I/O, and there are still several places where Unicode isn't fully supported, such as in filenames. .IP "Input and Output Layers" 4 .IX Item "Input and Output Layers" Perl knows when a filehandle uses Perl's internal Unicode encodings (\s-1UTF\-8\s0, or UTF-EBCDIC if in \s-1EBCDIC\s0) if the filehandle is opened with the \*(L":encoding(utf8)\*(R" layer. Other encodings can be converted to Perl's encoding on input or from Perl's encoding on output by use of the \&\*(L":encoding(...)\*(R" layer. See open. .Sp To indicate that Perl source itself is in \s-1UTF\-8\s0, use \f(CW\*(C`use utf8;\*(C'\fR. .ie n .IP """use utf8"" still needed to enable \s-1UTF\-8/UTF\-EBCDIC\s0 in scripts" 4 .el .IP "\f(CWuse utf8\fR still needed to enable \s-1UTF\-8/UTF\-EBCDIC\s0 in scripts" 4 .IX Item "use utf8 still needed to enable UTF-8/UTF-EBCDIC in scripts" As a compatibility measure, the \f(CW\*(C`use utf8\*(C'\fR pragma must be explicitly included to enable recognition of \s-1UTF\-8\s0 in the Perl scripts themselves (in string or regular expression literals, or in identifier names) on ASCII-based machines or to recognize UTF-EBCDIC on EBCDIC-based machines. \fBThese are the only times when an explicit \f(CB\*(C`use utf8\*(C'\fB is needed.\fR See utf8. .IP "BOM-marked scripts and \s-1UTF\-16\s0 scripts autodetected" 4 .IX Item "BOM-marked scripts and UTF-16 scripts autodetected" If a Perl script begins marked with the Unicode \s-1BOM\s0 (\s-1UTF\-16LE\s0, \s-1UTF16\-BE\s0, or \s-1UTF\-8\s0), or if the script looks like non-BOM-marked \s-1UTF\-16\s0 of either endianness, Perl will correctly read in the script as Unicode. (BOMless \s-1UTF\-8\s0 cannot be effectively recognized or differentiated from \&\s-1ISO\s0 8859\-1 or other eight-bit encodings.) .ie n .IP """use encoding"" needed to upgrade non\-Latin\-1 byte strings" 4 .el .IP "\f(CWuse encoding\fR needed to upgrade non\-Latin\-1 byte strings" 4 .IX Item "use encoding needed to upgrade non-Latin-1 byte strings" By default, there is a fundamental asymmetry in Perl's Unicode model: implicit upgrading from byte strings to Unicode strings assumes that they were encoded in \fI\s-1ISO\s0 8859\-1 (Latin\-1)\fR, but Unicode strings are downgraded with \s-1UTF\-8\s0 encoding. This happens because the first 256 codepoints in Unicode happens to agree with Latin\-1. .Sp See \*(L"Byte and Character Semantics\*(R" for more details. .SS "Byte and Character Semantics" .IX Subsection "Byte and Character Semantics" Beginning with version 5.6, Perl uses logically-wide characters to represent strings internally. .PP Starting in Perl 5.14, Perl-level operations work with characters rather than bytes within the scope of a \&\f(CW\*(C`use feature \*(Aqunicode_strings\*(Aq\*(C'\fR (or equivalently \&\f(CW\*(C`use 5.012\*(C'\fR or higher). (This is not true if bytes have been explicitly requested by \f(CW\*(C`use bytes\*(C'\fR, nor necessarily true for interactions with the platform's operating system.) .PP For earlier Perls, and when \f(CW\*(C`unicode_strings\*(C'\fR is not in effect, Perl provides a fairly safe environment that can handle both types of semantics in programs. For operations where Perl can unambiguously decide that the input data are characters, Perl switches to character semantics. For operations where this determination cannot be made without additional information from the user, Perl decides in favor of compatibility and chooses to use byte semantics. .PP When \f(CW\*(C`use locale\*(C'\fR is in effect (which overrides \&\f(CW\*(C`use feature \*(Aqunicode_strings\*(Aq\*(C'\fR in the same scope), Perl uses the semantics associated with the current locale. Otherwise, Perl uses the platform's native byte semantics for characters whose code points are less than 256, and Unicode semantics for those greater than 255. On \s-1EBCDIC\s0 platforms, this is almost seamless, as the \s-1EBCDIC\s0 code pages that Perl handles are equivalent to Unicode's first 256 code points. (The exception is that \&\s-1EBCDIC\s0 regular expression case-insensitive matching rules are not as as robust as Unicode's.) But on \s-1ASCII\s0 platforms, Perl uses US-ASCII (or Basic Latin in Unicode terminology) byte semantics, meaning that characters whose ordinal numbers are in the range 128 \- 255 are undefined except for their ordinal numbers. This means that none have case (upper and lower), nor are any a member of character classes, like \f(CW\*(C`[:alpha:]\*(C'\fR or \f(CW\*(C`\ew\*(C'\fR. (But all do belong to the \f(CW\*(C`\eW\*(C'\fR class or the Perl regular expression extension \f(CW\*(C`[:^alpha:]\*(C'\fR.) .PP This behavior preserves compatibility with earlier versions of Perl, which allowed byte semantics in Perl operations only if none of the program's inputs were marked as being a source of Unicode character data. Such data may come from filehandles, from calls to external programs, from information provided by the system (such as \f(CW%ENV\fR), or from literals and constants in the source text. .PP The \f(CW\*(C`utf8\*(C'\fR pragma is primarily a compatibility device that enables recognition of \s-1UTF\-\s0(8|EBCDIC) in literals encountered by the parser. Note that this pragma is only required while Perl defaults to byte semantics; when character semantics become the default, this pragma may become a no-op. See utf8. .PP If strings operating under byte semantics and strings with Unicode character data are concatenated, the new string will have character semantics. This can cause surprises: See \*(L"\s-1BUGS\s0\*(R", below. You can choose to be warned when this happens. See encoding::warnings. .PP Under character semantics, many operations that formerly operated on bytes now operate on characters. A character in Perl is logically just a number ranging from 0 to 2**31 or so. Larger characters may encode into longer sequences of bytes internally, but this internal detail is mostly hidden for Perl code. See perluniintro for more. .SS "Effects of Character Semantics" .IX Subsection "Effects of Character Semantics" Character semantics have the following effects: .IP "\(bu" 4 Strings\*(--including hash keys\*(--and regular expression patterns may contain characters that have an ordinal value larger than 255. .Sp If you use a Unicode editor to edit your program, Unicode characters may occur directly within the literal strings in \s-1UTF\-8\s0 encoding, or \s-1UTF\-16\s0. (The former requires a \s-1BOM\s0 or \f(CW\*(C`use utf8\*(C'\fR, the latter requires a \s-1BOM\s0.) .Sp Unicode characters can also be added to a string by using the \f(CW\*(C`\eN{U+...}\*(C'\fR notation. The Unicode code for the desired character, in hexadecimal, should be placed in the braces, after the \f(CW\*(C`U\*(C'\fR. For instance, a smiley face is \&\f(CW\*(C`\eN{U+263A}\*(C'\fR. .Sp Alternatively, you can use the \f(CW\*(C`\ex{...}\*(C'\fR notation for characters 0x100 and above. For characters below 0x100 you may get byte semantics instead of character semantics; see \*(L"The \*(R"Unicode Bug"". On \s-1EBCDIC\s0 machines there is the additional problem that the value for such characters gives the \s-1EBCDIC\s0 character rather than the Unicode one. .Sp Additionally, if you .Sp .Vb 1 \& use charnames \*(Aq:full\*(Aq; .Ve .Sp you can use the \f(CW\*(C`\eN{...}\*(C'\fR notation and put the official Unicode character name within the braces, such as \f(CW\*(C`\eN{WHITE SMILING FACE}\*(C'\fR. See charnames. .IP "\(bu" 4 If an appropriate encoding is specified, identifiers within the Perl script may contain Unicode alphanumeric characters, including ideographs. Perl does not currently attempt to canonicalize variable names. .IP "\(bu" 4 Regular expressions match characters instead of bytes. \*(L".\*(R" matches a character instead of a byte. .IP "\(bu" 4 Bracketed character classes in regular expressions match characters instead of bytes and match against the character properties specified in the Unicode properties database. \f(CW\*(C`\ew\*(C'\fR can be used to match a Japanese ideograph, for instance. .IP "\(bu" 4 Named Unicode properties, scripts, and block ranges may be used (like bracketed character classes) by using the \f(CW\*(C`\ep{}\*(C'\fR \*(L"matches property\*(R" construct and the \f(CW\*(C`\eP{}\*(C'\fR negation, \*(L"doesn't match property\*(R". See \*(L"Unicode Character Properties\*(R" for more details. .Sp You can define your own character properties and use them in the regular expression with the \f(CW\*(C`\ep{}\*(C'\fR or \f(CW\*(C`\eP{}\*(C'\fR construct. See \*(L"User-Defined Character Properties\*(R" for more details. .IP "\(bu" 4 The special pattern \f(CW\*(C`\eX\*(C'\fR matches a logical character, an \*(L"extended grapheme cluster\*(R" in Standardese. In Unicode what appears to the user to be a single character, for example an accented \f(CW\*(C`G\*(C'\fR, may in fact be composed of a sequence of characters, in this case a \f(CW\*(C`G\*(C'\fR followed by an accent character. \f(CW\*(C`\eX\*(C'\fR will match the entire sequence. .IP "\(bu" 4 The \f(CW\*(C`tr///\*(C'\fR operator translates characters instead of bytes. Note that the \f(CW\*(C`tr///CU\*(C'\fR functionality has been removed. For similar functionality see pack('U0', ...) and pack('C0', ...). .IP "\(bu" 4 Case translation operators use the Unicode case translation tables when character input is provided. Note that \f(CW\*(C`uc()\*(C'\fR, or \f(CW\*(C`\eU\*(C'\fR in interpolated strings, translates to uppercase, while \f(CW\*(C`ucfirst\*(C'\fR, or \f(CW\*(C`\eu\*(C'\fR in interpolated strings, translates to titlecase in languages that make the distinction (which is equivalent to uppercase in languages without the distinction). .IP "\(bu" 4 Most operators that deal with positions or lengths in a string will automatically switch to using character positions, including \&\f(CW\*(C`chop()\*(C'\fR, \f(CW\*(C`chomp()\*(C'\fR, \f(CW\*(C`substr()\*(C'\fR, \f(CW\*(C`pos()\*(C'\fR, \f(CW\*(C`index()\*(C'\fR, \f(CW\*(C`rindex()\*(C'\fR, \&\f(CW\*(C`sprintf()\*(C'\fR, \f(CW\*(C`write()\*(C'\fR, and \f(CW\*(C`length()\*(C'\fR. An operator that specifically does not switch is \f(CW\*(C`vec()\*(C'\fR. Operators that really don't care include operators that treat strings as a bucket of bits such as \&\f(CW\*(C`sort()\*(C'\fR, and operators dealing with filenames. .IP "\(bu" 4 The \f(CW\*(C`pack()\*(C'\fR/\f(CW\*(C`unpack()\*(C'\fR letter \f(CW\*(C`C\*(C'\fR does \fInot\fR change, since it is often used for byte-oriented formats. Again, think \f(CW\*(C`char\*(C'\fR in the C language. .Sp There is a new \f(CW\*(C`U\*(C'\fR specifier that converts between Unicode characters and code points. There is also a \f(CW\*(C`W\*(C'\fR specifier that is the equivalent of \&\f(CW\*(C`chr\*(C'\fR/\f(CW\*(C`ord\*(C'\fR and properly handles character values even if they are above 255. .IP "\(bu" 4 The \f(CW\*(C`chr()\*(C'\fR and \f(CW\*(C`ord()\*(C'\fR functions work on characters, similar to \&\f(CW\*(C`pack("W")\*(C'\fR and \f(CW\*(C`unpack("W")\*(C'\fR, \fInot\fR \f(CW\*(C`pack("C")\*(C'\fR and \&\f(CW\*(C`unpack("C")\*(C'\fR. \f(CW\*(C`pack("C")\*(C'\fR and \f(CW\*(C`unpack("C")\*(C'\fR are methods for emulating byte-oriented \f(CW\*(C`chr()\*(C'\fR and \f(CW\*(C`ord()\*(C'\fR on Unicode strings. While these methods reveal the internal encoding of Unicode strings, that is not something one normally needs to care about at all. .IP "\(bu" 4 The bit string operators, \f(CW\*(C`& | ^ ~\*(C'\fR, can operate on character data. However, for backward compatibility, such as when using bit string operations when characters are all less than 256 in ordinal value, one should not use \f(CW\*(C`~\*(C'\fR (the bit complement) with characters of both values less than 256 and values greater than 256. Most importantly, DeMorgan's laws (\f(CW\*(C`~($x|$y) eq ~$x&~$y\*(C'\fR and \f(CW\*(C`~($x&$y) eq ~$x|~$y\*(C'\fR) will not hold. The reason for this mathematical \fIfaux pas\fR is that the complement cannot return \fBboth\fR the 8\-bit (byte-wide) bit complement \fBand\fR the full character-wide bit complement. .IP "\(bu" 4 You can define your own mappings to be used in \f(CW\*(C`lc()\*(C'\fR, \&\f(CW\*(C`lcfirst()\*(C'\fR, \f(CW\*(C`uc()\*(C'\fR, and \f(CW\*(C`ucfirst()\*(C'\fR (or their double-quoted string inlined versions such as \f(CW\*(C`\eU\*(C'\fR). See User-Defined Case-Mappings for more details. .IP "\(bu" 4 And finally, \f(CW\*(C`scalar reverse()\*(C'\fR reverses by character rather than by byte. .SS "Unicode Character Properties" .IX Subsection "Unicode Character Properties" (The only time that Perl considers a sequence of individual code points as a single logical character is in the \f(CW\*(C`\eX\*(C'\fR construct, already mentioned above. Therefore \*(L"character\*(R" in this discussion means a single Unicode code point.) .PP Very nearly all Unicode character properties are accessible through regular expressions by using the \f(CW\*(C`\ep{}\*(C'\fR \*(L"matches property\*(R" construct and the \f(CW\*(C`\eP{}\*(C'\fR \*(L"doesn't match property\*(R" for its negation. .PP For instance, \f(CW\*(C`\ep{Uppercase}\*(C'\fR matches any single character with the Unicode \&\*(L"Uppercase\*(R" property, while \f(CW\*(C`\ep{L}\*(C'\fR matches any character with a General_Category of \*(L"L\*(R" (letter) property. Brackets are not required for single letter property names, so \f(CW\*(C`\ep{L}\*(C'\fR is equivalent to \f(CW\*(C`\epL\*(C'\fR. .PP More formally, \f(CW\*(C`\ep{Uppercase}\*(C'\fR matches any single character whose Unicode Uppercase property value is True, and \f(CW\*(C`\eP{Uppercase}\*(C'\fR matches any character whose Uppercase property value is False, and they could have been written as \&\f(CW\*(C`\ep{Uppercase=True}\*(C'\fR and \f(CW\*(C`\ep{Uppercase=False}\*(C'\fR, respectively. .PP This formality is needed when properties are not binary; that is, if they can take on more values than just True and False. For example, the Bidi_Class (see \&\*(L"Bidirectional Character Types\*(R" below), can take on several different values, such as Left, Right, Whitespace, and others. To match these, one needs to specify the property name (Bidi_Class), \s-1AND\s0 the value being matched against (Left, Right, etc.). This is done, as in the examples above, by having the two components separated by an equal sign (or interchangeably, a colon), like \&\f(CW\*(C`\ep{Bidi_Class: Left}\*(C'\fR. .PP All Unicode-defined character properties may be written in these compound forms of \f(CW\*(C`\ep{property=value}\*(C'\fR or \f(CW\*(C`\ep{property:value}\*(C'\fR, but Perl provides some additional properties that are written only in the single form, as well as single-form short-cuts for all binary properties and certain others described below, in which you may omit the property name and the equals or colon separator. .PP Most Unicode character properties have at least two synonyms (or aliases if you prefer): a short one that is easier to type and a longer one that is more descriptive and hence easier to understand. Thus the \*(L"L\*(R" and \*(L"Letter\*(R" properties above are equivalent and can be used interchangeably. Likewise, \&\*(L"Upper\*(R" is a synonym for \*(L"Uppercase\*(R", and we could have written \&\f(CW\*(C`\ep{Uppercase}\*(C'\fR equivalently as \f(CW\*(C`\ep{Upper}\*(C'\fR. Also, there are typically various synonyms for the values the property can be. For binary properties, \&\*(L"True\*(R" has 3 synonyms: \*(L"T\*(R", \*(L"Yes\*(R", and \*(L"Y\*(R"; and \*(L"False has correspondingly \*(R"F\*(L", \&\*(R"No\*(L", and \*(R"N\*(L". But be careful. A short form of a value for one property may not mean the same thing as the same short form for another. Thus, for the General_Category property, \*(R"L\*(L" means \*(R"Letter\*(L", but for the Bidi_Class property, \&\*(R"L\*(L" means \*(R"Left". A complete list of properties and synonyms is in perluniprops. .PP Upper/lower case differences in property names and values are irrelevant; thus \f(CW\*(C`\ep{Upper}\*(C'\fR means the same thing as \f(CW\*(C`\ep{upper}\*(C'\fR or even \f(CW\*(C`\ep{UpPeR}\*(C'\fR. Similarly, you can add or subtract underscores anywhere in the middle of a word, so that these are also equivalent to \f(CW\*(C`\ep{U_p_p_e_r}\*(C'\fR. And white space is irrelevant adjacent to non-word characters, such as the braces and the equals or colon separators, so \f(CW\*(C`\ep{ Upper }\*(C'\fR and \f(CW\*(C`\ep{ Upper_case : Y }\*(C'\fR are equivalent to these as well. In fact, white space and even hyphens can usually be added or deleted anywhere. So even \f(CW\*(C`\ep{ Up\-per case = Yes}\*(C'\fR is equivalent. All this is called \*(L"loose-matching\*(R" by Unicode. The few places where stricter matching is used is in the middle of numbers, and in the Perl extension properties that begin or end with an underscore. Stricter matching cares about white space (except adjacent to non-word characters), hyphens, and non-interior underscores. .PP You can also use negation in both \f(CW\*(C`\ep{}\*(C'\fR and \f(CW\*(C`\eP{}\*(C'\fR by introducing a caret (^) between the first brace and the property name: \f(CW\*(C`\ep{^Tamil}\*(C'\fR is equal to \f(CW\*(C`\eP{Tamil}\*(C'\fR. .PP Almost all properties are immune to case-insensitive matching. That is, adding a \f(CW\*(C`/i\*(C'\fR regular expression modifier does not change what they match. There are two sets that are affected. The first set is \&\f(CW\*(C`Uppercase_Letter\*(C'\fR, \&\f(CW\*(C`Lowercase_Letter\*(C'\fR, and \f(CW\*(C`Titlecase_Letter\*(C'\fR, all of which match \f(CW\*(C`Cased_Letter\*(C'\fR under \f(CW\*(C`/i\*(C'\fR matching. And the second set is \&\f(CW\*(C`Uppercase\*(C'\fR, \&\f(CW\*(C`Lowercase\*(C'\fR, and \f(CW\*(C`Titlecase\*(C'\fR, all of which match \f(CW\*(C`Cased\*(C'\fR under \f(CW\*(C`/i\*(C'\fR matching. This set also includes its subsets \f(CW\*(C`PosixUpper\*(C'\fR and \f(CW\*(C`PosixLower\*(C'\fR both of which under \f(CW\*(C`/i\*(C'\fR matching match \f(CW\*(C`PosixAlpha\*(C'\fR. (The difference between these sets is that some things, such as Roman numerals, come in both upper and lower case so they are \f(CW\*(C`Cased\*(C'\fR, but aren't considered letters, so they aren't \f(CW\*(C`Cased_Letter\*(C'\fRs.) .PP \fI\f(BIGeneral_Category\fI\fR .IX Subsection "General_Category" .PP Every Unicode character is assigned a general category, which is the \*(L"most usual categorization of a character\*(R" (from ). .PP The compound way of writing these is like \f(CW\*(C`\ep{General_Category=Number}\*(C'\fR (short, \f(CW\*(C`\ep{gc:n}\*(C'\fR). But Perl furnishes shortcuts in which everything up through the equal or colon separator is omitted. So you can instead just write \&\f(CW\*(C`\epN\*(C'\fR. .PP Here are the short and long forms of the General Category properties: .PP .Vb 1 \& Short Long \& \& L Letter \& LC, L& Cased_Letter (that is: [\ep{Ll}\ep{Lu}\ep{Lt}]) \& Lu Uppercase_Letter \& Ll Lowercase_Letter \& Lt Titlecase_Letter \& Lm Modifier_Letter \& Lo Other_Letter \& \& M Mark \& Mn Nonspacing_Mark \& Mc Spacing_Mark \& Me Enclosing_Mark \& \& N Number \& Nd Decimal_Number (also Digit) \& Nl Letter_Number \& No Other_Number \& \& P Punctuation (also Punct) \& Pc Connector_Punctuation \& Pd Dash_Punctuation \& Ps Open_Punctuation \& Pe Close_Punctuation \& Pi Initial_Punctuation \& (may behave like Ps or Pe depending on usage) \& Pf Final_Punctuation \& (may behave like Ps or Pe depending on usage) \& Po Other_Punctuation \& \& S Symbol \& Sm Math_Symbol \& Sc Currency_Symbol \& Sk Modifier_Symbol \& So Other_Symbol \& \& Z Separator \& Zs Space_Separator \& Zl Line_Separator \& Zp Paragraph_Separator \& \& C Other \& Cc Control (also Cntrl) \& Cf Format \& Cs Surrogate \& Co Private_Use \& Cn Unassigned .Ve .PP Single-letter properties match all characters in any of the two-letter sub-properties starting with the same letter. \&\f(CW\*(C`LC\*(C'\fR and \f(CW\*(C`L&\*(C'\fR are special: both are aliases for the set consisting of everything matched by \f(CW\*(C`Ll\*(C'\fR, \f(CW\*(C`Lu\*(C'\fR, and \f(CW\*(C`Lt\*(C'\fR. .PP \fI\f(BIBidirectional Character Types\fI\fR .IX Subsection "Bidirectional Character Types" .PP Because scripts differ in their directionality (Hebrew and Arabic are written right to left, for example) Unicode supplies these properties in the Bidi_Class class: .PP .Vb 1 \& Property Meaning \& \& L Left\-to\-Right \& LRE Left\-to\-Right Embedding \& LRO Left\-to\-Right Override \& R Right\-to\-Left \& AL Arabic Letter \& RLE Right\-to\-Left Embedding \& RLO Right\-to\-Left Override \& PDF Pop Directional Format \& EN European Number \& ES European Separator \& ET European Terminator \& AN Arabic Number \& CS Common Separator \& NSM Non\-Spacing Mark \& BN Boundary Neutral \& B Paragraph Separator \& S Segment Separator \& WS Whitespace \& ON Other Neutrals .Ve .PP This property is always written in the compound form. For example, \f(CW\*(C`\ep{Bidi_Class:R}\*(C'\fR matches characters that are normally written right to left. .PP \fI\f(BIScripts\fI\fR .IX Subsection "Scripts" .PP The world's languages are written in many different scripts. This sentence (unless you're reading it in translation) is written in Latin, while Russian is written in Cyrillic, and Greek is written in, well, Greek; Japanese mainly in Hiragana or Katakana. There are many more. .PP The Unicode Script property gives what script a given character is in, and the property can be specified with the compound form like \&\f(CW\*(C`\ep{Script=Hebrew}\*(C'\fR (short: \f(CW\*(C`\ep{sc=hebr}\*(C'\fR). Perl furnishes shortcuts for all script names. You can omit everything up through the equals (or colon), and simply write \f(CW\*(C`\ep{Latin}\*(C'\fR or \f(CW\*(C`\eP{Cyrillic}\*(C'\fR. .PP A complete list of scripts and their shortcuts is in perluniprops. .PP \fI\f(BIUse of \*(L"Is\*(R" Prefix\fI\fR .IX Subsection "Use of Is Prefix" .PP For backward compatibility (with Perl 5.6), all properties mentioned so far may have \f(CW\*(C`Is\*(C'\fR or \f(CW\*(C`Is_\*(C'\fR prepended to their name, so \f(CW\*(C`\eP{Is_Lu}\*(C'\fR, for example, is equal to \f(CW\*(C`\eP{Lu}\*(C'\fR, and \f(CW\*(C`\ep{IsScript:Arabic}\*(C'\fR is equal to \&\f(CW\*(C`\ep{Arabic}\*(C'\fR. .PP \fI\f(BIBlocks\fI\fR .IX Subsection "Blocks" .PP In addition to \fBscripts\fR, Unicode also defines \fBblocks\fR of characters. The difference between scripts and blocks is that the concept of scripts is closer to natural languages, while the concept of blocks is more of an artificial grouping based on groups of Unicode characters with consecutive ordinal values. For example, the \*(L"Basic Latin\*(R" block is all characters whose ordinals are between 0 and 127, inclusive; in other words, the \s-1ASCII\s0 characters. The \*(L"Latin\*(R" script contains some letters from this as well as several other blocks, like \*(L"Latin\-1 Supplement\*(R", \&\*(L"Latin Extended-A\*(R", etc., but it does not contain all the characters from those blocks. It does not, for example, contain the digits 0\-9, because those digits are shared across many scripts. The digits 0\-9 and similar groups, like punctuation, are in the script called \f(CW\*(C`Common\*(C'\fR. There is also a script called \f(CW\*(C`Inherited\*(C'\fR for characters that modify other characters, and inherit the script value of the controlling character. (Note that there are several different sets of digits in Unicode that are equivalent to 0\-9 and are matchable by \f(CW\*(C`\ed\*(C'\fR in a regular expression. If they are used in a single language only, they are in that language's script. Only sets are used across several languages are in the \&\f(CW\*(C`Common\*(C'\fR script.) .PP For more about scripts versus blocks, see UAX#24 \*(L"Unicode Script Property\*(R": .PP The Script property is likely to be the one you want to use when processing natural language; the Block property may occasionally be useful in working with the nuts and bolts of Unicode. .PP Block names are matched in the compound form, like \f(CW\*(C`\ep{Block: Arrows}\*(C'\fR or \&\f(CW\*(C`\ep{Blk=Hebrew}\*(C'\fR. Unlike most other properties, only a few block names have a Unicode-defined short name. But Perl does provide a (slight) shortcut: You can say, for example \f(CW\*(C`\ep{In_Arrows}\*(C'\fR or \f(CW\*(C`\ep{In_Hebrew}\*(C'\fR. For backwards compatibility, the \f(CW\*(C`In\*(C'\fR prefix may be omitted if there is no naming conflict with a script or any other property, and you can even use an \f(CW\*(C`Is\*(C'\fR prefix instead in those cases. But it is not a good idea to do this, for a couple reasons: .IP "1." 4 It is confusing. There are many naming conflicts, and you may forget some. For example, \f(CW\*(C`\ep{Hebrew}\*(C'\fR means the \fIscript\fR Hebrew, and \s-1NOT\s0 the \fIblock\fR Hebrew. But would you remember that 6 months from now? .IP "2." 4 It is unstable. A new version of Unicode may pre-empt the current meaning by creating a property with the same name. There was a time in very early Unicode releases when \f(CW\*(C`\ep{Hebrew}\*(C'\fR would have matched the \fIblock\fR Hebrew; now it doesn't. .PP Some people prefer to always use \f(CW\*(C`\ep{Block: foo}\*(C'\fR and \f(CW\*(C`\ep{Script: bar}\*(C'\fR instead of the shortcuts, whether for clarity, because they can't remember the difference between 'In' and 'Is' anyway, or they aren't confident that those who eventually will read their code will know that difference. .PP A complete list of blocks and their shortcuts is in perluniprops. .PP \fI\f(BIOther Properties\fI\fR .IX Subsection "Other Properties" .PP There are many more properties than the very basic ones described here. A complete list is in perluniprops. .PP Unicode defines all its properties in the compound form, so all single-form properties are Perl extensions. Most of these are just synonyms for the Unicode ones, but some are genuine extensions, including several that are in the compound form. And quite a few of these are actually recommended by Unicode (in ). .PP This section gives some details on all extensions that aren't synonyms for compound-form Unicode properties (for those, you'll have to refer to the Unicode Standard . .ie n .IP "\fB\fB""\ep{All}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{All}\fB\fR" 4 .IX Item "p{All}" This matches any of the 1_114_112 Unicode code points. It is a synonym for \&\f(CW\*(C`\ep{Any}\*(C'\fR. .ie n .IP "\fB\fB""\ep{Alnum}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{Alnum}\fB\fR" 4 .IX Item "p{Alnum}" This matches any \f(CW\*(C`\ep{Alphabetic}\*(C'\fR or \f(CW\*(C`\ep{Decimal_Number}\*(C'\fR character. .ie n .IP "\fB\fB""\ep{Any}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{Any}\fB\fR" 4 .IX Item "p{Any}" This matches any of the 1_114_112 Unicode code points. It is a synonym for \&\f(CW\*(C`\ep{All}\*(C'\fR. .ie n .IP "\fB\fB""\ep{ASCII}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{ASCII}\fB\fR" 4 .IX Item "p{ASCII}" This matches any of the 128 characters in the US-ASCII character set, which is a subset of Unicode. .ie n .IP "\fB\fB""\ep{Assigned}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{Assigned}\fB\fR" 4 .IX Item "p{Assigned}" This matches any assigned code point; that is, any code point whose general category is not Unassigned (or equivalently, not Cn). .ie n .IP "\fB\fB""\ep{Blank}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{Blank}\fB\fR" 4 .IX Item "p{Blank}" This is the same as \f(CW\*(C`\eh\*(C'\fR and \f(CW\*(C`\ep{HorizSpace}\*(C'\fR: A character that changes the spacing horizontally. .ie n .IP "\fB\fB""\ep{Decomposition_Type: Non_Canonical}""\fB\fR (Short: ""\ep{Dt=NonCanon}"")" 4 .el .IP "\fB\f(CB\ep{Decomposition_Type: Non_Canonical}\fB\fR (Short: \f(CW\ep{Dt=NonCanon}\fR)" 4 .IX Item "p{Decomposition_Type: Non_Canonical} (Short: p{Dt=NonCanon})" Matches a character that has a non-canonical decomposition. .Sp To understand the use of this rarely used property=value combination, it is necessary to know some basics about decomposition. Consider a character, say H. It could appear with various marks around it, such as an acute accent, or a circumflex, or various hooks, circles, arrows, \&\fIetc.\fR, above, below, to one side or the other, etc. There are many possibilities among the world's languages. The number of combinations is astronomical, and if there were a character for each combination, it would soon exhaust Unicode's more than a million possible characters. So Unicode took a different approach: there is a character for the base H, and a character for each of the possible marks, and these can be variously combined to get a final logical character. So a logical character\*(--what appears to be a single character\*(--can be a sequence of more than one individual characters. This is called an \*(L"extended grapheme cluster\*(R"; Perl furnishes the \f(CW\*(C`\eX\*(C'\fR regular expression construct to match such sequences. .Sp But Unicode's intent is to unify the existing character set standards and practices, and several pre-existing standards have single characters that mean the same thing as some of these combinations. An example is \s-1ISO\-8859\-1\s0, which has quite a few of these in the Latin\-1 range, an example being \*(L"\s-1LATIN\s0 \&\s-1CAPITAL\s0 \s-1LETTER\s0 E \s-1WITH\s0 \s-1ACUTE\s0\*(R". Because this character was in this pre-existing standard, Unicode added it to its repertoire. But this character is considered by Unicode to be equivalent to the sequence consisting of the character \&\*(L"\s-1LATIN\s0 \s-1CAPITAL\s0 \s-1LETTER\s0 E\*(R" followed by the character \*(L"\s-1COMBINING\s0 \s-1ACUTE\s0 \s-1ACCENT\s0\*(R". .Sp \&\*(L"\s-1LATIN\s0 \s-1CAPITAL\s0 \s-1LETTER\s0 E \s-1WITH\s0 \s-1ACUTE\s0\*(R" is called a \*(L"pre-composed\*(R" character, and its equivalence with the sequence is called canonical equivalence. All pre-composed characters are said to have a decomposition (into the equivalent sequence), and the decomposition type is also called canonical. .Sp However, many more characters have a different type of decomposition, a \&\*(L"compatible\*(R" or \*(L"non-canonical\*(R" decomposition. The sequences that form these decompositions are not considered canonically equivalent to the pre-composed character. An example, again in the Latin\-1 range, is the \*(L"\s-1SUPERSCRIPT\s0 \s-1ONE\s0\*(R". It is somewhat like a regular digit 1, but not exactly; its decomposition into the digit 1 is called a \*(L"compatible\*(R" decomposition, specifically a \&\*(L"super\*(R" decomposition. There are several such compatibility decompositions (see ), including one called \*(L"compat\*(R", which means some miscellaneous type of decomposition that doesn't fit into the decomposition categories that Unicode has chosen. .Sp Note that most Unicode characters don't have a decomposition, so their decomposition type is \*(L"None\*(R". .Sp For your convenience, Perl has added the \f(CW\*(C`Non_Canonical\*(C'\fR decomposition type to mean any of the several compatibility decompositions. .ie n .IP "\fB\fB""\ep{Graph}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{Graph}\fB\fR" 4 .IX Item "p{Graph}" Matches any character that is graphic. Theoretically, this means a character that on a printer would cause ink to be used. .ie n .IP "\fB\fB""\ep{HorizSpace}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{HorizSpace}\fB\fR" 4 .IX Item "p{HorizSpace}" This is the same as \f(CW\*(C`\eh\*(C'\fR and \f(CW\*(C`\ep{Blank}\*(C'\fR: a character that changes the spacing horizontally. .ie n .IP "\fB\fB""\ep{In=*}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{In=*}\fB\fR" 4 .IX Item "p{In=*}" This is a synonym for \f(CW\*(C`\ep{Present_In=*}\*(C'\fR .ie n .IP "\fB\fB""\ep{PerlSpace}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{PerlSpace}\fB\fR" 4 .IX Item "p{PerlSpace}" This is the same as \f(CW\*(C`\es\*(C'\fR, restricted to \s-1ASCII\s0, namely \f(CW\*(C`[\ \ef\en\er\et]\*(C'\fR. .Sp Mnemonic: Perl's (original) space .ie n .IP "\fB\fB""\ep{PerlWord}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{PerlWord}\fB\fR" 4 .IX Item "p{PerlWord}" This is the same as \f(CW\*(C`\ew\*(C'\fR, restricted to \s-1ASCII\s0, namely \f(CW\*(C`[A\-Za\-z0\-9_]\*(C'\fR .Sp Mnemonic: Perl's (original) word. .ie n .IP "\fB\fB""\ep{Posix...}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{Posix...}\fB\fR" 4 .IX Item "p{Posix...}" There are several of these, which are equivalents using the \f(CW\*(C`\ep\*(C'\fR notation for Posix classes and are described in \&\*(L"\s-1POSIX\s0 Character Classes\*(R" in perlrecharclass. .ie n .IP "\fB\fB""\ep{Present_In: *}""\fB\fR (Short: ""\ep{In=*}"")" 4 .el .IP "\fB\f(CB\ep{Present_In: *}\fB\fR (Short: \f(CW\ep{In=*}\fR)" 4 .IX Item "p{Present_In: *} (Short: p{In=*})" This property is used when you need to know in what Unicode version(s) a character is. .Sp The \*(L"*\*(R" above stands for some two digit Unicode version number, such as \&\f(CW1.1\fR or \f(CW4.0\fR; or the \*(L"*\*(R" can also be \f(CW\*(C`Unassigned\*(C'\fR. This property will match the code points whose final disposition has been settled as of the Unicode release given by the version number; \f(CW\*(C`\ep{Present_In: Unassigned}\*(C'\fR will match those code points whose meaning has yet to be assigned. .Sp For example, \f(CW\*(C`U+0041\*(C'\fR \*(L"\s-1LATIN\s0 \s-1CAPITAL\s0 \s-1LETTER\s0 A\*(R" was present in the very first Unicode release available, which is \f(CW1.1\fR, so this property is true for all valid \*(L"*\*(R" versions. On the other hand, \f(CW\*(C`U+1EFF\*(C'\fR was not assigned until version 5.1 when it became \*(L"\s-1LATIN\s0 \s-1SMALL\s0 \s-1LETTER\s0 Y \s-1WITH\s0 \s-1LOOP\s0\*(R", so the only \*(L"*\*(R" that would match it are 5.1, 5.2, and later. .Sp Unicode furnishes the \f(CW\*(C`Age\*(C'\fR property from which this is derived. The problem with Age is that a strict interpretation of it (which Perl takes) has it matching the precise release a code point's meaning is introduced in. Thus \&\f(CW\*(C`U+0041\*(C'\fR would match only 1.1; and \f(CW\*(C`U+1EFF\*(C'\fR only 5.1. This is not usually what you want. .Sp Some non-Perl implementations of the Age property may change its meaning to be the same as the Perl Present_In property; just be aware of that. .Sp Another confusion with both these properties is that the definition is not that the code point has been \fIassigned\fR, but that the meaning of the code point has been \fIdetermined\fR. This is because 66 code points will always be unassigned, and so the Age for them is the Unicode version in which the decision to make them so was made. For example, \f(CW\*(C`U+FDD0\*(C'\fR is to be permanently unassigned to a character, and the decision to do that was made in version 3.1, so \f(CW\*(C`\ep{Age=3.1}\*(C'\fR matches this character, as also does \f(CW\*(C`\ep{Present_In: 3.1}\*(C'\fR and up. .ie n .IP "\fB\fB""\ep{Print}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{Print}\fB\fR" 4 .IX Item "p{Print}" This matches any character that is graphical or blank, except controls. .ie n .IP "\fB\fB""\ep{SpacePerl}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{SpacePerl}\fB\fR" 4 .IX Item "p{SpacePerl}" This is the same as \f(CW\*(C`\es\*(C'\fR, including beyond \s-1ASCII\s0. .Sp Mnemonic: Space, as modified by Perl. (It doesn't include the vertical tab which both the Posix standard and Unicode consider white space.) .ie n .IP "\fB\fB""\ep{VertSpace}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{VertSpace}\fB\fR" 4 .IX Item "p{VertSpace}" This is the same as \f(CW\*(C`\ev\*(C'\fR: A character that changes the spacing vertically. .ie n .IP "\fB\fB""\ep{Word}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{Word}\fB\fR" 4 .IX Item "p{Word}" This is the same as \f(CW\*(C`\ew\*(C'\fR, including over 100_000 characters beyond \s-1ASCII\s0. .ie n .IP "\fB\fB""\ep{XPosix...}""\fB\fR" 4 .el .IP "\fB\f(CB\ep{XPosix...}\fB\fR" 4 .IX Item "p{XPosix...}" There are several of these, which are the standard Posix classes extended to the full Unicode range. They are described in \&\*(L"\s-1POSIX\s0 Character Classes\*(R" in perlrecharclass. .SS "User-Defined Character Properties" .IX Subsection "User-Defined Character Properties" You can define your own binary character properties by defining subroutines whose names begin with \*(L"In\*(R" or \*(L"Is\*(R". The subroutines can be defined in any package. The user-defined properties can be used in the regular expression \&\f(CW\*(C`\ep\*(C'\fR and \f(CW\*(C`\eP\*(C'\fR constructs; if you are using a user-defined property from a package other than the one you are in, you must specify its package in the \&\f(CW\*(C`\ep\*(C'\fR or \f(CW\*(C`\eP\*(C'\fR construct. .PP .Vb 3 \& # assuming property Is_Foreign defined in Lang:: \& package main; # property package name required \& if ($txt =~ /\ep{Lang::IsForeign}+/) { ... } \& \& package Lang; # property package name not required \& if ($txt =~ /\ep{IsForeign}+/) { ... } .Ve .PP Note that the effect is compile-time and immutable once defined. However, the subroutines are passed a single parameter, which is 0 if case-sensitive matching is in effect and non-zero if caseless matching is in effect. The subroutine may return different values depending on the value of the flag, and one set of values will immutably be in effect for all case-sensitive matches, and the other set for all case-insensitive matches. .PP Note that if the regular expression is tainted, then Perl will die rather than calling the subroutine, where the name of the subroutine is determined by the tainted data. .PP The subroutines must return a specially-formatted string, with one or more newline-separated lines. Each line must be one of the following: .IP "\(bu" 4 A single hexadecimal number denoting a Unicode code point to include. .IP "\(bu" 4 Two hexadecimal numbers separated by horizontal whitespace (space or tabular characters) denoting a range of Unicode code points to include. .IP "\(bu" 4 Something to include, prefixed by \*(L"+\*(R": a built-in character property (prefixed by \*(L"utf8::\*(R") or a user-defined character property, to represent all the characters in that property; two hexadecimal code points for a range; or a single hexadecimal code point. .IP "\(bu" 4 Something to exclude, prefixed by \*(L"\-\*(R": an existing character property (prefixed by \*(L"utf8::\*(R") or a user-defined character property, to represent all the characters in that property; two hexadecimal code points for a range; or a single hexadecimal code point. .IP "\(bu" 4 Something to negate, prefixed \*(L"!\*(R": an existing character property (prefixed by \*(L"utf8::\*(R") or a user-defined character property, to represent all the characters in that property; two hexadecimal code points for a range; or a single hexadecimal code point. .IP "\(bu" 4 Something to intersect with, prefixed by \*(L"&\*(R": an existing character property (prefixed by \*(L"utf8::\*(R") or a user-defined character property, for all the characters except the characters in the property; two hexadecimal code points for a range; or a single hexadecimal code point. .PP For example, to define a property that covers both the Japanese syllabaries (hiragana and katakana), you can define .PP .Vb 6 \& sub InKana { \& return <) \& [4] \ed \eD \es \eS \ew \eW \eX [:prop:] [:^prop:] \& [5] can use regular expression look\-ahead [a] or \& user\-defined character properties [b] to emulate set \& operations \& [6] \eb \eB \& [7] note that Perl does Full case\-folding in matching (but with \& bugs), not Simple: for example U+1F88 is equivalent to \& U+1F00 U+03B9, not with 1F80. This difference matters \& mainly for certain Greek capital letters with certain \& modifiers: the Full case\-folding decomposes the letter, \& while the Simple case\-folding would map it to a single \& character. \& [8] should do ^ and $ also on U+000B (\ev in C), FF (\ef), CR \& (\er), CRLF (\er\en), NEL (U+0085), LS (U+2028), and PS \& (U+2029); should also affect <>, $., and script line \& numbers; should not split lines within CRLF [c] (i.e. there \& is no empty line between \er and \en) \& [9] Linebreaking conformant with UAX#14 "Unicode Line Breaking \& Algorithm" is available through the Unicode::LineBreaking \& module. \& [10] UTF\-8/UTF\-EBDDIC used in Perl allows not only U+10000 to \& U+10FFFF but also beyond U+10FFFF .Ve .Sp [a] You can mimic class subtraction using lookahead. For example, what UTS#18 might write as .Sp .Vb 1 \& [{Greek}\-[{UNASSIGNED}]] .Ve .Sp in Perl can be written as: .Sp .Vb 2 \& (?!\ep{Unassigned})\ep{InGreekAndCoptic} \& (?=\ep{Assigned})\ep{InGreekAndCoptic} .Ve .Sp But in this particular example, you probably really want .Sp .Vb 1 \& \ep{GreekAndCoptic} .Ve .Sp which will match assigned characters known to be part of the Greek script. .Sp Also see the Unicode::Regex::Set module, it does implement the full UTS#18 grouping, intersection, union, and removal (subtraction) syntax. .Sp [b] '+' for union, '\-' for removal (set-difference), '&' for intersection (see \*(L"User-Defined Character Properties\*(R") .Sp [c] Try the \f(CW\*(C`:crlf\*(C'\fR layer (see PerlIO). .IP "\(bu" 4 Level 2 \- Extended Unicode Support .Sp .Vb 6 \& RL2.1 Canonical Equivalents \- MISSING [10][11] \& RL2.2 Default Grapheme Clusters \- MISSING [12] \& RL2.3 Default Word Boundaries \- MISSING [14] \& RL2.4 Default Loose Matches \- MISSING [15] \& RL2.5 Name Properties \- MISSING [16] \& RL2.6 Wildcard Properties \- MISSING \& \& [10] see UAX#15 "Unicode Normalization Forms" \& [11] have Unicode::Normalize but not integrated to regexes \& [12] have \eX but we don\*(Aqt have a "Grapheme Cluster Mode" \& [14] see UAX#29, Word Boundaries \& [15] see UAX#21 "Case Mappings" \& [16] missing loose match [e] .Ve .Sp [e] \f(CW\*(C`\eN{...}\*(C'\fR allows namespaces (see charnames). .IP "\(bu" 4 Level 3 \- Tailored Support .Sp .Vb 11 \& RL3.1 Tailored Punctuation \- MISSING \& RL3.2 Tailored Grapheme Clusters \- MISSING [17][18] \& RL3.3 Tailored Word Boundaries \- MISSING \& RL3.4 Tailored Loose Matches \- MISSING \& RL3.5 Tailored Ranges \- MISSING \& RL3.6 Context Matching \- MISSING [19] \& RL3.7 Incremental Matches \- MISSING \& ( RL3.8 Unicode Set Sharing ) \& RL3.9 Possible Match Sets \- MISSING \& RL3.10 Folded Matching \- MISSING [20] \& RL3.11 Submatchers \- MISSING \& \& [17] see UAX#10 "Unicode Collation Algorithms" \& [18] have Unicode::Collate but not integrated to regexes \& [19] have (?<=x) and (?=x), but look\-aheads or look\-behinds \& should see outside of the target substring \& [20] need insensitive matching for linguistic features other \& than case; for example, hiragana to katakana, wide and \& narrow, simplified Han to traditional Han (see UTR#30 \& "Character Foldings") .Ve .SS "Unicode Encodings" .IX Subsection "Unicode Encodings" Unicode characters are assigned to \fIcode points\fR, which are abstract numbers. To use these numbers, various encodings are needed. .IP "\(bu" 4 \&\s-1UTF\-8\s0 .Sp \&\s-1UTF\-8\s0 is a variable-length (1 to 4 bytes), byte-order independent encoding. For \s-1ASCII\s0 (and we really do mean 7\-bit \s-1ASCII\s0, not another 8\-bit encoding), \s-1UTF\-8\s0 is transparent. .Sp The following table is from Unicode 3.2. .Sp .Vb 1 \& Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte \& \& U+0000..U+007F 00..7F \& U+0080..U+07FF * C2..DF 80..BF \& U+0800..U+0FFF E0 * A0..BF 80..BF \& U+1000..U+CFFF E1..EC 80..BF 80..BF \& U+D000..U+D7FF ED 80..9F 80..BF \& U+D800..U+DFFF +++++++ utf16 surrogates, not legal utf8 +++++++ \& U+E000..U+FFFF EE..EF 80..BF 80..BF \& U+10000..U+3FFFF F0 * 90..BF 80..BF 80..BF \& U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF \& U+100000..U+10FFFF F4 80..8F 80..BF 80..BF .Ve .Sp Note the gaps marked by \*(L"*\*(R" before several of the byte entries above. These are caused by legal \s-1UTF\-8\s0 avoiding non-shortest encodings: it is technically possible to UTF\-8\-encode a single code point in different ways, but that is explicitly forbidden, and the shortest possible encoding should always be used (and that is what Perl does). .Sp Another way to look at it is via bits: .Sp .Vb 1 \& Code Points 1st Byte 2nd Byte 3rd Byte 4th Byte \& \& 0aaaaaaa 0aaaaaaa \& 00000bbbbbaaaaaa 110bbbbb 10aaaaaa \& ccccbbbbbbaaaaaa 1110cccc 10bbbbbb 10aaaaaa \& 00000dddccccccbbbbbbaaaaaa 11110ddd 10cccccc 10bbbbbb 10aaaaaa .Ve .Sp As you can see, the continuation bytes all begin with \*(L"10\*(R", and the leading bits of the start byte tell how many bytes there are in the encoded character. .Sp The original \s-1UTF\-8\s0 specification allowed up to 6 bytes, to allow encoding of numbers up to 0x7FFF_FFFF. Perl continues to allow those, and has extended that up to 13 bytes to encode code points up to what can fit in a 64\-bit word. However, Perl will warn if you output any of these as being non-portable; and under strict \s-1UTF\-8\s0 input protocols, they are forbidden. .Sp The Unicode non-character code points are also disallowed in \s-1UTF\-8\s0 in \&\*(L"open interchange\*(R". See \*(L"Non-character code points\*(R". .IP "\(bu" 4 UTF-EBCDIC .Sp Like \s-1UTF\-8\s0 but EBCDIC-safe, in the way that \s-1UTF\-8\s0 is ASCII-safe. .IP "\(bu" 4 \&\s-1UTF\-16\s0, \s-1UTF\-16BE\s0, \s-1UTF\-16LE\s0, Surrogates, and BOMs (Byte Order Marks) .Sp The followings items are mostly for reference and general Unicode knowledge, Perl doesn't use these constructs internally. .Sp Like \s-1UTF\-8\s0, \s-1UTF\-16\s0 is a variable-width encoding, but where \&\s-1UTF\-8\s0 uses 8\-bit code units, \s-1UTF\-16\s0 uses 16\-bit code units. All code points occupy either 2 or 4 bytes in \s-1UTF\-16:\s0 code points \&\f(CW\*(C`U+0000..U+FFFF\*(C'\fR are stored in a single 16\-bit unit, and code points \f(CW\*(C`U+10000..U+10FFFF\*(C'\fR in two 16\-bit units. The latter case is using \fIsurrogates\fR, the first 16\-bit unit being the \fIhigh surrogate\fR, and the second being the \fIlow surrogate\fR. .Sp Surrogates are code points set aside to encode the \f(CW\*(C`U+10000..U+10FFFF\*(C'\fR range of Unicode code points in pairs of 16\-bit units. The \fIhigh surrogates\fR are the range \f(CW\*(C`U+D800..U+DBFF\*(C'\fR and the \fIlow surrogates\fR are the range \f(CW\*(C`U+DC00..U+DFFF\*(C'\fR. The surrogate encoding is .Sp .Vb 2 \& $hi = ($uni \- 0x10000) / 0x400 + 0xD800; \& $lo = ($uni \- 0x10000) % 0x400 + 0xDC00; .Ve .Sp and the decoding is .Sp .Vb 1 \& $uni = 0x10000 + ($hi \- 0xD800) * 0x400 + ($lo \- 0xDC00); .Ve .Sp Because of the 16\-bitness, \s-1UTF\-16\s0 is byte-order dependent. \s-1UTF\-16\s0 itself can be used for in-memory computations, but if storage or transfer is required either \s-1UTF\-16BE\s0 (big-endian) or \s-1UTF\-16LE\s0 (little-endian) encodings must be chosen. .Sp This introduces another problem: what if you just know that your data is \s-1UTF\-16\s0, but you don't know which endianness? Byte Order Marks, or BOMs, are a solution to this. A special character has been reserved in Unicode to function as a byte order marker: the character with the code point \f(CW\*(C`U+FEFF\*(C'\fR is the \s-1BOM\s0. .Sp The trick is that if you read a \s-1BOM\s0, you will know the byte order, since if it was written on a big-endian platform, you will read the bytes \f(CW\*(C`0xFE 0xFF\*(C'\fR, but if it was written on a little-endian platform, you will read the bytes \f(CW\*(C`0xFF 0xFE\*(C'\fR. (And if the originating platform was writing in \s-1UTF\-8\s0, you will read the bytes \f(CW\*(C`0xEF 0xBB 0xBF\*(C'\fR.) .Sp The way this trick works is that the character with the code point \&\f(CW\*(C`U+FFFE\*(C'\fR is not supposed to be in input streams, so the sequence of bytes \f(CW\*(C`0xFF 0xFE\*(C'\fR is unambiguously \*(L"\s-1BOM\s0, represented in little-endian format\*(R" and cannot be \f(CW\*(C`U+FFFE\*(C'\fR, represented in big-endian format". .Sp Surrogates have no meaning in Unicode outside their use in pairs to represent other code points. However, Perl allows them to be represented individually internally, for example by saying \&\f(CW\*(C`chr(0xD801)\*(C'\fR, so that all code points, not just those valid for open interchange, are representable. Unicode does define semantics for them, such as their General Category is \*(L"Cs\*(R". But because their use is somewhat dangerous, Perl will warn (using the warning category \*(L"surrogate\*(R", which is a sub-category of \*(L"utf8\*(R") if an attempt is made to do things like take the lower case of one, or match case-insensitively, or to output them. (But don't try this on Perls before 5.14.) .IP "\(bu" 4 \&\s-1UTF\-32\s0, \s-1UTF\-32BE\s0, \s-1UTF\-32LE\s0 .Sp The \s-1UTF\-32\s0 family is pretty much like the \s-1UTF\-16\s0 family, expect that the units are 32\-bit, and therefore the surrogate scheme is not needed. \s-1UTF\-32\s0 is a fixed-width encoding. The \s-1BOM\s0 signatures are \&\f(CW\*(C`0x00 0x00 0xFE 0xFF\*(C'\fR for \s-1BE\s0 and \f(CW\*(C`0xFF 0xFE 0x00 0x00\*(C'\fR for \s-1LE\s0. .IP "\(bu" 4 \&\s-1UCS\-2\s0, \s-1UCS\-4\s0 .Sp Legacy, fixed-width encodings defined by the \s-1ISO\s0 10646 standard. \s-1UCS\-2\s0 is a 16\-bit encoding. Unlike \s-1UTF\-16\s0, \s-1UCS\-2\s0 is not extensible beyond \f(CW\*(C`U+FFFF\*(C'\fR, because it does not use surrogates. \s-1UCS\-4\s0 is a 32\-bit encoding, functionally identical to \s-1UTF\-32\s0 (the difference being that \&\s-1UCS\-4\s0 forbids neither surrogates nor code points larger than 0x10_FFFF). .IP "\(bu" 4 \&\s-1UTF\-7\s0 .Sp A seven-bit safe (non-eight-bit) encoding, which is useful if the transport or storage is not eight-bit safe. Defined by \s-1RFC\s0 2152. .SS "Non-character code points" .IX Subsection "Non-character code points" 66 code points are set aside in Unicode as \*(L"non-character code points\*(R". These all have the Unassigned (Cn) General Category, and they never will be assigned. These are never supposed to be in legal Unicode input streams, so that code can use them as sentinels that can be mixed in with character data, and they always will be distinguishable from that data. To keep them out of Perl input streams, strict \s-1UTF\-8\s0 should be specified, such as by using the layer \f(CW\*(C`:encoding(\*(AqUTF\-8\*(Aq)\*(C'\fR. The non-character code points are the 32 between U+FDD0 and U+FDEF, and the 34 code points U+FFFE, U+FFFF, U+1FFFE, U+1FFFF, ... U+10FFFE, U+10FFFF. Some people are under the mistaken impression that these are \*(L"illegal\*(R", but that is not true. An application or cooperating set of applications can legally use them at will internally; but these code points are \&\*(L"illegal for open interchange\*(R". Therefore, Perl will not accept these from input streams unless lax rules are being used, and will warn (using the warning category \*(L"nonchar\*(R", which is a sub-category of \*(L"utf8\*(R") if an attempt is made to output them. .SS "Beyond Unicode code points" .IX Subsection "Beyond Unicode code points" The maximum Unicode code point is U+10FFFF. But Perl accepts code points up to the maximum permissible unsigned number available on the platform. However, Perl will not accept these from input streams unless lax rules are being used, and will warn (using the warning category \&\*(L"non_unicode\*(R", which is a sub-category of \*(L"utf8\*(R") if an attempt is made to operate on or output them. For example, \f(CW\*(C`uc(0x11_0000)\*(C'\fR will generate this warning, returning the input parameter as its result, as the upper case of every non-Unicode code point is the code point itself. .SS "Security Implications of Unicode" .IX Subsection "Security Implications of Unicode" Read Unicode Security Considerations . Also, note the following: .IP "\(bu" 4 Malformed \s-1UTF\-8\s0 .Sp Unfortunately, the original specification of \s-1UTF\-8\s0 leaves some room for interpretation of how many bytes of encoded output one should generate from one input Unicode character. Strictly speaking, the shortest possible sequence of \s-1UTF\-8\s0 bytes should be generated, because otherwise there is potential for an input buffer overflow at the receiving end of a \s-1UTF\-8\s0 connection. Perl always generates the shortest length \s-1UTF\-8\s0, and with warnings on, Perl will warn about non-shortest length \s-1UTF\-8\s0 along with other malformations, such as the surrogates, which are not Unicode code points valid for interchange. .IP "\(bu" 4 Regular expression pattern matching may surprise you if you're not accustomed to Unicode. Starting in Perl 5.14, several pattern modifiers are available to control this, called the character set modifiers. Details are given in \*(L"Character set modifiers\*(R" in perlre. .PP As discussed elsewhere, Perl has one foot (two hooves?) planted in each of two worlds: the old world of bytes and the new world of characters, upgrading from bytes to characters when necessary. If your legacy code does not explicitly use Unicode, no automatic switch-over to characters should happen. Characters shouldn't get downgraded to bytes, either. It is possible to accidentally mix bytes and characters, however (see perluniintro), in which case \f(CW\*(C`\ew\*(C'\fR in regular expressions might start behaving differently (unless the \f(CW\*(C`/a\*(C'\fR modifier is in effect). Review your code. Use warnings and the \f(CW\*(C`strict\*(C'\fR pragma. .SS "Unicode in Perl on \s-1EBCDIC\s0" .IX Subsection "Unicode in Perl on EBCDIC" The way Unicode is handled on \s-1EBCDIC\s0 platforms is still experimental. On such platforms, references to \s-1UTF\-8\s0 encoding in this document and elsewhere should be read as meaning the UTF-EBCDIC specified in Unicode Technical Report 16, unless \s-1ASCII\s0 vs. \s-1EBCDIC\s0 issues are specifically discussed. There is no \f(CW\*(C`utfebcdic\*(C'\fR pragma or \&\*(L":utfebcdic\*(R" layer; rather, \*(L"utf8\*(R" and \*(L":utf8\*(R" are reused to mean the platform's \*(L"natural\*(R" 8\-bit encoding of Unicode. See perlebcdic for more discussion of the issues. .SS "Locales" .IX Subsection "Locales" See \*(L"Unicode and \s-1UTF\-8\s0\*(R" in perllocale .SS "When Unicode Does Not Happen" .IX Subsection "When Unicode Does Not Happen" While Perl does have extensive ways to input and output in Unicode, and a few other \*(L"entry points\*(R" like the \f(CW@ARGV\fR array (which can sometimes be interpreted as \s-1UTF\-8\s0), there are still many places where Unicode (in some encoding or another) could be given as arguments or received as results, or both, but it is not. .PP The following are such interfaces. Also, see \*(L"The \*(R"Unicode Bug"". For all of these interfaces Perl currently (as of 5.8.3) simply assumes byte strings both as arguments and results, or \s-1UTF\-8\s0 strings if the (problematic) \f(CW\*(C`encoding\*(C'\fR pragma has been used. .PP One reason that Perl does not attempt to resolve the role of Unicode in these situations is that the answers are highly dependent on the operating system and the file system(s). For example, whether filenames can be in Unicode and in exactly what kind of encoding, is not exactly a portable concept. Similarly for \f(CW\*(C`qx\*(C'\fR and \f(CW\*(C`system\*(C'\fR: how well will the \&\*(L"command-line interface\*(R" (and which of them?) handle Unicode? .IP "\(bu" 4 chdir, chmod, chown, chroot, exec, link, lstat, mkdir, rename, rmdir, stat, symlink, truncate, unlink, utime, \-X .IP "\(bu" 4 \&\f(CW%ENV\fR .IP "\(bu" 4 glob (aka the <*>) .IP "\(bu" 4 open, opendir, sysopen .IP "\(bu" 4 qx (aka the backtick operator), system .IP "\(bu" 4 readdir, readlink .ie n .SS "The ""Unicode Bug""" .el .SS "The ``Unicode Bug''" .IX Subsection "The Unicode Bug" The term, the \*(L"Unicode bug\*(R" has been applied to an inconsistency on \s-1ASCII\s0 platforms with the Unicode code points in the Latin\-1 Supplement block, that is, between 128 and 255. Without a locale specified, unlike all other characters or code points, these characters have very different semantics in byte semantics versus character semantics, unless \&\f(CW\*(C`use feature \*(Aqunicode_strings\*(Aq\*(C'\fR is specified. (The lesson here is to specify \f(CW\*(C`unicode_strings\*(C'\fR to avoid the headaches.) .PP In character semantics they are interpreted as Unicode code points, which means they have the same semantics as Latin\-1 (\s-1ISO\-8859\-1\s0). .PP In byte semantics, they are considered to be unassigned characters, meaning that the only semantics they have is their ordinal numbers, and that they are not members of various character classes. None are considered to match \f(CW\*(C`\ew\*(C'\fR for example, but all match \f(CW\*(C`\eW\*(C'\fR. .PP The behavior is known to have effects on these areas: .IP "\(bu" 4 Changing the case of a scalar, that is, using \f(CW\*(C`uc()\*(C'\fR, \f(CW\*(C`ucfirst()\*(C'\fR, \f(CW\*(C`lc()\*(C'\fR, and \f(CW\*(C`lcfirst()\*(C'\fR, or \f(CW\*(C`\eL\*(C'\fR, \f(CW\*(C`\eU\*(C'\fR, \f(CW\*(C`\eu\*(C'\fR and \f(CW\*(C`\el\*(C'\fR in regular expression substitutions. .IP "\(bu" 4 Using caseless (\f(CW\*(C`/i\*(C'\fR) regular expression matching .IP "\(bu" 4 Matching any of several properties in regular expressions, namely \f(CW\*(C`\eb\*(C'\fR, \&\f(CW\*(C`\eB\*(C'\fR, \f(CW\*(C`\es\*(C'\fR, \f(CW\*(C`\eS\*(C'\fR, \f(CW\*(C`\ew\*(C'\fR, \f(CW\*(C`\eW\*(C'\fR, and all the Posix character classes \&\fIexcept\fR \f(CW\*(C`[[:ascii:]]\*(C'\fR. .IP "\(bu" 4 In \f(CW\*(C`quotemeta\*(C'\fR or its inline equivalent \f(CW\*(C`\eQ\*(C'\fR, no characters code points above 127 are quoted in \s-1UTF\-8\s0 encoded strings, but in byte encoded strings, code points between 128\-255 are always quoted. .IP "\(bu" 4 User-defined case change mappings. You can create a \f(CW\*(C`ToUpper()\*(C'\fR function, for example, which overrides Perl's built-in case mappings. The scalar must be encoded in utf8 for your function to actually be invoked. .PP This behavior can lead to unexpected results in which a string's semantics suddenly change if a code point above 255 is appended to or removed from it, which changes the string's semantics from byte to character or vice versa. As an example, consider the following program and its output: .PP .Vb 11 \& $ perl \-le\*(Aq \& no feature \*(Aqunicode_strings\*(Aq; \& $s1 = "\exC2"; \& $s2 = "\ex{2660}"; \& for ($s1, $s2, $s1.$s2) { \& print /\ew/ || 0; \& } \& \*(Aq \& 0 \& 0 \& 1 .Ve .PP If there's no \f(CW\*(C`\ew\*(C'\fR in \f(CW\*(C`s1\*(C'\fR or in \f(CW\*(C`s2\*(C'\fR, why does their concatenation have one? .PP This anomaly stems from Perl's attempt to not disturb older programs that didn't use Unicode, and hence had no semantics for characters outside of the \&\s-1ASCII\s0 range (except in a locale), along with Perl's desire to add Unicode support seamlessly. The result wasn't seamless: these characters were orphaned. .PP Starting in Perl 5.14, \f(CW\*(C`use feature \*(Aqunicode_strings\*(Aq\*(C'\fR can be used to cause Perl to use Unicode semantics on all string operations within the scope of the feature subpragma. Regular expressions compiled in its scope retain that behavior even when executed or compiled into larger regular expressions outside the scope. (The pragma does not, however, affect the \f(CW\*(C`quotemeta\*(C'\fR behavior. Nor does it affect the deprecated user-defined case changing operations\*(--these still require a \s-1UTF\-8\s0 encoded string to operate.) .PP In Perl 5.12, the subpragma affected casing changes, but not regular expressions. See \*(L"lc\*(R" in perlfunc for details on how this pragma works in combination with various others for casing. .PP For earlier Perls, or when a string is passed to a function outside the subpragma's scope, a workaround is to always call \f(CW\*(C`utf8::upgrade($string)\*(C'\fR, or to use the standard module Encode. Also, a scalar that has any characters whose ordinal is above 0x100, or which were specified using either of the \&\f(CW\*(C`\eN{...}\*(C'\fR notations, will automatically have character semantics. .SS "Forcing Unicode in Perl (Or Unforcing Unicode in Perl)" .IX Subsection "Forcing Unicode in Perl (Or Unforcing Unicode in Perl)" Sometimes (see \*(L"When Unicode Does Not Happen\*(R" or \*(L"The \*(R"Unicode Bug"") there are situations where you simply need to force a byte string into \s-1UTF\-8\s0, or vice versa. The low-level calls utf8::upgrade($bytestring) and utf8::downgrade($utf8string[, \s-1FAIL_OK\s0]) are the answers. .PP Note that \fIutf8::downgrade()\fR can fail if the string contains characters that don't fit into a byte. .PP Calling either function on a string that already is in the desired state is a no-op. .SS "Using Unicode in \s-1XS\s0" .IX Subsection "Using Unicode in XS" If you want to handle Perl Unicode in \s-1XS\s0 extensions, you may find the following C APIs useful. See also \*(L"Unicode Support\*(R" in perlguts for an explanation about Unicode at the \s-1XS\s0 level, and perlapi for the \s-1API\s0 details. .IP "\(bu" 4 \&\f(CW\*(C`DO_UTF8(sv)\*(C'\fR returns true if the \f(CW\*(C`UTF8\*(C'\fR flag is on and the bytes pragma is not in effect. \f(CW\*(C`SvUTF8(sv)\*(C'\fR returns true if the \f(CW\*(C`UTF8\*(C'\fR flag is on; the bytes pragma is ignored. The \f(CW\*(C`UTF8\*(C'\fR flag being on does \fBnot\fR mean that there are any characters of code points greater than 255 (or 127) in the scalar or that there are even any characters in the scalar. What the \f(CW\*(C`UTF8\*(C'\fR flag means is that the sequence of octets in the representation of the scalar is the sequence of \s-1UTF\-8\s0 encoded code points of the characters of a string. The \f(CW\*(C`UTF8\*(C'\fR flag being off means that each octet in this representation encodes a single character with code point 0..255 within the string. Perl's Unicode model is not to use \s-1UTF\-8\s0 until it is absolutely necessary. .IP "\(bu" 4 \&\f(CW\*(C`uvchr_to_utf8(buf, chr)\*(C'\fR writes a Unicode character code point into a buffer encoding the code point as \s-1UTF\-8\s0, and returns a pointer pointing after the \s-1UTF\-8\s0 bytes. It works appropriately on \s-1EBCDIC\s0 machines. .IP "\(bu" 4 \&\f(CW\*(C`utf8_to_uvchr(buf, lenp)\*(C'\fR reads \s-1UTF\-8\s0 encoded bytes from a buffer and returns the Unicode character code point and, optionally, the length of the \s-1UTF\-8\s0 byte sequence. It works appropriately on \s-1EBCDIC\s0 machines. .IP "\(bu" 4 \&\f(CW\*(C`utf8_length(start, end)\*(C'\fR returns the length of the \s-1UTF\-8\s0 encoded buffer in characters. \f(CW\*(C`sv_len_utf8(sv)\*(C'\fR returns the length of the \s-1UTF\-8\s0 encoded scalar. .IP "\(bu" 4 \&\f(CW\*(C`sv_utf8_upgrade(sv)\*(C'\fR converts the string of the scalar to its \s-1UTF\-8\s0 encoded form. \f(CW\*(C`sv_utf8_downgrade(sv)\*(C'\fR does the opposite, if possible. \f(CW\*(C`sv_utf8_encode(sv)\*(C'\fR is like sv_utf8_upgrade except that it does not set the \f(CW\*(C`UTF8\*(C'\fR flag. \f(CW\*(C`sv_utf8_decode()\*(C'\fR does the opposite of \f(CW\*(C`sv_utf8_encode()\*(C'\fR. Note that none of these are to be used as general-purpose encoding or decoding interfaces: \f(CW\*(C`use Encode\*(C'\fR for that. \f(CW\*(C`sv_utf8_upgrade()\*(C'\fR is affected by the encoding pragma but \f(CW\*(C`sv_utf8_downgrade()\*(C'\fR is not (since the encoding pragma is designed to be a one-way street). .IP "\(bu" 4 \&\f(CWis_utf8_char(s)\fR returns true if the pointer points to a valid \s-1UTF\-8\s0 character. .IP "\(bu" 4 \&\f(CW\*(C`is_utf8_string(buf, len)\*(C'\fR returns true if \f(CW\*(C`len\*(C'\fR bytes of the buffer are valid \s-1UTF\-8\s0. .IP "\(bu" 4 \&\f(CW\*(C`UTF8SKIP(buf)\*(C'\fR will return the number of bytes in the \s-1UTF\-8\s0 encoded character in the buffer. \f(CW\*(C`UNISKIP(chr)\*(C'\fR will return the number of bytes required to UTF\-8\-encode the Unicode character code point. \f(CW\*(C`UTF8SKIP()\*(C'\fR is useful for example for iterating over the characters of a \s-1UTF\-8\s0 encoded buffer; \f(CW\*(C`UNISKIP()\*(C'\fR is useful, for example, in computing the size required for a \s-1UTF\-8\s0 encoded buffer. .IP "\(bu" 4 \&\f(CW\*(C`utf8_distance(a, b)\*(C'\fR will tell the distance in characters between the two pointers pointing to the same \s-1UTF\-8\s0 encoded buffer. .IP "\(bu" 4 \&\f(CW\*(C`utf8_hop(s, off)\*(C'\fR will return a pointer to a \s-1UTF\-8\s0 encoded buffer that is \f(CW\*(C`off\*(C'\fR (positive or negative) Unicode characters displaced from the \s-1UTF\-8\s0 buffer \f(CW\*(C`s\*(C'\fR. Be careful not to overstep the buffer: \&\f(CW\*(C`utf8_hop()\*(C'\fR will merrily run off the end or the beginning of the buffer if told to do so. .IP "\(bu" 4 \&\f(CW\*(C`pv_uni_display(dsv, spv, len, pvlim, flags)\*(C'\fR and \&\f(CW\*(C`sv_uni_display(dsv, ssv, pvlim, flags)\*(C'\fR are useful for debugging the output of Unicode strings and scalars. By default they are useful only for debugging\*(--they display \fBall\fR characters as hexadecimal code points\*(--but with the flags \f(CW\*(C`UNI_DISPLAY_ISPRINT\*(C'\fR, \&\f(CW\*(C`UNI_DISPLAY_BACKSLASH\*(C'\fR, and \f(CW\*(C`UNI_DISPLAY_QQ\*(C'\fR you can make the output more readable. .IP "\(bu" 4 \&\f(CW\*(C`foldEQ_utf8(s1, pe1, l1, u1, s2, pe2, l2, u2)\*(C'\fR can be used to compare two strings case-insensitively in Unicode. For case-sensitive comparisons you can just use \f(CW\*(C`memEQ()\*(C'\fR and \f(CW\*(C`memNE()\*(C'\fR as usual, except if one string is in utf8 and the other isn't. .PP For more information, see perlapi, and \fIutf8.c\fR and \fIutf8.h\fR in the Perl source code distribution. .SS "Hacking Perl to work on earlier Unicode versions (for very serious hackers only)" .IX Subsection "Hacking Perl to work on earlier Unicode versions (for very serious hackers only)" Perl by default comes with the latest supported Unicode version built in, but you can change to use any earlier one. .PP Download the files in the desired version of Unicode from the Unicode web site ). These should replace the existing files in \&\fIlib/unicore\fR in the Perl source tree. Follow the instructions in \&\fI\s-1README\s0.perl\fR in that directory to change some of their names, and then build perl (see \fI\s-1INSTALL\s0\fR). .PP It is even possible to copy the built files to a different directory, and then change \fIutf8_heavy.pl\fR in the directory \f(CW$Config{privlib}\fR to point to the new directory, or maybe make a copy of that directory before making the change, and using \f(CW@INC\fR or the \f(CW\*(C`\-I\*(C'\fR run-time flag to switch between versions at will (but because of caching, not in the middle of a process), but all this is beyond the scope of these instructions. .SH "BUGS" .IX Header "BUGS" .SS "Interaction with Locales" .IX Subsection "Interaction with Locales" See \*(L"Unicode and \s-1UTF\-8\s0\*(R" in perllocale .SS "Problems with characters in the Latin\-1 Supplement range" .IX Subsection "Problems with characters in the Latin-1 Supplement range" See \*(L"The \*(R"Unicode Bug"" .SS "Interaction with Extensions" .IX Subsection "Interaction with Extensions" When Perl exchanges data with an extension, the extension should be able to understand the \s-1UTF8\s0 flag and act accordingly. If the extension doesn't recognize that flag, it's likely that the extension will return incorrectly-flagged data. .PP So if you're working with Unicode data, consult the documentation of every module you're using if there are any issues with Unicode data exchange. If the documentation does not talk about Unicode at all, suspect the worst and probably look at the source to learn how the module is implemented. Modules written completely in Perl shouldn't cause problems. Modules that directly or indirectly access code written in other programming languages are at risk. .PP For affected functions, the simple strategy to avoid data corruption is to always make the encoding of the exchanged data explicit. Choose an encoding that you know the extension can handle. Convert arguments passed to the extensions to that encoding and convert results back from that encoding. Write wrapper functions that do the conversions for you, so you can later change the functions when the extension catches up. .PP To provide an example, let's say the popular Foo::Bar::escape_html function doesn't deal with Unicode data yet. The wrapper function would convert the argument to raw \s-1UTF\-8\s0 and convert the result back to Perl's internal representation like so: .PP .Vb 6 \& sub my_escape_html ($) { \& my($what) = shift; \& return unless defined $what; \& Encode::decode_utf8(Foo::Bar::escape_html( \& Encode::encode_utf8($what))); \& } .Ve .PP Sometimes, when the extension does not convert data but just stores and retrieves them, you will be able to use the otherwise dangerous \fIEncode::_utf8_on()\fR function. Let's say the popular \&\f(CW\*(C`Foo::Bar\*(C'\fR extension, written in C, provides a \f(CW\*(C`param\*(C'\fR method that lets you store and retrieve data according to these prototypes: .PP .Vb 2 \& $self\->param($name, $value); # set a scalar \& $value = $self\->param($name); # retrieve a scalar .Ve .PP If it does not yet provide support for any encoding, one could write a derived class with such a \f(CW\*(C`param\*(C'\fR method: .PP .Vb 12 \& sub param { \& my($self,$name,$value) = @_; \& utf8::upgrade($name); # make sure it is UTF\-8 encoded \& if (defined $value) { \& utf8::upgrade($value); # make sure it is UTF\-8 encoded \& return $self\->SUPER::param($name,$value); \& } else { \& my $ret = $self\->SUPER::param($name); \& Encode::_utf8_on($ret); # we know, it is UTF\-8 encoded \& return $ret; \& } \& } .Ve .PP Some extensions provide filters on data entry/exit points, such as DB_File::filter_store_key and family. Look out for such filters in the documentation of your extensions, they can make the transition to Unicode data much easier. .SS "Speed" .IX Subsection "Speed" Some functions are slower when working on \s-1UTF\-8\s0 encoded strings than on byte encoded strings. All functions that need to hop over characters such as \fIlength()\fR, \fIsubstr()\fR or \fIindex()\fR, or matching regular expressions can work \fBmuch\fR faster when the underlying data are byte-encoded. .PP In Perl 5.8.0 the slowness was often quite spectacular; in Perl 5.8.1 a caching scheme was introduced which will hopefully make the slowness somewhat less spectacular, at least for some operations. In general, operations with \s-1UTF\-8\s0 encoded strings are still slower. As an example, the Unicode properties (character classes) like \f(CW\*(C`\ep{Nd}\*(C'\fR are known to be quite a bit slower (5\-20 times) than their simpler counterparts like \f(CW\*(C`\ed\*(C'\fR (then again, there are hundreds of Unicode characters matching \f(CW\*(C`Nd\*(C'\fR compared with the 10 \s-1ASCII\s0 characters matching \f(CW\*(C`d\*(C'\fR). .SS "Problems on \s-1EBCDIC\s0 platforms" .IX Subsection "Problems on EBCDIC platforms" There are several known problems with Perl on \s-1EBCDIC\s0 platforms. If you want to use Perl there, send email to perlbug@perl.org. .PP In earlier versions, when byte and character data were concatenated, the new string was sometimes created by decoding the byte strings as \fI\s-1ISO\s0 8859\-1 (Latin\-1)\fR, even if the old Unicode string used \s-1EBCDIC\s0. .PP If you find any of these, please report them as bugs. .SS "Porting code from perl\-5.6.X" .IX Subsection "Porting code from perl-5.6.X" Perl 5.8 has a different Unicode model from 5.6. In 5.6 the programmer was required to use the \f(CW\*(C`utf8\*(C'\fR pragma to declare that a given scope expected to deal with Unicode data and had to make sure that only Unicode data were reaching that scope. If you have code that is working with 5.6, you will need some of the following adjustments to your code. The examples are written such that the code will continue to work under 5.6, so you should be safe to try them out. .IP "\(bu" 4 A filehandle that should read or write \s-1UTF\-8\s0 .Sp .Vb 3 \& if ($] > 5.007) { \& binmode $fh, ":encoding(utf8)"; \& } .Ve .IP "\(bu" 4 A scalar that is going to be passed to some extension .Sp Be it Compress::Zlib, Apache::Request or any extension that has no mention of Unicode in the manpage, you need to make sure that the \&\s-1UTF8\s0 flag is stripped off. Note that at the time of this writing (October 2002) the mentioned modules are not UTF\-8\-aware. Please check the documentation to verify if this is still true. .Sp .Vb 4 \& if ($] > 5.007) { \& require Encode; \& $val = Encode::encode_utf8($val); # make octets \& } .Ve .IP "\(bu" 4 A scalar we got back from an extension .Sp If you believe the scalar comes back as \s-1UTF\-8\s0, you will most likely want the \s-1UTF8\s0 flag restored: .Sp .Vb 4 \& if ($] > 5.007) { \& require Encode; \& $val = Encode::decode_utf8($val); \& } .Ve .IP "\(bu" 4 Same thing, if you are really sure it is \s-1UTF\-8\s0 .Sp .Vb 4 \& if ($] > 5.007) { \& require Encode; \& Encode::_utf8_on($val); \& } .Ve .IP "\(bu" 4 A wrapper for fetchrow_array and fetchrow_hashref .Sp When the database contains only \s-1UTF\-8\s0, a wrapper function or method is a convenient way to replace all your fetchrow_array and fetchrow_hashref calls. A wrapper function will also make it easier to adapt to future enhancements in your database driver. Note that at the time of this writing (October 2002), the \s-1DBI\s0 has no standardized way to deal with \s-1UTF\-8\s0 data. Please check the documentation to verify if that is still true. .Sp .Vb 10 \& sub fetchrow { \& # $what is one of fetchrow_{array,hashref} \& my($self, $sth, $what) = @_; \& if ($] < 5.007) { \& return $sth\->$what; \& } else { \& require Encode; \& if (wantarray) { \& my @arr = $sth\->$what; \& for (@arr) { \& defined && /[^\e000\-\e177]/ && Encode::_utf8_on($_); \& } \& return @arr; \& } else { \& my $ret = $sth\->$what; \& if (ref $ret) { \& for my $k (keys %$ret) { \& defined \& && /[^\e000\-\e177]/ \& && Encode::_utf8_on($_) for $ret\->{$k}; \& } \& return $ret; \& } else { \& defined && /[^\e000\-\e177]/ && Encode::_utf8_on($_) for $ret; \& return $ret; \& } \& } \& } \& } .Ve .IP "\(bu" 4 A large scalar that you know can only contain \s-1ASCII\s0 .Sp Scalars that contain only \s-1ASCII\s0 and are marked as \s-1UTF\-8\s0 are sometimes a drag to your program. If you recognize such a situation, just remove the \s-1UTF8\s0 flag: .Sp .Vb 1 \& utf8::downgrade($val) if $] > 5.007; .Ve .SH "SEE ALSO" .IX Header "SEE ALSO" perlunitut, perluniintro, perluniprops, Encode, open, utf8, bytes, perlretut, \*(L"${^UNICODE}\*(R" in perlvar ).