(split) LDP v3.24 -> v3.29 の定型的な変更内容を反映。

[linuxjm/LDP_man-pages.git] / draft / man7 / glob.7

.\" Copyright (c) 1998 Andries Brouwer
.\"
.\" This is free documentation; you can redistribute it and/or
.\" modify it under the terms of the GNU General Public License as
.\" published by the Free Software Foundation; either version 2 of
.\" the License, or (at your option) any later version.
.\"
.\" The GNU General Public License's references to "object code"
.\" and "executables" are to be interpreted as the output of any
.\" document formatting or typesetting system, including
.\" intermediate and printed output.
.\"
.\" This manual is distributed in the hope that it will be useful,
.\" but WITHOUT ANY WARRANTY; without even the implied warranty of
.\" MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
.\" GNU General Public License for more details.
.\"
.\" You should have received a copy of the GNU General Public
.\" License along with this manual; if not, write to the Free
.\" Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111,
.\" USA.
.\"
.\" 2003-08-24 fix for / by John Kristoff + joey
.\"
.\" Japanese Version Copyright (c) 1998 NAKANO Takeo all rights reserved.
.\" Translated Wed 12 Aug 1998 by NAKANO Takeo <nakano@apm.seikei.ac.jp>
.\" Updated 2003-09-28 by NAKANO Takeo
.\"
.\"WORD:        wild card pattern       ¥ï¥¤¥ë¥É¥«¡¼¥É¥Ñ¥¿¡¼¥ó
.\"WORD:        character class         ʸ»ú¥¯¥é¥¹
.\"WORD:        range                   Îΰè»ØÄê
.\"WORD:        complementation         Ê佸¹ç
.\"WORD:        regular expression      Àµµ¬É½¸½
.\"WORD:        bracket expression      ¥Ö¥é¥±¥Ã¥Èɽ¸½
.\"WORD:        collating sequence      ¾È¹ç½ç½ø
.\"WORD:        collating element       ¾È¹ç½ç½ø¤ÎÍ×ÁÇ
.\"WORD:        current locale          ¥«¥ì¥ó¥È¥í¥±¡¼¥ë
.\"WORD:        equivalence class       Åù²Á¥¯¥é¥¹
.\"
.TH GLOB 7 2003-08-24 "Linux" "Linux Programmer's Manual"
.\"O .SH NAME
.SH ̾Á°
.\"O glob \- Globbing pathnames
glob \- ¥Ñ¥¹Ì¾¤ò glob ¤¹¤ë
.\"O .SH DESCRIPTION
.SH ÀâÌÀ
.\"O Long ago, in UNIX V6, there was a program
.\"O .I /etc/glob
.\"O that would expand wildcard patterns.
.\"O Soon afterward this became a shell built-in.
ÀΡ¹ UNIX V6 ¤Ç¤Ï¡¢¥ï¥¤¥ë¥É¥«¡¼¥É¥Ñ¥¿¡¼¥ó¤òŸ³«¤¹¤ë
.I /etc/glob
¤È¸À¤¦¥×¥í¥°¥é¥à¤¬¤¢¤Ã¤¿¡£¤½¤Î¸å¤¹¤°¤Ë¡¢
¤³¤Îµ¡Ç½¤Ï¥·¥§¥ë¤ËÁȤ߹þ¤Þ¤ì¤ë¤è¤¦¤Ë¤Ê¤Ã¤¿¡£

.\"O These days there is also a library routine
.\"O .BR glob (3)
.\"O that will perform this function for a user program.
º£Æü¤Ç¤Ï¡¢¤³¤Îµ¡Ç½¤ò¥æ¡¼¥¶¡¼¥×¥í¥°¥é¥à¤«¤é¤â¼Â¹Ô¤Ç¤­¤ë¤è¤¦¡¢
.BR glob (3)
¤È¤¤¤¦¥é¥¤¥Ö¥é¥ê¥ë¡¼¥Á¥ó¤â¸ºß¤·¤Æ¤¤¤ë¡£

.\"O The rules are as follows (POSIX.2, 3.13).
glob ¤Îµ¬Â§¤ò°Ê²¼¤Ë½Ò¤Ù¤ë (POSIX.2 3.13)¡£
.\"O .SS "Wildcard Matching"
.SS ¥ï¥¤¥ë¥É¥«¡¼¥É¥Þ¥Ã¥Á
.\"O A string is a wildcard pattern if it contains one of the
.\"O characters \(aq?\(aq, \(aq*\(aq or \(aq[\(aq.
.\"O Globbing is the operation
.\"O that expands a wildcard pattern into the list of pathnames
.\"O matching the pattern.
.\"O Matching is defined by:
ʸ»úÎó¤Ë \(aq?\(aq, \(aq*\(aq, \(aq[\(aq ¤¬´Þ¤Þ¤ì¤Æ¤¤¤ë¤È¡¢
¤½¤ì¤Ï¥ï¥¤¥ë¥É¥«¡¼¥É¥Ñ¥¿¡¼¥ó¤È¤ß¤Ê¤µ¤ì¤ë¡£
¡Öglob ¤¹¤ë¡×¤È¤¤¤¦¤Î¤Ï¡¢¥ï¥¤¥ë¥É¥«¡¼¥É¥Ñ¥¿¡¼¥ó¤òŸ³«¤·¤Æ¡¢
¤½¤Î¥Ñ¥¿¡¼¥ó¤Ë¥Þ¥Ã¥Á¤¹¤ë¥Ñ¥¹Ì¾¤Î¥ê¥¹¥È¤òÆÀ¤ë¤³¤È¤Ç¤¢¤ë¡£
¥Þ¥Ã¥Á¤Ï°Ê²¼¤Î¤è¤¦¤ËÄêµÁ¤µ¤ì¤ë¡£

.\"O A \(aq?\(aq (not between brackets) matches any single character.
(¥Ö¥é¥±¥Ã¥È³°Éô¤Î) \(aq?\(aq ¤Ï¤¢¤é¤æ¤ëñ°ì¤Îʸ»ú¤Ë¥Þ¥Ã¥Á¤¹¤ë¡£

.\"O A \(aq*\(aq (not between brackets) matches any string,
.\"O including the empty string.
(¥Ö¥é¥±¥Ã¥È³°Éô¤Î) \(aq*\(aq ¤Ï¤¢¤é¤æ¤ëʸ»úÎó¤Ë¥Þ¥Ã¥Á¤¹¤ë¡£
¶õʸ»úÎó (empty string) ¤Ë¤â¥Þ¥Ã¥Á¤¹¤ë¡£
.PP
.\"O .B "Character classes"
.B "ʸ»ú¥¯¥é¥¹ (character class)"
.sp
.\"O An expression "\fI[...]\fP" where the first character after the
.\"O leading \(aq[\(aq is not an \(aq!\(aq matches a single character,
.\"O namely any of the characters enclosed by the brackets.
.\"O The string enclosed by the brackets cannot be empty;
.\"O therefore \(aq]\(aq can be allowed between the brackets, provided
.\"O that it is the first character.
.\"O (Thus, "\fI[][!]\fP" matches the
.\"O three characters \(aq[\(aq, \(aq]\(aq and \(aq!\(aq.)
"\fI[...]\fP" ¤È¸À¤¦É½µ­¤Ï¡¢ÀèƬ¤Î \(aq[\(aq ¤Ë³¤¯ºÇ½é¤Îʸ»ú¤¬ \(aq!\(aq ¤Ç
¤Ê¤±¤ì¤Ð¡¢¥Ö¥é¥±¥Ã¥È¤ÎÃæ¤Ë´Þ¤Þ¤ì¤Æ¤¤¤ëʸ»ú¤Î¤É¤ì¤«°ì¤Ä¤Ë¥Þ¥Ã¥Á¤¹¤ë¡£
¥Ö¥é¥±¥Ã¥È¤ÎÆâÉô¤Ë´Þ¤Þ¤ì¤ëʸ»úÎó¤Ï¶õ¤Ç¤¢¤Ã¤Æ¤Ï¤Ê¤é¤Ê¤¤¡£
¤·¤¿¤¬¤Ã¤Æ \(aq]\(aq ¤âºÇ½é¤Îʸ»ú¤Ë»ØÄꤹ¤ì¤Ð¥Ö¥é¥±¥Ã¥È¤ÎÆâÉô¤Ë´Þ¤á¤ë¤³¤È¤¬
¤Ç¤­¤ë (¤Ä¤Þ¤ê "\fI[][!]\fP" ¤Ï \(aq[\(aq, \(aq]\(aq, \(aq!\(aq ¤Î
3 ʸ»ú¤Î¤É¤ì¤«¤Ë¥Þ¥Ã¥Á¤¹¤ë)¡£
.PP
.\"O .B Ranges
.B "Îΰè»ØÄê (range)"
.sp
.\"O There is one special convention:
.\"O two characters separated by \(aq\-\(aq denote a range.
.\"O (Thus, "\fI[A\-Fa\-f0\-9]\fP"
.\"O is equivalent to "\fI[ABCDEFabcdef0123456789]\fP".)
.\"O One may include \(aq\-\(aq in its literal meaning by making it the
.\"O first or last character between the brackets.
.\"O (Thus, "\fI[]\-]\fP" matches just the two characters \(aq]\(aq and \(aq\-\(aq,
.\"O and "\fI[\-\-0]\fP" matches the
.\"O three characters \(aq\-\(aq, \(aq.\(aq, \(aq0\(aq, since \(aq/\(aq
.\"O cannot be matched.)
Æüì¤Êɽµ­Ë¡¤¬°ì¤Ä¸ºß¤¹¤ë¡£\(aq\-\(aq ¤ò¶´¤àÆó¤Ä¤Îʸ»ú¤ÏÎΰè»ØÄê¤È¤Ê¤ë¡£
(¤Ä¤Þ¤ê "\fI[A\-Fa\-f0\-9]\fP" ¤Ï "\fI[ABCDEFabcdef0123456789]\fP"
¤ÈÅù²Á¤È¤Ê¤ë¡£) \(aq\-\(aq ʸ»ú¤½¤Î¤â¤Î¤òÆþ¤ì¤¿¤¤¾ì¹ç¤Ï¡¢
¥Ö¥é¥±¥Ã¥È¤ÎÀèƬ¤Þ¤¿¤ÏºÇ¸å¤Îʸ»ú¤Ë»ØÄꤹ¤ì¤Ð¤è¤¤¡£
(¤Ä¤Þ¤ê "\fI[]\-]\fP" ¤ÏÆó¤Ä¤Îʸ»ú \(aq]\(aq ¤È \(aq\-\(aq
¤Ë¥Þ¥Ã¥Á¤·¡¢"\fI[\-\-0]\fP" ¤Ï \(aq\-\(aq, \(aq.\(aq, \(aq0\(aq ¤Î
3 ʸ»ú¤Ë¥Þ¥Ã¥Á¤¹¤ë¡£¤³¤Î´Ö¤Î \(aq/\(aq ¤Ë¤Ï¥Þ¥Ã¥Á¤·¤Ê¤¤¡£¸å½Ò¤ò»²¾È¡£)
.PP
.\"O .B Complementation
.B Ê佸¹ç (complementation)
.sp
.\"O An expression "\fI[!...]\fP" matches a single character, namely
.\"O any character that is not matched by the expression obtained
.\"O by removing the first \(aq!\(aq from it.
.\"O (Thus, "\fI[!]a\-]\fP" matches any
.\"O single character except \(aq]\(aq, \(aqa\(aq and \(aq\-\(aq.)
"\fI[!...]\fP" ¤È¸À¤¦É½µ­¤Ï¡¢¥Ö¥é¥±¥Ã¥È¤ÎÆâÉô¤Ë´Þ¤Þ¤ì¤Ê¤¤Ã±°ì¤Îʸ»ú¤Ë¥Þ¥Ã¥Á¤¹¤ë
(¤¿¤À¤·ÀèƬ¤Ë¤¢¤ë \(aq!\(aq ¤Ï½ü³°)¡£ (¤Ä¤Þ¤ê "\fI[!]a\-]\fP" ¤Ï
\(aq]\(aq, \(aqa\(aq, \(aq\-\(aq °Ê³°¤Î¤¹¤Ù¤Æ¤Îʸ»ú¤Î¡¢¤É¤ì¤«°ì¤Ä¤Ë¥Þ¥Ã¥Á¤¹¤ë¡£)

.\"O One can remove the special meaning of \(aq?\(aq, \(aq*\(aq and \(aq[\(aq by
.\"O preceding them by a backslash, or, in case this is part of
.\"O a shell command line, enclosing them in quotes.
.\"O Between brackets these characters stand for themselves.
.\"O Thus, "\fI[[?*\e]\fP" matches the
.\"O four characters \(aq[\(aq, \(aq?\(aq, \(aq*\(aq and \(aq\e\(aq.
¥Ð¥Ã¥¯¥¹¥é¥Ã¥·¥å \(aq\e\(aq ¤òÁ°ÃÖ¤¹¤ì¤Ð¡¢ \(aq?\(aq, \(aq*\(aq, \(aq[\(aq
¤ÏÄ̾ï¤Îʸ»ú¤È¤·¤Æ°·¤ï¤ì¤ë¡£
¤Þ¤¿¤Ï¥·¥§¥ë¤Î¥³¥Þ¥ó¥É¥é¥¤¥ó¤Î°ìÉô¤Ë»ØÄꤹ¤ë¾ì¹ç¤Ï¡¢
¥¯¥©¡¼¥È¤Ç°Ï¤Ã¤Æ¤âƱ¤¸¸ú²Ì¤¬ÆÀ¤é¤ì¤ë¡£¥Ö¥é¥±¥Ã¥È¤ÎÆâÉô¤Ç¤Ï¡¢
¤³¤ì¤é¤Îʸ»ú¤Ï¤½¤Îʸ»ú¼«¿È¤À¤±¤ò°ÕÌ£¤¹¤ë¡£
¤¹¤Ê¤ï¤Á "\fI[[?*\e]\fP" ¤Ï \(aq[\(aq, \(aq?\(aq, \(aq*\(aq, \(aq\e\(aq
¤Î¤É¤ì¤«°ìʸ»ú¤Ë¥Þ¥Ã¥Á¤¹¤ë¡£
.\"O .SS Pathnames
.SS "¥Ñ¥¹Ì¾ (pathname)"
.\"O Globbing is applied on each of the components of a pathname
.\"O separately.
.\"O A \(aq/\(aq in a pathname cannot be matched by a \(aq?\(aq or \(aq*\(aq
.\"O wildcard, or by a range like "\fI[.\-0]\fP".
.\"O A range cannot contain an
.\"O explicit \(aq/\(aq character; this would lead to a syntax error.
glob Æ°ºî¤Ï¡¢¥Ñ¥¹Ì¾¤Î¤½¤ì¤¾¤ì¤ÎÉôʬ¤ËÆÈΩ¤ËŬÍѤµ¤ì¤ë¡£
¥Ñ¥¹Ì¾¤Ë¸ºß¤¹¤ë \(aq/\(aq ¤Ï \(aq?\(aq ¤ä \(aq*\(aq ¥ï¥¤¥ë¥É¥«¡¼¥É¤Ë¤Ï¥Þ¥Ã¥Á¤·¤Ê¤¤¡£
¤Þ¤¿ "\fI[.\-0]\fP" ¤Î¤è¤¦¤ÊÎΰè»ØÄê¤Ë¤â¥Þ¥Ã¥Á¤·¤Ê¤¤¡£
Îΰè»ØÄê¤ÏÍÛ¤Ë \(aq/\(aq ʸ»ú¤ò´Þ¤à¤³¤È¤Ï¤Ç¤­¤Ê¤¤¡£¤³¤ì¤Ïʸˡ¥¨¥é¡¼¤È¤Ê¤ë¡£

.\"O If a filename starts with a \(aq.\(aq,
.\"O this character must be matched explicitly.
.\"O (Thus, \fIrm\ *\fP will not remove .profile, and \fItar\ c\ *\fP will not
.\"O archive all your files; \fItar\ c\ .\fP is better.)
\(aq.\(aq ¤Ç»Ï¤Þ¤ë¥Ñ¥¹Ì¾¤Ç¤Ï¡¢¤³¤Îʸ»ú¤ÏÍۤ˥ޥåÁ¤µ¤»¤Ê¤±¤ì¤Ð¤Ê¤é¤Ê¤¤¡£
(¤Ä¤Þ¤ê \fIrm\ *\fP ¤Ï .profile ¤òºï½ü¤·¤Ê¤¤¡£¤Þ¤¿ \fItar\ c\ *\fP
¤Ç¤Ï¤¹¤Ù¤Æ¤Î¥Õ¥¡¥¤¥ë¤Ï¥¢¡¼¥«¥¤¥Ö¤µ¤ì¤Ê¤¤¡£ \fItar\ c\ .\fP ¤ÎÊý¤¬Îɤ¤¡£)
.\"O .SS "Empty Lists"
.SS ¶õ¤Î¥ê¥¹¥È
.\"O The nice and simple rule given above: "expand a wildcard pattern
.\"O into the list of matching pathnames" was the original UNIX
.\"O definition.
.\"O It allowed one to have patterns that expand into
.\"O an empty list, as in
Àè¤ËÍ¿¤¨¤¿¡¢¤ï¤«¤ê¤ä¤¹¤¯´Êñ¤Ê¥ë¡¼¥ë¡¢
¡Ö¥ï¥¤¥ë¥É¥«¡¼¥É¥Ñ¥¿¡¼¥ó¤ò¥Þ¥Ã¥Á¤·¤¿¥Ñ¥¹Ì¾¤Î¥ê¥¹¥È¤ËŸ³«¤¹¤ë¡×¤È¸À¤¦¤Î¤Ï¡¢
¥ª¥ê¥¸¥Ê¥ë¤Î UNIX ¤Ë¤ª¤±¤ëÄêµÁ¤Ç¤¢¤Ã¤¿¡£
¤³¤ì¤Ï¥Ñ¥¿¡¼¥ó¤¬¶õ¤Î¥ê¥¹¥È¤ËŸ³«¤µ¤ì¤ë¤³¤È¤âµö²Ä¤µ¤ì¤Æ¤¤¤¿¡£
Î㤨¤Ð
.br
.nf
    xv \-wait 0 *.gif *.jpg
.fi
.\"O where perhaps no *.gif files are present (and this is not
.\"O an error).
¤Ë¤ª¤¤¤Æ¡¢*.gif ¥Õ¥¡¥¤¥ë¤¬Á´¤¯¤Ê¤¤¾ì¹ç¤Ç¤â¡¢
¤³¤ì¤Ï¶õ¤Î¥ê¥¹¥È¤ËŸ³«¤µ¤ì¤ë¤¿¤á¡¢¥¨¥é¡¼¤Ë¤Ê¤é¤Ê¤¤¡£
.\"O However, POSIX requires that a wildcard pattern is left
.\"O unchanged when it is syntactically incorrect, or the list of
.\"O matching pathnames is empty.
.\"O With
.\"O .I bash
.\"O one can force the classical behavior by setting
.\"O .IR allow_null_glob_expansion=true .
¤·¤«¤· POSIX ¤Ç¤Ï¡¢Ê¸Ë¡Åª¤ËÀµ¤·¤¯¤Ê¤¤¥Ñ¥¿¡¼¥ó¤ä¡¢
¥Þ¥Ã¥Á¤¬¤Ê¤«¤Ã¤¿¥Ñ¥¿¡¼¥ó¤Ï¡¢
¤½¤Î¤Þ¤ÞÊѹ¹¤µ¤ì¤º¤Ë»Ä¤µ¤ì¤ë¤³¤È¤Ë¤Ê¤Ã¤Æ¤¤¤ë¡£
.I bash
¤Ç¤Ï
.I allow_null_glob_expansion=true
¤ò»ØÄꤹ¤ë¤³¤È¤Ç¡¢°ÊÁ°¤Î¿¶¤ëÉñ¤¤¤ËÀßÄꤹ¤ë¤³¤È¤¬¤Ç¤­¤ë¡£

.\"O (Similar problems occur elsewhere.
.\"O E.g., where old scripts have
(ƱÍͤÎÌäÂê¤ÏÊ̤ΤȤ³¤í¤Ç¤âµ¯¤³¤Ã¤Æ¤¤¤ë¡£Î㤨¤Ð¡¢¸Å¤¤¥¹¥¯¥ê¥×¥È¤Ë¤ª¤±¤ë
.br
.nf
    rm \`find . \-name "*~"\`
.fi
.\"O new scripts require
¤Î¤è¤¦¤Êµ­½Ò¤Ï¡¢¿·¤·¤¤¥¹¥¯¥ê¥×¥È¤Ç¤Ï
.br
.nf
    rm \-f nosuchfile \`find . \-name "*~"\`
.fi
.\"O to avoid error messages from
.\"O .I rm
.\"O called with an empty argument list.)
¤Î¤è¤¦¤Ë¤·¤Ê¤±¤ì¤Ð¤Ê¤é¤Ê¤¤¡£¤µ¤â¤Ê¤¤¤È
.I rm
¤ò°ú¤­¿ô¥ê¥¹¥È¤Ê¤·¤Ç¸Æ¤Ó½Ð¤¹²ÄǽÀ­¤¬¤¢¤ê¡¢
¥¨¥é¡¼¥á¥Ã¥»¡¼¥¸¤¬½Ð¤Æ¤·¤Þ¤¦¡£)
.\"O .SH NOTES
.SH Ãí°Õ
.\"O .SS Regular expressions
.SS Àµµ¬É½¸½
.\"O Note that wildcard patterns are not regular expressions,
.\"O although they are a bit similar.
.\"O First of all, they match
.\"O filenames, rather than text, and secondly, the conventions
.\"O are not the same: for example, in a regular expression \(aq*\(aq means zero or
.\"O more copies of the preceding thing.
¥ï¥¤¥ë¥É¥«¡¼¥É¥Ñ¥¿¡¼¥ó¤ÏÀµµ¬É½¸½¤È¿¾¯»÷¤Æ¤¤¤ë¤¬¡¢¤·¤«¤·¤³¤Îξ¼Ô¤Ï°Û¤Ê¤ë¡£
¤Þ¤ºÂè°ì¤Ë¡¢Á°¼Ô¤¬¥Õ¥¡¥¤¥ë̾¤Ë¥Þ¥Ã¥Á¤¹¤ë¤Î¤ËÂФ·¤Æ¡¢
¸å¼Ô¤Ï¥Æ¥­¥¹¥È¤Ë¥Þ¥Ã¥Á¤¹¤ë¡£ÂèÆó¤Ë¡¢¥ë¡¼¥ë¤âƱ¤¸¤Ç¤Ï¤Ê¤¤¡£
Î㤨¤ÐÀµµ¬É½¸½¤Ë¤ª¤±¤ë \(aq*\(aq ¤Ï¡¢
Á°ÃÖ¤µ¤ì¤¿Ê¸»ú¤Î 0 °Ê¾å¤Î·«¤êÊÖ¤·¤òɽ¤¹¡£

.\"O Now that regular expressions have bracket expressions where
.\"O the negation is indicated by a \(aq^\(aq, POSIX has declared the
.\"O effect of a wildcard pattern "\fI[^...]\fP" to be undefined.
Àµµ¬É½¸½¤Ë¤â¥Ö¥é¥±¥Ã¥Èɽ¸½¤Ï¤¢¤ë¤¬¡¢ÈÝÄê¤Ï \(aq^\(aq ¤Ç¤Ê¤µ¤ì¤ë¡£
POSIX ¤Ç¤Ï¥ï¥¤¥ë¥É¥«¡¼¥É¥Ñ¥¿¡¼¥ó¤Ë¤ª¤±¤ë "\fI[^...]\fP" ¤ò̤ÄêµÁ¤Ç¤¢¤ë¤È¤·¤Æ¤¤¤ë¡£
.\"O .SS Character classes and Internationalization
.SS ʸ»ú¥¯¥é¥¹¤È¹ñºÝ²½
.\"O Of course ranges were originally meant to be ASCII ranges,
.\"O so that "\fI[\ \-%]\fP" stands for "\fI[\ !"#$%]\fP" and "\fI[a\-z]\fP" stands
.\"O for "any lowercase letter".
.\"O Some UNIX implementations generalized this so that a range X\-Y
.\"O stands for the set of characters with code between the codes for
.\"O X and for Y.
.\"O However, this requires the user to know the
.\"O character coding in use on the local system, and moreover, is
.\"O not convenient if the collating sequence for the local alphabet
.\"O differs from the ordering of the character codes.
Îΰè»ØÄê¤Ï¡¢¤â¤È¤â¤È¤Ï¤â¤Á¤í¤ó ASCII ¤Ë¤ª¤±¤ë½ç½øʤӤò°ÕÌ£¤·¤Æ¤¤¤¿¡£
¤·¤¿¤¬¤Ã¤Æ "\fI[\ \-%]\fP" ¤Ï "\fI[\ !"#$%]\fP" ¤Î°ÕÌ£¤Ç¤¢¤ê¡¢
"\fI[a\-z]\fP" ¤Ï¡Ö¤¹¤Ù¤Æ¤Î¾®Ê¸»ú¡×¤Î°ÕÌ£¤Ç¤¢¤Ã¤¿¡£
UNIX ¤Î¼ÂÁõ¤ÎÃæ¤Ë¤Ï¡¢¤³¤ì¤ò³ÈÄ¥¤·¤¿¤â¤Î¤¬Â¸ºß¤·¡¢
¤½¤³¤Ç¤Ï X\-Y ¤È¤¤¤¦Îΰè»ØÄê¤Ï¡¢X ¤Î¥³¡¼¥É¤È
Y ¤Î¥³¡¼¥É¤Ë¶´¤Þ¤ì¤¿¥³¡¼¥É¤ò»ý¤Äʸ»ú¤¹¤Ù¤Æ¤òɽ¤¹¤è¤¦¤Ë¤Ê¤Ã¤Æ¤¤¤¿¡£
¤·¤«¤·¡¢¤³¤ì¤Ë¤Ï¥æ¡¼¥¶¡¼¤¬¥í¡¼¥«¥ë¤Ê¥·¥¹¥Æ¥à¤Ë¤ª¤±¤ë
ʸ»ú¥³¡¼¥É¤òÃΤé¤Ê¤±¤ì¤Ð¤Ê¤é¤º¡¢
¤µ¤é¤Ë¥í¡¼¥«¥ë¤Ê¥¢¥ë¥Õ¥¡¥Ù¥Ã¥È¤ËÂФ¹¤ë¾È¹ç½ç½ø
(collating sequence) ¤¬Ê¸»ú¥³¡¼¥É¤Î½ç½ø¤È°Û¤Ê¤Ã¤Æ¤¤¤ë¾ì¹ç¤Ë¤ÏÉÔÊؤǤ¢¤Ã¤¿¡£
(ÌõÃí: collating sequence ¤Ë´Ø¤·¤Æ¤Ï
.BR regex (7)
¤ò»²¾È¤·¤Æ²¼¤µ¤¤¡£)
.\"O Therefore, POSIX extended the bracket notation greatly,
.\"O both for wildcard patterns and for regular expressions.
.\"O In the above we saw three types of items that can occur in a bracket
.\"O expression: namely (i) the negation, (ii) explicit single characters,
.\"O and (iii) ranges.
.\"O POSIX specifies ranges in an internationally
.\"O more useful way and adds three more types:
¤·¤¿¤¬¤Ã¤Æ POSIX ¤Ç¤Ï¡¢¥ï¥¤¥ë¥É¥«¡¼¥É¥Ñ¥¿¡¼¥ó¤ÈÀµµ¬É½¸½¤ÎÁÐÊý¤Ë¤ª¤¤¤Æ¡¢
¥Ö¥é¥±¥Ã¥Èɽµ­¤òÂçÉý¤Ë³ÈÄ¥¤·¤Æ¤¤¤ë¡£
¤³¤ì¤Þ¤Ç²æ¡¹¤Ï¡¢¥Ö¥é¥±¥Ã¥Èɽµ­¤Ë¤Ï»°¤Ä¤ÎÍ×ÁǤ¬´Þ¤Þ¤ì¤¦¤ë¤³¤È¤ò¸«¤Æ¤­¤¿¡£
¤¹¤Ê¤ï¤Á (i) ÈÝÄê¡¢(ii) ñ°ì¤Îʸ»ú¡¢(iii) Îΰè»ØÄê¡¢¤Î»°¤Ä¤Ç¤¢¤ë¡£
POSIX ¤Ç¤Ï¡¢Îΰè»ØÄê¤ò¤è¤ê¹ñºÝ²½¤ËÊØÍø¤Ê¤è¤¦¤ËÄêµÁ¤·¤Æ¤ª¤ê¡¢
¤Þ¤¿»°¤Ä¤Î¥¿¥¤¥×¤ò¥Ö¥é¥±¥Ã¥Èɽµ­¤ÎÍ×ÁǤȤ·¤ÆÄɲ䷤Ƥ¤¤ë¡£

.\"O (iii) Ranges X\-Y comprise all characters that fall between X
.\"O and Y (inclusive) in the currect collating sequence as defined
.\"O by the
.\"O .B LC_COLLATE
.\"O category in the current locale.
(iii) Îΰè»ØÄê X\-Y ¤Ï X ¤È Y ¤Ë¶´¤Þ¤ì¤¿ (ξü´Þ¤à) ¤¹¤Ù¤Æ¤Îʸ»ú¤ò°ÕÌ£¤¹¤ë¡£
¤³¤Î¤È¤­¡¢¥«¥ì¥ó¥È¥í¥±¡¼¥ë (current locale) ¤Î
.B LC_COLLATE
¥«¥Æ¥´¥ê¤ÇÄêµÁ¤µ¤ì¤Æ¤¤¤ë¾È¹ç½ç½ø¤¬ÍѤ¤¤é¤ì¤ë¡£

.\"O (iv) Named character classes, like
(iv) ̾Á°ÉÕ¤­Ê¸»ú¥¯¥é¥¹: °Ê²¼¤Î¤è¤¦¤Ê¤â¤Î¤Ç¤¢¤ë¡£
.nf

[:alnum:]  [:alpha:]  [:blank:]  [:cntrl:]
[:digit:]  [:graph:]  [:lower:]  [:print:]
[:punct:]  [:space:]  [:upper:]  [:xdigit:]

.fi
.\"O so that one can say "\fI[[:lower:]]\fP" instead of "\fI[a\-z]\fP", and have
.\"O things work in Denmark, too, where there are three letters past \(aqz\(aq
.\"O in the alphabet.
.\"O These character classes are defined by the
.\"O .B LC_CTYPE
.\"O category
.\"O in the current locale.
¤³¤ì¤òÍѤ¤¤ì¤Ð "\fI[a\-z]\fP" ¤ÎÂå¤ï¤ê¤Ë "\fI[[:lower:]]\fP" ¤Î¤è¤¦¤Ê»ØÄ꤬¤Ç¤­¤ë¡£
¤Þ¤¿¥Ç¥ó¥Þ¡¼¥¯¤Î¤è¤¦¤Ë¡¢¥¢¥ë¥Õ¥¡¥Ù¥Ã¥È¤Î \(aqz\(aq °Ê¹ß¤Ë
3 ¤Ä¤Îʸ»ú¤¬Â¸ºß¤¹¤ë¤è¤¦¤Ê¾ì¹ç¤Ç¤â¡¢Æ±¤¸¤è¤¦¤ÊÆ°ºî¤¬´üÂԤǤ­¤ë¡£
¤³¤ì¤é¤Îʸ»ú¥¯¥é¥¹¤Ï¥«¥ì¥ó¥È¥í¥±¡¼¥ë¤Î
.B LC_CTYPE
¥«¥Æ¥´¥ê¤ÇÄêµÁ¤µ¤ì¤Æ¤¤¤ë¡£

.\"O (v) Collating symbols, like "\fI[.ch.]\fP" or "\fI[.a-acute.]\fP",
.\"O where the string between "\fI[.\fP" and "\fI.]\fP" is a collating
.\"O element defined for the current locale.
.\"O Note that this may
.\"O be a multicharacter element.
(v) ¾È¹ç½ç½ø¤Ë¤ª¤±¤ë¥·¥ó¥Ü¥ë: "\fI[.ch.]\fP" ¤ä "\fI[.a-acute.]\fP" ¤Î¤è¤¦¤Ë "\fI[.\fP"
¤È "\fI.]\fP" ¤Ç¶´¤Þ¤ì¤¿Ê¸»úÎó¤Ï¡¢¥«¥ì¥ó¥È¥í¥±¡¼¥ë¤ÇÄêµÁ¤µ¤ì¤¿¾È¹ç½ç½ø¤ÎÍ×ÁǤȤʤ롣
¤¢¤ë°ì¤Ä¤ÎÍ×ÁǤ¬Ê£¿ô¤Îʸ»ú¤«¤é¤Ê¤ë¾ì¹ç¤â¤¢¤ê¤¦¤ë¤³¤È¤ËÃí°Õ¡£

.\"O (vi) Equivalence class expressions, like "\fI[=a=]\fP",
.\"O where the string between "\fI[=\fP" and "\fI=]\fP" is any collating
.\"O element from its equivalence class, as defined for the
.\"O current locale.
.\"O For example, "\fI[[=a=]]\fP" might be equivalent
.\"O .\" FIXME . the accented 'a' characters are not rendering properly
.\"O .\" mtk May 2007
.\"O to "\fI[aaaaa]\fP" (warning: Latin-1 here), that is,
.\"O to "\fI[a[.a-acute.][.a-grave.][.a-umlaut.][.a-circumflex.]]\fP".
(vi) Åù²Á¥¯¥é¥¹É½¸½ (equivalence class expressions): "\fI[=a=]\fP"
¤Î¤è¤¦¤Ë "\fI[=\fP" ¤È "\fI=]\fP" ¤È¤Ç¶´¤Þ¤ì¤¿Ê¸»úÎó¤Ç¤¢¤ë¡£
¤³¤ì¤ÏÅù²Á¥¯¥é¥¹¤Î¥á¥ó¥Ð¡¼¤Ç¤¢¤ë¾È¹ç½ç½ø¤ÎÍ×ÁǤ¹¤Ù¤Æ¤Ë¤Ê¤ë¡£
Åù²Á¥¯¥é¥¹¤Ï¥«¥ì¥ó¥È¥í¥±¡¼¥ë¤ÇÄêµÁ¤µ¤ì¤Æ¤¤¤ë¤â¤Î¤Ë¤Ê¤ë¡£
Î㤨¤Ð¡¢"\fI[[=a=]]\fP" ¤Ï
"\fI[a[.a\-acute.][.a\-grave.][.a\-umlaut.][.a\-circumflex.]]\fP"
¤ÈÅù²Á¤Ç¤¢¤ë
(Latin-1 ɽµ­¤Ç¤Ï [a\e`{a}\e'{a}\e"{a}\e^{a}] ¤âƱ¤¸¡£
[ÌõÃí] ÆüËܸì¤Î roff ¥Ú¡¼¥¸¤Ç¤Ï latin1 ¥³¡¼¥É¤¬½Ð¤Ê¤¤¤Î¤Ç¡¢
¤³¤³¤Ç¤Ï TeX ɽµ­¤Çµ­ºÜ)¡£
.\"O .SH "SEE ALSO"
.SH ´ØÏ¢¹àÌÜ
.BR sh (1),
.BR fnmatch (3),
.BR glob (3),
.BR locale (7),
.BR regex (7)