Import translated manuals from JM CVS Repository.

[linuxjm/jm.git] / manual / gnumaniak / draft / man1 / tr.1

.\" You may copy, distribute and modify under the terms of the LDP General
.\" Public License as specified in the LICENSE file that comes with the
.\" gnumaniak distribution
.\"
.\" The author kindly requests that no comments regarding the "better"
.\" suitability or up-to-date notices of any info documentation alternative
.\" is added without contacting him first.
.\"
.\" (C) 2002 Ragnar Hojland Espinosa <ragnar@ragnar-hojland.com>
.\"
.\"     GNU tr man page
.\"     man pages are NOT obsolete!
.\"     <ragnar@ragnar-hojland.com>
.\"
.\" Japanese Version Copyright (c) 2000 NAKANO Takeo all rights reserved.
.\" Translated Sun 12 Mar 2000 by NAKANO Takeo <nakano@apm.seikei.ac.jp>
.\"
.TH TR 1 "7 October 2002" "GNU textutils 2.1"
.\"O .SH NAME
.\"O \fBtr\fR \- translate, squeeze, and/or delete characters
.SH ̾Á°
tr \- ʸ»ú¤ÎÊÑ´¹¡¦ºï½ü¤ä¡¢Ï¢Â³¤¹¤ëʸ»ú¤Î°µ½Ì¤ò¹Ô¤¦
.\"O .SH SYNOPSIS
.SH ½ñ¼°
.B tr
.RB [ \-cdst\ ]
.RB [ \-\-complement ]
.RB [ \-\-delete ]
.RB [ \-\-squeeze\-repeats ]
.RB [ \-\-truncate\-set1 ]
.\"nakano error!
.\" .IR FILE1 " [" FILE2 ]
.IR SET1 " [" SET2 ]

.BR tr " [" \-\-help "] [" \-\-version ]
.\"O .SH DESCRIPTION
.SH ÀâÌÀ
.\"O .B tr
.\"O copies standard input to standard output, performing one of the
.\"O following operations:
.B tr
¤Ïɸ½àÆþÎϤòɸ½à½ÐÎϤ˥³¥Ô¡¼¤·¡¢¤½¤ÎºÝ¤Ë°Ê²¼¤Î¤¦¤Á¤Î¤¤¤º¤ì¤«¤ò¹Ô¤¦¡£
.\"O .IP
.\"O \fB\(bu\fR translate, and optionally squeeze repeated characters in the result
.IP \(bu 2
ʸ»ú¤òÊÑ´¹¤·¡¢¥ª¥×¥·¥ç¥ó¤Ç»ØÄꤹ¤ì¤Ð¤µ¤é¤ËϢ³¤¹¤ëʸ»ú¤ò°µ½Ì
(squeeze) ¤¹¤ë¡£
.\"O .IP
.\"O \fB\(bu\fR squeeze repeated characters
.IP \(bu
Ϣ³¤¹¤ëʸ»ú¤ò°µ½Ì¤¹¤ë¡£
.\"O .IP
.\"O \fB\(bu\fR delete characters
.IP \(bu
ʸ»ú¤òºï½ü¤¹¤ë¡£
.\"O .IP
.\"O \fB\(bu\fR delete characters, then squeeze repeated characters from the result
.IP \(bu
ʸ»ú¤òºï½ü¤·¤Æ¤«¤é¡¢Ï¢Â³¤¹¤ëʸ»ú¤ò°µ½Ì¤¹¤ë¡£
.PP
.\"O The \fISET1\fR and (if given) \fISET2\fR arguments define ordered sets of
.\"O characters, referred to below as \fISET1\fR and \fISET2\fR.  These sets are
.\"O the characters of the input that \fBtr\fR operates on.  The \fB\-c\fR option
.\"O replaces \fISET1\fR with its complement (all of the characters that are not
.\"O in \fISET1\fR).
\fISET1\fP ¤ª¤è¤Ó (Í¿¤¨¤é¤ì¤Æ¤¤¤ì¤Ð) \fISET2\fP °ú¤­¿ô¤Ï¡¢
ʸ»ú¤Î½ç½øÉÕ¤­¥»¥Ã¥È¤òÄêµÁ¤¹¤ë¡£
.\"nakano: °ÕÌ£¤Ê¤·(^^;
.\" °Ê²¼¤ÎÀâÌÀ¤Ç¤Ï¤³¤ì¤ò \fISET1\fP ¤ª¤è¤Ó \fISET2\fP ¤È¸Æ¤Ö¡£
¤³¤ì¤é¤Îʸ»ú¥»¥Ã¥È¤¬ÆþÎϤ˸ºß¤¹¤ë¤È¡¢
.B tr
¤Ï¤½¤Îʸ»ú¤ËÂФ·¤ÆÆ°ºî¤ò¹Ô¤¦¡£
.\"nakano: °Ê²¼¤Ï¾éĹ¤Êµ¤¤¬...
.\" .I \-c
.\" ¥ª¥×¥·¥ç¥ó¤Ï \fISET\fP ¤ò¤½¤ÎÊ佸¹ç
.\" (\fISET1\fP ¤Ë´Þ¤Þ¤ì¤Ê¤¤Ê¸»ú¤¹¤Ù¤Æ) ¤ËÃÖ´¹¤¹¤ë¡£
.\"O .SH OPTIONS
.SH ¥ª¥×¥·¥ç¥ó
.TP
.B \-c, \-\-complement
.\"O Replace \fISET1\fR with its complement (that is, all the characters that are
.\"O not in \fISET1\fR.
\fISET1\fR ¤ò¡¢ \fISET1\fR ¼«¿È¤ÎÊ佸¹ç
(\fISET1\fR ¤Ë´Þ¤Þ¤ì¤Ê¤¤Ê¸»ú¤¹¤Ù¤Æ) ¤ÇÃÖ¤­´¹¤¨¤ë¡£
.TP
.B \-d, \-\-delete
.\"O Delete any input characters that are in \fISET1\fR, do not translate.
\fISET1\fP ¤Ë¤¢¤ëÆþÎÏʸ»ú¤òºï½ü¤¹¤ë¡£ÊÑ´¹¤·¤Ê¤¤¡£
.TP
.B \-s, \-\-squeeze\-repeats
.\"O Replace sequence of characters with one; this is done after translating or
.\"O deleting. It replaces each input sequence of a repated character that is in
.\"O \fISET1\fR with a single occurrence of that character.
Ʊ¤¸Ê¸»ú¤Î·«¤êÊÖ¤·¤ò 1 ʸ»ú¤ËÃÖ¤­´¹¤¨¤ë¡£
¤³¤ÎÁàºî¤ÏÊÑ´¹¤äºï½ü¤Î¤¢¤È¤Ë¹Ô¤ï¤ì¤ë¡£
\fISET1\fP ¤Ë¤¢¤ëʸ»ú¤Î·«¤êÊÖ¤·¤ò¡¢¤½¤Îʸ»ú 1 ʸ»ú¤ËÃÖ¤­´¹¤¨¤ë¡£
.TP
.B \-t, \-\-truncate\-set1
.\"O Before doing any translations, first truncate \fISET1\fR to length of
.\"O \fISET2\fR.
ÊÑ´¹¤ò¹Ô¤¦Á°¤Ë¡¢¤Þ¤º \fISET1\fP ¤ò \fISET2\fP ¤ÎŤµ¤ËÀÚ¤êµÍ¤á¤ë¡£
.TP
.B "\-\-help"
.\"O Print a usage message on standard output and exit successfully.
ɸ½à½ÐÎϤ˻ÈÍÑÊýË¡¤Î¥á¥Ã¥»¡¼¥¸¤ò½ÐÎϤ·¤ÆÀµ¾ï½ªÎ»¤¹¤ë¡£
.TP
.B "\-\-version"
.\"O Print version information on standard output then exit successfully.
ɸ½à½ÐÎϤ˥С¼¥¸¥ç¥ó¾ðÊó¤ò½ÐÎϤ·¤ÆÀµ¾ï½ªÎ»¤¹¤ë¡£
.\"O .SH SPECIFYING SETS
.SH ʸ»ú¥»¥Ã¥È¤Î»ØÄê
.\"O The format of the \fISET1\fR and \fISET2\fR arguments resembles the format of
.\"O regular expressions; however, they are not regular expressions, only
.\"O lists of characters.  Most characters simply represent themselves in
.\"O these strings, but the strings can contain the shorthands listed below,
.\"O for convenience.  Some of them can be used only in \fISET1\fR or \fISET2\fR,
.\"O as noted below.
\fISET1\fP ¤ª¤è¤Ó \fISET2\fP °ú¤­¿ô¤Î¥Õ¥©¡¼¥Þ¥Ã¥È¤Ï
Àµµ¬É½¸½¤È»÷¤¿¤â¤Î¤Ë¤Ê¤Ã¤Æ¤¤¤ë¡£¤·¤«¤·Àµµ¬É½¸½¤½¤Î¤â¤Î¤Ç¤Ï¤Ê¤¯¡¢
¼Â¤Ïʸ»ú¤ÎʤӤ˲᤮¤Ê¤¤¡£
¤³¤ì¤é¤Îʸ»úÎó¤ÎÃæ¤Ëµ­¤µ¤ì¤¿Ê¸»ú¤Ï¡¢¤Û¤È¤ó¤É¤Î¾ì¹ç¤Ï¤½¤Îʸ»ú¼«¿È¤ò»Ø¤¹¡£
ʸ»úÎó¤Ë¤ÏÊص¹¾å°Ê²¼¤Î¤è¤¦¤Êû½Ì·Á¤òÍѤ¤¤ë¤³¤È¤â¤Ç¤­¤ë¡£
¤¿¤À¤· \fISET1\fP ¤Þ¤¿¤Ï \fISET2\fP ¤Î¤¤¤º¤ì¤«¤Ë¤·¤«
ÍѤ¤¤ë¤³¤È¤Î¤Ç¤­¤Ê¤¤¤â¤Î¤â¤¢¤ë (¤³¤Î¤è¤¦¤Ê¤â¤Î¤Ë¤ÏÃíµ­¤¬¤µ¤ì¤Æ¤¤¤ë)¡£
.\"O .SS Backslash Escapes
.SS ¥Ð¥Ã¥¯¥¹¥é¥Ã¥·¥å¤òÍѤ¤¤¿¥¨¥¹¥±¡¼¥×
.RS
.TP
.B \ea
Control\-G
.TP
.B \eb
Control\-H
.TP
.B \ef
Control\-L
.TP
.B \en
Control\-J
.TP
.B \er
Control\-M
.TP
.B \et
Control\-I
.TP
.B \ev
Control\-K
.TP
.B \e\fIOOO
.\"O The character with the value given by \fIOOO\fR, which is 1 to 3 octal digits
\fIOOO\fP (1¡Á3 ·å¤Î 8 ¿Ê¿ô) ¤Ë¤è¤Ã¤Æ»ØÄꤵ¤ì¤ëʸ»ú
.TP
.B \e\e
.\"O A backslash
¥Ð¥Ã¥¯¥¹¥é¥Ã¥·¥åʸ»ú
.PP
.\"O .SS Ranges
.SS ÈÏ°Ï»ØÄê
.\"O The notation `\fIM\fR\-\fIN\fR' expands to all of the characters from 
.\"O \fiM\fR through \fIN\fR, in ascending order.  \fIM\fR should collate before
.\"O \fIN\fR; if it doesn't, an error results.  As an example, `0\-9' is the same as
.\"O `0123456789'.  Although GNU \fBtr\fR does not support the System V
.\"O syntax that uses square brackets to enclose ranges, translations
.\"O specified in that format will still work as long as the brackets
.\"O in \fISTRING1\fR correspond to identical brackets in \fISTRING2\fR.
`\fIM\fR\-\fIN\fR' ¤È¤¤¤Ã¤¿É½µ­¤Ï¡¢
\fIM\fP ¤«¤é \fIN\fP ¤Þ¤Ç¤Î¤¹¤Ù¤Æ¤Îʸ»ú¤ò¾º½ç¤ËŸ³«¤·¤¿Ê¸»úÎó¤Ë¤Ê¤ë¡£
\fIM\fP ¤Ï¾È¹ç½ç½øÃæ¤Ç \fIN\fP ¤ÎÁ°¤Ë¤Ê¤±¤ì¤Ð¤Ê¤é¤º¡¢
¤³¤ì¤ËÈ¿¤·¤¿¾ì¹ç¤Ï¥¨¥é¡¼¤È¤Ê¤ë¡£
Î㤨¤Ð `0\-9' ¤Ï`0123456789' ¤ò»ØÄꤷ¤¿¤Î¤ÈƱ¤¸¤³¤È¤Ë¤Ê¤ë¡£
System V ÈǤÎ
.B tr
¤Ç¤ÏÈϰϤò»ØÄꤹ¤ëºÝ¤Ë³Ñ³ç¸Ì `[]' ¤òÍѤ¤¤ë¤¬¡¢
GNU ÈÇ
.BR tr
¤Ç¤Ï¤³¤Î·Á¼°¤Ï¥µ¥Ý¡¼¥È¤·¤Æ¤¤¤Ê¤¤¡£
¤¿¤À¤·¤³¤Î·Á¼°¤Ë¤ª¤±¤ëÊÑ´¹»ØÄê¤Ï
\fISET1\fP ¤È \fISET2\fP ¤Î´Ö¤Ç³Ñ³ç¸Ì¤¬Âбþ¤·¤Æ¤¤¤ì¤Ð¤Á¤ã¤ó¤ÈÆ°ºî¤¹¤ë¡£
.\"O .SS Repeated Characters
.SS ·«¤êÊÖ¤·Ê¸»ú
.\"O The notation `\fB[\fIC\fB*\fIN\fB]\fR' in \fISET2\fR expands to \fIN\fR
.\"O copies of character \fIC\fR.  Thus, `[y*6]' is the same as `yyyyyy'.  The
.\"O notation `[C*]' in \fISTRING2\fR expands to as many copies of C as are
.\"O needed to make \fISET2\fR as long as \fISET1\fR.  If \fIN\fR begins with
.\"O `0', it is interpreted in octal, otherwise in decimal.
\fISET2\fP ¤Ë¤ª¤±¤ë `\fB[\fIC\fP*\fIN\fB]\fR' ¤È¤¤¤Ã¤¿É½µ­¤Ï
\fIN\fP ¸Ä¤Îʸ»ú \fIC\fP ¤ËŸ³«¤µ¤ì¤ë¡£
¤·¤¿¤¬¤Ã¤Æ `[y*6]' ¤Ï `yyyyyy' ¤Ë¤Ê¤ë¡£
\fISET2\fP ¤Ë¤ª¤±¤ë `[\fIC\fP*]' ¤Ï \fISET2\fP ¤ò \fISET1\fP
¤ÈƱ¤¸Ä¹¤µ¤Ë¤¹¤ë¤¿¤á¤ËɬÍפʸĿô¤Îʸ»ú \fIC\fP ¤ËŸ³«¤µ¤ì¤ë¡£
\fIN\fP ¤¬ 0 ¤Ç¤Ï¤¸¤Þ¤ë¾ì¹ç¤Ï 8 ¿Ê¿ô¤È¤ß¤Ê¤µ¤ì¡¢
¤½¤ì°Ê³°¤Î¾ì¹ç¤Ï 10 ¿Ê¿ô¤È¤ß¤Ê¤µ¤ì¤ë¡£
.\"O .SS Character Classes
.SS ʸ»ú¥¯¥é¥¹
.\"O The notation `\fB[:\fICLASS\fB:]\fR' expands to all of the
.\"O characters in the
.\"O (predefined) class \fICLASS\fR.  The characters expand in no particular
.\"O order, except for the \fBupper\fR and \fBlower\fR classes, which expand in
.\"O ascending order.  When the \fB\-\-delete\fR and \fB\-\-squeeze\-repeats\fR
.\"O options are both given, any character class can be used in \fISET2\fR.
.\"O Otherwise, only the character classes  \fBlower\fR and \fBupper\fR are
.\"O accepted in \fISET2\fR, and then only if the corresponding character class
.\"O (\fBupper\fR and \fBlower\fR, respectively) is specified in the same
.\"O relative position in \fISET1\fR.  Doing this specifies case conversion.  The
.\"O class names are given below; an error results when an invalid class name is
.\"O given. 
`\fB[:\fICLASS\fB:]\fR' ¤È¤¤¤Ã¤¿É½µ­¤Ï¡¢
¤¢¤é¤«¤¸¤áÄêµÁ¤µ¤ì¤¿Ê¸»ú¥¯¥é¥¹ \fICLASS\fP ¤Î¤¹¤Ù¤Æ¤Îʸ»ú¤ËŸ³«¤µ¤ì¤ë¡£
ʸ»ú¥¯¥é¥¹¤Ë¤ÏÆÃÄê¤Î½ç½ø¤ÏÄêµÁ¤µ¤ì¤Æ¤¤¤Ê¤¤¡£
¤¿¤À¤· `upper' ¤È `lower' ¥¯¥é¥¹¤Î¤ß¤ÏÎã³°¤Ç¡¢¾º½ç¤ËŸ³«¤µ¤ì¤ë¡£
.BR \-\-delete  " (" \-d )
¤ª¤è¤Ó
.BR \-\-squeeze\-repeats " (" \-s )
¤ÎξÊý¤¬»ØÄꤵ¤ì¤¿¾ì¹ç¤Ï¡¢
\fISET2\fP ¤Ë¤Ï¤É¤ó¤Êʸ»ú¥¯¥é¥¹¤Ç¤âÍѤ¤¤ë¤³¤È¤¬¤Ç¤­¤ë¡£
¤½¤ì°Ê³°¤Î¾ì¹ç¤Ï¡¢ \fISET2\fP ¤Ë»ØÄê¤Ç¤­¤ë¤Î¤Ï `lower' ¤ª¤è¤Ó `upper' ¤Î¤ß¤Ç¡¢
¤·¤«¤â¤½¤ì¤¾¤ì¤ËÂбþ¤¹¤ë¥¯¥é¥¹
(`lower' ¤Ë¤Ï `upper'¡¢ `upper' ¤Ë¤Ï `lower') ¤¬
\fISET1\fP ¤ÎƱ¤¸°ÌÃ֤˻ØÄꤵ¤ì¤Æ¤¤¤Ê¤±¤ì¤Ð¤Ê¤é¤Ê¤¤¡£
¤³¤Î¾ì¹ç¤ÏÂçʸ»ú¾®Ê¸»ú¤¬ÊÑ´¹¤µ¤ì¤ë¡£
ʸ»ú¥¯¥é¥¹Ì¾¤ò°Ê²¼¤Ë¼¨¤¹¡£
¤³¤ì¤é°Ê³°¤Î̾Á°¤¬»ØÄꤵ¤ì¤¿¾ì¹ç¤Ë¤Ï¥¨¥é¡¼¤È¤Ê¤ë¡£
.RS
.TP
.B alnum
.\"O Letters and digits
±Ñʸ»ú¤È¿ô»ú
.TP
.B alpha
.\"O Letters
±Ñʸ»ú
.TP
.B blank
.\"O Horizontal whitespace
¿åÊ¿Êý¸þ¤Î¶õÇò (Horizontal whitespace)
.TP
.B cntrl
.\"O Control characters
À©¸æʸ»ú
.TP
.B digit
.\"O Digits
¿ô»ú
.TP
.B graph
.\"O Printable characters, not including space
°õºþ¤Ç¤­¤ëʸ»ú (¶õÇò¤Ï´Þ¤Þ¤Ê¤¤)
.TP
.B lower
.\"O Lowercase letters
±Ñ¾®Ê¸»ú
.TP
.B print
.\"O Printable characters, including space
°õºþ¤Ç¤­¤ëʸ»ú (¶õÇò¤â´Þ¤à)
.TP
.B punct
.\"O Punctuation characters
¶çÆÉÅÀ
.TP
.B space
.\"O Horizontal or vertical whitespace
¿åÊ¿¤ª¤è¤Ó¿âľÊý¸þ¤Î¶õÇò
.TP
.B upper
.\"O Uppercase letters
±ÑÂçʸ»ú
.TP
.B xdigit
.\"O Hexadecimal digits
16 ¿Ê¿ô¤Îʸ»ú
.PP
.\"O .SS Equivalence Classes
.SS Åù²Á¥¯¥é¥¹
.\"O The syntax `\fB[=\fIC\fB=]\fR' expands to all of the characters that are
.\"O equivalent to \fIC\fR, in no particular order.  Equivalence classes are
.\"O a relatively recent invention intended to support non-English alphabets.
.\"O But there seems to be no standard way to define them or determine their
.\"O contents.  Therefore, they are not fully implemented in GNU \fBtr\fR; each
.\"O character's equivalence class consists only of that character, which is of
.\"O no particular use.
`\fB[=\fIC\fB=]\fR' ¤È¤¤¤Ã¤¿É½µ­¤Ï¡¢
\fIC\fP ¤ÈÅù²Á¤Êʸ»ú¤ËŸ³«¤µ¤ì¤ë¡£¤³¤Î¤È¤­½ç½ø¤ÏÄêµÁ¤µ¤ì¤Ê¤¤¡£
Åù²Á¥¯¥é¥¹¤Ï¡¢±Ñ¸ì°Ê³°¤Î¥¢¥ë¥Õ¥¡¥Ù¥Ã¥È¤ò¥µ¥Ý¡¼¥È¤¹¤ë¤¿¤á¤Ë
ºÇ¶áȯÌÀ¤µ¤ì¤¿¤â¤Î¤Ç¤¢¤ë¡£
¤·¤«¤·¸½ºß¤Î¤È¤³¤í¤³¤ì¤é¤ÎÄêµÁ¤äÆâÍÆ»ØÄê¤Ëɸ½àŪ¤ÊÊýË¡¤Ï¤Ê¤¤¤è¤¦¤Ç¤¢¤ë¡£
¤·¤¿¤¬¤Ã¤Æ GNU ÈÇ
.BR tr
¤Ç¤â´°Á´¤Ë¤Ï¼ÂÁõ¤µ¤ì¤Æ¤¤¤Ê¤¤¡£
¤½¤ì¤¾¤ì¤Îʸ»ú¤ÎÅù²Á¥¯¥é¥¹¤Ï¤½¤Îʸ»ú¤«¤é¤Î¤ß¹½À®¤µ¤ì¤Æ¤¤¤ë¡£
¤·¤¿¤¬¤Ã¤Æ¤³¤Îµ¡Ç½¤Ï¸½ºß¤Î¤È¤³¤íÌò¤ËΩ¤¿¤Ê¤¤¡£
.\"O .SH TRANSLATING
.SH ÊÑ´¹
.\"O \fBtr\fR performs translation when \fISET1\fR and \fISET2\fR are both given and
.\"O \fB\-\-delete\fR is not specified.  \fRtr\fN translates each character
.\"O of its input that is in \fISET1\fR to the corresponding character in \fISET2\fR.
.\"O Characters not in \fISET1\fR are passed through unchanged.  When a character
.\"O appears more than once in \fISET1\fR and the corresponding characters in \fISET2\fR
.\"O are not all the same, only the final one is used.  For example, these
.\"O two commands are equivalent:
.B tr
¤Ï \fISET1\fP ¤È \fISET2\fP ¤ÎξÊý¤¬»ØÄꤵ¤ì¤Æ¤¤¤Æ¡¢
.BR \-\-delete " (" \-d )
¥ª¥×¥·¥ç¥ó¤¬»ØÄꤵ¤ì¤Æ¤¤¤Ê¤¤¾ì¹ç¤Ë¤ÏÊÑ´¹¤ò¹Ô¤¦¡£
.B tr
¤ÏÆþÎϤΤ¦¤Á \fISET1\fR ¤Ë¤¢¤ëʸ»ú¤ò
\fISET2\fR ¤ÎÂбþ¤¹¤ë°ÌÃ֤ˤ¢¤ëʸ»ú¤ËÊÑ´¹¤¹¤ë¡£
\fISET1\fR ¤Ë̵¤¤Ê¸»ú¤Ï¤½¤Î¤Þ¤Þ½ÐÎϤµ¤ì¤ë¡£
¤â¤·Ê¸»ú¤¬ \fISET1\fR ¤Ë 2 ²ó°Ê¾å¸½¤ï¤ì¡¢
\fISET2\fR ¤ÎÂбþ¤¹¤ëʸ»ú¤¬¤¹¤Ù¤ÆƱ¤¸¤Ç¤Ê¤¤¾ì¹ç¤Ë¤Ï¡¢
ºÇ¸å¤Îʸ»ú¤À¤±¤¬»È¤ï¤ì¤ë¡£Î㤨¤Ð°Ê²¼¤ÎÆó¤Ä¤Î¥³¥Þ¥ó¥É¤ÏÅù²Á¤Ç¤¢¤ë:
.sp
.RS
.nf
tr aaa xyz
tr a z
.fi
.RE
.sp
.\"O A common use of \fBtr\fR is to convert lowercase characters to
.\"O uppercase.  This can be done in many ways.  Here are three of them:
.B tr
¤ÎÎɤ¯¤¢¤ë»È¤¤Êý¤È¤·¤Æ¤Ï¡¢±Ñ¾®Ê¸»ú¤ÎÂçʸ»ú¤Ø¤ÎÊÑ´¹¤¬¤¢¤²¤é¤ì¤ë¡£
¤¤¤í¤¤¤í¤Ê¤ä¤êÊý¤¬²Äǽ¤Ç¤¢¤ë¤¬¡¢°Ê²¼¤ËÎã¤ò 3 ¤Ä¼¨¤¹¡£
.sp
.RS
.nf
tr abcdefghijklmnopqrstuvwxyz ABCDEFGHIJKLMNOPQRSTUVWXYZ
tr a\-z A\-Z
tr '[:lower:]' '[:upper:]'
.fi
.RE
.sp
.\"O When \fBtr\fR is performing translation, \fISET1\fR and \fISET2\fR typically have
.\"O the same length.  If \fISET1\fR is shorter than \fISET2\fR, the extra characters at
.\"O the end of \fISET2\fR are ignored.
.B tr
¤¬ÊÑ´¹¤ò¹Ô¤¦ºÝ¤Ë¤Ï¡¢ \fISET1\fR ¤È \fISET2\fR ¤ÏÄ̾ïƱ¤¸Ä¹¤µ¤Ç¤Ê¤±¤ì¤Ð¤Ê¤é¤Ê¤¤¡£
\fISET1\fR ¤¬ \fISET2\fR ¤è¤êû¤¤¾ì¹ç¤Ë¤Ï¡¢ \fISET2\fR ¤ÎÈøÉô¤Î;¤ê¤Ï̵»ë¤µ¤ì¤ë¡£

.\"O On the other hand, making \fISET1\fR longer than \fISET2\fR is not portable;
.\"O POSIX.2 says that the result is undefined.  In this situation, BSD \fBtr\fR
.\"O pads \fISET2\fR to the length of \fISET1\fR by repeating the last character
.\"O of \fISET2\fR as many times as necessary.  System V \fBtr\fR truncates
.\"O \fBSET1\fR to the length of \fBSET2\fR.
µÕ¤Ë \fISET1\fR ¤ò \fISET2\fR ¤è¤êŤ¯»ØÄꤹ¤ë¤È¡¢Æ°ºî¤Ïͽ¬¤Ç¤­¤Ê¤¯¤Ê¤ë¡£
POSIX.2 ¤Ë¤è¤ì¤Ð¡¢·ë²Ì¤ÏÄêµÁ¤µ¤ì¤Ê¤¤¡£
¤³¤Î¤è¤¦¤Ê¾ì¹ç¡¢ BSD ÈÇ
.B tr
¤Ç¤Ï \fISET2\fR ¤Î­¤ê¤Ê¤¤Éôʬ¤òºÇ¸å¤Îʸ»ú¤ÇËä¤á¡¢ \fISET1\fR ¤ÈƱ¤¸Ä¹¤µ¤Ë¤¹¤ë¡£
¤Þ¤¿ System V ÈÇ
.B tr
¤Ï \fISET1\fR ¤ÎÈøÉô¤òÀÚ¤ê¼Î¤Æ¤Æ \fISET2\fR ¤ÈƱ¤¸Ä¹¤µ¤Ë¤¹¤ë¡£

.\"O By default, GNU \fBtr\fR handles this case like BSD \fBtr\fR.  When the
.\"O \fB\-\-truncate\-set1\fR option is given, GNU \fBtr\fR handles this case
.\"O like the System V \fBtr\fR instead.  This option is ignored for operations
.\"O other than translation. Acting like System V \fBtr\fR in this case breaks the
.\"O relatively common BSD idiom:
¥Ç¥Õ¥©¥ë¥È¤Ç¤Ï GNU ÈÇ
.B tr
¤Ï BSD ÈÇ
.B tr
¤ÈƱ¤¸¤è¤¦¤Ë¿¶¤ëÉñ¤¦¡£
¤Þ¤¿ \-\-truncate\-set1 (\fI\-t\fP) ¥ª¥×¥·¥ç¥ó¤¬»ØÄꤵ¤ì¤¿¾ì¹ç¤Ë¤Ï¡¢
GNU ÈÇ
.B tr
¤Ï System V ÈÇ
.B tr
¤Î¤è¤¦¤Ë¿¶¤ëÉñ¤¦¡£¤³¤Î¥ª¥×¥·¥ç¥ó¤ÏÊÑ´¹°Ê³°¤ÎÅÀ¤Ë´Ø¤·¤Æ¤Ï̵»ë¤µ¤ì¤ë¡£
System V ÈÇ
.B tr
Ū¤Ê¿¶¤ëÉñ¤¤¤Î²¼¤Ç¤Ï¡¢ BSD ÈǤÎͭ̾¤ÊÍÑË¡:
.sp
.RS
.nf
tr \-cs A\-Za\-z0\-9 '\e012'
.fi
.RE
.sp
.\"O because it converts only zero bytes (the first element in the
.\"O complement of \fISET1\fR), rather than all non-alphanumerics, to newlines.
¤Ï»È¤¨¤Ê¤¯¤Ê¤ë¡£¤³¤ì¤Ï¥¢¥ë¥Õ¥¡¥Ù¥Ã¥Èʸ»ú¤ª¤è¤Ó¿ô»ú°Ê³°¤Îʸ»ú¤ò²þ¹Ôʸ»ú
¤ËÃÖ´¹¤¹¤ë¤â¤Î¤Ç¤¢¤ë¤¬¡¢ System V ¤Ç¤ÏÆâÍƤ¬ 0 ¤Î¥Ð¥¤¥È
(\fISET1\fR ¤ÎÊ佸¹ç¤ÎÀèƬÍ×ÁÇ) ¤Î¤ß¤òÊÑ´¹¤¹¤ë¤³¤È¤Ë¤Ê¤ë¡£
.\"O .SH SQUEEZING REPEATS AND DELETING
.SH ·«¤êÊÖ¤·¤Î°µ½Ì¤Èºï½ü
.\"O When given just the \fB\-\-delete\fR option, \fBtr\fR removes any input
.\"O characters that are in \fISET1\fR.
.BR \-\-delete " (" \-d )
¥ª¥×¥·¥ç¥ó¤À¤±¤¬»ØÄꤵ¤ì¤¿¾ì¹ç¤Ï¡¢
.B tr
¤ÏÆþÎϤΤ¦¤Á \fISET1\fR ¤Ë¤¢¤ëʸ»ú¤òºï½ü¤¹¤ë¡£

.\"O When given just the \fB\-\-squeeze\-repeats\fR option, \fBtr\fR replaces
.\"O each input sequence of a repeated character that is in \fISET1\fR with a
.\"O single occurrence of that character.
.BR \-\-squeeze\-repeats " (" \-s )
¥ª¥×¥·¥ç¥ó¤À¤±¤¬»ØÄꤵ¤ì¤¿¾ì¹ç¤Ë¤Ï¡¢
.B tr
¤ÏÆþÎϤΤ¦¤Á¡¢ \fISET1\fR ¤Ë¤¢¤ëʸ»ú¤Î·«¤êÊÖ¤·¤ò¡¢¤½¤Îʸ»ú 1 ʸ»ú¤ËÃÖ¤­´¹¤¨¤ë¡£

.\"O When given both \fB\-\-delete\fR and \fB\-\-squeeze\-repeats\fR, \fBtr\fR
.\"O first performs any deletions using \fISET1\fR, then squeezes repeats from any
.\"O remaining characters using \fISET2\fR.
.BR \-\-delete " ¤È " \-\-squeeze\-repeats
¤ÎξÊý¤Î¥ª¥×¥·¥ç¥ó¤¬»ØÄꤵ¤ì¤¿¾ì¹ç¤Ë¤Ï¡¢
.B tr
¤Ï¤Þ¤º \fISET1\fR ¤Ë¤¢¤ëʸ»ú¤òºï½ü¤·¡¢
»Ä¤ê¤Î¤¦¤Á \fISET2\fR ¤Ë¤¢¤ëʸ»ú¤Î·«¤êÊÖ¤·¤ò°µ½Ì¤¹¤ë¡£

.\"O The \fB\-\-squeeze\-repeats\fR option may also be used when translating, in
.\"O which case tr' first performs translation, then squeezes repeats from
.\"O any remaining characters using \fISET2\fR.
.B \-\-squeeze\-repeats
¥ª¥×¥·¥ç¥ó¤ÏÊÑ´¹¤È¶¦¤ËÍѤ¤¤ë¤³¤È¤â¤Ç¤­¤ë¡£¤³¤Î¾ì¹ç¤Ë¤Ï
.B tr
¤Ï¤Þ¤ºÊÑ´¹¤ò¹Ô¤¤¡¢»Ä¤ê¤Îʸ»ú¤Î¤¦¤Á \fISET2\fR ¤Ë¤¢¤ë¤â¤Î¤ò°µ½Ì¤¹¤ë¡£

.\"O Here are some examples to illustrate various combinations of options:
°Ê²¼¤Ë¤³¤ì¤é¤Î¥ª¥×¥·¥ç¥ó¤òÁȤ߹ç¤ï¤»¤¿ÍøÍÑÎã¤ò¼¨¤¹:

.\"O \fB\(bu\fR Remove all zero bytes
\fB\(bu\fR ÆâÍƤ¬ 0 ¤Î¥Ð¥¤¥È¤òºï½ü¤¹¤ë:
.sp
.RS
.nf
tr \-d '\e000'
.fi
.RE
.sp
.\"O \fB\(bu\fR Put all words on lines by themselves.  This converts all
.\"O non-alphanumeric characters to newlines, then squeezes each string
.\"O of repeated newlines into a single newline
\fB\(bu\fR ¤¹¤Ù¤Æ¤Îñ¸ì¤ò¤½¤ì¼«¿È¤Î¤ß¤«¤é¤Ê¤ë¹Ô¤ËÊÑ´¹¤¹¤ë¡£
¤Ä¤Þ¤ê¥¢¥ë¥Õ¥¡¥Ù¥Ã¥Èʸ»ú¡¢¿ô»ú°Ê³°¤Îʸ»ú¤ò²þ¹Ôʸ»ú¤ËÊÑ´¹¤·¡¢
²þ¹Ôʸ»ú¤Î·«¤êÊÖ¤·¤ò°ì¤Ä¤Î²þ¹Ôʸ»ú¤Ë°µ½Ì¤¹¤ë:
.sp
.RS
.nf
tr \-cs '[a\-zA\-Z0\-9]' '[\en*]'
.fi
.RE
.sp
.\"O \fB\(bu\fR Convert each sequence of repeated newlines to a single newline
\fB\(bu\fR ²þ¹Ôʸ»ú¤Î·«¤êÊÖ¤·¤ò°ì¤Ä¤Î²þ¹Ôʸ»ú¤ËÊÑ´¹¤¹¤ë¡£
.sp
.RS
.nf
tr \-s '\en'
.fi
.RE
.sp
.\"O \fB\(bu\fR Find doubled occurrences of words in a document.  For example,
.\"O people often write "the the" with the duplicated words separated by a
.\"O newline.  The bourne shell script below works first by converting each
.\"O sequence of punctuation and blank characters to a single newline.  That puts
.\"O each "word" on a line by itself.  Next it maps all uppercase characters to
.\"O lower case, and finally it runs \fBuniq\fR(1) with the \fB\-d\fR option to
.\"O print out only the words that were adjacent duplicates.
\fB\(bu\fR ʸ½ñÃæ¤ÎƱ¤¸Ã±¸ì¤ÎϢ³¤ò¸«¤Ä¤±¤ë¡£Î㤨¤Ð
"the the" ¤Î¤è¤¦¤Ë¡¢²þ¹Ô¤ò¶´¤ó¤Çñ¸ì¤ò¥À¥Ö¤Ã¤Æ½ñ¤¤¤Æ¤·¤Þ¤¦¤³¤È¤¬¤è¤¯¤¢¤ë¡£
°Ê²¼¤Î bourne ¥·¥§¥ë¥¹¥¯¥ê¥×¥È¤Ï¡¢¤Þ¤º¶çÆÉÅÀ¤ä¶õÇò¤ÎϢ³¤ò²þ¹Ô¤ËÊÑ´¹¤¹¤ë¡£
¤³¤ì¤Ë¤è¤Ã¤Æ¡¢³Æ¡Öñ¸ì¡×¤Ï¡¢¤½¤ì¤¾¤ì¼«Ê¬¼«¿È¤À¤±¤«¤é¤Ê¤ë¹Ô¤ËÃÖ¤«¤ì¤ë¡£
¼¡¤ËÂçʸ»ú¤ò¾®Ê¸»ú¤ËÊÑ´¹¤·¡¢ºÇ¸å¤Ë
.BR uniq (1)
¤ò
.B \-d
¥ª¥×¥·¥ç¥ó¤Çµ¯Æ°¤·¤Æ¡¢Ï¢Â³¤·¤¿Æ±¤¸Ã±¸ì¤À¤±¤òɽ¼¨¤¹¤ë¡£
.sp
.RS
.nf
#!/bin/sh
cat "$@" \e
| tr \-s '[:punct:][:blank:]' '\en' \e
| tr '[:upper:]' '[:lower:]' \e
| uniq \-d
.fi
.RE
.sp                                                          
.\"O .SH COMPATIBILITY
.SH °Ü¿¢À­
.\"O Setting the environment variable \fBPOSIXLY_CORRECT\fR turns off the
.\"O following warning and error messages, for strict compliance with
.\"O POSIX.2.  Otherwise, the following diagnostics are issued:
´Ä¶­ÊÑ¿ô
.B POSIXLY_CORRECT
¤ò»ØÄꤹ¤ë¤È¡¢ POSIX.2 ¤È¤Î¸·Ì©¤Ê¸ß´¹À­¤ò¤È¤ë¤¿¤á¤Ë¡¢
·Ù¹ð¤ª¤è¤Ó¥¨¥é¡¼¥á¥Ã¥»¡¼¥¸¤Î¤¤¤¯¤Ä¤«¤¬½ÐÎϤµ¤ì¤Ê¤¯¤Ê¤ë¡£
»ØÄꤷ¤Ê¤±¤ì¤Ð¡¢°Ê²¼¤Î¤è¤¦¤Ê¾ì¹ç¤Ë¥á¥Ã¥»¡¼¥¸¤¬½Ð¤ë¡£
.\"O .IP
.\"O 1. When the \fB\-\-delete\fR option is given but \fB\-\-squeeze\-repeats\fR is
.\"O not, and \fISET2\fR is given, GNU \fBtr\fR by default prints a usage message
.\"O and exits, because \fISET2\fR would not be used.  The POSIX specification
.\"O says that \fISET2\fR must be ignored in this case.  Silently ignoring
.\"O arguments is a bad idea.
.IP 1. 3
.B \-\-delete
¥ª¥×¥·¥ç¥ó¤¬»ØÄꤵ¤ì¤Æ¤¤¤ë¤¬
.B \-\-squeeze\-repeats
¤¬»ØÄꤵ¤ì¤Æ¤ª¤é¤º¡¢¤«¤Ä \fISET2\fP ¤¬Í¿¤¨¤é¤ì¤Æ¤¤¤ë¾ì¹ç¤Ë¤Ï¡¢GNU ÈÇ
.B tr
¤Ï¥Ç¥Õ¥©¥ë¥È¤Ç¤Ï»ÈÍÑË¡¤Î¥á¥Ã¥»¡¼¥¸¤òɽ¼¨¤·¤Æ½ªÎ»¤¹¤ë¡£
¤Ê¤¼¤Ê¤é¤³¤Î¤È¤­ \fISET2\fP ¤ÏÍøÍѤµ¤ì¤Ê¤¤¤«¤é¤Ç¤¢¤ë¡£
¤·¤«¤· POSIX µ¬³Ê¤Ë¤è¤ì¤Ð¡¢
¤³¤Î¾ì¹ç¤Ï \fISET2\fP ¤Ïñ¤Ë̵»ë¤µ¤ì¤Ê¤±¤ì¤Ð¤Ê¤é¤Ê¤¤¡£
°ú¤­¿ô¤òÌۤäÆ̵»ë¤¹¤ë¤Î¤ÏÎɤ¯¤Ê¤¤»ÅÍͤÀ¤È»×¤¦¤Î¤À¤¬¡£
.\"O .IP
.\"O 2. When an ambiguous octal escape is given.  For example, `\400' is
.\"O actually \40' followed by the digit `0', because the value 400
.\"O octal does not fit into a single byte.
.IP 2.
¤¢¤¤¤Þ¤¤¤Ê 8 ¿Ê¿ô¥¨¥¹¥±¡¼¥×¤¬»ØÄꤵ¤ì¤¿¾ì¹ç¡£Î㤨¤Ð `\e400' ¤Ï¼ÂºÝ¤Ë
¤Ï \e40 ¤È¿ô»ú 0 ¤ÎʤӤȲò¼á¤µ¤ì¤ë¡£
¤Ê¤¼¤Ê¤é 8 ¿Ê¿ô¤Î 400 ¤ËÂбþ¤¹¤ë¥Ð¥¤¥È¤Ï¸ºß¤·¤Ê¤¤¤«¤é¤Ç¤¢¤ë¡£
.PP       
.\"O GNU \fBtr\fR does not provide complete BSD or System V compatibility.
.\"O For example, it is impossible to disable interpretation of the POSIX
.\"O constructs `[:alpha:]', `[=c=]', and `[c*10]'.  Also, GNU \fBtr\fR does not
.\"O delete zero bytes automatically, unlike traditional Unix versions,
.\"O which provide no way to preserve zero bytes.            
GNU ÈÇ
.B tr
¤Ï BSD ¤ä System V ¤È¸ß´¹¤Ç¤Ï¤Ê¤¤¡£
Îã¤òµó¤²¤ì¤Ð¡¢ POSIX ÄêµÁ¤Ë¤¢¤ë `[:alpha:]', `[=c=]', `[c*10]'
¤È¤¤¤Ã¤¿»ØÄê¤Î²ò¼á¤ò»ß¤á¤µ¤»¤ë¤è¤¦¤Ê¥ª¥×¥·¥ç¥ó¤Ï¸ºß¤·¤Ê¤¤¡£
¤Þ¤¿ GNU ÈÇ
.B tr
¤ÏÆâÍƤ¬ 0 ¤Î¥Ð¥¤¥È¤ò¼«Æ°Åª¤Ë¤Ïºï½ü¤·¤Ê¤¤¡£
°ìÊý¤³¤ì¤Þ¤Ç¤Î UNIX ÈǤǤϡ¢¤³¤Î¥Ð¥¤¥È¤ò»Ä¤¹ÊýË¡¤ÎÊý¤¬Â¸ºß¤·¤Ê¤¤¡£
.\"O .SH NOTES
.SH Ãí°Õ
.\"O Report bugs to bug-textutils@gnu.org.
.\"O .br
.\"O Man page by Ragnar Hojland Espinosa <ragnar@ragnar-hojland.com>
¥×¥í¥°¥é¥à¤Î¥Ð¥°¤Ë¤Ä¤¤¤Æ¤Ï bug-textutils@gnu.org ¤ËÊó¹ð¤·¤Æ¤¯¤À¤µ¤¤¡£
.br
man ¥Ú¡¼¥¸¤Ï Ragnar Hojland Espinosa <ragnar@ragnar-hojland.com> ¤¬ºîÀ®¤·¤Þ¤·¤¿¡£