[unix-history] / man / man1 / tr.1

.th TR I 9/24/73
.sh NAME
tr \*- transliterate
.sh SYNOPSIS
.bd tr
[
.bd \*-cds
] [ string1 [ string2 ] ]
.sh DESCRIPTION
.it Tr
copies the standard input to the standard output with 
substitution or deletion of selected characters.
Input characters found in 
.it string1
are mapped into the corresponding characters of
.it string2.
If 
.it string2
is short, it is padded with corresponding characters from
.it string1.
Any combination of the options
.bd \*-cds
may be used.
.bd \*-c
complements the set of characters in
.it string1
with respect to the universe of characters
whose ascii codes are 001 through 377 octal.
.bd \*-d
deletes all input characters not in
.it string1.
.bd \*-s
squeezes all strings of repeated output characters that are
in 
.it string2
to single characters.
.s3
The following abbreviation conventions may be used
to introduce ranges of characters or repeated characters into
the strings:
.s3
\fB[\fIa\*|\fB\*-\fIb\fB\*|]\fR
stands for the string of characters whose ascii codes run
from character
.it a
to character
.it b.
.s3
\fB[\fIa\fB\*|*\fIn\fB\*|]\fR,
where
.it n
is an integer or empty, stands for \fIn\fR-fold 
repetition of character
.it a.
.it n
is taken to be octal or decimal according as its
first digit is or is not zero.
A zero or missing
.it n
is taken to be huge;
this facility is useful for padding
.it string2.
.s3
The escape character `\\' may be used as in
.it sh
to remove special meaning from any character in a string.
In addition,
`\\' followed by 1, 2 or 3 octal digits stands for the
character whose ascii code is given by those digits.
.s3
The following example creates a list of all
the words in `file1' one per line in `file2',
where a word is taken to be a maximal string of alphabetics.
The strings are quoted
to protect the special characters from interpretation by the Shell;
012 is the ascii code for newline.
.s3
.ti +8
tr \*-cs "[A\*-Z][a\*-z]" "[\\012*]" <file1 >file2
.sh "SEE ALSO"
sh(I), ed(I), ascii(VII)
.sh BUGS
Won't handle ascii NUL.
.br
Also, Kernighan's Lemma can really bite you;
try looking for strings which have \\ and * in them.
Commit	Line	Data
48b67403 DM	1	.th TR I 9/24/73
	2	.sh NAME
	3	tr \*- transliterate
	4	.sh SYNOPSIS
	5	.bd tr
	6	[
	7	.bd \*-cds
	8	] [ string1 [ string2 ] ]
	9	.sh DESCRIPTION
	10	.it Tr
	11	copies the standard input to the standard output with
	12	substitution or deletion of selected characters.
	13	Input characters found in
	14	.it string1
	15	are mapped into the corresponding characters of
	16	.it string2.
	17	If
	18	.it string2
	19	is short, it is padded with corresponding characters from
	20	.it string1.
	21	Any combination of the options
	22	.bd \*-cds
	23	may be used.
	24	.bd \*-c
	25	complements the set of characters in
	26	.it string1
	27	with respect to the universe of characters
	28	whose ascii codes are 001 through 377 octal.
	29	.bd \*-d
	30	deletes all input characters not in
	31	.it string1.
	32	.bd \*-s
	33	squeezes all strings of repeated output characters that are
	34	in
	35	.it string2
	36	to single characters.
	37	.s3
	38	The following abbreviation conventions may be used
	39	to introduce ranges of characters or repeated characters into
	40	the strings:
	41	.s3
	42	\fB[\fIa\\|\fB\-\fIb\fB\*\|]\fR
	43	stands for the string of characters whose ascii codes run
	44	from character
	45	.it a
	46	to character
	47	.it b.
	48	.s3
	49	\fB[\fIa\fB\\|\fIn\fB\*\|]\fR,
	50	where
	51	.it n
	52	is an integer or empty, stands for \fIn\fR-fold
	53	repetition of character
	54	.it a.
	55	.it n
	56	is taken to be octal or decimal according as its
	57	first digit is or is not zero.
	58	A zero or missing
	59	.it n
	60	is taken to be huge;
	61	this facility is useful for padding
	62	.it string2.
	63	.s3
	64	The escape character `\\' may be used as in
65	.it sh
66	to remove special meaning from any character in a string.
67	In addition,
68	`\\' followed by 1, 2 or 3 octal digits stands for the
69	character whose ascii code is given by those digits.
70	.s3
71	The following example creates a list of all
72	the words in `file1' one per line in `file2',
73	where a word is taken to be a maximal string of alphabetics.
74	The strings are quoted
75	to protect the special characters from interpretation by the Shell;
76	012 is the ascii code for newline.
77	.s3
78	.ti +8
79	tr \-cs "[A\-Z][a\-z]" "[\\012]" <file1 >file2
80	.sh "SEE ALSO"
81	sh(I), ed(I), ascii(VII)
82	.sh BUGS
83	Won't handle ascii NUL.
84	.br
85	Also, Kernighan's Lemma can really bite you;
86	try looking for strings which have \\ and * in them.