GREP(C) UNIX System V
Name
grep, egrep, fgrep - searches a file for a pattern
Syntax
grep [ -bchlnsvy ] [ -e expression ] [ files ]
egrep [ -bchlnv ] [ -e expression ] [ files ]
fgrep [ -bclnvxy ] [ -f expfile ] [ files ]
Description
Commands of the grep family search the input files (or
standard input if no files are specified) for lines matching
a pattern. Normally, each matching line is copied to the
standard output. If more than one file is being searched,
the name of the file in which each match occurs is also
written to the standard output along with the matching line
(unless the -h option is used, see below).
grep patterns are limited regular expressions in the style
of ed(C). grep uses a compact nondeterministic algorithm.
egrep patterns are full regular expressions; it uses a fast
deterministic algorithm that sometimes needs exponential
space. fgrep patterns are fixed strings. fgrep is fast and
compact. The following options are recognized:
-v All lines but those matching are displayed.
-x Displays only exact matches of an entire line. (fgrep
only.)
-c Only a count of matching lines is displayed.
-l Only the names of files with matching lines are
displayed, separated by newlines.
-h Prevents the name of the file containing the matching
line from being prepended to that line. Used when
searching multiple files. (This option works with
grep and egrep only.)
-n Each line is preceded by its relative line number in
the file.
-b Each line is preceded by the block number on which it
was found. This is sometimes useful in locating disk
block numbers by context.
-s Suppresses error messages produced for nonexistent or
unreadable files. (grep only). Note that the -s
option will not suppress error messages generated by
the -f option.
-y Turns on matching of letters of either case in the
input so that case is insignificant. Conversion
between uppercase and lowercase letters is dependent
on the locale setting. -y does not work with egrep.
-e expression orstrings
Same as a simple expression argument, but useful when
the expression begins with a dash (-).
-f expfile
The regular expression for grep or egrep, or strings
list for fgrep is taken from the expfile.
In all cases (except with -h) the filename is output if
there is more than one input file. Care should be taken
when using the characters $, *, [, ^, |, (, ), and \ in
expression, because they are also meaningful to the shell.
It is safest to enclose the entire expression or strings
argument in single quotation marks. For example:
grep '[Ss]omeone' text.file
This command would find all lines containing the word
``someone'' in the file text.file, whether the initial ``s''
is uppercase or lowercase.
Multiple strings can be specified in fgrep without using a
separate strings file by using the quoting conventions of
the shell to imbed newlines in the string argument. For
example, if you were using the Bourne shell (sh(C)) you
might enter the following on the command line:
fgrep 'Someone
someone' text.file
This would have the same effect as the grep example above.
See the csh(C) manual page for ways to imbed newlines in a
string when using csh(C).
egrep accepts regular expressions as in ed(C), with the
addition of the following:
- A regular expression followed by a plus sign (+)
matches one or more occurrences of the regular
expression.
- A regular expression followed by a question mark (?)
matches 0 or 1 occurrences of the regular expression.
- Two regular expressions separated by a vertical bar (|)
or by a newline match strings that are matched by
either regular expression.
- A regular expression may be enclosed in parentheses ()
for grouping. For example:
egrep '([Ss]ome|[Aa]ny)one' text.file
This example displays all lines in text.file containing the
words ``someone'' or ``anyone'', whether or not they are
spelled with initial capital letters. Without the
parentheses, this example would display all lines containing
the words ``some'' or ``anyone'' (because the vertical bar
(|) operator is of lower precedence than concatenation, see
below).
Because of the algorithm used, egrep does not support
extended ranges as in ed(C): Ranges like [a-z] are
interpreted on the basis of the machine's collating
sequence, not the collating sequence defined by the locale.
grep supports col(C) extended ranges.
The \( and \) operators, supported by ed(C), are not
supported by egrep.
The order of precedence of operators is [], then *?+, then
concatenation, then backslash (\) with newline or vertical
bar (|).
See Also
coltbl(M), ed(C), locale(M), sed(C), sh(C)
Diagnostics
Exit status is 0 if any matches are found, 1 if no matches
are found, and 2 for syntax errors or inaccessible files.
Notes
Ideally there should be only one grep, but there isn't a
single algorithm that spans a wide enough range of space-
time tradeoffs.
Lines are limited to 256 characters. Longer lines are
truncated.
When using grep with the -y option, the search is not made
totally case insensitive in character ranges specified
within brackets.
Standards Conformance
egrep, fgrep and grep are conformant with:
AT&T SVID Issue 2, Select Code 307-127;
and The X/Open Portability Guide II of January 1987.
(printed 8/28/89) GREP(C)