.\"===========================================================================
.\" rgrep ­ a recursive highlighting grep program, this manpage was writen by
.\" "Boris D. Beletsky" <borik@isracom.co.il> copyright(c) 1996
.\" This manpage may be freely distrebuted as part of GNU Debian Linux
.\"===========================================================================
.TH RGREP 1 "OCT 1996" Debian "User Manuals"
.SH NAME
rgrep \- a recursive, highlighting grep program
.SH SYNOPSIS
.B rgrep 
.I [ options ]
.I pattern
.I [ file ]
.B ...
.SH DESCRIPTION
.I rgrep,
unlike 
.I grep(1) 
and 
.I egrep(1) 
rgrep has the ability to recursively descend
directories. The traditional way of performing this kind of search
on Unix
systems utilizes the 
.I find(1) 
command in conjunction with 
.I grep(1). 
However,
this results in very poor performance.
.SH COMMAND LINE OPTIONS
.LP
.I -?
.RS
additional help (use '-?' to avoid shell expansion on some systems)
.RE
.I -c
.RS
count matches
.RE
.I -h
.RS
highlight match (ANSI compatable terminal assumed)
.RE
.I -H
.RS
Output match instead of entire line containing match
.RE
.I -i
.RS
ignore case
.RE
.I -l
.RS 
list filename only
.RE
.I -n
.RS
print line number of match
.RE
.I -F
.RS
follow links
.RE
.I -r
.RS
recursively scan through directory tree
.RE
.I -N
.RS
Do NOT perform a recursive search
.RE
.I -R 'pat'
.RS
like '-r' except that only those files matching 'pat' are checked
.RE
.I -v
.RS
print only lines that do NOT match the specified pattern
.RE
.I -x 'ext'
.RS
checks only files with extension given by 'ext'.
.RE
.I -D
.RS
Print all directories that would be searched.  This option is for
debugging purposes only.  No file is grepped with this
option.
.RE
.I -W 'len'
.RS
lines are 'len' characters long (not newline terminated).
.RE
.LP
.SH SUPPORTED REGULAR EXPRESSIONS:
.LP
.I .
.RS
match any character except newline
.RE
.I \\\d
.RS
match any digit
.RE
.I \\\\e
.RS
match ESC char
.RE
.I *
.RS
matches zero or more occurences of previous RE
.RE
.I +
.RS
matches one or more occurences of previous RE
.RE
.I ?
.RS
matches zero or one occurence of previous RE
.RE
.I ^
.RS
matches beginning of line
.RE
.I $
.RS
matches end of line
.RE
.I [ ... ]            
.RS
matches any single character between brackets.
For example, 
.I [-02468] 
matches 
.I '-' 
or any even digit.
and 
.I [-0-9a-z] 
matches 
.I '-' 
and any digit between 
.I 0 
and 
.I 9
as well as letters 
.I a 
through 
.I z.
.RE
.LP
.I \\\\{ ... \\\\}
.LP
.I \\\\( ... \\\\)
.LP
.I \\\\1, \\\\2, ..., \\\\9    
.RS
matches match
specified by nth 
.I '\\\\( ... \\\\)' 
expression.
For example,
.I '\\\\([\ \\\\t][a-zA-Z]+\\\\)\\\\1[\ \\\\t]'
matches any
word repeated consecutively.
.RE
.LP
.SH EXAMPLES
Look in all files with a 'c' extension in current directory and all
its
subdirectories looking for matches of 'int ' at the beginning of a
line,
printing the line containing the match with its line number: (two
methods)
.LP
.I rgrep -n -R '*.c' '^int ' .
.LP
.I rgrep -n -x c '^int ' .
.LP	
Highlight all matches of repeated words in file 'paper.tex':
.LP
.I rgrep -h 
.I '[\ \\\\t]\\\\([a-zA-Z]+\\\\)[\ \\\\t]+\\\\1[\ \\\\t\\\\n]' paper.tex
.LP
.I rgrep -h '^\\\\([a-zA-Z]+\\\\)[\ \\\\t]+\\\\1[\ \\\\t\\\\n]' paper.tex
.LP
(Note that this version of rgrep requires two
passes for this example)
.LP
Search through all files EXCEPT .o and .a file below /usr/src/linux
looking
for the string 'mouse' without regard to case:
.LP
.I rgrep -i -R '*.[^ao]' mouse /usr/src/linux
.LP     
Search a fixed record length FITS file for the keyword EXTNAME:
.LP
.I rgrep -W80 ^EXTNAME file.fits
.LP
(Note that the regular expression 
.I '^[A-Z]+' 
will dump all fits headers.)
.SH AUTHOR
.LP
.RS
.I \
"""John E. Davis""" <davis@space.mit.edu>
.RE
.LP
-- This manpage was translated to troff by
.LP
"Boris D. Beletsky"
<borik@isracom.co.il>