[hamete.org] Regm

Documentation

The output of regm --help :

regm version 1.0 15/07/01

Filter and extract messages from mailboxes with regular expressions.

installation:
        - change defaults at the begining of the script to fit your 
          needs, it's GPL code.
        - put it with your python scripts (need python 2.0) and do a 
          ln -s regm.py ~/bin/regm
bugs:
- parse encoded attachment
- only tested on Linux
- not usable as a library

usage:
regm [options] [search string] [input files]

exemples of use:
regm hello ~/mail/* > ~/tmp/mbox
regm -v -f ~/tmp/mbox -b hello -o -b bye ~/mail/*
regm -vx -f ~/tmp/mbox 'hello||bye'  ~/mail/*
regm -v -f ~/tmp/mbox -n -b hello ~/mail/*
regm -v -f ~/tmp/mbox -h '^to:.*@free.fr' ~/mail/*
regm -vx 't:@free.fr' > ~/tmp/mbox
regm -vN -f ~/tmp/mbox -h '^to:.*@free.fr' ~/mail/*
regm -xm 'f:joe&&hello&&!t:@free.fr'
regm -p '%40d  sub: %s\n %5B\n\n' "" ~/mail/inbox
regm -v -p '%a\n\n' '' ~/news/fr.misc.bavardages.dinosaures
regm -p '%D %T\n' "" ~/mail/sent
regm -U '' duplicate.mbox > clean.mbox

Default file path for mbox is ~/mail/*

All research string are regex, in expert mode (-x) you can use "||" 

"&&" and "!" operator in the same string as separator between different
regex.

Each output message is absolutely left unmodified by the filter.

options:
-h string       search string in message header
-b string       search string in message body
-n              negation of the following -h or -b 
-N              global negation (invert the filter output)
-o              "or" (between -h or -b). Default is a "and"
-f file         output file. there is a warning if output file exists.
                Default output is stdout.
-x              xpert mode
                no -h or -b option, use of && || !, the search string
                must be after all options (and before input path)
                's:' is '^Subject: .*'
                'd:' is '^Date: .*'
                'f:' is '^From: .*'
                'e:' is '^Sender: .*'
                't:' is '^To: .*'
                'c:' is '^Cc: .*'
                'r:' is '^Reply-To: .*'
                'i:' is '^Message-ID: .*'
                'g:' is '^References: .*'
                'a:' is '^Approved: .*'
                'x:' is '^X-Loop: .*'
                'n:' is '^newsgroups: .*'
                'h:' is equivalent to the -h option, default is 
                        searching in body
                
-u              case sensitive. Default is case insensitive
-p string       output format, syntax : \n \t and 
                %[number][sdfetcrixnagBDFETCR] for subject, date, from,... 
                B is body, other uppercase are for stripped mail 
                addresses and D for date with "%D" format
-m              direct launch of mutt on the result (using temp output
                file)
-U              discard duplicate messages (with same message-id)
-D string       change output date format of -p "%D" option
-q              quiet
-v              verbosity
--help          this help
--version       version
Files

regm-1.0.tar.gz, the archive (only contains regm.py, GPL license and .sig)
regm.py, the actual script
regm.py.sig, gpg file signature