Re: [sup-talk] printing emails (but not attachments)

From: John Bent <johnbent@lanl.gov>
To: sup-talk <sup-talk@rubyforge.org>
Subject: Re: [sup-talk] printing emails (but not attachments)
Date: Wed, 17 Mar 2010 13:25:02 -0600	[thread overview]
Message-ID: <1268853590-sup-9103@guava.lanl.gov> (raw)
In-Reply-To: <1268341771-sup-2631@guava.lanl.gov>

[-- Attachment #1: Type: text/plain, Size: 1254 bytes --]

Excerpts from John Bent's message of Thu Mar 11 14:11:21 -0700 2010:
> I've been piping emails into muttprint which works well unless there's
> a binary attachment.  When there's a binary attachment, muttprint
> prints it and it usually takes tens and tens of pages.  Whoops.
> 
> Anyone else figure out how to print emails but not attachments?
> 
Well, I'm sure I've once again reinvented the wheel but I wrote my own
script to print emails without printing binary attachments.  It reads an
email, pipes the headers to muttprint, then pipes text/plain and
text/html through w3m into muttprint, for everything else it pipes "#
Attachment: filename (type)."  It requires w3m and muttprint to be in
the user's path.

I'm sorry that it's horribly in violation of sup philosophy but I wrote
it in python. :(  Sorry but my ruby is just barely good enough to do
'if' statements in my hooks.

It's attached in case anyone else wants it.  I'm scared I really did
just reinvent the wheel, otherwise I'd add it to the wiki.  To print an
email from sup using this, hit '|' in thread-view-mode to open a pipe
and then just type sup-print (assuming you've put this attachment in
your path).  I believe all the python imports it uses come standard.
-- 
Thanks,

John 

[-- Attachment #2: sup-print --]
[-- Type: application/octet-stream, Size: 1963 bytes --]

#! /usr/bin/env python

import email
import sys
from subprocess import Popen,PIPE

# read in the email message from an arg or stdin
def get_message():
    try: 
        fp = open(sys.argv[1])
    except IndexError:
        fp = sys.stdin 
    return email.message_from_file(fp)

# open a pipe to muttprint
def open_muttprint():
    try:
        mutt = Popen("muttprint",shell=True,stdin=PIPE)
    except OSError, e:
        print "popen muttprint failed:", e
        sys.exit()
    return mutt

# write all the headers to muttprint
def write_header( msg, mutt ):
    for key in msg.keys( ):
        val = msg.__getitem__(key)
        pipe_mutt( "%s: %s\n" % ( key, val ), mutt )
    pipe_mutt( "\n", mutt ) # one extra newline in case text abuts headers

# utility for piping text to mutt
# useful for debugging, just select which one 
def pipe_mutt( text, mutt ):
    mutt.stdin.write(text)
    #sys.stdout.write(text)

# write all the text or html parts to muttprint
# otherwise write an attachment notation
def write_part( part, mutt ):
    type = part.get_content_type()
    if type == "text/plain" or type == "text/html":
        command = "w3m -dump -T %s" % type
        try:
            ch = Popen(command,shell=True,stdin=PIPE,stdout=PIPE,stderr=PIPE)
        except OSError, e:
            print >>sys.stderr, "Execution failed:", e
            sys.exit()
        ch.stdin.write(str(part.get_payload()))
        ch.stdin.close()
        out = ch.stdout.read().strip()
        err = ch.stderr.read()
        ch.wait()
        if len(err):
            print "Error with w3m: %s" % err
            sys.exit()
        pipe_mutt( out + "\n", mutt )
    elif part.get_filename() is not None:
        pipe_mutt("# Attachment: %s (%s)\n" % (part.get_filename(),type),mutt)

if __name__ == '__main__':
    msg  = get_message( )
    mutt = open_muttprint()
    write_header( msg, mutt )
    for part in msg.walk(): write_part( part, mutt )
    mutt.stdin.close()

[-- Attachment #3: Type: text/plain, Size: 140 bytes --]

_______________________________________________
sup-talk mailing list
sup-talk@rubyforge.org
http://rubyforge.org/mailman/listinfo/sup-talk