kadin2048/pidgintoeml.py

## pidgintoeml.py
#!/usr/bin/env python

#  Turn an Pidgin HTML chatlog into a Thunderbird-compatible .eml file
#  so that it can be imported into Gmail for archive purposes.
#
#  Syntax: $ python pidgintoeml.py pidginlogfile.html [outputfilename.eml]
#
#  Version: 2011-09-28
#

import sys
import datetime
import os.path

from email.MIMEMultipart import MIMEMultipart
from email.MIMEText import MIMEText

# Debug flag; set to False to suppress information
global debug
debug = False

def main():
    # Use the first argument as the input file
    try:
        if sys.argv[1]:
            filename = sys.argv[1]
    except IndexError:
        sys.stderr.write("No input file specified.\n")
        return 1

    # Second arg, if present, is the ouput file
    try:
        outfilename = sys.argv[2]
        # Note that this will throw IndexError if not present
    except IndexError:
        # which we catch here
        outfilename = filename + '.eml'

    # DEBUG
    if debug:
        print "-- Reading from " + filename
        print "-- Writing to " + outfilename

    # Test to see if the output file already exists (processed already)
    if os.path.isfile(outfilename):
        sys.stderr.write("Output file " + outfilename + " already exists. Terminating.\n")
        return 1

    # Open the in and out files
    try:
        fi = open(filename, 'r')   # fi is a file object
        fo = open(outfilename, 'w')
    except IOError:
        sys.stderr.write("IO Error while opening files.\n")
        return 1

    # Create a message object
    msg_base = MIMEMultipart('mixed')

    if filename.split('.')[-1] == 'html':
        # For probable Pidgin logs (ending in .html)...

        # Process the first line of the input file to determine the eml headers
        determineHTMLLogHeaders( fi.readline(), msg_base )
        fi.seek(0) # reset file object

        if debug:
            print "-- Headers after parsing first line are..."
            for key, value in msg_base.items():
                print key + ": " + value

        # Create the HTML payload using the entire file
        doctype = '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">\n'
        ht = doctype + fi.read()#.encode('utf-8')

        if debug:
            print "-- HTML body is of type:"
            print type(ht)

        msghtml = MIMEText(ht, 'html')#, 'utf-8')

        # Attach the HTML to the root
        msg_base.attach(msghtml)

    if filename.split('.')[-1] == 'chatlog':
        # For XML formatted chat logs
        sys.stderr.write("XML chatlog processing not implemented.  Terminating.\n")
        return 1

    #if filename.split('.')[-1] ==

    if debug:
        print "-- Ready to flatten and write message..."

    # Write out the message
    fo.write( msg_base.as_string() )

    if debug:
        print "-- Complete."

    return 0


def determineHTMLLogHeaders(firstline, msg_base):
    # Start by just looking at the <title> element
    title = firstline[firstline.find("<title>")+7:firstline.find("</title>")]

    if debug:
        print "<title>: " + title

    # Determine the 'From' address of the chat
    #  TODO: This would be better done with a regexp but I was lazy
    msg_base['From'] = title[title.find("Conversation with ")+18:title.find(" at ")]

    if debug:
        print "-- From is: " + msg_base['From']

    # Determine the 'To' address
    msg_base['To'] = title[title.find(" on ")+4:]

    if debug:
        print "-- To is: " + msg_base['To']

    # Now we have to deal with the date.  This is messy.
    logdate = title[title.find(" at ")+4:title.find(" on ")]
    # Turn it into a datetime object
    d = datetime.datetime.strptime(logdate, '%m/%d/%Y %I:%M:%S %p')
    # Then write it out to RFC822 format
    #  TODO: This is a naive/stupid way of handling timezone!
    msg_base['Date'] = d.strftime("%a, %d %b %Y %H:%M:%S" + " -0500 (EST)")

    # And the message subject
    msg_base['Subject'] = title[0:title.find(" on ")]


if __name__ == "__main__":
    sys.exit( main() )  # program return value is main()'s return value
	#!/usr/bin/env python

	# Turn an Pidgin HTML chatlog into a Thunderbird-compatible .eml file
	# so that it can be imported into Gmail for archive purposes.
	#
	# Syntax: $ python pidgintoeml.py pidginlogfile.html [outputfilename.eml]
	#
	# Version: 2011-09-28
	#

	import sys
	import datetime
	import os.path

	from email.MIMEMultipart import MIMEMultipart
	from email.MIMEText import MIMEText

	# Debug flag; set to False to suppress information
	global debug
	debug = False

	def main():
	# Use the first argument as the input file
	try:
	if sys.argv[1]:
	filename = sys.argv[1]
	except IndexError:
	sys.stderr.write("No input file specified.\n")
	return 1

	# Second arg, if present, is the ouput file
	try:
	outfilename = sys.argv[2]
	# Note that this will throw IndexError if not present
	except IndexError:
	# which we catch here
	outfilename = filename + '.eml'

	# DEBUG
	if debug:
	print "-- Reading from " + filename
	print "-- Writing to " + outfilename

	# Test to see if the output file already exists (processed already)
	if os.path.isfile(outfilename):
	sys.stderr.write("Output file " + outfilename + " already exists. Terminating.\n")
	return 1

	# Open the in and out files
	try:
	fi = open(filename, 'r') # fi is a file object
	fo = open(outfilename, 'w')
	except IOError:
	sys.stderr.write("IO Error while opening files.\n")
	return 1

	# Create a message object
	msg_base = MIMEMultipart('mixed')

	if filename.split('.')[-1] == 'html':
	# For probable Pidgin logs (ending in .html)...

	# Process the first line of the input file to determine the eml headers
	determineHTMLLogHeaders( fi.readline(), msg_base )
	fi.seek(0) # reset file object

	if debug:
	print "-- Headers after parsing first line are..."
	for key, value in msg_base.items():
	print key + ": " + value

	# Create the HTML payload using the entire file
	doctype = '<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">\n'
	ht = doctype + fi.read()#.encode('utf-8')

	if debug:
	print "-- HTML body is of type:"
	print type(ht)

	msghtml = MIMEText(ht, 'html')#, 'utf-8')

	# Attach the HTML to the root
	msg_base.attach(msghtml)

	if filename.split('.')[-1] == 'chatlog':
	# For XML formatted chat logs
	sys.stderr.write("XML chatlog processing not implemented. Terminating.\n")
	return 1

	#if filename.split('.')[-1] ==

	if debug:
	print "-- Ready to flatten and write message..."

	# Write out the message
	fo.write( msg_base.as_string() )

	if debug:
	print "-- Complete."

	return 0


	def determineHTMLLogHeaders(firstline, msg_base):
	# Start by just looking at the <title> element
	title = firstline[firstline.find("<title>")+7:firstline.find("</title>")]

	if debug:
	print "<title>: " + title

	# Determine the 'From' address of the chat
	# TODO: This would be better done with a regexp but I was lazy
	msg_base['From'] = title[title.find("Conversation with ")+18:title.find(" at ")]

	if debug:
	print "-- From is: " + msg_base['From']

	# Determine the 'To' address
	msg_base['To'] = title[title.find(" on ")+4:]

	if debug:
	print "-- To is: " + msg_base['To']

	# Now we have to deal with the date. This is messy.
	logdate = title[title.find(" at ")+4:title.find(" on ")]
	# Turn it into a datetime object
	d = datetime.datetime.strptime(logdate, '%m/%d/%Y %I:%M:%S %p')
	# Then write it out to RFC822 format
	# TODO: This is a naive/stupid way of handling timezone!
	msg_base['Date'] = d.strftime("%a, %d %b %Y %H:%M:%S" + " -0500 (EST)")

	# And the message subject
	msg_base['Subject'] = title[0:title.find(" on ")]



	if __name__ == "__main__":
	sys.exit( main() ) # program return value is main()'s return value