Convert Remember the Milk tasks feed to Emacs org-mode file

rtm2org.py

#!/usr/bin/env python

# rtm2org.py
#
# Convert Remember the Milk tasks feed to Emacs org-mode file
#
# requires Python >= 2.7 (for ElementTree.iter())
#
# Find your RTM Atom feed here:
#   https://www.rememberthemilk.com/atom/<username>
#
# Assumes you organize your RTM todos by tags, not by lists (should be
# easily modifiable to do the latter).  Outputs an org-mode hierarchy
# organized by first tag, like so:
#
#     * <first tag> :<first tag>:
#     ** TODO [<priority] Item 1
#     ** TODO Item 2 :<othertags>:
#     ** TODO Item 3
#     <note title>
#     <note content>

from collections import defaultdict

import os.path
import sys
import xml.etree.ElementTree as ET

_atom_ns  = 'http://www.w3.org/2005/Atom'
_xhtml_ns = 'http://www.w3.org/1999/xhtml'

_priority_map = {'none': None, '1': 'A', '2': 'B', '3': 'C'}

class Rtm2Org(object):

    def __init__(self):
        self._by_tag = defaultdict(list)

    def parseAtom(self, source):
        tree = ET.parse(source)
        for entry in tree.iter('{%s}entry' % _atom_ns):
            title = entry.find('{%s}title' % _atom_ns).text
            priority, tags, note = None, None, None
            for span in entry.iter('{%s}span' % _xhtml_ns):
                if span.get('class') == 'rtm_priority_value':
                    priority = _priority_map.get(span.text, span.text)
                elif span.get('class') == 'rtm_tags_value':
                    if tags == 'none':
                        tags = []
                    else:
                        tags = [x.strip() for x in span.text.split(',')]
            for div in entry.iter('{%s}div' % _xhtml_ns):
                # First line of a note appears in one XML element as the
                # note's "title", and the remainder appears in a second
                # XML element as the note's "content", so concatenate them
                # BUG: I think RTM supports more than one note per item;
                #      should really put them all in a list and output
                #      as Org subitems or plain list
                if div.get('class') == 'rtm_note_title_container':
                    span = div.find('{%s}span' % _xhtml_ns)
                    note = span.text
                elif div.get('class') == 'rtm_note_content':
                    note += '\n' + div.text
            self._by_tag[tags[0]].append((title, priority, tags, note))

    def writeOrg(self, stream=sys.stdout):
        print >> stream, '#+STARTUP: content indent hidestars'
        for tag in sorted(self._by_tag.keys()):
            print >> stream
            print >> stream, '* %s :%s:' % (tag, tag)
            for title, priority, tags, note in self._by_tag[tag]:
                priority = ' [#%s]' % priority if priority else ''
                tags     = filter(lambda x: x != tag, tags)
                tags     = ' :%s:' % ':'.join(tags) if tags else ''
                print >> stream, '** TODO%s %s%s' % (priority, title, tags)
                if note:
                    print >> stream, note

def main():
    args = sys.argv[1:]
    if len(args) != 1:
        print >> sys.stderr, \
            "usage: %s <file>" % os.path.basename(sys.argv[0])
        sys.exit(1)
    rtm2org = Rtm2Org()
    rtm2org.parseAtom(args[0])
    rtm2org.writeOrg()

if __name__ == '__main__':
    main()