Convert evernote (enex) files to markdown

enex2md.py

#!/usr/bin/env python
# coding=utf-8
#
# convert evernote enex files to markdown
# depends on https://github.com/claytron/ever2simple
# assumes enex files are named as CATEG-foo-bar.enex

import datetime
import glob
import json
import os
import re
import subprocess

opath = os.path
ENEX_DIR = opath.expanduser('~/tmp/evernotes_enex')
DST_DIR = opath.expanduser('~/tmp/evernotes_md')


def tmp_filename(prefix='tmp_file'):
    return '{}.{}'.format(prefix, os.getpid())


def convert2json(fname):
    '''
    ever2simple -o /tmp/xxx -f json fname
    load json conten from /tmp/xxx
    remove /tmp/xxx
    '''
    tmp_file = opath.join('/tmp', tmp_filename('enex'))
    args = ['ever2simple', '-o', tmp_file, '-f json', fname]
    subprocess.check_call(args, shell=True)
    with open(tmp_file) as fp:
        content = json.load(fp)
    os.unlink(tmp_file)
    return content[0]


def parse_date(text):
    ''' convert 'May 24 2013 00:44:45' to datetime object
    '''
    return datetime.datetime.strptime(text, '%b %d %Y %H:%M:%S')


def date2yymmdd(date):
    return '{}{:02d}{:02d}'.format(date.year, date.month, date.day)


def tags2str(tags):
    tt = ', '.join(map(lambda x: '"' + x + '"', tags))
    return '[ {} ]'.format(tt)


def make_header(title, cdate, categ, tags):
        txt  = '+++\n'
        txt += 'title = "{}"\n'.format(title)
        txt += 'date = "{}"\n'.format(cdate.isoformat())
        txt += 'categories = [ "{}" ]\n'.format(categ)
        txt += 'tags = {}\n'.format(tags2str(tags))
        txt += '\n'
        txt += '+++\n\n'
        return txt


def make_body(content):
    body = re.sub(r'\ \ \n', '\n', content)
    body = re.sub(r'\n\*\*\n', '\n\n', body)
    body = re.sub(r'\n\n+', '\n\n', body)
    body = re.sub(r'&', '\&', body)
    body = re.sub(r'(-)*\>', '`-->`', body)
    body = re.sub(r'\&lt;(-)*', '`<--`', body)
    return body.encode('utf-8')


def save_md(json_data, name, dest='.'):
    if not opath.isdir(dest):
        os.mkdir(dest)
    name_parts = name.split('-')
    title = ' '.join(name_parts[1:]).capitalize()
    categ = name_parts[0]
    tags = json_data['tags']
    if categ in tags and len(tags) > 1:
        tags.remove(categ)
    cdate = parse_date(json_data['createdate'])
    file_name = date2yymmdd(cdate) + '-' + name + '.md'
    full_name = opath.join(dest, file_name)
    with open(full_name, 'w') as fp:
        fp.write(make_header(title, cdate, categ, tags))
        fp.write(make_body(json_data['content']))


list_of_enex = glob.glob(opath.join(ENEX_DIR, '*.enex'))

for each_file in list_of_enex:
    json_data = convert2json(each_file)
    base_name, _ext = opath.splitext(opath.basename(each_file))
    save_md(json_data, base_name, dest=DST_DIR)