dir2rdf.py

#!/usr/bin/python

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

try:
  from urllib.parse import quote
except ImportError:
  from urllib import quote

ns = {
 'rdf': 'http://www.w3.org/1999/02/22-rdf-syntax-ns#',
 'z': 'http://www.zotero.org/namespaces/export#',
 'dc': 'http://purl.org/dc/elements/1.1/',
 'link': 'http://purl.org/rss/1.0/modules/link/',
 'dcterms': 'http://purl.org/dc/terms/',
}
for prefix, url in ns.items():
  ET.register_namespace(prefix, url)

extensions = ['.' + ext for ext in sys.argv[1:]]
print(extensions)

def elementID(path):
  return '#' + quote(path, safe='')

rdf = ET.Element('{%s}RDF' % (ns['rdf'],))
for path, dirs, files in os.walk("."):
  path = path[2:] # removes './'

  if path != '':
    coll = ET.SubElement(rdf, '{%s}Collection' % (ns['z'],),)
    coll.set('{%s}about' % (ns['rdf'],), elementID(path))

    ET.SubElement(coll, '{%s}title' % (ns['dc'],)).text = os.path.basename(path)

    elt = ET.SubElement(coll, '{%s}type' % (ns['rdf'],))
    elt.set('{%s}resource' % (ns['rdf'],), 'http://www.zotero.org/namespaces/export#Collection')

    for d in dirs:
      elt = ET.SubElement(coll, '{%s}hasPart' % (ns['dcterms'],))
      elt.set('{%s}resource' % (ns['rdf'],), elementID(os.path.join(path, d)))

    for f in files:
      if not os.path.splitext(f)[1] in extensions: continue
      elt = ET.SubElement(coll, '{%s}hasPart' % (ns['dcterms'],))
      elt.set('{%s}resource' % (ns['rdf'],), elementID(os.path.join(path, f)))

  for f in files:
    if not os.path.splitext(f)[1] in extensions: continue
    att = ET.SubElement(rdf, '{%s}Attachment' % (ns['z'],))
    att.set('{%s}about' % (ns['rdf'],), elementID(os.path.join(path, f)))

    ET.SubElement(att, '{%s}itemType' % (ns['z'],)).text = 'attachment'

    elt = ET.SubElement(att, '{%s}resource' % (ns['rdf'],))
    elt.set('{%s}resource' % (ns['rdf'],), os.path.join(path, f))

    ET.SubElement(att, '{%s}title' % (ns['dc'],)).text = f
    #ET.SubElement(att, '{%s}type' % (ns['link'],)).text = 

with open('attachments.rdf', 'wb') as f:
  ET.ElementTree(rdf).write(f)