ISO metadata keywords analyzer

iso_keywords_analyzer.py

# scan a directory of ISO metadata files
# and generate a keyword frequency count

from glob import glob
import os
from pprint import pprint
import sys

from lxml import etree
from owslib.iso import MD_Metadata

if len(sys.argv) < 2:
    print('Usage: %s <dirpath>' % sys.argv[0])
    sys.exit(1)

keyword_counts = {}

for md_file in glob('%s%s*.xml' % (sys.argv[1], os.sep)):
    md = MD_Metadata(etree.parse(md_file))
    for keywords in md.identification.keywords:
        for keyword in keywords['keywords']:
            if keyword in keyword_counts:
                keyword_counts[keyword] += 1
            else:
                keyword_counts[keyword] = 1

print(pprint(sorted(keyword_counts.items(),
                    key=lambda x: x[1],
                    reverse=True)))