def call_api(action, cache=True, **kwargs):
""" Invoke the given method of wikidata APIs with the given parameters
"""
kwargs['format'] = 'json'
kwargs['action'] = action
resp = io.get_and_cache(WIKIDATA_API_URL, use_cache=cache, params=kwargs)
return json.loads(resp)
def call_api(action, cache=True, **kwargs):
""" Invoke the given method of wikidata APIs with the given parameters
"""
kwargs['format'] = 'json'
kwargs['action'] = action
resp = io.get_and_cache(WIKIDATA_API_URL, use_cache=cache, params=kwargs)
return json.loads(resp)
def main(out_file, use_cache, step):
finished = False
offset = 0
person = {}
while not finished:
print 'processed %d records' % offset
url = query_url.format(limit=step, offset=offset)
result = json.loads(get_and_cache(url, use_cache))
offset += step
finished = len(result['results']['bindings']) < step
# adjacent records can refer to the same person, so group all properties in sets
for i, data in enumerate(result['results']['bindings']):
url = data.pop('person')['value']
if person and person['url'] != url:
out_file.write(serialize_person(person))
out_file.write('\n')
person = {}
person['url'] = url
for key, value in data.iteritems():
value = value['value']
if key == 'gender':
value = value.split('/')[-1]
if key in person:
person[key].add(value)
else:
person[key] = {value}
def who_was_who(ctx):
out_file = ctx.obj.pop('out_file')
use_cache = ctx.obj.pop('cache')
url = 'https://archive.org/download/whowaswhocompani01londuoft/' \
'whowaswhocompani01londuoft_djvu.txt'
text = get_and_cache(url, use_cache)
parse_and_save(text, ur'([A-Z]+, ([0-9A-Z][. \-a-z]+)+),[^,]+[,;]', out_file, url)
def who_was_who(ctx):
out_file = ctx.obj.pop('out_file')
use_cache = ctx.obj.pop('cache')
url = 'https://archive.org/download/whowaswhocompani01londuoft/' \
'whowaswhocompani01londuoft_djvu.txt'
text = get_and_cache(url, use_cache)
parse_and_save(text, ur'([A-Z]+, ([0-9A-Z][. \-a-z]+)+),[^,]+[,;]',
out_file, url)
def american_bio(ctx):
out_file = ctx.obj.pop('out_file')
use_cache = ctx.obj.pop('cache')
for volume in xrange(1, 11):
print 'Volume', volume
volume_url = 'https://archive.org/download/biographicaldict{volume:02d}johnuoft/' \
'biographicaldict{volume:02d}johnuoft_djvu.txt'.format(volume=volume)
vol = get_and_cache(volume_url, use_cache)
parse_and_save(vol, ur'([A-Z]+, [A-Z][a-z]+ ?),[^,]+,', out_file, volume_url)
def american_bio(ctx):
out_file = ctx.obj.pop('out_file')
use_cache = ctx.obj.pop('cache')
for volume in xrange(1, 11):
print 'Volume', volume
volume_url = 'https://archive.org/download/biographicaldict{volume:02d}johnuoft/' \
'biographicaldict{volume:02d}johnuoft_djvu.txt'.format(volume=volume)
vol = get_and_cache(volume_url, use_cache)
parse_and_save(vol, ur'([A-Z]+, [A-Z][a-z]+ ?),[^,]+,', out_file,
volume_url)
def wikidata_id_from_wikipedia_url(wiki_url):
title = urlparse(wiki_url).path[len('/wiki/'):]
data = json.loads(io.get_and_cache(
'https://en.wikipedia.org/w/api.php?action=query&prop=pageprops&format=json&titles=' + title
))
ids = [
page['pageprops']['wikibase_item']
for pid, page in data['query']['pages'].iteritems()
if pid >= 0 and 'wikibase_item' in page.get('pageprops', {})
]
if not ids:
logger.debug('failed to reconcile uri %s with a wikidata page')
return None
elif len(ids) > 1:
logger.debug('uri %s was reconciled to items %s, picking the first one',
', '.join(ids))
return ids[0]
def wikidata_id_from_wikipedia_url(wiki_url):
title = urlparse(wiki_url).path[len('/wiki/'):]
data = json.loads(
io.get_and_cache(
'https://en.wikipedia.org/w/api.php?action=query&prop=pageprops&format=json&titles='
+ title))
ids = [
page['pageprops']['wikibase_item']
for pid, page in data['query']['pages'].iteritems()
if pid >= 0 and 'wikibase_item' in page.get('pageprops', {})
]
if not ids:
logger.debug('failed to reconcile uri %s with a wikidata page')
return None
elif len(ids) > 1:
logger.debug(
'uri %s was reconciled to items %s, picking the first one',
', '.join(ids))
return ids[0]
