import json
import re
from urllib.request import FancyURLopener
+from wikidata.client import Client
+from catalogue.constants import WIKIDATA
+
class URLOpener(FancyURLopener):
@property
>>> get_wikimedia_data('https://commons.wikimedia.org/wiki/File:Valdai_IverskyMon_asv2018_img47.jpg')
{'title': 'Valdai IverskyMon asv2018 img47', 'author': 'A.Savin', 'source_url': 'https://commons.wikimedia.org/wiki/File:Valdai_IverskyMon_asv2018_img47.jpg', 'download_url': 'https://upload.wikimedia.org/wikipedia/commons/4/43/Valdai_IverskyMon_asv2018_img47.jpg', 'license_url': 'http://artlibre.org/licence/lal/en', 'license_name': 'FAL'}
+ >>> get_wikimedia_data('https://commons.wikimedia.org/wiki/File:Pymonenko_A_boy_in_a_straw_hat.jpg')
+ {'title': 'Chłopiec w słomkowym kapeluszu', 'author': 'Mykola Pymonenko', 'source_url': 'https://commons.wikimedia.org/wiki/File:Pymonenko_A_boy_in_a_straw_hat.jpg', 'download_url': 'https://upload.wikimedia.org/wikipedia/commons/9/9b/Pymonenko_A_boy_in_a_straw_hat.jpg', 'license_url': 'https://pl.wikipedia.org/wiki/Domena_publiczna', 'license_name': 'domena publiczna'}
+
"""
file_name = url.rsplit('/', 1)[-1].rsplit(':', 1)[-1]
d = json.loads(URLOpener().open('https://commons.wikimedia.org/w/api.php?action=query&titles=File:{}&prop=imageinfo&iiprop=url|user|extmetadata&iimetadataversion=latest&format=json'.format(file_name)).read().decode('utf-8'))
+
d = list(d['query']['pages'].values())[0]['imageinfo'][0]
ext = d['extmetadata']
'author': d['user'],
'source_url': d['descriptionurl'],
'download_url': d['url'],
- 'license_url': ext['LicenseUrl']['value'],
+ 'license_url': ext.get('LicenseUrl', {}).get('value', ''),
'license_name': ext['LicenseShortName']['value'],
}
-
+
+ # There are Wikidata links in ObjectName sametimes. Let's use it.
+ wikidata_match = re.search(r'wikidata\.org/wiki/(Q\d+)', meta['title'])
+ if wikidata_match is not None:
+ qitem = wikidata_match.group(1)
+ client = Client()
+ entity = client.get(qitem)
+ meta['title'] = entity.label.get('pl', str(entity.label))
+ author = entity.get(client.get(WIKIDATA.CREATOR))
+ meta['author'] = author.label.get('pl', str(author.label))
+
+ if meta['license_name'] == 'Public domain':
+ meta['license_name'] = 'domena publiczna'
+ meta['license_url'] = 'https://pl.wikipedia.org/wiki/Domena_publiczna'
+
+
return meta
def get_mnw_data(url):
"""
- >>> get_mnw_data('https://cyfrowe.mnw.art.pl/pl/katalog/794032')
- {'title': 'Pejzaż z podwójnym świerkiem', 'author': 'nieznany, Altdorfer, Albrecht (ca 1480-1538)', 'source_url': 'https://cyfrowe.mnw.art.pl/pl/katalog/794032', 'download_url': 'https://cyfrowe-cdn.mnw.art.pl/upload/multimedia/49/58/49583b3e9b23e2d25f372fe6021ae220.jpg', 'license_url': 'https://pl.wikipedia.org/wiki/Domena_publiczna', 'license_name': 'DOMENA PUBLICZNA'}
-
+ >>> get_mnw_data('https://cyfrowe.mnw.art.pl/pl/katalog/511078')
+ {'title': 'Chłopka (Baba ukraińska)', 'author': 'Krzyżanowski, Konrad (1872-1922)', 'source_url': 'https://cyfrowe.mnw.art.pl/pl/katalog/511078', 'download_url': 'https://cyfrowe-cdn.mnw.art.pl/upload/multimedia/32/60/3260ae1704cc530cc62befa9b7d58cbd.jpg', 'license_url': 'https://pl.wikipedia.org/wiki/Domena_publiczna', 'license_name': 'domena publiczna'}
"""
nr = url.rsplit('/', 1)[-1]
d = list(
license_name = d['copyrights.0.name']
if license_name == 'DOMENA PUBLICZNA':
license_name = 'domena publiczna'
- license_url = ''
+ license_url = 'https://pl.wikipedia.org/wiki/Domena_publiczna'
return {
'title': d['title'],