Catalogue improvements.

[redakcja.git] / src / cover / utils.py

diff --git a/src/cover/utils.py b/src/cover/utils.py
index 3bf3e9f..56fb24a 100644 (file)
--- a/src/cover/utils.py
+++ b/src/cover/utils.py
@@ -6,6 +6,11 @@ from io import StringIO
 import json
 import re
 from urllib.request import FancyURLopener
+from django.conf import settings
+import requests
+from wikidata.client import Client
+from catalogue.constants import WIKIDATA
+
 
 class URLOpener(FancyURLopener):
     @property
@@ -71,13 +76,9 @@ def get_flickr_data(url):
 
 
 def get_wikimedia_data(url):
-    """
-    >>> 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'}
-
-    """
     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']
 
@@ -86,19 +87,29 @@ def get_wikimedia_data(url):
         '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'}
-
-    """
     nr = url.rsplit('/', 1)[-1]
     d = list(
         csv.DictReader(
@@ -111,29 +122,61 @@ def get_mnw_data(url):
     )[0]
 
     authors = []
-    i = 0
-    while f'authors.{i}.name' in d:
-        authors.append(d[f'authors.{i}.name'])
+    i = 1
+    while f'twórca/wytwórnia {i}' in d:
+        authors.append(d[f'twórca/wytwórnia {i}'])
         i += 1
 
-    license_url = d['copyrights.0.link']
-    license_name = d['copyrights.0.name']
+    license_url = ''
+    license_name = d['klasyfikacja praw autorskich 1']
     if license_name == 'DOMENA PUBLICZNA':
         license_name = 'domena publiczna'
-        license_url = ''
+        license_url = 'https://pl.wikipedia.org/wiki/Domena_publiczna'
         
     return {
-        'title': d['title'],
+        'title': d['nazwa/tytuł'],
         'author': ', '.join(authors),
         'source_url': url,
         'download_url': 'https://cyfrowe-cdn.mnw.art.pl/upload/multimedia/{}.{}'.format(
-            d['image.filePath'],
-            d['image.extension'],
+            d['ścieżka wizerunku'],
+            d['rozszerzenie pliku wizerunku'],
         ),
         'license_url': license_url,
         'license_name': license_name,
     }
 
+def get_rawpixel_data(url):
+    photo_id = re.search(r'/(\d+)/', url).group(1)
+
+    s = requests.Session()
+    cookies = settings.RAWPIXEL_SESSION
+
+    token = s.post(
+            'https://www.rawpixel.com/api/v1/user/session',
+            cookies=cookies
+            ).json()['token']
+
+    h = {'X-CSRF-Token': token, 'Accept': 'application/json'}
+
+    data = s.get(
+        f'https://www.rawpixel.com/api/v1/image/data/{photo_id}',
+        headers=h, cookies=cookies).json()
+    download_url = s.post(
+        f'https://www.rawpixel.com/api/v1/image/download/{photo_id}/original',
+        headers=h, cookies=cookies
+    ).json()['downloadUrl']
+
+    title = data['metadata']['title'].rsplit('|', 1)[0].strip()
+
+    return {
+        'title': title,
+        'author': ', '.join(data['metadata']['artist_names']),
+        'source_url': data['url'],
+        'download_url': download_url,
+        'license_url': data['metadata']['licenseUrl'],
+        'license_name': data['metadata']['license'],
+    }
+
 
 def get_import_data(url):
     if re.match(r'(https?://)?(www\.|secure\.)?flickr\.com/', url):
@@ -142,3 +185,5 @@ def get_import_data(url):
         return get_wikimedia_data(url)
     if re.match(r'(https?://)?cyfrowe\.mnw\.art\.pl/', url):
         return get_mnw_data(url)
+    if re.match(r'(https?://)?www\.rawpixel\.com/', url):
+        return get_rawpixel_data(url)