fix for bad documents.

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

diff --git a/src/cover/utils.py b/src/cover/utils.py
old mode 100755 (executable)
new mode 100644 (file)
index 51aee19..6815d29
--- a/src/cover/utils.py
+++ b/src/cover/utils.py
@@ -1,19 +1,21 @@
-# -*- coding: utf-8 -*-
-#

 # This file is part of FNP-Redakcja, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 # This file is part of FNP-Redakcja, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #

+import csv
+from io import StringIO

 import json
 import re
 import json
 import re

-from urllib import FancyURLopener
-
-from django.contrib.sites.models import Site
+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
     def version(self):
 
 
 class URLOpener(FancyURLopener):
     @property
     def version(self):

-        return 'FNP Redakcja (http://%s)' % Site.objects.get_current()
+        return 'FNP Redakcja'

 
 
 class FlickrError(Exception):
 
 
 class FlickrError(Exception):


@@ -71,3 +73,120 @@ def get_flickr_data(url):
         'title': title,
         'download_url': download_url,
     }
         'title': title,
         'download_url': download_url,
     }

+
+
+def get_wikimedia_data(url):
+    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']
+
+    meta = {
+        'title': ext['ObjectName']['value'],
+        'author': d['user'],
+        'source_url': d['descriptionurl'],
+        'download_url': d['url'],
+        '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))
+        if author is not None:
+            meta['author'] = author.label.get('pl', str(author.label))
+        else:
+            meta['author'] = ''
+
+    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):
+    nr = url.rsplit('/', 1)[-1]
+    d = list(
+        csv.DictReader(
+            StringIO(
+                URLOpener().open(
+                    'https://cyfrowe-api.mnw.art.pl/api/object/{}/csv'.format(nr)
+                ).read().decode('utf-8')
+            )
+        )
+    )[0]
+
+    authors = []
+    i = 1
+    while f'twórca/wytwórnia {i}' in d:
+        authors.append(d[f'twórca/wytwórnia {i}'])
+        i += 1
+
+    license_url = ''
+    license_name = d['klasyfikacja praw autorskich 1']
+    if license_name == 'DOMENA PUBLICZNA':
+        license_name = 'domena publiczna'
+        license_url = 'https://pl.wikipedia.org/wiki/Domena_publiczna'
+        
+    return {
+        'title': d['nazwa/tytuł'],
+        'author': ', '.join(authors),
+        'source_url': url,
+        'download_url': 'https://cyfrowe-cdn.mnw.art.pl/upload/multimedia/{}.{}'.format(
+            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):
+        return get_flickr_data(url)
+    if re.match(r'(https?://)?(commons|upload)\.wikimedia\.org/', 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)