fix for null source urls

[redakcja.git] / apps / cover / forms.py

diff --git a/apps/cover/forms.py b/apps/cover/forms.py
index e6ffdff..5cc316b 100755 (executable)
--- a/apps/cover/forms.py
+++ b/apps/cover/forms.py
@@ -3,13 +3,14 @@
 # This file is part of FNP-Redakcja, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
-import re
-from urllib2 import urlopen
 from django import forms
 from django.utils.translation import ugettext_lazy as _, ugettext
 from cover.models import Image
 from django.utils.text import mark_safe
 
+from cover.utils import get_flickr_data, FlickrError
+
+
 class ImageAddForm(forms.ModelForm):
     class Meta:
         model = Image
@@ -27,8 +28,8 @@ class ImageAddForm(forms.ModelForm):
                 pass
             else:
                 raise forms.ValidationError(mark_safe(
-                    ugettext('Image <a href="%(url)s">already in repository</a>.'
-                        ) % {'url': img.get_absolute_url()}))
+                    ugettext('Image <a href="%(url)s">already in repository</a>.')
+                    % {'url': img.get_absolute_url()}))
         return cl
 
     def clean(self):
@@ -37,6 +38,7 @@ class ImageAddForm(forms.ModelForm):
             raise forms.ValidationError('No image specified')
         return cleaned_data
 
+
 class ImageEditForm(forms.ModelForm):
     """Form used for editing a Book."""
     class Meta:
@@ -48,61 +50,23 @@ class ReadonlyImageEditForm(ImageEditForm):
     """Form used for not editing an Image."""
 
     def __init__(self, *args, **kwargs):
-        ret = super(ReadonlyImageEditForm, self).__init__(*args, **kwargs)
+        super(ReadonlyImageEditForm, self).__init__(*args, **kwargs)
         for field in self.fields.values():
             field.widget.attrs.update({"readonly": True})
-        return ret
 
     def save(self, *args, **kwargs):
-        raise AssertionError, "ReadonlyImageEditForm should not be saved."
+        raise AssertionError("ReadonlyImageEditForm should not be saved.")
 
 
 class FlickrForm(forms.Form):
     source_url = forms.URLField(label=_('Flickr URL'))
 
     def clean_source_url(self):
-        def normalize_html(html):
-            return re.sub('[\t\n]', '', html)
-    
         url = self.cleaned_data['source_url']
-        m = re.match(r'(https?://)?(www\.|secure\.)?flickr\.com/photos/(?P<author>[^/]+)/(?P<img>\d+)/?', url)
-        if not m:
-            raise forms.ValidationError("It doesn't look like Flickr URL.")
-        author_slug, img_id = m.group('author'), m.group('img')
-        base_url = "https://www.flickr.com/photos/%s/%s/" % (author_slug, img_id)
-
         try:
-            html = normalize_html(urlopen(url).read().decode('utf-8'))
-        except:
-            raise forms.ValidationError('Error reading page.')
-        match = re.search(r'<a href="([^"]*)" rel="license cc:license">Some rights reserved</a>', html)
-        try:
-            assert match
-            license_url = match.group(1)
-            self.cleaned_data['license_url'] = license_url
-            re_license = re.compile(r'https?://creativecommons.org/licenses/([^/]*)/([^/]*)/.*')
-            m = re_license.match(license_url)
-            assert m
-            self.cleaned_data['license_name'] = 'CC %s %s' % (m.group(1).upper(), m.group(2))
-        except AssertionError:
-            raise forms.ValidationError('Error reading license name.')
-
-        m = re.search(r'"ownername":"([^"]*)', html)
-        if m:
-            self.cleaned_data['author'] = "%s@Flickr" % m.group(1)
-        else:
-            raise forms.ValidationError('Error reading author name.')
-
-        m = re.search(r'<h1[^>]*>(.*?)</h1>', html)
-        if not m:
-            raise forms.ValidationError('Error reading image title.')
-        self.cleaned_data['title'] = m.group(1)
-
-        url_size = base_url + "sizes/o/"
-        html = normalize_html(urlopen(url_size).read().decode('utf-8'))
-        m = re.search(r'<div id="allsizes-photo">\s*<img src="([^"]*)"', html)
-        if m:
-            self.cleaned_data['download_url'] = m.group(1)
-        else:
-            raise forms.ValidationError('Error reading image URL.')
-        return base_url
+            flickr_data = get_flickr_data(url)
+        except FlickrError as e:
+            raise forms.ValidationError(e)
+        for field_name in ('license_url', 'license_name', 'author', 'title', 'download_url'):
+            self.cleaned_data[field_name] = flickr_data[field_name]
+        return flickr_data['source_url']