librarian update

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

diff --git a/apps/cover/forms.py b/apps/cover/forms.py
index 67a0c53..e4c949c 100755 (executable)
--- a/apps/cover/forms.py
+++ b/apps/cover/forms.py
@@ -3,11 +3,13 @@
 # This file is part of FNP-Redakcja, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
+import json
 import re
 from urllib2 import urlopen
 from django import forms
-from django.utils.translation import ugettext_lazy as _
+from django.utils.translation import ugettext_lazy as _, ugettext
 from cover.models import Image
+from django.utils.text import mark_safe
 
 class ImageAddForm(forms.ModelForm):
     class Meta:
@@ -15,10 +17,20 @@ class ImageAddForm(forms.ModelForm):
 
     def __init__(self, *args, **kwargs):
         super(ImageAddForm, self).__init__(*args, **kwargs)
-        self.fields['file'].required = self.fields['download_url'].required = False
+        self.fields['file'].required = False
 
     def clean_download_url(self):
-        return self.cleaned_data['download_url'] or None
+        cl = self.cleaned_data['download_url'] or None
+        if cl is not None:
+            try:
+                img = Image.objects.get(download_url=cl)
+            except Image.DoesNotExist:
+                pass
+            else:
+                raise forms.ValidationError(mark_safe(
+                    ugettext('Image <a href="%(url)s">already in repository</a>.'
+                        ) % {'url': img.get_absolute_url()}))
+        return cl
 
     def clean(self):
         cleaned_data = super(ImageAddForm, self).clean()
@@ -26,7 +38,6 @@ 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:
@@ -35,7 +46,7 @@ class ImageEditForm(forms.ModelForm):
 
 
 class ReadonlyImageEditForm(ImageEditForm):
-    """Form used for not editing a Book."""
+    """Form used for not editing an Image."""
 
     def __init__(self, *args, **kwargs):
         ret = super(ReadonlyImageEditForm, self).__init__(*args, **kwargs)
@@ -52,6 +63,7 @@ class FlickrForm(forms.Form):
 
     def clean_source_url(self):
         def normalize_html(html):
+            return html
             return re.sub('[\t\n]', '', html)
     
         url = self.cleaned_data['source_url']
@@ -65,34 +77,33 @@ class FlickrForm(forms.Form):
             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)
+        match = re.search(r'<a href="([^"]*)"[^>]* rel="license ', html)
         try:
             assert match
             license_url = match.group(1)
             self.cleaned_data['license_url'] = license_url
-            re_license = re.compile(r'http://creativecommons.org/licenses/([^/]*)/([^/]*)/.*')
+            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)
+        m = re.search(r'<a[^>]* class="owner-name [^>]*>([^<]*)<', 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)
+        m = re.search(r'<h1[^>]*>(.*?)</h1>', html, re.S)
         if not m:
             raise forms.ValidationError('Error reading image title.')
-        self.cleaned_data['title'] = m.group(1)
+        self.cleaned_data['title'] = m.group(1).strip()
 
-        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:
+        m = re.search(r'modelExport: (\{.*\})', html)
+        try:
+            assert m
+            self.cleaned_data['download_url'] = 'https:' + json.loads(m.group(1))['photo-models'][0]['sizes']['o']['url']
+        except (AssertionError, ValueError, IndexError, KeyError):
             raise forms.ValidationError('Error reading image URL.')
         return base_url