fix for null source urls

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

diff --git a/apps/cover/forms.py b/apps/cover/forms.py
index fab0bb6..5cc316b 100755 (executable)
--- a/apps/cover/forms.py
+++ b/apps/cover/forms.py
@@ -3,16 +3,42 @@
 # 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 re
-from urllib2 import urlopen

 from django import forms
 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 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
 
 
 class ImageAddForm(forms.ModelForm):
     class Meta:
         model = Image
 

+    def __init__(self, *args, **kwargs):
+        super(ImageAddForm, self).__init__(*args, **kwargs)
+        self.fields['file'].required = False
+
+    def clean_download_url(self):
+        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()
+        if not cleaned_data.get('download_url', None) and not cleaned_data.get('file', None):
+            raise forms.ValidationError('No image specified')
+        return cleaned_data
+
+

 class ImageEditForm(forms.ModelForm):
     """Form used for editing a Book."""
     class Meta:
 class ImageEditForm(forms.ModelForm):
     """Form used for editing a Book."""
     class Meta:


@@ -21,16 +47,15 @@ class ImageEditForm(forms.ModelForm):
 
 
 class ReadonlyImageEditForm(ImageEditForm):
 
 
 class ReadonlyImageEditForm(ImageEditForm):

-    """Form used for not editing a Book."""
+    """Form used for not editing an Image."""

 
     def __init__(self, *args, **kwargs):
 
     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})
         for field in self.fields.values():
             field.widget.attrs.update({"readonly": True})

-        return ret

 
     def save(self, *args, **kwargs):
 
     def save(self, *args, **kwargs):

-        raise AssertionError, "ReadonlyImageEditForm should not be saved."
+        raise AssertionError("ReadonlyImageEditForm should not be saved.")

 
 
 class FlickrForm(forms.Form):
 
 
 class FlickrForm(forms.Form):


@@ -38,44 +63,10 @@ class FlickrForm(forms.Form):
 
     def clean_source_url(self):
         url = self.cleaned_data['source_url']
 
     def clean_source_url(self):
         url = self.cleaned_data['source_url']

-        m = re.match(r'(https?://)?(www.)?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 = 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:
         try:

-            assert match
-            license_url = match.group(1)
-            self.cleaned_data['license_url'] = license_url
-            re_license = re.compile(r'http://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'<strong class="username">By <a href="[^"]*">([^<]*)</a></strong>', 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 = 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']