Automatically search for the next match after single replace, fixes #3384

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

diff --git a/apps/cover/forms.py b/apps/cover/forms.py
index 754d697..e6ffdff 100755 (executable)
--- a/apps/cover/forms.py
+++ b/apps/cover/forms.py
@@ -6,12 +6,37 @@
 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
 
 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:
@@ -20,7 +45,7 @@ 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)
@@ -33,18 +58,21 @@ class ReadonlyImageEditForm(ImageEditForm):
 
 
 class FlickrForm(forms.Form):
-    source_url = forms.URLField()
+    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.)?flickr.com/photos/(?P<author>[^/]+)/(?P<img>\d+)', 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 = urlopen(url).read().decode('utf-8')
+            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)
@@ -52,14 +80,14 @@ class FlickrForm(forms.Form):
             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'<strong class="username">By <a href="[^"]*">([^<]*)</a></strong>', html)
+        m = re.search(r'"ownername":"([^"]*)', html)
         if m:
             self.cleaned_data['author'] = "%s@Flickr" % m.group(1)
         else:
@@ -71,7 +99,7 @@ class FlickrForm(forms.Form):
         self.cleaned_data['title'] = m.group(1)
 
         url_size = base_url + "sizes/o/"
-        html = urlopen(url_size).read().decode('utf-8')
+        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)