Fix resource updating for harvested datasets

Starting from 2.2, resource_update calls package_show before updating
the resource via a package_update call. The dict passed had the harvest
extras (eg harvest_object_id) added which made the update call fails due
to duplicated extra keys. To fix it we now remove any harvest extras
on after_show if there is a 'for_edit' property on the context.

ckanext/harvest/plugin.py

@@ -95,7 +95,21 @@ class Harvest(p.SingletonPlugin, DefaultDatasetForm):
                     .filter(HarvestObject.current==True) \
                     .first()
 
-            # validate is false is passed only on indexing.
+            if context.get('for_edit'):
+                # If we are editing the dataset, check if the harvest extras
+                # are there, and if so, remove them. This can happen eg when
+                # doing resource_update, which calls package_show
+                data_dict['extras'][:] = [e for e in data_dict.get('extras', [])
+                                          if not e['key']
+                                          in ('harvest_object_id', 'harvest_source_id', 'harvest_source_title',)]
+
+
+            # We only want to add these extras at index time so they are part
+            # of the cached data_dict used to display, search results etc. We
+            # don't want them added when editing the dataset, otherwise we get
+            # duplicated key errors.
+            # The only way to detect indexing right now is checking that
+            # validate is set to False.
             if harvest_object and not context.get('validate', True):
                 for key, value in [
                     ('harvest_object_id', harvest_object.id),