# HG changeset patch
# User Sylvain Thénault <sylvain.thenault@logilab.fr>
# Date 1318333832 -7200
# Node ID 2ad26cc3b5c62208adc67158d68559395d1a4351
# Parent  664d52cb936da8627a5077206b219a6716c3fd39
[datafeed parser] only update an entity attributes when needed. Closes #1989142

This may avoid error with some entities which have immutable attributes
and don't check something actually changed.

Also, avoid modifying the cached item representation, this may cause pb
when an item is seen several times during an import.

diff -r 664d52cb936d -r 2ad26cc3b5c6 sobjects/parsers.py
--- a/sobjects/parsers.py	Tue Oct 11 10:59:35 2011 +0200
+++ b/sobjects/parsers.py	Tue Oct 11 13:50:32 2011 +0200
@@ -72,7 +72,7 @@
     typeddict = {}
     for rschema in eschema.subject_relations():
         if rschema.final and rschema in stringdict:
-            if rschema == 'eid':
+            if rschema in ('eid', 'cwuri', 'cwtype', 'cwsource'):
                 continue
             attrtype = eschema.destination(rschema)
             value = stringdict[rschema]
@@ -200,8 +200,8 @@
         * `rels` is for relations and structured as
            {role: {relation: [(related item, related rels)...]}
         """
-        entity = self.extid2entity(str(item.pop('cwuri')),  item.pop('cwtype'),
-                                   cwsource=item.pop('cwsource'), item=item)
+        entity = self.extid2entity(str(item['cwuri']),  item['cwtype'],
+                                   cwsource=item['cwsource'], item=item)
         if entity is None:
             return None
         if entity.eid in self._processed_entities:
@@ -209,10 +209,16 @@
         self._processed_entities.add(entity.eid)
         if not (self.created_during_pull(entity) or self.updated_during_pull(entity)):
             self.notify_updated(entity)
-            item.pop('eid')
-            # XXX check modification date
             attrs = extract_typed_attrs(entity.e_schema, item)
-            entity.set_attributes(**attrs)
+            # check modification date and compare attribute values to only
+            # update what's actually needed
+            entity.complete(tuple(attrs))
+            mdate = attrs.get('modification_date')
+            if not mdate or mdate > entity.modification_date:
+                attrs = dict( (k, v) for k, v in attrs.iteritems()
+                              if v != getattr(entity, k))
+                if attrs:
+                    entity.set_attributes(**attrs)
         self.process_relations(entity, rels)
         return entity