Remove batching of inconsistent records.

Printing all discovered inconsistent records to stdout as a batch contradicts
with various things, such as pyDKB's file mode and the possibility of
controlling the workflow with Apache Kafka.

Create an output message with _id and _type for each inconsistent record.
Still exit with code 1 if at least one inconsistent record was found,
0 otherwise.

Utils/Dataflow/071_esConsistency/consistency.py

@@ -42,7 +42,7 @@ def log(msg, prefix='DEBUG'):
 
 
 INDEX = None
-FOUND_DIFF = []
+FOUND_DIFF = False
 
 
 def load_config(fname):
@@ -166,8 +166,10 @@ def process(stage, message):
     if data != es_data:
         log('Document (%s, %d) differs between Oracle and ES: Oracle:%s ES:%s'
             % (_type, _id, data, es_data), 'WARN')
+        out_message = JSONMessage({'_type': _type, '_id': _id})
+        stage.output(out_message)
         global FOUND_DIFF
-        FOUND_DIFF.append((_type, _id))
+        FOUND_DIFF = True
     else:
         log('Document (%s, %d) is up to date in ES' % (_type, _id), 'INFO')
 
@@ -218,8 +220,6 @@ def main(args):
     if exit_code == 0 and FOUND_DIFF:
         exit_code = 1
 
-    print FOUND_DIFF
-
     exit(exit_code)