Cleaning indices before (re)indexing, in order to make it happen for real!

lib/index_cleaner.py

-from elasticsearch import Elasticsearch
+from elasticsearch import Elasticsearch, NotFoundError
 
 def index_cleaner( cfg, content_type, content_id=None ):
 
@@ -17,13 +17,16 @@ def index_cleaner( cfg, content_type, content_id=None ):
 
         try:
             res = source_es.indices.delete(source_index)
+        except NotFoundError:
+            pass
         except Exception as e:
             print(e)
             successful = False
 
         try:
             rep = destin_es.indices.delete(destin_index)
-        #print(rep)
+        except NotFoundError:
+            pass
         except Exception as e:
             print(e)
             successful = False
@@ -36,16 +39,18 @@ def index_cleaner( cfg, content_type, content_id=None ):
 
         try:
             res = source_es.delete(index=source_index, doc_type='_doc', id=content_id)
+        except NotFoundError:
+                pass
         except Exception as e:
             print(e)
             successful = False
-            pass
 
         try:
-            res = source_es.delete(index=destin_index, doc_type='_doc', id=content_id)
+            res = destin_es.delete(index=destin_index, doc_type='_doc', id=content_id)
+        except NotFoundError:
+            pass
         except Exception as e:
             print(e)
             successful = False
-            pass
 
     return successful

lib/reindexer.py

@@ -5,7 +5,7 @@ from pprint import pprint
 from .elastic_mapping_template_pages import template as template_pages
 from .elastic_mapping_template_posts import template as template_posts
 
-def reindexer(cfg, content_type):
+def reindexer(cfg, content_type, content_id=None):
 
   template = eval('template_%s' % content_type)
 
@@ -63,6 +63,11 @@ def reindexer(cfg, content_type):
             }
           }
 
+  if content_id != None:
+    body['source']['query'] = {"term": {"_id": content_id}}
+
+  print(body)
+
 
   # waiting for source index to be stable
   count1 = 0
@@ -79,6 +84,8 @@ def reindexer(cfg, content_type):
 
   #if rep['failures'] == []:
   if 'task' in rep:
-      return cfg['reindexer']['destination_url'] + '/_tasks/' + rep['task']
+      task_url = cfg['reindexer']['destination_url'] + '/_tasks/' + rep['task']
+      print("Task URL: %s" % task_url)
+      return task_url
   else:
       return False

main.py

@@ -55,9 +55,8 @@ def add_content( cfg, content_type, data ):
 
     # N.B. : pages -> page; posts -> post, that's why we remove the last letter from content_type
     content_id = data[ content_type[0:len(content_type)-1] ]['current']['id']
-    print('Here', content_id)
 
-    print('Getting content with id=%s...' % content_id)
+    print('Getting content with id = %s...' % content_id)
 
     try:
         content = get_content(cfg, content_type, content_id)
@@ -66,14 +65,19 @@ def add_content( cfg, content_type, data ):
         return False
 
     processed_content = eval('process_%s' % content_type)([content])
-    successful = index_docs(cfg, content_type, processed_content)
 
+    successful = index_cleaner(cfg, content_type, content_id)
+    if not successful:
+        raise Exception('Something went wrong. Exiting...')
+        exit(1)
+
+    successful = index_docs(cfg, content_type, processed_content)
     if not successful:
         raise Exception('Something went wrong. Exiting...')
         exit(1)
 
     print('Reindexing...')
-    task_url = reindexer(cfg, content_type)
+    task_url = reindexer(cfg, content_type, content_id)
     #print(task_url)
 
     if not successful: