[FIX] SQL fix to select field1 only.

analysis/cooccurrences.py

@@ -66,13 +66,6 @@ def do_cooc(corpus=None
 
     doc_id = cache.NodeType['Document'].id
    
-    hyperdata_id = session.query(Hyperdata).filter(Hyperdata.name=='source').first().id
-    test_query = (session.query(NodeHyperdataNgram)
-                 .join(Node, Node.id == NodeHyperdataNgram.node_id)
-                 .filter(Node.parent_id==corpus.id, Node.type_id==doc_id)
-                 .filter(NodeHyperdataNgram.hyperdata_id==hyperdata_id)
-                 )
-    
     #print([n for n in test_query])
     if isMonopartite :
         NodeNgramX = aliased(NodeNgram)
@@ -91,9 +84,8 @@ def do_cooc(corpus=None
         cooc_query = (session.query(NodeHyperdataNgram.ngram_id, NodeNgramY.ngram_id, cooc_score)
                  .join(Node, Node.id == NodeHyperdataNgram.node_id)
                  .join(NodeNgramY, NodeNgramY.node_id == Node.id)
-                 .join(NodeHyperdata, NodeHyperdata.node_id==Node.id)
-                 .join(Hyperdata, Hyperdata.id == NodeHyperdata.hyperdata_id)
-                 .filter(Node.parent_id==corpus.id, Node.type_id==doc_id)
+                 .join(Hyperdata, Hyperdata.id == NodeHyperdataNgram.hyperdata_id)
+                 .filter(Node.parent_id == corpus.id, Node.type_id == doc_id)
                  .filter(Hyperdata.name == field1)
                     )
 

parsing/corpustools.py

@@ -147,7 +147,8 @@ def parse_resources(corpus, user=None, user_id=None):
     hyperdata_set = set()
     hyperdata_ngrams = set()
     node_hyperdata_ngrams = set()
-    for field in ['source', 'authors', 'journal']:
+    #for field in ['source', 'authors', 'journal']:
+    for field in ['journal', 'authors']:
         hyperdata_set.add(session.query(Hyperdata.id).filter(Hyperdata.name==field).first()[0])
     
     #print("hyperdata_set", hyperdata_set)
@@ -155,9 +156,11 @@ def parse_resources(corpus, user=None, user_id=None):
     for key, values in node_hyperdata_lists.items():
         #print('here', key, values)
         bulk_insert(Node_Hyperdata, ['node_id', 'hyperdata_id', 'value_'+key], values)
-        if key == 'string':
+        if key == 'string' :
             for value in values:
+                print('value', value)
                 if value[1] in hyperdata_set:
+                    print('value_1', value[1])
                     for val in value[2].split(', '):
                         hyperdata_ngrams.add((val, len(val.split(' '))))
                         node_hyperdata_ngrams.add((value[0], value[1], val))