[FIX] labels for nodes.

analysis/cooccurrences.py

@@ -3,7 +3,8 @@ from sqlalchemy import literal_column
 from sqlalchemy.orm import aliased
 from sqlalchemy.sql import func
 
-from gargantext_web.db import Node, NodeNgram, NodeNgramNgram, NodeNodeNgram, NodeHyperdata, Hyperdata
+from gargantext_web.db import Node, Ngram, NodeNgram, NodeNgramNgram, \
+        NodeNodeNgram, NodeHyperdata, Hyperdata
 from gargantext_web.db import session, cache, get_or_create_node, bulk_insert
 from analysis.lists import WeightedMatrix, UnweightedList, Translations
 
@@ -11,6 +12,7 @@ def cooc(corpus=None
          , field_X=None, field_Y=None
          , miam_id=None, stop_id=None, group_id=None
          , cvalue_id=None
+         , n_min=2, n_max=None
          , start=None, end=None
          , limit=1000):
     '''
@@ -57,8 +59,30 @@ def cooc(corpus=None
              .join(NodeNgramY, NodeNgramY.node_id == Node.id)
              .filter(Node.parent_id==corpus.id, Node.type_id==doc_id)
                 )
-
-
+    
+# Size of the ngrams between n_min and n_max
+    if n_min is not None or n_max is not None:
+        NgramX = aliased(Ngram)
+        NgramY = aliased(Ngram)
+
+        cooc_query = (cooc_query
+             .join(NgramX, NgramX.id == NodeNgramX.ngram_id)
+             .join(NgramY, NgramY.id == NodeNgramY.ngram_id)
+            )
+
+    if n_min is not None:
+        cooc_query = (cooc_query
+             .filter(NgramX.n >= n_min)
+             .filter(NgramY.n >= n_min)
+            )
+
+    if n_max is not None:
+        cooc_query = (cooc_query
+             .filter(NgramX.n >= n_min)
+             .filter(NgramY.n >= n_min)
+            )
+
+# Cooc between the dates start and end
     if start is not None:
         Start=aliased(NodeHyperdata)
         StartFormat = aliased(Hyperdata)
@@ -79,11 +103,12 @@ def cooc(corpus=None
                       )
 
 
+# Cooc is symetric, take only the main cooccurrences and cut at the limit
     cooc_query = (cooc_query.filter(Node.parent_id == corpus.id, Node.type_id == doc_id)
              .filter(NodeNgramX.ngram_id < NodeNgramY.ngram_id)
 
-             .group_by(Node.id, NodeNgramX.ngram_id, NodeNgramY.ngram_id)
-             .order_by(func.count())
+             .group_by(NodeNgramX.ngram_id, NodeNgramY.ngram_id)
+             .order_by(desc(func.count()))
 
              .limit(limit)
              )
@@ -91,6 +116,7 @@ def cooc(corpus=None
     matrix = WeightedMatrix(cooc_query)
     #print(matrix)
 
+# Select according some scores
     if cvalue_id is not None :
         #miam = get_or_create_node(nodetype='Cvalue', corpus=corpus)
         cvalue_list = UnweightedList(session.query(NodeNodeNgram.ngram_id)

analysis/functions.py

@@ -34,7 +34,6 @@ def create_blacklist(user, corpus):
 def create_synonymes(user, corpus):
     pass
 
-
 size = 1000
 
 def create_whitelist(user, corpus_id, size=size, count_min=2, miam_id=None):
@@ -170,7 +169,9 @@ def create_cooc(user=None, corpus_id=None, whitelist=None, size=size, year_start
     return cooc.id
 
 def get_cooc(request=None, corpus=None, cooc_id=None, type='node_link', size=size):
-
+    '''
+    get_ccoc : to compute the graph.
+    '''
     matrix = defaultdict(lambda : defaultdict(float))
     ids    = dict()
     labels = dict()
@@ -185,16 +186,15 @@ def get_cooc(request=None, corpus=None, cooc_id=None, type='node_link', size=siz
     
     # data deleted each time
     session.query(NodeNgramNgram).filter(NodeNgramNgram.node_id==cooc_id).delete()
-    #cooc_id = cooc(corpus=corpus, miam_id=miam_id, stop_id=stop_id, limit=size)
     cooc_id = cooc(corpus=corpus, miam_id=miam_id, group_id=group_id, stop_id=stop_id, limit=size)
-    #cooc_id = cooc(corpus=corpus, miam_id=miam_id, limit=size)
 
     #print([n for n in session.query(NodeNgramNgram).filter(NodeNgramNgram.node_id==cooc_id).all()])
     for cooccurrence in session.query(NodeNgramNgram).filter(NodeNgramNgram.node_id==cooc_id).all():
         #print(cooccurrence)
         # print(cooccurrence.ngramx.terms," <=> ",cooccurrence.ngramy.terms,"\t",cooccurrence.score)
-        labels[cooccurrence.ngramx_id] = session.query(Ngram.id).filter(Ngram.id == cooccurrence.ngramx_id).first()[0]
-        labels[cooccurrence.ngramy_id] = session.query(Ngram.id).filter(Ngram.id == cooccurrence.ngramy_id).first()[0]
+        # TODO clean this part, unuseful
+        labels[cooccurrence.ngramx_id] = cooccurrence.ngramx_id #session.query(Ngram.id).filter(Ngram.id == cooccurrence.ngramx_id).first()[0]
+        labels[cooccurrence.ngramy_id] = cooccurrence.ngramy_id #session.query(Ngram.id).filter(Ngram.id == cooccurrence.ngramy_id).first()[0]
 
         matrix[cooccurrence.ngramx_id][cooccurrence.ngramy_id] = cooccurrence.score
         matrix[cooccurrence.ngramy_id][cooccurrence.ngramx_id] = cooccurrence.score
@@ -205,7 +205,6 @@ def get_cooc(request=None, corpus=None, cooc_id=None, type='node_link', size=siz
         weight[cooccurrence.ngramx_id] = weight.get(cooccurrence.ngramx_id, 0) + cooccurrence.score
         weight[cooccurrence.ngramy_id] = weight.get(cooccurrence.ngramy_id, 0) + cooccurrence.score
 
-
     x = pd.DataFrame(matrix).fillna(0)
     y = pd.DataFrame(matrix).fillna(0)
 
@@ -280,7 +279,6 @@ def get_cooc(request=None, corpus=None, cooc_id=None, type='node_link', size=siz
                 #node,type(labels[node])
                 G.node[node]['pk'] = ids[node]
                 G.node[node]['label']   = session.query(Ngram.terms).filter(Ngram.id==node).first()
-                # G.node[node]['pk']      = ids[str(node)]
                 G.node[node]['size']    = weight[ids[node]]
                 G.node[node]['group']   = partition[node]
                 # G.add_edge(node, "cluster " + str(partition[node]), weight=3)