Merge branch 'unstable' of ssh://delanoe.org:1979/gargantext into samuel

gargantext_web/db.py

@@ -135,7 +135,7 @@ class ModelCache(dict):
             self.preload()
 
     def __missing__(self, key):
-        print(key)
+        #print(key)
         conditions = [
             (column == str(key))
             for column in self._columns

gargantext_web/views_optimized.py

@@ -2,7 +2,7 @@ from django.shortcuts import redirect
 from django.shortcuts import render
 from django.http import Http404, HttpResponse, HttpResponseRedirect, HttpResponseForbidden
 
-from sqlalchemy import func
+from sqlalchemy import func, and_, or_
 from sqlalchemy.orm import aliased
 
 from collections import defaultdict
@@ -48,22 +48,33 @@ def project(request, project_id):
     # Let's find out about the children nodes of the project
     ChildrenNode = aliased(Node)
     # This query is giving you the wrong number of docs from the pubmedquerier (x 5)
-    #  ... sqlalchemy.func is the guilty
+    #  ... sqlalchemy.func by Resource.type_id is the guilty
+    # ISSUE L51
     corpus_query = (session
-        .query(Node.id, Node.name, Resource.type_id, func.count(ChildrenNode.id))
-        .outerjoin(ChildrenNode, ChildrenNode.parent_id == Node.id)
-        .outerjoin(Node_Resource, Node_Resource.node_id == Node.id)
-        .outerjoin(Resource, Resource.id == Node_Resource.resource_id)
+        .query(Node.id, Node.name, func.count(ChildrenNode.id))
+        #.query(Node.id, Node.name, Resource.type_id, func.count(ChildrenNode.id))
+        #.join(Node_Resource, Node_Resource.node_id == Node.id)
+        #.join(Resource, Resource.id == Node_Resource.resource_id)
         .filter(Node.parent_id == project.id)
         .filter(Node.type_id == cache.NodeType['Corpus'].id)
-        .group_by(Node.id, Node.name, Resource.type_id)
+        .filter(and_(ChildrenNode.parent_id  == Node.id, ChildrenNode.type_id  == cache.NodeType['Document'].id))
+        .group_by(Node.id, Node.name)
         .order_by(Node.name)
+        .all()
     )
     corpora_by_resourcetype = defaultdict(list)
     documents_count_by_resourcetype = defaultdict(int)
     corpora_count = 0
     corpusID_dict = {}
-    for corpus_id, corpus_name, resource_type_id, document_count in corpus_query:
+    for corpus_id, corpus_name, document_count in corpus_query:
+        
+        # Not optimized GOTO ISSUE L51
+        resource_type_id = (session.query(Resource.type_id)
+                                   .join(Node_Resource, Node_Resource.resource_id == Resource.id)
+                                   .join(Node, Node.id == Node_Resource.node_id )
+                                   .filter(Node.id==corpus_id)
+                                   .first())[0]
+        
         if not corpus_id in corpusID_dict:
             if resource_type_id is None:
                 resourcetype_name = '(no resource)'

init/init_gargantext.py

@@ -4,11 +4,13 @@ import os
 os.environ.setdefault("DJANGO_SETTINGS_MODULE", "gargantext_web.settings")
 os.environ.setdefault("DJANGO_HSTORE_GLOBAL_REGISTER", "False")
 
-
 # We're gonna use all the models!
 
-from node.models import *
+# Django models
+from node import models
 
+# SQLA models
+from gargantext_web.db import *
 
 # Reset: all data
 #
@@ -42,7 +44,7 @@ metadata = {
     'journal': 'string',
 }
 for name, type in metadata.items():
-    Metadata(name=name, type=type).save()
+    models.Metadata(name=name, type=type).save()
 
 
 # Integration: languages
@@ -66,12 +68,16 @@ french  = Language.objects.get(iso2='fr')
 # Integration: users
 
 print('Initialize users...')
-try:
-    me = User.objects.get(username='alexandre')
-except:
-    me = User(username='alexandre')
-    me.save()
-
+me = models.User.objects.get_or_create(username='alexandre')
+gargantua = models.User.objects.get_or_create(username='gargantua')
+node_root = Node(user_id=gargantua.id, type_id=cache.NodeType['Root'].id, name='Root')
+node_stem = Node(user_id=gargantua.id, type_id=cache.NodeType['Stem'].id, name='Stem', parent_id=node_root.id)
+node_lem = Node(user_id=gargantua.id, type_id=cache.NodeType['Lem'].id, name='Lem', parent_id=node_root.id)
+
+session.add(node_root)
+session.add(node_stem)
+session.add(node_lem)
+session.commit()
 
 # Integration: node types
 
@@ -87,7 +93,7 @@ node_types = [
         ]
 
 for node_type in node_types:
-    NodeType.objects.get_or_create(name=node_type)
+    models.NodeType.objects.get_or_create(name=node_type)
 
 # Integration: resource types
 
@@ -97,7 +103,8 @@ resources = [
         'pubmed', 'isi', 'ris', 'europress_french', 'europress_english']
 
 for resource in resources:
-    ResourceType.objects.get_or_create(name=resource)
+    models.ResourceType.objects.get_or_create(name=resource)
+
 
 
 # TODO 
@@ -145,7 +152,6 @@ for resource in resources:
 
 
 
-from gargantext_web.db import *
 
 # Instantiante table NgramTag:
 f = open("part_of_speech_labels.txt", 'r')

templates/project.html

@@ -327,7 +327,7 @@
 					console.log("enabling "+"#"+value.id)
 					$("#"+value.id).attr('onclick','getGlobalResults(this);');
 					// $("#submit_thing").prop('disabled' , false)
-					$("#submit_thing").html("Process a 100 sample!")
+					$("#submit_thing").html("Process a 1000 sample!")
 
 		            thequeries = data
 		            var N=0,k=0;