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

annotations/static/annotations/document.js

@@ -2,10 +2,13 @@
   'use strict';
 
   var annotationsAppDocument = angular.module('annotationsAppDocument', ['annotationsAppHttp']);
-
   annotationsAppDocument.controller('DocController',
     ['$scope', '$rootScope', '$timeout', 'NgramListHttpService', 'DocumentHttpService',
     function ($scope, $rootScope, $timeout, NgramListHttpService, DocumentHttpService) {
+
+      // dataLoading = signal pour afficher wait
+      $scope.dataLoading = true ;
+
       $rootScope.documentResource = DocumentHttpService.get(
         {'docId': $rootScope.docId},
         function(data, responseHeaders) {
@@ -27,6 +30,7 @@
             function(data) {
               $rootScope.annotations = data[$rootScope.corpusId.toString()][$rootScope.docId.toString()];
               $rootScope.lists = data[$rootScope.corpusId.toString()].lists;
+              $scope.dataLoading = false ;
             },
             function(data) {
               console.error("unable to get the list of ngrams");
@@ -34,6 +38,7 @@
           );
 
       });
+
       // TODO setup article pagination
       $scope.onPreviousClick = function () {
         DocumentHttpService.get($scope.docId - 1);

annotations/templates/annotations/main.html

@@ -86,6 +86,16 @@
                 <li class="list-group-item small"><span class="badge">date</span>{[{publication_date}]}</li>
               </ul>
             </div>
+
+            <div ng-if="dataLoading">
+                Loading text...
+                <br>
+                  <center>
+                    <img width="10%" src="{% static 'img/ajax-loader.gif'%}"></img>
+                  </center>
+                <br>
+            </div>
+
             <div ng-if="abstract_text != null">
                 <span class="badge">abstract</span>
             </div>

annotations/urls.py

@@ -2,9 +2,15 @@ from django.conf.urls import patterns, url
 from annotations import views
 
 
+# /!\ urls patterns here are *without* the trailing slash
+
 urlpatterns = patterns('',
+    # json:title,id,authors,journal,
+    #      publication_date
+    #      abstract_text,full_text
     url(r'^document/(?P<doc_id>[0-9]+)$', views.Document.as_view()), # document view
     url(r'^corpus/(?P<corpus_id>[0-9]+)/document/(?P<doc_id>[0-9]+)$', views.NgramList.as_view()), # the list associated with an ngram
     url(r'^lists/(?P<list_id>[0-9]+)/ngrams/(?P<ngram_ids>[0-9,\+]+)+$', views.NgramEdit.as_view()),
+    # POST (fixed 2015-12-16)
     url(r'^lists/(?P<list_id>[0-9]+)/ngrams/create$', views.NgramCreate.as_view()), #
 )

annotations/views.py

@@ -13,7 +13,7 @@ from rest_framework.exceptions import APIException
 from rest_framework.authentication import SessionAuthentication, BasicAuthentication
 
 from node.models import Node
-from gargantext_web.db import session, cache, Node, NodeNgram
+from gargantext_web.db import session, cache, Node, NodeNgram, Ngram
 from ngram.lists import listIds, listNgramIds
 from gargantext_web.db import get_or_create_node
 
@@ -138,6 +138,8 @@ class NgramCreate(APIView):
     def post(self, request, list_id):
         """
         create NGram in a given list
+        
+        example: request.data = {'text': 'phylogeny'}
         """
         list_id = int(list_id)
         # format the ngram's text
@@ -161,6 +163,7 @@ class NgramCreate(APIView):
         ngram_id = ngram.id
         # create the new node_ngram relation
         # TODO check existing Node_Ngram ?
+        # £TODO ici indexation
         node_ngram = NodeNgram(node_id=list_id, ngram_id=ngram_id, weight=1.0)
         session.add(node_ngram)
         session.commit()

exec.py

@@ -11,7 +11,9 @@ from ngram.mapList import compute_mapList
 from gargantext_web.db import NodeNgram
 from admin.utils import WorkflowTracking
 from ngram.importExport import exportNgramList, importNgramList
+from analysis.periods import phylo_clusters
 
+from ngram.occurrences import compute_occs
 
 
 def ngram_workflow(corpus, n=5000):
@@ -50,13 +52,17 @@ def ngram_workflow(corpus, n=5000):
 #    update_state.processing_(corpus, "TF-IDF local score")
 #    compute_tfidf(corpus)
     # update_state.processing_(corpus, "OCCS local score")
-    # compute_occs(corpus)
-    update_state.processing_(corpus, "0")
+    compute_occs(corpus)
+    #update_state.processing_(corpus, "0")
 
 if __name__ == "__main__":
     node_id = sys.argv[1] 
     corpus=session.query(Node).filter(Node.id==node_id).first()
-    exportNgramList(corpus, "list.csv")
-    #importNgramList(corpus, "list.csv")
+    
     #ngram_workflow(corpus)
     
+    #exportNgramList(corpus, "list.csv")
+    #importNgramList(corpus, "list.csv")
+
+    phylo_clusters(corpus, range(2012,2016))
+

init/stop_lists/en.txt

-
  ­
  -
 ­
@@ -462,3 +461,16 @@ your
 yours
 yourself
 yourselves
+Francis group
+© taylor
+copyright taylor
+copyright © taylor
+copyright ©
+springer science+business media
+sage publications
+oxford university press
+© springer international publishing switzerland
+john wiley
+elsevier ltd
+© fpi
+elsevier inc

init_accounts.py

@@ -14,16 +14,18 @@ def notify_user(username, email, password):
     Votre login est: %s
     Votre mot de passe est : %s
 
-    Nous restons votre disposition pour tout complément d'information.
+    En janvier prochain, il y aura une formation Gargantext (gratuite).
+    Inscription obligatoire pour les dernière places:
+    http://iscpif.fr/event/formation-gargantext/
+
+    Nous restons à votre disposition pour tout complément d'information.
     Cordialement
     --
         L'équipe de Gargantext (CNRS)
 
     ''' % (username, password)
 
-    send_mail('[Gargantext] Votre compte', message, 'alexandre.delanoe@mines-paristech.fr', [email], fail_silently=False )
-    #send_mail('[Gargantext] Votre compte', message, 'alexandre.delanoe@mines-paristech.fr', [email], ['alexandre@delanoe.org'] )
-
+    send_mail('[Gargantext] Votre accès à la plateforme', message, 'alexandre.delanoe@iscpif.fr', [email], fail_silently=False )
 
     # add option for mass sending email
 

parsing/FileParsers/IsiFileParser.py

@@ -15,6 +15,7 @@ class IsiFileParser(RisFileParser):
             b"TI":  {"type": "hyperdata", "key": "title", "separator": " "},
             b"AU":  {"type": "hyperdata", "key": "authors", "separator": ", "},
             b"DI":  {"type": "hyperdata", "key": "doi"},
+            b"SO":  {"type": "hyperdata", "key": "journal"},
             b"PY":  {"type": "hyperdata", "key": "publication_year"},
             b"PD":  {"type": "hyperdata", "key": "publication_month"},
             b"LA":  {"type": "hyperdata", "key": "language_fullname"},

parsing/FileParsers/RisFileParser.py

@@ -19,6 +19,7 @@ class RisFileParser(FileParser):
             b"TI":  {"type": "hyperdata", "key": "title", "separator": " "},
             b"ST":  {"type": "hyperdata", "key": "subtitle", "separator": " "},
             b"AU":  {"type": "hyperdata", "key": "authors", "separator": ", "},
+            b"T2":  {"type": "hyperdata", "key": "journal"},
             b"UR":  {"type": "hyperdata", "key": "doi"},
             b"PY":  {"type": "hyperdata", "key": "publication_year"},
             b"PD":  {"type": "hyperdata", "key": "publication_month"},

parsing/FileParsers/ZoteroFileParser.py

@@ -12,6 +12,7 @@ class ZoteroFileParser(RisFileParser):
             b"ER":  {"type": "delimiter"},
             b"TI":  {"type": "hyperdata", "key": "title", "separator": " "},
             b"AU":  {"type": "hyperdata", "key": "authors", "separator": ", "},
+            b"T2":  {"type": "hyperdata", "key": "journal"},
             b"UR":  {"type": "hyperdata", "key": "doi"},
             b"DA":  {"type": "hyperdata", "key": "publication_date_to_parse"},
             b"PY":  {"type": "hyperdata", "key": "publication_year"},

parsing/NgramsExtractors/NgramsExtractor.py

 # from ..Taggers import NltkTagger
 from ..Taggers import TurboTagger
 import nltk
+from re import sub
 
 
 """Base class for all ngrams extractors.
@@ -33,9 +34,21 @@ class NgramsExtractor:
     Returns a list of the ngrams found in the given text.
     """
     def extract_ngrams(self, contents):
-        tagged_tokens = list(self.tagger.tag_text(contents))
+        clean_contents = self._prepare_text(contents)
+
+        # ici tagging
+        tagged_tokens = list(self.tagger.tag_text(clean_contents))
+
         if len(tagged_tokens):
             grammar_parsed = self._grammar.parse(tagged_tokens)
             for subtree in grammar_parsed.subtrees():
                 if subtree.label() == self._label:
                     yield subtree.leaves()
+
+    @staticmethod
+    def _prepare_text(text_contents):
+        """
+        Clean the text for better POS tagging
+        """
+        # strip xml tags
+        return sub(r"<[^>]{0,45}>","",text_contents)