Merge branch 'unstable' into celery

init/init.py

@@ -104,30 +104,15 @@ except Exception as error:
 
 # In[33]:
 
-try:
-    typePubmed = ResourceType.objects.get(name='pubmed')
-    typeIsi    = ResourceType.objects.get(name='isi')
-    typeRis    = ResourceType.objects.get(name='ris')
-    typePresseFrench = ResourceType.objects.get(name='europress_french')
-    typePresseEnglish = ResourceType.objects.get(name='europress_english')
-
-except Exception as error:
-    print(error)
+from parsing.parsers_config import parsers
 
-    typePubmed = ResourceType(name='pubmed')
-    typePubmed.save()
+ResourceType.objects.all().delete()
 
-    typeIsi    = ResourceType(name='isi')
-    typeIsi.save()
-    
-    typeRis    = ResourceType(name='ris')
-    typeRis.save()
-    
-    typePresseFrench = ResourceType(name='europress_french')
-    typePresseFrench.save()
-    
-    typePresseEnglish = ResourceType(name='europress_english')
-    typePresseEnglish.save()
+for key in parsers.keys():
+    try:
+        ResourceType.objects.get_or_create(name=key)
+    except Exception as error:
+        print("Ressource Error: ", error)
 
 
 # In[34]:

init/init.sh

@@ -4,6 +4,7 @@ psql -d gargandb -f init.sql
 
 sleep 2
 ../manage.py syncdb
+
 psql -d gargandb -f init2.sql
 
 

init/init2.sql

+
+ALTER TABLE ONLY node_node ALTER COLUMN date SET DEFAULT CURRENT_DATE ;
+
+

init/init_gargantext.py

@@ -54,16 +54,13 @@ import pycountry
 Language.objects.all().delete()
 for language in pycountry.languages:
     if 'alpha2' in language.__dict__:
-        Language(
+        models.Language(
             iso2 = language.alpha2,
             iso3 = language.bibliographic,
             fullname = language.name,
             implemented = 1 if language.alpha2 in ['en', 'fr'] else 0,
         ).save()
 
-english = Language.objects.get(iso2='en')
-french  = Language.objects.get(iso2='fr')
-
 
 # Integration: users
 
@@ -99,56 +96,10 @@ for node_type in node_types:
 
 print('Initialize resource...')
 
-resources = [
-        'pubmed', 'isi', 'ris', 'europress_french', 'europress_english']
-
-for resource in resources:
-    models.ResourceType.objects.get_or_create(name=resource)
-
-
-
-# TODO 
-# here some tests
-# add a new project and some corpora to test it
-
-
-# Integration: project
-#
-#print('Initialize project...')
-#try:
-#    project = Node.objects.get(name='Bees project')
-#except:
-#    project = Node(name='Bees project', type=typeProject, user=me)
-#    project.save()
-#
-
-# Integration: corpus
-
-#print('Initialize corpus...')
-#try:
-#    corpus_pubmed = Node.objects.get(name='PubMed corpus')
-#except:
-#    corpus_pubmed = Node(parent=project, name='PubMed corpus', type=typeCorpus, user=me)
-#    corpus_pubmed.save()
-#
-#print('Initialize resource...')
-#corpus_pubmed.add_resource(
-#    # file='./data_samples/pubmed.zip',
-#    #file='./data_samples/pubmed_2013-04-01_HoneyBeesBeeBees.xml',
-#    file='/srv/gargantext_lib/data_samples/pubmed.xml',
-#    type=typePubmed,
-#    user=me
-#)
-#
-#for resource in corpus_pubmed.get_resources():
-#    print('Resource #%d - %s - %s' % (resource.id, resource.digest, resource.file))
-#    
-## print('Parse corpus #%d...' % (corpus_pubmed.id, ))
-# corpus_pubmed.parse_resources(verbose=True)
-# print('Extract corpus #%d...' % (corpus_pubmed.id, ))
-# corpus_pubmed.children.all().extract_ngrams(['title',])
-# print('Parsed corpus #%d.' % (corpus_pubmed.id, ))
+from parsing.parsers_config import parsers
 
+for parser in parsers.keys():
+    models.ResourceType.objects.get_or_create(name=parser)
 
 
 

node/models.py

@@ -29,6 +29,7 @@ from celery import current_app
 import os
 import subprocess
 
+from parsing.parsers_config import parsers
 
 # Some usefull functions
 # TODO: start the function name with an underscore (private)
@@ -194,15 +195,19 @@ class Node(CTENode):
         print("= = = = = = = = =  = =\n")
         for node_resource in self.node_resource.filter(parsed=False):
             resource = node_resource.resource
-            parser = defaultdict(lambda:FileParser.FileParser, {
-                'istext'    : ISText,
-                'pubmed'    : PubmedFileParser,
-                'isi'       : IsiFileParser,
-                'ris'       : RisFileParser,
-                'europress' : EuropressFileParser,
-                'europress_french'  : EuropressFileParser,
-                'europress_english' : EuropressFileParser,
-            })[resource.type.name]()
+            parser = defaultdict(lambda:FileParser.FileParser, parsers
+#                    {
+#                'istext'            : ISText,
+#                'pubmed'            : PubmedFileParser,
+#                'isi'               : IsiFileParser,
+#                'ris'               : RisFileParser,
+#                'RIS (Jstor)'       : JstorFileParser,
+#                'europress'         : EuropressFileParser,
+#                'europress_french'  : EuropressFileParser,
+#                'europress_english' : EuropressFileParser,
+#            }
+                    
+                    )[resource.type.name]()
             metadata_list += parser.parse(str(resource.file))
         type_id = NodeType.objects.get(name='Document').id
         langages_cache = LanguagesCache()

parsing/FileParsers/IsiFileParser.py

 from .RisFileParser import RisFileParser
 
 
+
 class IsiFileParser(RisFileParser):
     
     _parameters = {

parsing/FileParsers/JstorFileParser.py

+from .RisFileParser import RisFileParser
+
+
+class JstorFileParser(RisFileParser):
+
+    _parameters = {
+        b"ER":  {"type": "delimiter"},
+        b"TI":  {"type": "metadata", "key": "title", "separator": " "},
+        b"AU":  {"type": "metadata", "key": "authors", "separator": ", "},
+        b"UR":  {"type": "metadata", "key": "doi"},
+        b"Y1":  {"type": "metadata", "key": "publication_year"},
+        b"PD":  {"type": "metadata", "key": "publication_month"},
+        b"LA":  {"type": "metadata", "key": "language_iso2"},
+        b"AB":  {"type": "metadata", "key": "abstract", "separator": " "},
+        b"WC":  {"type": "metadata", "key": "fields"},
+    }
+
+

parsing/FileParsers/__init__.py

 from .RisFileParser import RisFileParser
 from .IsiFileParser import IsiFileParser
+from .JstorFileParser import JstorFileParser
 from .PubmedFileParser import PubmedFileParser
 from .EuropressFileParser import EuropressFileParser
 from .ISText import ISText

parsing/corpustools.py

@@ -7,7 +7,6 @@ from math import log
 
 from gargantext_web.db import *
 
-from .FileParsers import *
 
 
 
@@ -30,15 +29,7 @@ class DebugTime:
 
 # keep all the parsers in a cache
 class Parsers(defaultdict):
-
-    _parsers = {
-        'pubmed'            : PubmedFileParser,
-        'isi'               : IsiFileParser,
-        'ris'               : RisFileParser,
-        'europress'         : EuropressFileParser,
-        'europress_french'  : EuropressFileParser,
-        'europress_english' : EuropressFileParser,
-    }
+    from .parsers_config import parsers as _parsers
 
     def __missing__(self, key):
         if key not in self._parsers:

parsing/parsers_config.py

+from .FileParsers import *
+
+parsers = {
+        'Pubmed (xml format)'               : PubmedFileParser,
+        'Web of Science (ISI format)'       : IsiFileParser,
+        'Scopus ou Zotero (RIS format)'     : RisFileParser,
+        'Jstor (RIS format)'                : JstorFileParser,
+        #'Europress'                        : EuropressFileParser,
+        'Europress (French)'                : EuropressFileParser,
+        'Europress (English)'               : EuropressFileParser,
+        
+    }
+