improves rating

SmartDataAnalytics · Apr 16, 2021 · f49cee4 · f49cee4
1 parent 96df908
commit f49cee4
Show file tree

Hide file tree

Showing 3 changed files with 28 additions and 13 deletions.
diff --git a/migration/openresearch/eventcorpus.py b/migration/openresearch/eventcorpus.py
@@ -9,7 +9,6 @@ class EventCorpus(object):
     Towards a gold standard event corpus ...
     '''
 
-
     def __init__(self,debug=False):
         '''
         Constructor
@@ -24,7 +23,6 @@ def fromWikiUser(self,wikiUser):
         self.eventList.debug=self.debug
         self.eventList.fromCache(wikiUser)
 
-
         self.eventSeriesList=EventSeriesList()
         self.eventSeriesList.debug=self.debug
         self.eventSeriesList.fromCache(wikiUser)

diff --git a/migration/ormigrate/eventSeriesFixer.py b/migration/ormigrate/eventSeriesFixer.py
@@ -4,6 +4,7 @@
 @author: wf
 '''
 from ormigrate.fixer import PageFixer
+from ormigrate.rating import Rating
 
 class EventSeriesProvenanceFixer(PageFixer):
     '''
@@ -15,4 +16,17 @@ def __init__(self):
         '''
         Constructor
         '''
+
+    @classmethod
+    def getRating(cls,eventRecord):
+        hasDblp=eventRecord.get('dblpSeries') is not  None
+        hasWikidata=eventRecord.get('wikidataId') is not None
+        if hasDblp and hasWikidata:
+            return Rating(1,Rating.ok,'Gold Standard series')
+        if hasDblp:
+            return Rating(4,Rating.ok,'Wikdata id missing for dblp series')
+        if hasWikidata:
+            return Rating(3,Rating.ok,'Wikdata only series')
+        return Rating(5,Rating.invalid,'Series provenance data missing')
+
 
diff --git a/migration/tests/testIssue168.py b/migration/tests/testIssue168.py
@@ -17,7 +17,7 @@ class TestIssue168(unittest.TestCase):
 
 
     def setUp(self):
-        self.debug=False
+        self.debug=True
         pass
 
 
@@ -39,19 +39,22 @@ def testEventsWithSeries(self):
         '''
         eventCorpus=self.getEventCorpus(debug=self.debug)
 
-    def checkRatedLod(self,lod,errors):
+    def checkRatedLod(self,lod,errors,columns,showPainsAbove=8):
         if len(errors)>0:
             for error in errors:
                 print(error)
         self.assertEqual(0,len(errors))
-        counter=Counter()
-        for record in lod:
-            rating=record["acronym"]
+        for column in columns:
+            counter=Counter()
+            for record in lod:
+                rating=record[column]
+                if self.debug:
+                    if rating.pain>=showPainsAbove:
+                        print (rating)
+                counter[rating.pain]+=1
             if self.debug:
-                if rating.pain>=5:
-                    print (rating)
-            counter[rating.pain]+=1
-        print(counter.most_common(50))
+                print (f"rating results for {column}:")
+                print(counter.most_common(50))
         pass
 
     def testRatingCallback(self):
@@ -60,9 +63,9 @@ def testRatingCallback(self):
         '''
         eventCorpus=self.getEventCorpus(debug=self.debug)
         lod,errors=eventCorpus.eventList.getRatedLod(Event.rateMigration)
-        self.checkRatedLod(lod, errors)
+        self.checkRatedLod(lod, errors,['acronym','ordinal'])
         lod,errors=eventCorpus.eventSeriesList.getRatedLod(EventSeries.rateMigration)
-        self.checkRatedLod(lod, errors)
+        self.checkRatedLod(lod, errors,['provenance'])
 
 
 if __name__ == "__main__":