Make import robust against duplicate kataster numbers

core/views_caves.py

@@ -68,7 +68,8 @@ def numericalcmp(x, y):
 
 def caveKey(x):
     """python3 function for sort.
-    Note that cave kataster numbers are not generally integers
+    Note that cave kataster numbers are not generally integers.
+    This needs to be fixed make a decent sort order.
     """ 
     return x.kataster_number
 

parsers/caves.py

@@ -122,6 +122,7 @@ def readcave(filename):
         url = getXML(cavecontents, "url", maxItems = 1, context = context)
         entrances = getXML(cavecontents, "entrance", context = context)
         if len(non_public) == 1 and len(slugs) >= 1 and len(official_name) == 1 and len(areas) >= 1 and len(kataster_code) == 1 and len(kataster_number) == 1 and len(unofficial_number) == 1 and len(explorers) == 1 and len(underground_description) == 1 and len(equipment) == 1 and len(references) == 1 and len(survey) == 1 and len(kataster_status) == 1 and len(underground_centre_line) == 1 and len(notes) == 1 and len(length) == 1 and len(depth) == 1 and len(extent) == 1 and len(survex_file) == 1 and len(description_file ) == 1 and len(url) == 1 and len(entrances) >= 1:
+            try:
                 c, state = models_caves.Cave.objects.update_or_create(non_public = {"True": True, "False": False, "true": True, "false": False,}[non_public[0]],
                          official_name = official_name[0],
                          kataster_code = kataster_code[0],
@@ -142,6 +143,20 @@ def readcave(filename):
                          description_file = description_file[0],
                          url = url[0],
                          filename = filename)
+            except:
+                # need to cope with duplicates
+                print(" ! FAILED to get only one cave when updating using: "+filename)
+                kaves = models_caves.Cave.objects.all().filter(kataster_number=kataster_number[0])
+                for k in kaves:
+                    message = " ! - DUPLICATES in db. kataster:"+ str(k.kataster_number) + ", slug:" + str(k.slug()) 
+                    models.DataIssue.objects.create(parser='caves', message=message)
+                    print(message)
+                for k in kaves:
+                    if k.slug() != None:
+                        print(" ! - OVERWRITING this one: slug:"+ str(k.slug()))
+                        k.notes = "DUPLICATE kataster number found on import. Please fix\n" + k.notes
+                        c = k
+                
             for area_slug in areas:
                 area = models_caves.Area.objects.filter(short_name = area_slug)
                 if area:
@@ -157,7 +172,7 @@ def readcave(filename):
                               slug = slug,
                               primary = primary)
                 except:
-                    message = " ! Can't find text (slug): %s, skipping %s" % (slug, context)
+                    message = " ! Cave update/create failure: %s, skipping file %s" % (slug, context)
                     models.DataIssue.objects.create(parser='caves', message=message)
                     print(message)
                     
@@ -169,7 +184,7 @@ def readcave(filename):
                     entrance = models_caves.Entrance.objects.get(entranceslug__slug = slug)
                     ce = models_caves.CaveAndEntrance.objects.update_or_create(cave = c, entrance_letter = letter, entrance = entrance)
                 except:
-                    message = " ! Entrance text (slug) %s missing %s" % (slug, context)
+                    message = " ! Entrance setting failure, slug: %s letter: %s" % (slug, letter)
                     models.DataIssue.objects.create(parser='caves', message=message)
                     print(message)
                 

urls.py

@@ -23,7 +23,7 @@ actualurlpatterns = patterns('',
     
     url(r'^troggle$',            views_other.frontpage,      name="frontpage"),
     
-    url(r'^caves/?$',    views_caves.caveindex,      name="caveindex"),
+    url(r'^caves$',    views_caves.caveindex,      name="caveindex"),
     url(r'^people/?$',   views_logbooks.personindex, name="personindex"),
 
     url(r'^newqmnumber/?$',              views_other.ajax_QM_number, ),