Stop harvesting job if the package extent could not be saved. Still shows an ugly SA exception

ckanext/harvest/commands/harvester.py

@@ -118,10 +118,20 @@ class Harvester(CkanCommand):
         packages = [extra.package \
                     for extra in \
                     Session.query(PackageExtra).filter(PackageExtra.key == 'bbox-east-long').all()]
-        for package in packages:
-            save_extent(package)
 
-        print "Done. Extents generated for %i packages" % len(packages)
+        error = False
+        for package in packages:
+            try:
+                save_extent(package)
+            except:
+                errors = True
+ 
+        if error:
+            msg = "There was an error saving the package extent. Have you set up the package_extent table in the DB?"
+        else:
+            msg = "Done. Extents generated for %i packages" % len(packages)
+
+        print msg
 
     def run_harvester(self, *args, **kwds):
         from pylons.i18n.translation import _get_translator

ckanext/harvest/controllers/harvesting.py

@@ -224,8 +224,7 @@ class HarvestingJobController(object):
             # Create new package from data.
             package = self._create_package_from_data(package_data)
             if package.extras.get('bbox-east-long'):
-                save_extent(package)
-
+                self.save_package_extent(package)
             log.info("Created new package ID %s with GEMINI guid %s", package.id, gemini_guid)
             harvested_doc = HarvestedDocument(
                 content=content,
@@ -241,8 +240,7 @@ class HarvestingJobController(object):
         else:
             package = self._create_package_from_data(package_data, package = package)
             if package.extras.get('bbox-east-long'):
-                save_extent(package)
-   
+                self.save_package_extent(package)
             log.info("Updated existing package ID %s with existing GEMINI guid %s", package.id, gemini_guid)
             harvested_doc.content = content
             harvested_doc.source = self.job.source
@@ -252,6 +250,13 @@ class HarvestingJobController(object):
             assert gemini_guid == package.documents[0].guid
             return package
 
+    def save_package_extent(self,package):
+        try:
+            save_extent(package)
+        except:
+           log.error("There was an error saving the package extent. Have you set up the package_extent table in the DB?")
+           raise Exception
+
     def get_content(self, url):
         try:
             http_response = urllib2.urlopen(url)

ckanext/harvest/lib/__init__.py

@@ -106,8 +106,7 @@ def save_extent(package,extent=False):
 
         Session.commit()
         log.info(msg, package.id)
+        return package
     except:
         log.error('An error occurred when saving the extent for package %s',package.id)
-    finally:
-        return package
-
+        raise Exception