Merge pull request #272 from moj-analytical-services/dbrix_fix

possible fix for databricks issue

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "splink"
-version = "2.1.4"
+version = "2.1.5"
 description = "Implementation of Fellegi-Sunter's canonical model of record linkage in Apache Spark, including EM algorithm to estimate parameters"
 authors = ["Robin Linacre <[email protected]>", "Sam Lindsay", "Theodore Manassis"]
 license = "MIT"

splink/cluster.py

@@ -82,9 +82,15 @@ def _check_graphframes_installation(spark):
         if key in config_keys:
             if "graphframes" in value:
                 graphframe_jar_registered = True
+
+    databricks_backend = False
+    for key, value in all_config:
+        if "databricks" in key:
+            databricks_backend = True
 
     if not graphframe_jar_registered:
-        raise Exception(graphframes_jar_missing_message)
+        if not databricks_backend:
+            raise Exception(graphframes_jar_missing_message)
 
     from pyspark.sql import Row