Merge pull request #1137 from moj-analytical-services/partial_fix_for…

…_issue_1136 Fix clustering in issue 1136
moj-analytical-services · Mar 22, 2023 · ea33709 · ea33709
2 parents c4ac546 + 1f95830
commit ea33709
Show file tree

Hide file tree

Showing 3 changed files with 3 additions and 3 deletions.
diff --git a/pyproject.toml b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "splink"
-version = "3.7.0"
+version = "3.7.1"
 description = "Fast probabilistic data linkage at scale"
 authors = ["Robin Linacre <[email protected]>", "Sam Lindsay", "Theodore Manassis", "Tom Hepworth", "Andy Bond"]
 license = "MIT"

diff --git a/splink/__init__.py b/splink/__init__.py
@@ -1 +1 @@
-__version__ = "3.7.0"
+__version__ = "3.7.1"
diff --git a/splink/spark/spark_linker.py b/splink/spark/spark_linker.py
@@ -331,7 +331,7 @@ def _break_lineage_and_repartition(self, spark_df, templated_name, physical_name
             r"__splink__df_concat_with_tf",
             r"__splink__df_predict",
             r"__splink__df_tf_.+",
-            r"__splink__df_representatives",
+            r"__splink__df_representatives.+",
             r"__splink__df_neighbours",
             r"__splink__df_connected_components_df",
         ]