From 3d69f33c847b39fd9eb8de3a3cf93d5535a2438f Mon Sep 17 00:00:00 2001 From: ikanellos Date: Tue, 16 May 2023 17:34:42 +0300 Subject: [PATCH] Fix selection of columns in graph creation --- .../oozie_app/create_openaire_ranking_graph.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/dhp-workflows/dhp-impact-indicators/src/main/resources/eu/dnetlib/dhp/oa/graph/impact_indicators/oozie_app/create_openaire_ranking_graph.py b/dhp-workflows/dhp-impact-indicators/src/main/resources/eu/dnetlib/dhp/oa/graph/impact_indicators/oozie_app/create_openaire_ranking_graph.py index 50d2cd99b..3d131933d 100644 --- a/dhp-workflows/dhp-impact-indicators/src/main/resources/eu/dnetlib/dhp/oa/graph/impact_indicators/oozie_app/create_openaire_ranking_graph.py +++ b/dhp-workflows/dhp-impact-indicators/src/main/resources/eu/dnetlib/dhp/oa/graph/impact_indicators/oozie_app/create_openaire_ranking_graph.py @@ -131,8 +131,10 @@ cites_df = spark.read.json(graph_folder + "/relation")\ & (F.col('dataInfo.deletedbyinference') == "false")\ & (F.col('dataInfo.invisible') == "false"))\ .drop('dataInfo.deletedbyinference').drop('dataInfo.invisible')\ + .drop('deletedbyinference').drop('invisible')\ .repartition(num_partitions, 'citing').drop('relClass')\ .withColumn('collected_lower', F.expr('transform(value, x -> lower(x))'))\ + .drop('collectedfrom.value')\ .drop('value')\ .where( (F.array_contains(F.col('collected_lower'), "opencitations"))