From 3c38f7ba6f53f735e4eb23370cc3d09eedcc808a Mon Sep 17 00:00:00 2001 From: ikanellos Date: Tue, 16 May 2023 17:32:53 +0300 Subject: [PATCH] Fix selection of columns in graph creation --- .../oozie_app/create_openaire_ranking_graph.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) 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 182fd9309..50d2cd99b 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 @@ -132,8 +132,8 @@ cites_df = spark.read.json(graph_folder + "/relation")\ & (F.col('dataInfo.invisible') == "false"))\ .drop('dataInfo.deletedbyinference').drop('dataInfo.invisible')\ .repartition(num_partitions, 'citing').drop('relClass')\ - .withColumn('collected_lower', F.expr('transform(collectedfrom.value, x -> lower(x))'))\ - .drop('collectedfrom.value')\ + .withColumn('collected_lower', F.expr('transform(value, x -> lower(x))'))\ + .drop('value')\ .where( (F.array_contains(F.col('collected_lower'), "opencitations")) | (F.array_contains(F.col('collected_lower'), "crossref"))