Fix selection of columns in graph creation

This commit is contained in:
Ilias Kanellos 2023-05-16 17:34:42 +03:00
parent 3c38f7ba6f
commit 3d69f33c84
1 changed files with 2 additions and 0 deletions

View File

@ -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"))