From b243aad2a9604746b0678299f4c91517ce496fde Mon Sep 17 00:00:00 2001 From: Courtney Holcomb Date: Fri, 3 Nov 2023 17:09:18 -0700 Subject: [PATCH] Update docstring for visit_combine_metrics_node --- metricflow/plan_conversion/dataflow_to_sql.py | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/metricflow/plan_conversion/dataflow_to_sql.py b/metricflow/plan_conversion/dataflow_to_sql.py index 4e9417ad03..87b850b937 100644 --- a/metricflow/plan_conversion/dataflow_to_sql.py +++ b/metricflow/plan_conversion/dataflow_to_sql.py @@ -906,12 +906,10 @@ def visit_combine_metrics_node(self, node: CombineMetricsNode) -> SqlDataSet: """Join computed metric datasets together to return a single dataset containing all metrics. This node may exist in one of two situations: when metrics need to be combined in order to produce a single - dataset with all required inputs for a derived metric (in which case the join type is INNER), or when - metrics need to be combined in order to produce a single dataset of output for downstream consumption by - the end user, in which case we will use FULL OUTER JOIN. + dataset with all required inputs for a derived metric, or when metrics need to be combined in order to produce + a single dataset of output for downstream consumption by the end user. - In the case of a multi-data-source FULL OUTER JOIN the join key will be a coalesced set of all previously - seen dimension values. For example: + The join key will be a coalesced set of all previously seen dimension values. For example: FROM ( ... ) subq_9