I have a table that has a record type column called metrics that has the following struct
{"metric_type":"record_value", "metric_name": "source_table", metric_value": "ABC"}
I can do the following
SELECT col, col2, col3, metrics_value
FROM `my_dataset.mytable` t.
UNNEST(metrics) AS metrics_value
This displays data in BigQuery UI with metrics as {"key":"value"}. However, I need to have these columns added to SELECT statement as per below
SELECT col, col2, col3, metrics.source_table, SUM(metrics.row_count) AS row_count
FROM `dataset.table`
GROUP BY 1,2
I have looked around but cannot find an example how to do it..
appreciate any advice.