'Wrong column id showing up in join stage of BigQuery console (Bigquery SQL Plan)

While executing a SQL query on BigQuery console, getting a wrong column-id (which didn't show up while scanning the tables) in the join operator of Join+ stage.

As seen from the attached screenshot, there is scan on first table (stage: S00:Input) where the column-names are referring to ids -> $1, $2, $3. And there is scan on second table (stage: S01:Input) where the column-names are referring to ids -> $20, $21, $22. And there is third scan on a table (stage: S05:Join+) where the column-names are referring to ids -> $10, $11, $12.

And in the same stage (stage: S05:Join+), there are two 'JOIN' operators performing INNER HASH JOIN. In the first join, if we notice there is first join (INNER HASH JOIN EACH WITH ALL ON $33 = $1) based on two column-ids $33 and $1. But there is no column in the above table scans which is referencing to column-id $33.

Screenshot: BigQuery Bug Screenshot

I'm not sure whether this could be a bug from Google BigQuery side or am I missing anything here. BigQuery Team, could you please have a look into this

cc: Google BigQuery Team.

Thanks in Advance.

enter image description here



Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution Source