Amazon PIVOT bug? "column does not exist"

I'm trying to use the Amazon PIVOT feature and am encountering what appears to be a bug. See minimal query and output below:

> WITH t1 AS (select 1 as a) select 1 from t1 PIVOT (sum(a) for a in (1)) WHERE ('NaN'::FLOAT = 0);
ERROR:  Query unsupported due to an internal error.
DETAIL:  SQL reparse error.
CONTEXT:  column "nan" does not exist in unnamed_pivot1.

The literals in the where clause don't have anything to do with the pivot, so the error message does not seem accurate. Is this indeed a bug? Please let me know if I can provide any further information. Thanks!

主題

分析

標籤

Amazon Redshift

語言

English

mdrach

已提問 2 年前檢視次數 1030 次

1 個回答

最新
最多得票
最多評論

這些答案是否有幫助？支持正確答案，以幫助社區從您的知識中受益。

-1

The literal in the where clause is the problem. The query can't be parsed because there isn't a column named "nan".

Are you trying to check for null with "nan"?

專家

Jon Roberts

已回答 2 年前

mdrach
2 年前
Sorry, not sure I understand your question. Here's a query that gives the output that I expect. The WHERE clause has nothing to do with the rest of the query in the OP or in this example.

WITH t1 AS (select 1 as a) select 1 from t1 PIVOT (sum(a) for a in (1)) WHERE ('hello' = 'world'); ?column? ---------- (0 rows)

Amazon PIVOT bug? "column does not exist"

相關內容