Failed queries when there's no data yet

0

I have this recurring problem. I construct a web service (lambda + api gateway) that is running some query in timestream, but that query fails because there doesn't happen to be any data in the table that has that column. Because of the way the schema in Timestream is dynamic if you select "someColumn" and there's no data that includes that column yet the query will fail.

You can work around this by selecting * but that's generally a bad idea. In the case of timestream, it also causes you to select more data than you might want, which costs you money.

I think what I'd prefer would be an option that says if this column doesn't exist just don't return it, but don't declare the entire query invalid. Is that possible?

In the short term, I get around this by writing a garbage data record to the table. This isn't great, and I'm not sure what is going to happen when that garbage data gets expired - will Timestream still think the "column" exists?

Is there a better workaround for this? For example, can I somehow pre-populate the schema?

profile picture
wz2b
已提問 6 個月前檢視次數 184 次
1 個回答
0
AWS
支援工程師
Kyle_B
已回答 5 個月前

您尚未登入。 登入 去張貼答案。

一個好的回答可以清楚地回答問題並提供建設性的意見回饋,同時有助於提問者的專業成長。

回答問題指南