Feature Request: Alternative to ON DUPLICATE KEY UPDATE? about ydb

[dcherednik]

Hello.
Probably UPSERT can help with your task. https://ydb.tech/en/docs/yql/reference/syntax/upsert_into

[vladd11]

Hello. Probably UPSERT can help with your task. https://ydb.tech/en/docs/yql/reference/syntax/upsert_into

No, it can't.

UPSERT INTO users(id, phone, sms_code, sms_code_expiration)
VALUES (?id, ?phone, ?sms_code, ?sms_code_expiration);

will update id column of row, but I need to do this only if row doesn't exist.

[fomichev3000]

UPSERT INTO users(id, phone, sms_code, sms_code_expiration)
VALUES (?id, ?phone, ?sms_code, ?sms_code_expiration);

will update id column of row, but I need to do this only if row doesn't exist.
Technically, updating id with the same value leads to just updating values. To my mind result would be the same, either you call INSERT INTO ... ON DUPLICATE KEY <update all fields> or UPSERT ... <specify all fields>. Do you see any controversial example?

[uh-zuh]

INSERT INTO table1 (id, value, create_time)
VALUES ($id, $value, $create_time)
ON DUPLICATE KEY
UPDATE value = $value;

but UPSERT will modify create_time too.