thswlsqls commented on PR #39102: URL: https://github.com/apache/shardingsphere/pull/39102#issuecomment-5349152491
@terrymanu Done, with one change of approach worth flagging. Keeping `SYSTEM_USER` in `UNPARENTHESIZED_FUNCTION_NAMES` cannot satisfy the quoted-column boundary, because `isUnparenthesizedFunction` matches the unwrapped identifier before any owner is read. The token is now admitted through `specialFunction` as a niladic value, dropped from `unreservedWord`, and the `SQL92FunctionOption` entry is reverted; `visitSpecialFunction` already yields the function segment, so the visitor is unchanged. A new `SQL92BinderIT` covers both boundaries through the binder: bare `SELECT SYSTEM_USER` binds as the niladic function, and `SELECT t."SYSTEM_USER" FROM t_system_user t` keeps full column bound info, which fails if the option entry comes back. That second case needed a `t_system_user` table in the shared binder fixture; the whole binder IT is green (103 cases), as is the SQL92 parser IT (210 cases). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
