rtadepalli opened a new pull request, #34358: URL: https://github.com/apache/arrow/pull/34358
### Rationale for this change The `AvaticaConnection` class within the Calcite library [executes](https://github.com/apache/calcite-avatica/blob/b57eb7cd31a90d3f46b65c13832b398be5b0dad9/core/src/main/java/org/apache/calcite/avatica/AvaticaConnection.java#L357) the `prepareStatement(...)` function on a DB connection in the following order: 1. Calls `prepare()` on the `MetaImpl` class, which in our case is `ArrowFlightMetaImpl`. This function executes a command and returns a prepared statement. 2. Calls `newPreparedStatement` on the JDBC factory class, which in our case is `ArrowFlightJdbcFactory`. This function today creates an entirely new `PreparedStatement` without checking the statement cache, thereby causing 2 executions. This PR fixes the issue by checking whether or not the `StatementHandle` has been recorded before, and if so, uses the results of the previously executed command. If not, it creates a brand new `PreparedStatement`. My understanding is that `ConcurrentHashMap` (which is what is being used to keep a mapping of statement handles to prepared statements) is eventually consistent, so there _may_ be cases where the prepared statement execution runs twice. ### What changes are included in this PR? Check to see if the `PreparedStatement` has run before, and if so, use that instead of executing the command to contact the server again. ### Are these changes tested? Existing tests should cover this case. Namely, `ArrowFlightPreparedStatementTest` I _think_ should cover all cases. ### Are there any user-facing changes? There are no user facing changes. -- 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: github-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
