Message-ID: <gh-pgjdbc-pgjdbc-2648-rc2131794070@github.com>
From: "vlsi (@vlsi)" <noreply+vlsi@github.com>
To: "pgjdbc/pgjdbc" <noreply+pgjdbc-pgjdbc@github.com>
Date: Fri, 06 Jun 2025 08:49:27 +0000
Subject: Re: [pgjdbc/pgjdbc] PR #2648: Fix getObject(int, class)
In-Reply-To: <gh-pgjdbc-pgjdbc-2648@github.com>
References: <gh-pgjdbc-pgjdbc-2648@github.com>
List-Id: <gh-pgjdbc-pgjdbc.github.com>
X-GitHub-Author-Login: vlsi
X-GitHub-Comment-Id: 2131794070
X-GitHub-Comment-Type: review_comment
X-GitHub-Commit: 55b49295abfee54d3d5a930d9339cbf7d20a583b
X-GitHub-Issue: 2648
X-GitHub-Line: 665
X-GitHub-Path: pgjdbc/src/main/java/org/postgresql/jdbc/PgCallableStatement.java
X-GitHub-Repo: pgjdbc/pgjdbc
X-GitHub-Type: review_comment
X-GitHub-Url: https://github.com/pgjdbc/pgjdbc/pull/2648#discussion_r2131794070
Content-Type: text/plain; charset=utf-8

(on pgjdbc/src/main/java/org/postgresql/jdbc/PgCallableStatement.java:665)

This conversion does not look right.
I believe we should mimic `PgResultSet#getObject(int, Class)` so we use proper conversion methods for various types.
For instance, if a call returns `Timestamp`, and the user asks `getObject(1, String.class)` we should probably return a string-based representation rather than failing with `ClassCastException`.

The same goes for `getObject(1, LocalDateTime.class)`.
If we ever support more classes, we'd better add the relevant tests for it rather than release a feature and get a lot of complains on `ClassCastException` later