Message-ID: <gh-postgresql-interfaces-psqlodbc-178-c4334883792@github.com>
From: "jarvis24young (@jarvis24young)" <noreply+jarvis24young@github.com>
To: "postgresql-interfaces/psqlodbc" <noreply+postgresql-interfaces-psqlodbc@github.com>
Date: Tue, 28 Apr 2026 11:41:35 +0000
Subject: Re: [postgresql-interfaces/psqlodbc] PR #178: Reject overlong cursor names
In-Reply-To: <gh-postgresql-interfaces-psqlodbc-178@github.com>
References: <gh-postgresql-interfaces-psqlodbc-178@github.com>
List-Id: <gh-postgresql-interfaces-psqlodbc.github.com>
X-GitHub-Author-Login: jarvis24young
X-GitHub-Comment-Id: 4334883792
X-GitHub-Comment-Type: issue_comment
X-GitHub-Issue: 178
X-GitHub-Repo: postgresql-interfaces/psqlodbc
X-GitHub-Type: comment
X-GitHub-Url: https://github.com/postgresql-interfaces/psqlodbc/pull/178#issuecomment-4334883792
Content-Type: text/plain; charset=utf-8

Thanks for the review. I updated the regression test to also verify the diagnostic state for the rejected overlong cursor name. The test now builds the invalid name as SQL_MAX_CURSOR_NAME_LEN + 1, calls SQLGetDiagRec() after SQLSetCursorName() fails, and requires SQLSTATE 34000 before continuing with the existing valid cursor-name path.

Re-tested in WSL with the unixODBC black-box path and ASan/UBSan/gcov build:

```text
TAP version 13
1..1
ok 1 - cursor-name
```