Message-ID: <gh-postgresql-interfaces-psqlodbc-51-c2395006277@github.com>
From: "Hunaid2000 (@Hunaid2000)" <noreply+Hunaid2000@github.com>
To: "postgresql-interfaces/psqlodbc" <noreply+postgresql-interfaces-psqlodbc@github.com>
Date: Sat, 05 Oct 2024 10:10:20 +0000
Subject: Re: [postgresql-interfaces/psqlodbc] issue #51: SQL_DESC_OCTET_LENGTH regression test difference
In-Reply-To: <gh-postgresql-interfaces-psqlodbc-51@github.com>
References: <gh-postgresql-interfaces-psqlodbc-51@github.com>
List-Id: <gh-postgresql-interfaces-psqlodbc.github.com>
X-GitHub-Author-Login: Hunaid2000
X-GitHub-Comment-Id: 2395006277
X-GitHub-Comment-Type: issue_comment
X-GitHub-Edited-At: 2024-10-05T10:13:01Z
X-GitHub-Issue: 51
X-GitHub-Repo: postgresql-interfaces/psqlodbc
X-GitHub-Type: comment
X-GitHub-Url: https://github.com/postgresql-interfaces/psqlodbc/issues/51#issuecomment-2395006277
Content-Type: text/plain; charset=utf-8

It basically depends on client encoding, I guess.
I get 20 on Windows and 40 on Ubuntu. The length is set in `pgtype_attr_buffer_length` function.
 
https://github.com/postgresql-interfaces/psqlodbc/blob/59728b929b54b3b230aead909895483b5b2f35a1/pgtypes.c#L1098-L1118

It depends on `coef = conn->mb_maxbyte_per_char;` which is set on the line `629` in this block.
https://github.com/postgresql-interfaces/psqlodbc/blob/59728b929b54b3b230aead909895483b5b2f35a1/connection.c#L615-L632

In case of Ubuntu:
```
postgres=# show client_encoding;
 client_encoding
-----------------
 UTF8
(1 row)
```
https://github.com/postgresql-interfaces/psqlodbc/blob/59728b929b54b3b230aead909895483b5b2f35a1/multibyte.c#L211-L236

which maps to 4 here: `pg_mb_maxlen(self->ccsc);`. As a result, we get 10 * 4 = 40.
For Windows, I believe its just `prec * WCLEN` so 10 * 2 = 20.

@df7cb In your case, I guess `pg_mb_maxlen` returns 1.