SQLColAttribute using SQL_DESC_OCTET_LENGTH doesn't return the maximum length

[monetdb-team]

Date: 2010-09-27 18:28:27 +0200
From: @skinkie
To: clients devs <>
Version: 11.1.1 (Mar2011) [obsolete]

Last updated: 2011-03-28 17:31:39 +0200

Comment 14940

Date: 2010-09-27 18:28:27 +0200
From: @skinkie

User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US) AppleWebKit/534.7 (KHTML, like Gecko) Chrome/7.0.517.8 Safari/534.7
Build Identifier:

Following up http://bugs.monetdb.org/show_bug.cgi?id=2660

I investigated what call should provide the client interface with the maximum bytes of the column. The right [1] return values are:

1. The maximum storage size ever possible (MSSQL)
2. The maximum storage size used (MySQL)

MonetDB seems to return the same values for:
SQL_DESC_DISPLAY_SIZE (also used in SQLBindCol)
SQL_DESC_LENGTH (not tested)
SQL_DESC_OCTET_LENGTH (equal to Display Size)

As Sjoerd pointed out in [2] the return value here should be the display size times 6. Or (preferably) the maximum storage size for the greatest field.

[1] http://www.tech-archive.net/Archive/Data/microsoft.public.data.odbc/2004-06/0011.html
[2] http://bugs.monetdb.org/show_bug.cgi?id=2660

Reproducible: Always

Comment 15410

Date: 2011-01-24 14:13:29 +0100
From: @grobian

So, this bug is just about returning 6x SQL_DESC_DISPLAY_SIZE as SQL_DESC_OCTET_LENGTH?

If so, care to cook a patch?

Comment 15447

Date: 2011-02-05 16:01:26 +0100
From: @skinkie

Created attachment 52
Attempt to times 6 the octet length.

The attached patch is an attempt to only do it for 'from' MonetDB.

Attached file: octet_length_times_six.diff (text/plain, 2696 bytes)
Description: Attempt to times 6 the octet length.

Comment 15454

Date: 2011-02-14 14:56:59 +0100
From: @sjoerdmullender

Changeset 0be9028dca99 made by Sjoerd Mullender sjoerd@acm.org in the MonetDB repo, refers to this bug.

For complete details, see http//devmonetdborg/hg/MonetDB?cmd=changeset;node=0be9028dca99

Changeset description:

SQL_DESC_OCTET_LENGTH should give the size in bytes required to copy the data.
This fixes bug #2677.

Comment 15455

Date: 2011-02-14 14:58:54 +0100
From: @sjoerdmullender

Thanks for the patch. I have adapted it slightly and checked it in.
The adaptation is mainly for the closing NULL byte which should also be taken into account.
I'm assuming I can now close the bug.

Comment 15647

Date: 2011-03-28 17:31:39 +0200
From: @sjoerdmullender

The Mar2011 version has been released.