Newsgroups: php.internals
Path: news.php.net
Xref: news.php.net php.internals:43964
Mailing-List: contact internals-help@lists.php.net; run by ezmlm
Received-SPF: pass (pb1.pair.com: domain pobox.com designates 208.72.237.25 as permitted sender)
Message-ID: <4A0CA36D.7020207@pobox.com>
Date: Thu, 14 May 2009 16:04:13 -0700
User-Agent: Thunderbird 2.0.0.21 (Macintosh/20090302)
MIME-Version: 1.0
To: internals@lists.php.net
CC: Georg Richter <georg@php.net>, Andrey Hristov <andrey@php.net>
References: <49F3E835.6060009@pobox.com>
In-Reply-To: <49F3E835.6060009@pobox.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: [PATCH] mysqli #35203 / #48065 Eliminate special case for
 calling procedures in mysqli
From: schwern@pobox.com (Michael G Schwern)

Ping?

There's a fully formed patch here, with tests, to fix a mysqli bug.  I haven't
gotten any feedback.

Here's the original message with the patch.
http://news.php.net/php.internals/43773


Michael G Schwern wrote:
> This is a patch against 5.2.9 to fix mysqli::query so a user can call stored
> procedures the same as they do any other statement.  No more multi_query() and
> next_result() work arounds necessary to avoid a "Commands out of sync" error.
> 
> I note this has been rejected several times in the tracker as not being a bug.
>  I feel its a clear encapsulation violation making the user special case a
> query just because it's calling a stored procedure.  Aside from just being
> very inconvenient, some API wrappers around mysqli, Drupal 6 for example,
> leave the user with no way to get at mysqli::multi_query().
> 
> Anyhow, here's a complete patch with a test.  The technique and code is lifted
> from Perl's DBD::mysql driver, which you can see here as
> mysql_st_free_result_sets()
> http://cpansearch.perl.org/src/CAPTTOFU/DBD-mysql-4.011/dbdimp.c
> 
> Before each query it frees any results still hanging around on the connection,
> which is essentially what a user has to do.  I'm not really a C programmer so
> I left most of the code as is, do/while loop and all.
> 
> I'm not sure how mysqli normally handles errors so I just went with a printf()
> and return trusting that you'll fix that up.  It causes test 057 to fail
> because that test deliberately generates an out of sync error, which my code
> diligently prints.  So that should go away with fixed error handling.
> 
> Thanks,
> Schwern
> 
> 


-- 
ROCKS FALL! EVERYONE DIES!
	http://www.somethingpositive.net/sp05032002.shtml