Re: cvs: php-src(PHP_5_3) /ext/mysqlnd mysqlnd.h mysqlnd_wireprotocol.c

On Mon, Sep 15, 2008 at 4:07 PM, Andrey Hristov <andrey@php.net> wrote: > andrey Mon Sep 15 14:07:44 2008 UTC > > Modified files: (Branch: PHP_5_3) > /php-src/ext/mysqlnd mysqlnd.h mysqlnd_wireprotocol.c > Log: > MFH:Compile-in a string-to-int conversion for INT columns by default, > controlled by a ini setting. > > > > > http://cvs.php.net/viewvc.cgi/php-src/ext/mysqlnd/mysqlnd.h?r1=1.3.2.16&r2=1.3.2.17&diff_format=u > Index: php-src/ext/mysqlnd/mysqlnd.h > diff -u php-src/ext/mysqlnd/mysqlnd.h:1.3.2.16 php-src/ext/mysqlnd/mysqlnd.h:1.3.2.17 > --- php-src/ext/mysqlnd/mysqlnd.h:1.3.2.16 Tue Jun 24 11:01:38 2008 > +++ php-src/ext/mysqlnd/mysqlnd.h Mon Sep 15 14:07:44 2008 > @@ -18,19 +18,19 @@ > +----------------------------------------------------------------------+ > */ > > -/* $Id: mysqlnd.h,v 1.3.2.16 2008/06/24 11:01:38 andrey Exp $ */ > +/* $Id: mysqlnd.h,v 1.3.2.17 2008/09/15 14:07:44 andrey Exp $ */ > > #ifndef MYSQLND_H > #define MYSQLND_H > > -#define MYSQLND_VERSION "mysqlnd 5.0.4-dev - 080501 - $Revision: 1.3.2.16 $" > +#define MYSQLND_VERSION "mysqlnd 5.0.4-dev - 080501 - $Revision: 1.3.2.17 $" > #define MYSQLND_VERSION_ID 50004 > > /* This forces inlining of some accessor functions */ > -#define MYSQLND_USE_OPTIMISATIONS 0 > +#define MYSQLND_USE_OPTIMISATIONS 1 > > > -/* #define MYSQLND_STRING_TO_INT_CONVERSION */ > +#define MYSQLND_STRING_TO_INT_CONVERSION > /* > This force mysqlnd to do a single (or more depending on ammount of data) > non-blocking read() calls before sending a command to the server. Useful > > http://cvs.php.net/viewvc.cgi/php-src/ext/mysqlnd/mysqlnd_wireprotocol.c?r1=1.4.2.17&r2=1.4.2.18&diff_format=u > Index: php-src/ext/mysqlnd/mysqlnd_wireprotocol.c > diff -u php-src/ext/mysqlnd/mysqlnd_wireprotocol.c:1.4.2.17 > php-src/ext/mysqlnd/mysqlnd_wireprotocol.c:1.4.2.18 > --- php-src/ext/mysqlnd/mysqlnd_wireprotocol.c:1.4.2.17 Fri Sep 12 15:08:41 2008 > +++ php-src/ext/mysqlnd/mysqlnd_wireprotocol.c Mon Sep 15 14:07:44 2008 > @@ -1564,7 +1564,7 @@ > ZVAL_LONG(*current_field, v); > } else { > uint64 v = (uint64) atoll((char *) p); atoll is not portable, you may add some define in the headers for that. I can do it while fixing the windows builds. The same applies for the places where you check for SIZEOF_LONG for the max value of int64, you may prefer a MAX_INT64_SIZE somewhere in the header and test the actual size of int64 instead :) Please let me know if you like me to do it or if you have pending commits, Cheers, -- Pierre http://blog.thepimp.net | http://www.libgd.org