From: Warren Young
Date: February 8 2008 2:29am
Subject: Re: RFC: What to do about the integer type mess?
List-Archive: http://lists.mysql.com/plusplus/7440
Message-Id: <47ABBE6C.2090205@etr-usa.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Chris Morgan wrote:
> 
> I'd vote for using the C99 defines. 

Thanks for the moral support. :)

> does this preclude using 'long long'? 

No, that's handled through MySQL++'s ulonglong and longlong types now.

(If we go with the stdint.h patch, these would perhaps go away in favor 
of int64_t and uint64_t, but that depends on what the various platforms 
have in their stdint.h.  So far, I've only looked at glibc on CentOS 5. 
  We might have to keep these typedefs.)

Basically, you run into trouble when the type you want to use isn't in 
mysql_type_info::types[], a static array defined at the top of 
lib/type_info.cpp.  More precisely, it supports the types you see there, 
plus anything we typedef against in lib/sql_types.h.  So for example, 
std::string is supported, because we use that for sql_mediumtext.