[omniORB] omni4 beta2 bugg report

[Duncan Grisby]

On Thursday 29 August, "zhb" wrote:

> 1. I found thart wstring of omni4 beta2 wound translate ISO8859-1 to
> UTF-16 when get data and translate UTF-16 to ISO8859-1 when send
> data,this is error when client using other codeset ,eg. GB2312 for
> chinese.

You are not using wstring the way it's intended. You are creating
wstring literals of GB2312 mapped into ISO 8859-1. You can't do that.
Your string is fundamentally _not_ ISO 8859-1, so you can't treat it
as such. You must translate your GB2312 strings into their proper
Unicode representation. To do literals, you will have to use lots of
\u escapes to get the non-ASCII characters in.

If you write an omniORB codeset implementation for GB2312, you will be
able to use GB2312 directly in strings, and have the ORB translate it
for you. (I will implement it for you for a fee, if you want.)

>   2. When I compile omni4 beta2 in gcc 3.1,it will occur a warning.
> When I see the warning ,It occured beacuse omniORB4 beta2 using
> <iostream.h> ,not standard c++ <iostream>

The ORB library does not use iostream at all. omniNames and the
examples do use iostream.h, because they have to remain compatible
with pre-standard compilers.

Cheers,

Duncan.

-- 
 -- Duncan Grisby         --
  -- duncan@grisby.org     --
   -- http://www.grisby.org --