[orbit-cpp-cactus] Status report 16/04/02 (fwd)

[ERDI Gergo <cactus cactus rulez org>]

Resending because this was possibly lost in the mail.gnome.org problems

-- 
   .--= ULLA! =---------------------.   `We are not here to give users what
   \     http://cactus.rulez.org     \   they want'  -- RMS, at GUADEC 2001
    `---= cactus@cactus.rulez.org =---'
C.A.C.T.U.S.: Cybernetic Artificial Construct Trained for Ultimate Sabotage

---------- Forwarded message ----------
Date: Tue, 16 Apr 2002 11:53:26 +0200 (CEST)
From: ERDI Gergo <cactus@cactus.rulez.org>
To: orbit-list@gnome.org
Subject: [orbit-cpp-cactus] Status report 16/04/02

Hi,

So after finding that evil is-fixed bug in ORBit2, I was finally able to
continue working on array support. A reminder from my previous report:

>         Atomic types (incl. boolean) -- tested
>         Objects -- not really tested as arguments/return values
>         Enumerations -- not tested
>         Strings -- tested

Since then, I've implemented typedefs, but haven't tested them
throughoutly (atomic and arrays seem to work OK). As for arrays,
arrays of both fixed-length and variable-length work (tested with longs
and strings). Next will be more testing of arrays and support for
exceptions containing arrays.

Remember, you can track it in the orbit-cpp-cactus branch of the ORBit
module. It has less features than the main C++ branch, but the features
that work are more reliable since there is no more code assuming 
'reinterpret_cast<CObject*>(&CPPObject)' is valid (which is the main
problem plaguing the main C++ branch).

-- 
   .--= ULLA! =---------------------.   `We are not here to give users what
   \     http://cactus.rulez.org     \   they want'  -- RMS, at GUADEC 2001
    `---= cactus@cactus.rulez.org =---'
Inscription found at Pompeii: Nero MCCCXXXVII VkrIpVII kIddXIII est.