Subject RE: Rock>Java>Threads>Hardplace
Author Mark O'Donohue
>
>
> Message: 13
> Date: Mon, 27 Nov 2000 01:28:28 -0400
> From: "Claudio Valderrama C." <cvalde@...>
> Subject: RE: Rock>Java>Threads>Hardplace
>
>
>> -----Original Message-----
>> From: Jim Starkey [mailto:jas@...]
>> Sent: Domingo 26 de Noviembre de 2000 12:15
>>
>> I think the order of battle would be this:
>>
>> 1. Change all engine .c files to .cpp files
>> 2. Clean up the million warnings
>> 3. Upgrade (i.e. replace) current exception handling
>> 4. Gradually convert major subsystems to objects.
>>
>> I suspect that the first three phases would take under a week.
>> A little experimentation will suffice.
>
> That's an optimistic Wolf. About 1, changing extension of files to be
> recognized by the compiler as C++ won't be so slow. <g> The only problem is
> that AFAIK not all compilers use CPP as the extension, hope they are
> configurable. About 2, probably several problems will be caught immediately
> by the C++ compiler and the code changed easily but I expect that some of
> the warnings require thought. We want to eliminate the cause of the warning,
> not the warning itself. About 3, I'm surprised you think setjump/longjump
> can be replaced quickly by try/throw/catch... maybe I need experience with
> this task to have a decent estimation. About 4... better don't give an
> opinion.
>


Sometimes you just have to bite and start chewing hard.

Do we capture a snapshot and release a "before image" now? or just
start wading in?


Cheers

Mark

(Jerk indeed - what a pleasant bedside manner)

--
Your database needs YOU!
http://firebird.sourceforge.net