Hello,

On 01.12.2010 00:50, Mark Miesfeld wrote:
> The other thing though is, I don't want to start down the path of any
> big changes in APService.cpp for 4.1.0.
>
> If you need to piggy back off of the run as foreground stuff to get
> things working nicely on MAC for 4.1.0, then go ahead.
>
> In that case don't merge that code back into trunk.  For the next
> release after 4.1.0, maybe you and Rainer can come up with a cleaner
> way of doing this on AIX and Mac OS X.
>
> In trunk, I don't see any point of trying to make one function do two
> completely opposite things.  I don't see any point of having a
> function whose name should be runAsDaemon_or_DoNotRunAsDaemon().
> <grin>
>
OK,
I can have a look at that over the holidays (on trunk)...
The whole thing looks a bit like a Rube Goldberg machine :-))
> --
> Mark Miesfeld
Bye
  Rainer
> On Tue, Nov 30, 2010 at 3:40 PM, Mark Miesfeld <[email protected]> wrote:
>> On Tue, Nov 30, 2010 at 3:02 PM, CVBruce <[email protected]> wrote:
>>
>>> rxapi was only a partial success in 4.0.1.  It ran, but the system tried to 
>>> respawn a new server every 10 seconds.
>>>  ...
>>> This is caused by rxapi 'daemonizing' itself.
>>> ...
>>> >From the "Getting Started with Launchd documentation at Apple:
>>>
>>> Jobs run from launchd should not duplicate launchd functionality;
>> Okay, from some comments that Rainer put in APIService.cpp:
>>
>>    // - if this is AIX we check if the rxapi daemon was sarted via SRC
>>    //   - if the daemon was started via SRC we do not morph - the SRC
>> handles this
>>
>> it seems that AIX started via SRC and Apple's Launchd have similar
>> requirements.  Maybe Rainer could make the AIX installation package
>> dependent on being started via SRC.  (But I don't even know what SRC
>> is.)
>>
>> In APIService.cpp, rather than put in all kinds of #ifdefs in main()
>> interspersed through the code, I believe there should be two different
>> functions.  Then in main() you would have, in pseudo code something
>> like:
>>
>> int main(int argc, char *argv[])
>> {
>> #if defined(AIX) || defined(MAC)
>>    return non_daemon_func()
>> #else
>>    return daemon_func()
>> #endif
>> }
>>
>> Then in the 2 different functions, don't use any #ifdefs.  That way
>> the code would be a lot more readable.  Rick might have (probably has)
>> some even better ideas than mine.
>>
>> Maybe there could be different platform files?  If things are really
>> so different on either Mac OS X or AIX, maybe there should be an
>> APIService.cpp file for Mac OS X, just like there is a APIService.cpp
>> file for Windows.  We're not trying to use #ifdefs to make
>> APIService.cpp one file on both Windows and Linux.
>>
>> The APIService.cpp file as it is now, is a bit of a mess.  I want to
>> go in the opposite direction of making it more of a mess.
>>
>> --
>> Mark Miesfeld
>>
> ------------------------------------------------------------------------------
> Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
> Tap into the largest installed PC base & get more eyes on your game by
> optimizing for Intel(R) Graphics Technology. Get started today with the
> Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
> http://p.sf.net/sfu/intelisp-dev2dev
> _______________________________________________
> Oorexx-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/oorexx-devel
>
>


------------------------------------------------------------------------------
Increase Visibility of Your 3D Game App & Earn a Chance To Win $500!
Tap into the largest installed PC base & get more eyes on your game by
optimizing for Intel(R) Graphics Technology. Get started today with the
Intel(R) Software Partner Program. Five $500 cash prizes are up for grabs.
http://p.sf.net/sfu/intelisp-dev2dev
_______________________________________________
Oorexx-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/oorexx-devel

Reply via email to