Re: Serial patch for POSE 3.0a2



On Sat, 20 Nov 1999, Vadim Strizhevsky wrote:

> I didn't look into the details of the patch, but patched 3.0a3 seems
> to work much better than the original. Othwerwise it was droping and
> not connecting constantly. 

The essential problem was having to start gpilotd _after_ pressing
hotsync. That pissed me off.

> I believe this is called "Inheritence by Code Replacement" (TM)  :-)

Yeah, remember that from my Theory Of Objects courses. Mentioned in the
same chapter as "Obscure OO usage" ...

> What they doing is just have one class Platform, with no derive
...
> code-wise cleaner then #ifdef for platform each function in the same
> file.

Mmm.... You don't convince me. the main is in SrUnix anyways, so why not
have a global Platform *platform;, which they new PlatformUnix(). You'll
still avoid #ifdefs, since you'll just compile the BuildXXX you want, and
SrcShared/*c access a "extern Platform *platform"

Nor do I believe the staticness of the methods will begain that much.

Ah well, this is probably quite off-topic...

/dev/eskil
---



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]