[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: KTop: Request for porting; ready for KDE 1.1
From:       Alex Zepeda <garbanzo () hooked ! net>
Date:       1999-01-31 21:26:18
[Download RAW message or body]

On Sun, 31 Jan 1999, Chris Schlaeger wrote:

> To be a little bit more precise: The OS specific code is spread across
> at least 6 files. None has more than 300 lines, more than 1/3rd blank
> or comment lines. Makes less than 200 lines of code for each
> module. Since the code pieces for each platform are not #ifdefed into
> each other but each in it's own block I don't see your point.

It's messy. It's like sticking kapp and klocale into one file.  Personally
I'd suggest what kcminfo does/did and have platform specific files
included.

> I haven't found a call to getLocale() in my code. You are probably
> talking about a i18n() marco call. I have 101 of them! How about a
> more precise bug report if you don't want to debug it yourself?

Oh I already fixed it thank you.  You called i18n (which calls getLocale)
before your KApp existed.  Oops.  And _who_ said I didn't want to debug it
myself.

> I haven't found a single call to abort(). So I tried the assign()
> calls. All of them are in situations where it makes no sense to
> continue. There might be nicer ways to terminate an application but
> since all I have to worry about are open files and allocated memory I
> see no reason not to use assert().

assert(), calls abort which calls kill.... which kills the program..
Which as far as I'm concerned is stupid.  If the program can't run, it
should let the user know in a more informative way than "abort trap..".

> Why is 
> 
> #ifdef HAVE_SYS_STAT_H
> #include <sys/stat.h>
> #endif
> 
> a problem? Especially if you put a 
> 
> #ifdef __FreeBSD__
> 
> around it?

Fine if you don't mind a bunch of #ifdef'd headers but IMO that just looks
bad, 'sides that's a tad redundant.

> BTW, Hans Petter is already working on a FreeBSD version, if that is
> the platform you are concerned about.

I imagine he is, but one can't be too sure.

- alex

[prev in list] [next in list] [prev in thread] [next in thread]