
|
[openbeosstorage]
||
[Date Prev]
[04-2002 Date Index]
[Date Next]
||
[Thread Prev]
[04-2002 Thread Index]
[Thread Next]
[openbeosstorage] Re: BSymLink tests and check_*_name() added
- From: Ingo Weinhold <bonefish@xxxxxxxxxxxxxxx>
- To: openbeosstorage@xxxxxxxxxxxxx
- Date: Fri, 26 Apr 2002 03:21:44 +0200 (MET DST)
On Wed, 24 Apr 2002, Tyler Dauwalder wrote:
> > BTW, I have an idea for reuse of base class tests, which is rather simple:
> > The test class hierarchy follows the hierarchy of the tested classes and
> > the tests in question do not create objects of the base class, but instead
> > let objects be created by virtual functions.
> > The derived test classes just need to add the base class tests (or let
> > them be added by a base class method) and override the virtual functions.
> > Ehh, that's all... as promised, simple.
>
> That sounds good. Let me know how it turns out.
It actually works. :-)
After coding and debugging the whole day, the test suite runs smoothly
again. Don't wonder where the changes are, I haven't commited them yet.
I'm way too tired to get anything checked in now.
> > Tyler, I won't touch EntryTest, as I don't want to steal your work. ;-)
> > ... at least not until I'm done with the forementioned and you force me to
> > do so. ;-)
>
> Yes, well if I wasn't forcing you to work all the time like I do, we
> wouldn't be making any progress at all, now would we? :-)
Yeah, sometimes I'm just unable move my lazy butt. ;-)
> > Exceptionally it looks like I will have quite some time tomorrow, so I'm
> > looking forward to get things done very soon. :-)
>
> Honestly, you're welcome to work on anything of mine if you have time and
> interest. Unless you can get R1 finished before the quarter's over, there'll
> still be *something* left that I can work on.
It's a long time until the quarter is over. ;-)
If you've really that little time, I'd take over the BEntry tests and bug
fixes and leave the doxygen stuff for you. Alright?
CU, Ingo
PS: Vincent, could you provide a header and a source file with empty
method implementations -- you know, what I mean, just `return B_ERROR'
bodies.
This would be nice, because e.g. BStatable and one of the
find_directory() functions use it.
|

|