Not sure to understand all those "high level dimensions", I'm probably lost in 
the 4th one... but it seems option 1. is our choice: "create a test file for 
each interface".

Seb


Le Monday 06 October 2008 22:10:17 Rob Hamerling, vous avez écrit :
> Joep Suijs wrote:
> > Until now, I assumed we'd only need two files to compose a test:
> > - a board file, which tells the type of pic, it's configuration (like
> > type and frequency of the oscillator) and it's connections.
> > - a test file, which tests a specific function.
> >
> > But in some cases, there are more options. Some interfaces, like i2c
> > and serial, have multiple libaries. So this gives us 3 files: the
> > board, optional: the interface and then the test. How do we deal with
> > this?
> >
> > 1. create a test file for each interface
> > This gives more files with similar code. Not ideal and a bit
> > error-prone but not that bad.
> >
> > 2. adapt the test file when you want an other interface
> > Quick and dirty (mostly dirty). Not suitable for automated tasks.
> >
> > 3. have the to-be superscript support more then 2 files to compose a
> > test (3 or maybe more?).
> > Looks at the best option at first glance. But it also means we can not
> > define tests as simple 2-dimensional (test x on board y). The benefits
> > might not add up to the costs of the complexity induced.
>
> The 'intermediate' libraries (prerequisites for other libraries) will
> generally be for relatively simple common 'lower layer' services and
> should have been tested independently. Thus these libs themselves can be
> considered transparent (not take place in the test of the higher level
> library). But of course the interface between higher level and lower
> level library is part of the test of the higher level library.
> This means we have always 'two-dimensional' test setups: highest level
> library and hardware.
>
> Regards, Rob.



-- 
Sébastien LELONG
http://www.sirloon.net
http://sirbot.org

--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"jallib" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at 
http://groups.google.com/group/jallib?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to