On Mon, Dec 15, 2008 at 02:09:07PM -0800, Alan Irwin wrote:
> On 2008-12-15 20:12-0000 Andrew Ross wrote:
>
> > Alan,
> >
> > Sounds like an excellent idea. I was thinking of something similar
> > myself. If you are happy to put something together I'll help with the
> > work of propagating it out to the other languages.
>
> Thanks for that encouragement. I made an initial version of C example 31
> (revision 9126) which for now only tests plgvpw. It ctests fine. I also
> checked that the error message was expressed properly when there was a
> mismatch of one of the variables. Could you have a look and fill in a few
> more set/get pairs to make sure the example will scale well to lots of
> set/get pairs? I ask this because I know your C skills are better than
> mine. I will take responsibility for any set/get pairs that you don't do.
>
> I could not avoid using plinit (in order to use plwind as the "setter" for
> plgvpw). However, instead of viewing that as a negative, I decided to view
> that as a positive feature since it tests the capability of our various
> device drivers to make a valid plot file with minimal content. For example,
> I just now checked that x31c01.svg validates properly at validator.w3.org.
> Of course that generated file is also required to make the compare ctest
> work properly for, e.g., finding unimplemented versions of example 31. Sure
> enough, the compare ctest output now has lots of 31's splashed around, but
> fixing that by propagating 31 to the other languages will obviously have to
> wait until we fill out all the set/get pairs in the C example.
Alan,
Thanks for this. I've worked some more on this and we now have a fairly
complete test of the common API in the examples. The list of functions
documented in api.xml but unused in the examples is now
plaxes (called by plbox - I'm not quite sure what the difference is!)
plbin (called directly by plhist so tested indirectly)
plclear (could be added into one of the existing examples, although it
only really makes sense for interactive drivers)
plfamadv (only makes sense for familying drivers)
plsdimap (documented as not a general purpose function?)
plsmem (only makes sense for mem driver so hard to test generally)
plstart (requires specifying a device, so hard to test generally)
This is a big improvement. If people are happy with the changes then I
will start propagating them to other languages.
Andrew
------------------------------------------------------------------------------
SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
The future of the web can't happen without you. Join us at MIX09 to help
pave the way to the Next Web now. Learn more and register at
http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
_______________________________________________
Plplot-devel mailing list
Plplot-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/plplot-devel
