[petsc-dev] plans for testing examples?

[Barry Smith]

  Ok, do you have somewhere documented the current Python API for indicating information about runs (what you now generate with your tool from the makfile)?

   I'm ready to put it into a few examples to test it out.

    Barry


On Sep 17, 2012, at 11:03 AM, Jed Brown <jedbrown at mcs.anl.gov> wrote:

> On Mon, Sep 17, 2012 at 10:35 AM, Barry Smith <bsmith at mcs.anl.gov> wrote:
>  Almost there. We need a unique way of mapping from each test to a unique key that can be used to look in your big old output database.  It would be best if this key came automatically, unfortunately the command line options are not unique to a run and if we did a hash on the source code that would change with every trivial change to the source code.  I don't want to use the filename and or directory name because that makes moving examples around more painful.  Hash the help string AND the command line options together is that close enough to unique?
> 
> The problem is that different command line options _intentionally_ produce the same output. The SHA1 is intended to be unique identification of the output, not unique identification of the input. Simply hashing the output file is an easy way to do that.
> 
> Note also that command line options occasionally change while doing exactly the same thing. I think explicit listing of the output SHA1 is a reasonable thing. It's very similar to the problem we have of uniquely identifying commits with distributed SCM.
> 
> Note that the object store I'm describing is basically the same as Git's object store, where the "pointers" into it are stored in our source code.
> 
> http://git-scm.com/book/en/Git-Internals-Git-Objects
> 
> Note that any object not currently pointed at becomes eligible for garbage collection.
>  
> 
>    Note also: in the same place in the example code we provide the "chunk of python for registering tests" we can provide a "chunk of python to register information about the example": what dimension, what equation, what type of solver (TS, SNES, KSP) what type of DM etc thus generate automatically a decent searchable database of examples.
> 
> Yes
>  
>  By using "a chunk of python" rather than defining some "doctext" language we can avoid any need to parse.