Issue #1664: Unit Test support for mapserver

[Frank Warmerdam]

Howard Butler wrote:
> 3.  There is already an initiative to more fully develop a C API for 
> MapServer that MapScript etc can take advantage of.  The unittesting 
> initiative must be in alignment with this.

Howard,

This is a good point.  I was intending to suggest that unit tests be
primarily be aimed at the MapServer C API rather than being too
intertwined into internal functions, and structures.  Of course, we
don't exactly have a public API defined yet, but we are making some
progress in this regard.  Loosely speaking the public API is stuff
that is made accessable in Mapscript (which is why mapscript is usually
an adequate testing environment).

> 5.  A dependency on cunit for regular users is a deal breaker.

I don't think Umberto was suggesting this as a requirement.  Just that
a big red warning would be issued if cunit was not available (thus
disabling the unit tests) though I could have been confused.

 From my brief exposure to cppunit in the FDO product, it wasn't clear
to me that cppunit (the C++ equiv) was realling adding alot of value
over something we might do ourselves without outside requirements.
But then I didn't dig too deeply into how it all worked.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent