[hatari-devel] Hatari screen test (was: Re: fixing errors reported by GCC 10 -fanalyzer)

[Thomas Huth <th.huth@xxxxxxxxx>]

Am Wed, 13 May 2020 19:34:09 +0200
schrieb Thomas Huth <th.huth@xxxxxxxxx>:

> Am Wed, 13 May 2020 18:59:19 +0200
> schrieb Nicolas Pomarède <npomarede@xxxxxxxxxxxx>:
[...] 
> > it depends on the case, most of the time interpreting the results
> > means looking at the screen and  checking if some pixels are aligned
> > or in different colors (depending on which video test is performed),
> > which is not usable with natfeats.  
> 
> We could maybe create a screenshot at the end and then compare if the
> result is the same as a reference picture?

After finding an old fullscreen routine on my hard disk, I've now
implemented such a regression test: The program is run, then the test
script triggers a screenshot via the Hatari command fifo, and then the
screenshot is compared against a reference picture (using the
"identify" command from ImageMagick/GraphicsMagick). Please give it a
try, I hope it works in all environments (except SDL1 builds - screen-
shots are not possible here, I only get a black screen, but since we
plan to get rid of SDL1 anyway in the near future, I do not care too
much about testing that anymore).

 Thomas