[FFmpeg-devel] [PATCH 2/2] dca: add new fate tests based on the dcadec-samples test suite

[Hendrik Leppkes]

On Tue, Feb 2, 2016 at 12:24 AM, Michael Niedermayer
<michael at niedermayer.cc> wrote:
> On Mon, Feb 01, 2016 at 09:34:12PM +0100, Hendrik Leppkes wrote:
>> On Mon, Feb 1, 2016 at 9:22 PM, Hendrik Leppkes <h.leppkes at gmail.com> wrote:
>> > On Mon, Feb 1, 2016 at 9:20 PM, Hendrik Leppkes <h.leppkes at gmail.com> wrote:
>> >> ---
>> >>  tests/fate/dca.mak                       | 41 ++++++++++++++++++++++++++++++++
>> >>  tests/ref/fate/dca-xll_51_16_192_768_0   |  1 +
>> >>  tests/ref/fate/dca-xll_51_16_192_768_1   |  1 +
>> >>  tests/ref/fate/dca-xll_51_24_48_768      |  1 +
>> >>  tests/ref/fate/dca-xll_51_24_48_none     |  1 +
>> >>  tests/ref/fate/dca-xll_71_24_48_768_0    |  1 +
>> >>  tests/ref/fate/dca-xll_71_24_48_768_1    |  1 +
>> >>  tests/ref/fate/dca-xll_71_24_96_768      |  1 +
>> >>  tests/ref/fate/dca-xll_x96_51_24_96_1509 |  1 +
>> >>  tests/ref/fate/dca-xll_xch_61_24_48_768  |  1 +
>> >>  10 files changed, 50 insertions(+)
>> >>  create mode 100644 tests/ref/fate/dca-xll_51_16_192_768_0
>> >>  create mode 100644 tests/ref/fate/dca-xll_51_16_192_768_1
>> >>  create mode 100644 tests/ref/fate/dca-xll_51_24_48_768
>> >>  create mode 100644 tests/ref/fate/dca-xll_51_24_48_none
>> >>  create mode 100644 tests/ref/fate/dca-xll_71_24_48_768_0
>> >>  create mode 100644 tests/ref/fate/dca-xll_71_24_48_768_1
>> >>  create mode 100644 tests/ref/fate/dca-xll_71_24_96_768
>> >>  create mode 100644 tests/ref/fate/dca-xll_x96_51_24_96_1509
>> >>  create mode 100644 tests/ref/fate/dca-xll_xch_61_24_48_768
>> >>
>> >
>> > As an extra info, all new samples and the pcm references for the lossy
>> > tests are in total only 1MB, so quite tiny for 19 new fate tests that
>> > greatly increase coverage.
>> >
>>
>> For even more coverage, could also add downmix tests using all the
>> existing samples (down to 6 and 2 respectively).
>
> consideing this is just 1-2mb or something, iam in favor of it
> didnt test though
>

Adding a bunch of identical pcm files seems rather wasteful, so I'm
currently favoring the approach of simply testing both downmix modes
on all of the XLL files which only have a md5 check and not a
reference file, and testing downmix on a few lossy samples where its
actually doing something.
That should still cover a wide variety of code paths.

-  Hendrik