Re: testsuite under wine

[NightStrike]

On Wed, Jan 11, 2023 at 1:41 PM NightStrike <nightstrike@gmail.com> wrote:
>
>
>
> On Tue, Jan 10, 2023, 04:19 NightStrike <nightstrike@gmail.com> wrote:
>>
>> On Mon, Jan 9, 2023 at 11:05 AM Jacek Caban <jacek@codeweavers.com> wrote:
>> > Sure, I'd be interested how well my latest patch works for NightStrike.
>> > It should give us overview about how much conhost interferes with test
>> > results in practice when we get easy stuff out of the way.
>>
>> Generally, more tests run and pass, but most still fail for different
>> reasons.  IOW, the cursor hiding escapes are gone, but the EOLs still
>> don't match until we modify the tests themselves.
>
>
> Quick update that I found more tests that fail with what I assume are Wine 
> escape codes, K and C. (Ie ^[[K).  I can't figure out under what context they 
> appear, and I don't know what they are for.

With Jacek's latest patch:

            __________________________________________________
           /    PASS   FAIL  ?PASS  ?FAIL  UNSUP  UNRES UNTEST
           |--------------------------------------------------
    acats  |       0      0      0      0      0      0      0
      gcc  |  164293   1652      6   1150   3213    810      0 !E! !W!
      gdc  |    4167   1658      0      0   1553    121      0
 gfortran  |   66811    210      8    257    247      0      0 !W!
      gm2  |    1168   6673      0      0      0   2065      0
     gnat  |    3360     24      0     23     14      0      0 !W!
      g++  |  226182   1314      4   2346  10900     89      0 !E! !W!
libatomic  |      53      1      0      0      0      0      0 !W!
libstdc++  |   14563     50     26     74    835      1      0 !W!
  obj-c++  |    1502      8      1      1     79      0      0 !W!
     objc  |    1882      9      0      6     70     10      0 !E!
     rust  |    4148    432      0     39      0      0      0 !W!
           |--------------------------------------------------
           |  488129  12031     45   3896  16911   3096      0
           \__________________________________________________

We're making good progress here!

(Note that the numbers above for rust are for the gcc master branch,
but every line ending failure is fixed on the rust dev branch, leaving
only 1 actual bug)