Matwey V. Kornilov
Matwey V. Kornilov
Hi, I see the same issue with 0.24.3 on ppc64 (big endian architecture). ``` [ 955s] 60/79 Test #60: test_xnpy .........................***Failed 0.01 sec [ 955s] [doctest] doctest version is "2.4.8"...
> Hey @matwey, > > thank you so much for your PR! I've left some initial comments inline and would be happy to help you get this merged. > >...
Ok, I think now I am done with the second part of comments.
You are right. I added some documentation. I am not sure about wording though, so please say if something is missed or not clear enough.
Any progress here?
genicam dump is here: [flir.xml.gz](https://github.com/AravisProject/aravis/files/11618366/flir.xml.gz)
> Hi, > > Thanks for the report. > > It looks like the Genicam data are incorrect. Default cachable for a MaskedInt is WriteThrough. LineMode_Output_Avl and LineMode_Input_Avl are read...
The following is currently still failing for x86 arch for 0.24.7: ``` [ 933s] [doctest] doctest version is "2.4.11" [ 933s] [doctest] run with "--help" for options [ 933s] Built...
Not sure how to reimplement `test_xbuilder.cpp`. The failure appears when gcc is used (it seems to use FPU instead of SSE for floating point).
> Maybe we can use floating point comparisons / assertions instead I think it is enough to make the step be power of 2, like `0.25f`