You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Jason Shirk (@lzybkr) perhaps you could help us out here. Do we need a "dumber" PSHost implementation specifically for the non-interactive use case? I had to tunnel through to our RawUserInterface in 9a5c010 to make our Pester tests run on AppVeyor.
What do the Windows hosts do for this kind of thing?
I'm not sure how much help I can be w/o specifics, but I noticed one possible issue - new Size(0,0) seems like it has the potential to cause problems. I'm guess we need to assume a non-zero width somewhere to decide how to wrap lines or whatever.
Jason Shirk (@lzybkr) You were on the right track! As soon as we were able to revert our use of WindowWidth/Height back to BufferWidth/Height, these tests all started passing. Hurray for .NET Core getting updated.
Activity
andyleejordan commentedon Jan 23, 2016
I believe there are now 12 tests that fail due to lack of a TTY.
We do not know how to fix this, it heavily involves the implementation of the
PSHost(and itsUserInterfaceand that'sRawUserInterface).[-]Fix 7 non-interactive failing Pester tests[/-][+]Fix 12 non-interactive failing Pester tests[/+]andyleejordan commentedon Jan 23, 2016
Jason Shirk (@lzybkr) perhaps you could help us out here. Do we need a "dumber"
PSHostimplementation specifically for the non-interactive use case? I had to tunnel through to ourRawUserInterfacein 9a5c010 to make our Pester tests run on AppVeyor.What do the Windows hosts do for this kind of thing?
lzybkr commentedon Jan 25, 2016
I'm not sure how much help I can be w/o specifics, but I noticed one possible issue - new Size(0,0) seems like it has the potential to cause problems. I'm guess we need to assume a non-zero width somewhere to decide how to wrap lines or whatever.
andyleejordan commentedon Feb 3, 2016
Jason Shirk (@lzybkr) You were on the right track! As soon as we were able to revert our use of
WindowWidth/Heightback toBufferWidth/Height, these tests all started passing. Hurray for .NET Core getting updated.