I don't exactly recall the details of ExtractIO, in fact it was contributed by someone else (either Joel Burger or Brian Schroeder; neither of which use SBV any longer I think). However, since I'm getting test failures I'm hesitant to incorporate this particular change. (I did merge other parts of your commit.)
You can replicate the failure with:
make test TGT=exceptionRemote1
It appears the change to ExtractIO you made changes the exception-processing order, which isn't terribly surprising but leads to confusing behavior.
I'll leave this here so you can investigate yourself to see perhaps you didn't need this change exactly? Having a clean test-suite is important and when an exception happens it's good to have the report more accurate. (If you run the above test you'll see the change in behavior; new error message is much more cryptic.)
Hi,
I ran the test-suite on your pull request. Alas, I got test failures. I traced it down to the change you made to
ExtractIO
. I had to revert it thusly:https://github.com/LeventErkok/sbv/commit/3c04d86032556053c749e49d9ccb24df52e2d1ac
I don't exactly recall the details of
ExtractIO
, in fact it was contributed by someone else (either Joel Burger or Brian Schroeder; neither of which use SBV any longer I think). However, since I'm getting test failures I'm hesitant to incorporate this particular change. (I did merge other parts of your commit.)You can replicate the failure with:
It appears the change to
ExtractIO
you made changes the exception-processing order, which isn't terribly surprising but leads to confusing behavior.I'll leave this here so you can investigate yourself to see perhaps you didn't need this change exactly? Having a clean test-suite is important and when an exception happens it's good to have the report more accurate. (If you run the above test you'll see the change in behavior; new error message is much more cryptic.)