Changes between Initial Version and Version 1 of Ticket #14211, comment 21


Ignore:
Timestamp:
05/05/15 07:17:53 (6 years ago)
Author:
jdemeyer
Comment:

Legend:

Unmodified
Added
Removed
Modified
  • Ticket #14211, comment 21

    initial v1  
    22> Interesting. The C++ layer outputs an error message, and the Python layer deals with the error too, because presumably the error from calling `PyHash` from the C++ layer is still pending.
    33
    4 Well, behaviour shouldn't depend on what "presumably" happens. I admit I don't know the proper way to deal with this (a C++ exception and `PyErr_Occurred() != NULL`), but it's not right like this.
     4Well, behaviour shouldn't depend on what "presumably" happens (does the traceback look sensible? If not, you're certainly doing it wrong). I admit I don't know the proper way to deal with this (a C++ exception and `PyErr_Occurred() != NULL`), but it's not right like this.
    55
    66Besides that, I would never print the C++ exception.