[issue15365] Traceback reporting can fail if IO cannot be imported

Changes by Eric Snow : -- nosy: +eric.snow ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: http://mail.python.o

[issue15365] Traceback reporting can fail if IO cannot be imported

Changes by Kristján Valur Jónsson : -- status: open -> closed ___ Python tracker ___ ___ Python-bugs-list mailing list Unsubscribe: h

[issue15365] Traceback reporting can fail if IO cannot be imported

Roundup Robot added the comment: New changeset 54524897fafc by Kristján Valur Jónsson in branch 'default': Issue #15365: Make traceback reporting ignore any errors when printing out http://hg.python.org/cpython/rev/54524897fafc -- nosy: +python-dev

[issue15365] Traceback reporting can fail if IO cannot be imported

Antoine Pitrou added the comment: > The trouble is, any errors in the traceback reporting will be ignored > anyway, since there is no way to report them :) Which is a good reason to ignore them, isn't it? ;) -- ___ Python tracker

[issue15365] Traceback reporting can fail if IO cannot be imported

Kristján Valur Jónsson added the comment: Certainly. But the submission filter would have caught that. The question is: Is it acceptible to simply ignore errors from this api when printing tracebacks? There are other errors that can be returned as well. The trouble is, any errors in the trac

[issue15365] Traceback reporting can fail if IO cannot be imported

Antoine Pitrou added the comment: The patch looks simple enough, but you should use spaces, not tabs. -- nosy: +amaury.forgeotdarc, pitrou stage: -> patch review ___ Python tracker ___

[issue15365] Traceback reporting can fail if IO cannot be imported

New submission from Kristján Valur Jónsson : Reporting an error early in the python startup, before importing is properly initialized, can be tricky. For example, here: if (PyImport_ImportFrozenModule("_frozen_importlib") <= 0) { Py_FatalError("Py_Initialize: can't import _frozen_importlib");