On Fri, Sep 19, 2008 at 9:30 PM, Jean-Paul Calderone <exarkun@divmod.com> wrote:
On Fri, 19 Sep 2008 18:26:05 +0200, Amaury Forgeot d'Arc <amauryfa@gmail.com> wrote:
Hello Maciej,
Maciej Fijalkowski wrote:
Hello,
I'm a little clueless about exact semantics of following snippets:
http://paste.pocoo.org/show/85698/
is this fine? or shall I fill the bug? (the reason to ask is because a) django is relying on this b) pypy implements it differently)
Note that python 3.0 has a different behaviour; in the first sample, it prints: A (<class 'NameError'> ... B (<class 'ZeroDivisionError'>, ...
See the subtle differences between http://docs.python.org/dev/library/sys.html#sys.exc_info http://docs.python.org/dev/3.0/library/sys.html#sys.exc_info
The second example changes its behavior, too. It gives back the NameError from the exc_info call. I'm having a hard time reconciling this with the Python 3.0 documentation. Can you shed some light?
Jean-Paul
I think in python 2.x it's at least against the principle of least surprise. It should not behave differently. The behavior of python 3 though it's even against docs :-/