Author ncoghlan
Recipients adaptivelogic, eric.snow, gvanrossum, martius, ncoghlan, pitrou, rbcollins, rhettinger, vstinner, yselivanov
Date 2015-01-26.22:46:08
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1422312369.24.0.735240338728.issue17911@psf.upfronthosting.co.za>
In-reply-to
Content
On the triple-or-value front, the main case I see for considering the "you don't need exception state triples any more" argument already lost is the fact exc_info still returns a triple, and __exit__ methods still accept them as unpacked arguments. The C level exception handling APIs are also all still exception state triple based. In this view, __traceback__ is primarily about accessing *chained* tracebacks, while the top-level APIs still focus on exception state triples.

This perspective also has the virtue of confining use of the Python 3 only exception traceback attribute to code that is taking advantage of a Python 3 only feature (i.e. exception chaining). For chaining independent exception state handling, working with exception state triples then remains the "one obvious way to do it", a way which is conveniently also source compatible between Python 2 and 3.
History
Date User Action Args
2015-01-26 22:46:09ncoghlansetrecipients: + ncoghlan, gvanrossum, rhettinger, pitrou, vstinner, rbcollins, eric.snow, yselivanov, adaptivelogic, martius
2015-01-26 22:46:09ncoghlansetmessageid: <1422312369.24.0.735240338728.issue17911@psf.upfronthosting.co.za>
2015-01-26 22:46:09ncoghlanlinkissue17911 messages
2015-01-26 22:46:08ncoghlancreate