Author gregory.p.smith
Recipients Arfrever, bpoaugust, brian.curtin, eric.araujo, flox, gregory.p.smith, jldm, ncoghlan, ned.deily, python-dev, r.david.murray, terry.reedy, tim.golden, vstinner
Date 2015-02-24.08:49:24
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1424767765.3.0.499752786027.issue10197@psf.upfronthosting.co.za>
In-reply-to
Content
A side effect of the changes made within are that getstatusoutput() on POSIX systems now returns a different value for status.

The old implementation present in Python 2 and Python 3.3 before this patch returned the raw waitpid() status result as the status value.  ie: getstatusoutput("exit 1")[0] == 256.  the lower 8 bits were reserved for the signal number the process died with, if any.

Now it returns the sanitized subprocess style returncode: positive numbers are the process exit code (so the above example returns 1) and negative numbers are the negative signal number the process died with.

I prefer the new behavior, but this API change is not documented anywhere that I can find.
History
Date User Action Args
2015-02-24 08:49:25gregory.p.smithsetrecipients: + gregory.p.smith, terry.reedy, ncoghlan, vstinner, tim.golden, ned.deily, eric.araujo, Arfrever, r.david.murray, brian.curtin, flox, jldm, python-dev, bpoaugust
2015-02-24 08:49:25gregory.p.smithsetmessageid: <1424767765.3.0.499752786027.issue10197@psf.upfronthosting.co.za>
2015-02-24 08:49:25gregory.p.smithlinkissue10197 messages
2015-02-24 08:49:24gregory.p.smithcreate