Author lemburg
Recipients christian.heimes, cstratak, doko, lemburg, matrixise, pmpp, r.david.murray, vstinner
Date 2020-11-25.09:01:44
SpamBayes Score -1.0
Marked as misclassified Yes
Message-id <1606294905.04.0.494132938241.issue28468@roundup.psfhosted.org>
In-reply-to
Content
I'm -1 on adding any kind of replacement for platform.linux_distribution() tp Python's stdlib. The experiment has failed and we should acknowledge this.

The main reason why it failed was the Linux distros keep inventing new ways to identify themselves, sometimes misuse existing mechanisms to maintain compatibility (e.g. as a Ubuntu or RedHat based OS) or ship with two different release files.

At the time we removed the API, we said that people should use PyPI package to get this information, since those can much more easily be maintained with the ever changing patterns distributions use or invent.

The distro package is one package and if it doesn't satisfy the needs, simply create a new one which does.

Note that the original implementation had a mechanism to read such release files. Even the names of those files changed every few years. People appear to have settled on "os-release" nowadays, since using "<distro>-release" was found to create too much confusion, but this is not fixed and when it comes to parsing the contents, I'm pretty sure that there a subtle differences in how to interpret them between distros.

Then the LSB standardized this and wanted everyone to use lsb_release. Guess what... The tool is not even installed anymore by default on recent OpenSUSE releases. Experience shows that these things change too often to make the stdlib a good place to maintain support for this.
History
Date User Action Args
2020-11-25 09:01:45lemburgsetrecipients: + lemburg, doko, vstinner, christian.heimes, r.david.murray, pmpp, matrixise, cstratak
2020-11-25 09:01:45lemburgsetmessageid: <1606294905.04.0.494132938241.issue28468@roundup.psfhosted.org>
2020-11-25 09:01:44lemburglinkissue28468 messages
2020-11-25 09:01:44lemburgcreate