➜
This issue tracker has been migrated to GitHub,
and is currently read-only.
For more information,
see the GitHub FAQs in the Python's Developer Guide.
Created on 2018-01-23 17:16 by vstinner, last changed 2022-04-11 14:58 by admin. This issue is now closed.
| Pull Requests | |||
|---|---|---|---|
| URL | Status | Linked | Edit |
| PR 5287 | closed | vstinner, 2018-01-23 21:08 | |
| PR 5288 | closed | vstinner, 2018-01-23 23:55 | |
| Messages (6) | |||
|---|---|---|---|
| msg310511 - (view) | Author: STINNER Victor (vstinner) *  |
Date: 2018-01-23 17:16 | |
Currently, sys.platform and os.name are equal to 'linux' on Android. While Android uses the Linux kernel, the operating system is not a regular Linux. The libc (bionic) is very different than the regular glibc, the filesystem is organized very differently, etc. I propose to change sys.platform and os.name to 'android' when running on Android. |
|||
| msg310516 - (view) | Author: STINNER Victor (vstinner) * |
Date: 2018-01-23 17:38 | |
Wait, os.name is 'posix' on Android. That's fine in fact. Only sys.platform should be updated. |
|||
| msg310537 - (view) | Author: STINNER Victor (vstinner) * |
Date: 2018-01-24 00:12 | |
The idea was first discussed on python-dev: https://mail.python.org/pipermail/python-dev/2018-January/151874.html me: "It seems like sys.platform == 'android' would be more appropriate since Android is not Linux: different libc, different filesystems, etc." https://mail.python.org/pipermail/python-dev/2018-January/151887.html Brett Cannon: "I've had a similar thought myself.". |
|||
| msg310543 - (view) | Author: STINNER Victor (vstinner) * |
Date: 2018-01-24 00:42 | |
A similar change was proposed by pmp-p on MicroPython: https://github.com/micropython/micropython/pull/3564 His use case is to use ffi module on Android (see the PR for more information), currently ffi fails to locate libraries. |
|||
| msg310594 - (view) | Author: STINNER Victor (vstinner) * |
Date: 2018-01-24 13:37 | |
I propose to leave MACHDEP unchanged: keep MACHDEP="linux". Can it create inconsistencies? Is it an issue in practice? My short term goal is to use sys.platform == 'android' has the reference test to check if we are running on Android. |
|||
| msg318102 - (view) | Author: STINNER Victor (vstinner) * |
Date: 2018-05-29 21:47 | |
Sadly, I failed to find anyone to help me on this issue and I lost interest on supporting Android, so I just close the issue. Sorry. |
|||
| History | |||
|---|---|---|---|
| Date | User | Action | Args |
| 2022-04-11 14:58:57 | admin | set | github: 76818 |
| 2018-05-29 21:47:52 | vstinner | set | status: open -> closed resolution: out of date messages: + msg318102 stage: patch review -> resolved |
| 2018-01-31 04:24:57 | xgdomingo | set | nosy:
+ xgdomingo |
| 2018-01-25 12:38:38 | xdegaye | set | nosy:
- xdegaye |
| 2018-01-24 13:37:33 | vstinner | set | messages: + msg310594 |
| 2018-01-24 00:44:12 | vstinner | set | nosy:
+ yan12125 |
| 2018-01-24 00:42:58 | vstinner | set | messages: + msg310543 |
| 2018-01-24 00:38:05 | vstinner | set | nosy:
+ doko |
| 2018-01-24 00:12:46 | vstinner | set | nosy:
+ brett.cannon |
| 2018-01-24 00:12:39 | vstinner | set | messages: + msg310537 |
| 2018-01-23 23:55:21 | vstinner | set | pull_requests: + pull_request5134 |
| 2018-01-23 21:08:03 | vstinner | set | keywords:
+ patch stage: patch review pull_requests: + pull_request5133 |
| 2018-01-23 19:39:08 | pmpp | set | nosy:
+ pmpp |
| 2018-01-23 17:38:53 | vstinner | set | messages:
+ msg310516 title: Android: set sys.platform and os.name to android -> Android: set sys.platform to android |
| 2018-01-23 17:16:49 | vstinner | create | |