View previous topic :: View next topic |
Author |
Message |
bammbamm808 Guru
Joined: 08 Dec 2002 Posts: 548 Location: Hawaii
|
Posted: Sun Nov 29, 2015 4:24 am Post subject: Python-3.4.3 fail in install phase |
|
|
Python-3.4.3 install phase fail:
* ERROR: dev-lang/python-3.4.3::gentoo failed (install phase):
* python_do* and python_new* helpers are banned in EAPIs older than 5.
*
* Call stack:
* ebuild.sh, line 93: Called src_install
* environment, line 4126: Called python_domodule 'epython.py'
* environment, line 3312: Called die
* The specific snippet of code:
* die "python_do* and python_new* helpers are banned in EAPIs older than 5.";
Anyone else getting this? Where do I begin with this? _________________ MSI MAG B550 Tomahawk
Ryzen 3900x
32Gb Samsung B-die (16GB dual rank x2) DDR4 @ 3200MHz, cl14
Geforce RTX 2070S 8GB
Samsung m.2 NVME pcie-3.0
Etc.... |
|
Back to top |
|
|
khayyam Watchman
Joined: 07 Jun 2012 Posts: 6227 Location: Room 101
|
Posted: Sun Nov 29, 2015 4:59 am Post subject: |
|
|
bammbamm808 ...
That particular ebuild is EAPI=4 ... but uses 'python_domodule()' inherited from python-utils-r1.eclass ... which has the following:
python-utils-r1.eclass: | python_domodule() {
debug-print-function ${FUNCNAME} "${@}"
[[ ${EPYTHON} ]] || die 'No Python implementation set (EPYTHON is null).'
if [[ ${EAPI:-0} == [01234] ]]; then
die "python_do* and python_new* helpers are banned in EAPIs older than 5."
fi |
... so, it dies.
I expect it'll be the outcome of some hasty commit ... reverted as and when the issue is reported.
best ... khay |
|
Back to top |
|
|
zoomersam n00b
Joined: 29 Nov 2015 Posts: 1
|
Posted: Sun Nov 29, 2015 5:43 pm Post subject: |
|
|
Problem also exists with upgrading 2.7 python:
Quote: | * ERROR: dev-lang/python-2.7.10-r1::gentoo failed (install phase):
* python_do* and python_new* helpers are banned in EAPIs older than 5.
*
* Call stack:
* ebuild.sh, line 133: Called src_install
* environment, line 4178: Called python_domodule 'epython.py'
* environment, line 3352: Called die
* The specific snippet of code:
* die "python_do* and python_new* helpers are banned in EAPIs older than 5."; |
I've previously updated on another gentoo instance to this ebuild correctly. I suppose this error caused by this: https://www.mail-archive.com/gentoo-commits@lists.gentoo.org/msg169167.html |
|
Back to top |
|
|
khayyam Watchman
Joined: 07 Jun 2012 Posts: 6227 Location: Room 101
|
Posted: Sun Nov 29, 2015 6:32 pm Post subject: |
|
|
zoomersam wrote: | Problem also exists with upgrading 2.7 python: |
zoomersam ... yes, and there is even another thread where 2.7 is the effected package.
No surprise as to who committed it.
best ... khay |
|
Back to top |
|
|
trigggl Apprentice
Joined: 26 Aug 2007 Posts: 250 Location: Arkansas
|
Posted: Sun Nov 29, 2015 7:21 pm Post subject: |
|
|
So...run ~ until then?
EDIT: Nevermind
I'm not going to wade into this.
Quote: | The following mask changes are necessary to proceed:
(see "package.unmask" in the portage(5) man page for more details)
# required by =dev-lang/python-3.4.3-r4 (argument)
# /usr/portage/profiles/package.mask:
# Michał Górny <mgorny@gentoo.org> (11 Nov 2015)
# Cleaned up Python versions masked for testing:
# - python-config-X.Y compatibility removed,
# - python[23] choice is now stored in config file rather than symlink,
# - eselect-python reworked to reuse python-exec and wrap all execs,
# - ABIFLAGS reintroduced for 3.3+.
# Resulting API/ABI change can break reverse dependencies, especially
# if upstream hardcodes paths or library names.
=dev-lang/python-3.4.3-r4
# required by =dev-lang/python-2.7.10-r4 (argument)
# /usr/portage/profiles/package.mask:
# Michał Górny <mgorny@gentoo.org> (11 Nov 2015)
# Cleaned up Python versions masked for testing:
# - python-config-X.Y compatibility removed,
# - python[23] choice is now stored in config file rather than symlink,
# - eselect-python reworked to reuse python-exec and wrap all execs,
# - ABIFLAGS reintroduced for 3.3+.
# Resulting API/ABI change can break reverse dependencies, especially
# if upstream hardcodes paths or library names.
=dev-lang/python-2.7.10-r4 |
_________________ Greg |
|
Back to top |
|
|
trigggl Apprentice
Joined: 26 Aug 2007 Posts: 250 Location: Arkansas
|
Posted: Sun Nov 29, 2015 7:47 pm Post subject: |
|
|
khayyam wrote: | zoomersam wrote: | Problem also exists with upgrading 2.7 python: |
zoomersam ... yes, and there is even another thread where 2.7 is the effected package.
No surprise as to who committed it.
best ... khay |
Yeah, let's not bother to test it first. It's not like python is important or anything. _________________ Greg |
|
Back to top |
|
|
herr_tichy n00b
Joined: 09 Aug 2012 Posts: 11 Location: NRW, Germany
|
Posted: Sun Nov 29, 2015 8:54 pm Post subject: |
|
|
Sooo... is anything happening regarding that bug or am I just screwed now? |
|
Back to top |
|
|
khayyam Watchman
Joined: 07 Jun 2012 Posts: 6227 Location: Room 101
|
Posted: Sun Nov 29, 2015 9:21 pm Post subject: |
|
|
herr_tichy wrote: | Sooo... is anything happening regarding that bug or am I just screwed now? |
herr_tichy ... it is reported, see bug 567106 ... otherwise, you can correct the eclass in a local overlay.
trigggl wrote: | Yeah, let's not bother to test it first. It's not like python is important or anything. |
trigggl ... it's a change in an eclass, these by nature effect many packages, each of which would need testing. Anyhow, yes, its the sort of change that shouldn't have got through to the tree ...
best ... khay |
|
Back to top |
|
|
trigggl Apprentice
Joined: 26 Aug 2007 Posts: 250 Location: Arkansas
|
Posted: Sun Nov 29, 2015 10:32 pm Post subject: |
|
|
khayyam wrote: | herr_tichy wrote: | Sooo... is anything happening regarding that bug or am I just screwed now? |
herr_tichy ... it is reported, see bug 567106 ... otherwise, you can correct the eclass in a local overlay. |
Is correcting the eclass an easy thing to do? _________________ Greg |
|
Back to top |
|
|
khayyam Watchman
Joined: 07 Jun 2012 Posts: 6227 Location: Room 101
|
Posted: Mon Nov 30, 2015 7:27 am Post subject: |
|
|
trigggl wrote: | khayyam wrote: | herr_tichy ... it is reported, see bug 567106 ... otherwise, you can correct the eclass in a local overlay. |
Is correcting the eclass an easy thing to do? |
trigggl ... from the commit it looks fairly trivial ... if you want to go that route you might simply revert the change.
best ... khay |
|
Back to top |
|
|
|