Bug 721522 – app-portage/elogviewer-3.0-r2 stabilisation

• Product: Gentoo Linux
• Status: RESOLVED FIXED

[Description Andreas Sturmlechner 2020-05-07 17:45:04 UTC]

Any particular reason why that did not happen yet?

Tools proj, please add arches if ready, there is no stable version that is py37 compatible.

Please also check all other packages *you* maintain that depend on python.

[Comment 1 NATTkA bot 2020-05-09 02:04:33 UTC]

Sanity check failed:

> app-portage/elogviewer-3.0-r1
>   depend ppc stable profile default/linux/powerpc/ppc32/17.0 (10 total)
>     dev-python/PyQt5[gui,python_targets_python3_6(-),widgets]
>     dev-python/PyQt5[gui,python_targets_python3_7(-),widgets]
>   rdepend ppc stable profile default/linux/powerpc/ppc32/17.0 (10 total)
>     dev-python/PyQt5[gui,python_targets_python3_6(-),widgets]
>     dev-python/PyQt5[gui,python_targets_python3_7(-),widgets]

[Comment 2 Brian Dolbec (RETIRED) 2020-05-09 02:58:24 UTC]

I am getting a segfault due to an unexpected line format it is parsing.

The code is looking for a line format as follows:

ERROR: prepare

But I have a log with the line:

ERROR: dev-python/sqlalchemy-migrate-0.13.0::gentoo failed (prepare phase):

which produces the following traceback (NOTE: I added a print statement to help debug):

brian@storm ~/Dev/git/gentoo $ elogviewer 
****: INFO: setup

****: INFO: prepare

****: ERROR: prepare

****: ERROR: dev-python/sqlalchemy-migrate-0.13.0::gentoo failed (prepare phase):

Traceback (most recent call last):
  File "/usr/lib/python-exec/python3.7/elogviewer", line 219, in setEditorData
    editor.setHtml(model.itemFromIndex(index).html())
  File "/usr/lib/python-exec/python3.7/elogviewer", line 524, in html
    text = self._elog.html
  File "/usr/lib/python-exec/python3.7/elogviewer", line 193, in html
    parser.parse(line)
  File "/usr/lib/python-exec/python3.7/elogviewer", line 336, in parse
    self._results.append(self.state.parse(line))
  File "/usr/lib/python-exec/python3.7/elogviewer", line 261, in parse
    eclass, stage = line.split(":")
ValueError: too many values to unpack (expected 2)
Aborted
brian@storm ~/Dev/git/gentoo $ 



I will work out a fix and update the ebuild.

cc'ing upstream

[Comment 3 Larry the Git Cow 2020-05-09 03:48:53 UTC]

The bug has been referenced in the following commit(s):

https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=a7d3a01ae62b55d4e4f0ada9b8b5a9466565371d

commit a7d3a01ae62b55d4e4f0ada9b8b5a9466565371d
Author:     Brian Dolbec <[e-mail]>
AuthorDate: 2020-05-09 03:46:35 +0000
Commit:     Brian Dolbec <[e-mail]>
CommitDate: 2020-05-09 03:48:38 +0000

    app-portage/elogviewer: Revision bump 3.0, add patch
    
    Add a patch to fix a HeaderState.parse segfault for a newer line format
    Bug: https://bugs.gentoo.org/721522
    
    Package-Manager: Portage-2.3.99, Repoman-2.3.22
    Signed-off-by: Brian Dolbec <[e-mail]>

 app-portage/elogviewer/elogviewer-3.0-r2.ebuild    | 69 ++++++++++++++++++++++
 .../elogviewer/files/elogviewer-3.0-segfault.patch | 34 +++++++++++
 2 files changed, 103 insertions(+)

[Comment 4 Brian Dolbec (RETIRED) 2020-05-09 03:59:36 UTC]

Patched and pushed 3.0-r2

amd64 stable

Mathias, I also added an early empty line return as a just in case.

Also, I don't know where else it was getting the headerstate.parse() info, but on my system, any return from this function duplicated what it had there already.

You can view the patch in the package's files directory.

[Comment 5 Sam James 2020-05-09 04:14:57 UTC]

Please resolve the sanity check issues.

[Comment 6 NATTkA bot 2020-05-09 04:16:36 UTC]

Sanity check failed:

> app-portage/elogviewer-3.0-r1
>   depend ppc stable profile default/linux/powerpc/ppc32/17.0 (10 total)
>     dev-python/PyQt5[gui,python_targets_python3_6(-),widgets]
>     dev-python/PyQt5[gui,python_targets_python3_7(-),widgets]
>   rdepend ppc stable profile default/linux/powerpc/ppc32/17.0 (10 total)
>     dev-python/PyQt5[gui,python_targets_python3_6(-),widgets]
>     dev-python/PyQt5[gui,python_targets_python3_7(-),widgets]

[Comment 7 Brian Dolbec (RETIRED) 2020-05-09 04:42:42 UTC]

ppc arch team. I am failing to find anything causing the sanity check fail for ppc32.

[Comment 8 NATTkA bot 2020-05-09 04:44:36 UTC]

Sanity check failed:

> dev-python/PyQt5-5.14.2
>   depend ppc stable profile default/linux/powerpc/ppc32/17.0 (10 total)
>     >=dev-python/PyQt5-sip-4.19.20:=[-python_single_target_python2_7(-),-python_single_target_python3_6(-),-python_single_target_python3_7(-),-python_single_target_python3_8(-),python_targets_python2_7(-),python_targets_python3_6(-),python_targets_python3_7(-)]
>     >=dev-qt/qtsensors-5.12:5
>     >=dev-qt/qtwebchannel-5.12:5
>     >=dev-qt/qtwebsockets-5.12:5
>     dev-qt/qtwebkit:5[printsupport]
>   rdepend ppc stable profile default/linux/powerpc/ppc32/17.0 (10 total)
>     >=dev-python/PyQt5-sip-4.19.20:=[-python_single_target_python2_7(-),-python_single_target_python3_6(-),-python_single_target_python3_7(-),-python_single_target_python3_8(-),python_targets_python2_7(-),python_targets_python3_6(-),python_targets_python3_7(-)]
>     >=dev-qt/qtsensors-5.12:5
>     >=dev-qt/qtwebchannel-5.12:5
>     >=dev-qt/qtwebsockets-5.12:5
>     dev-qt/qtwebkit:5[printsupport]

[Comment 9 Brian Dolbec (RETIRED) 2020-05-09 04:45:57 UTC]

ah, hornets nest of stabilizations... :(

[Comment 10 Andreas Sturmlechner 2020-05-09 06:59:29 UTC]

(In reply to Brian Dolbec from comment #7)
> ppc arch team. I am failing to find anything causing the sanity check fail
> for ppc32.

dev-python/PyQt5 is not stable on ppc, and neither is app-portage/elogviewer.

[Comment 11 NATTkA bot 2020-05-09 07:00:34 UTC]

All sanity-check issues have been resolved

[Comment 12 Agostino Sarubbo 2020-05-12 06:39:37 UTC]

x86 stable