lp:~brendan-donegan/checkbox/xls_report

Branch merges

Propose for merging

No branches dependent on this one.

Merged into lp:checkbox at revision 2955

Zygmunt Krynicki (community): Needs Information on 2014-04-28

Daniel Manrique (community): Approve on 2014-04-28

Diff: 76 lines (+32/-0)

3 files modified

checkbox-gui/checkbox-gui/qml/SubmissionDialog.qml (+2/-0)
checkbox-gui/gui-engine/gui-engine.cpp (+27/-0)
checkbox-gui/gui-engine/gui-engine.h (+3/-0)

api

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Bug #1309544: [RFE] Need xls test report generated for Lenovo projects

High

Fix Released

Link a bug report

Related blueprints

2953. By Brendan Donegan on 2014-04-28

checkbox-gui: Export submission as XLSX file as well

Signed-off-by: Brendan Donegan <email address hidden>

2952. By Zygmunt Krynicki on 2014-04-28

"automatic merge by tarmac [r=zkrynicki][bug=][author=zkrynicki]"

2951. By Daniel Manrique on 2014-04-27

providers:checkbox: removed stray square bracket in qemu-system-arm params

2950. By Daniel Manrique on 2014-04-26

"Converts all the jobs in graphics.txt to use the graphics_card_resource to build per-GPU tests.

A few jobs were not converted because in my opinion they make no sense to run per card.

Whitelists using graphics tests in lp:checkbox were updated to use the new jobs.

Of note, see how all the "generator" jobs were named in such a way that they can be invoked with a single regex. I did it this way to avoid essentially wrapping the entire job file in a giant local job (this caused problems with the blank lines that delimit each sub-job). Also if needed, having individual generators for each job gives more control.

Also, remember the per-GPU jobs are prefixed with an index, so we can cluster jobs to run all jobs for one card first, and for the second one next.

All whitelists using graphics tests (CDTS is the example that comes to mind) need updating to account for the new job structure.

Also, note that as they are, these tests can't be run with checkbox-gui because of bug https://bugs.launchpad.net/checkbox-gui/+bug/1309128.

One way to test this is to merge this branch, then develop the checkbox and plainbox-resource-generic providers, then run:

plainbox run --dry-run -i 2013.com.canonical.certification::graphics/generator_.* -i 3013.com.canonical.certification::graphics/1_.*

this will run the local jobs to generate all the sub jobs and then run the jobs matching the first graphic card (which won't run since we specified --dry-run, but at least the fact that they were generated can be verified). [r=zkrynicki][bug=][author=roadmr]"

2949. By dann frazier on 2014-04-26

"automatic merge by tarmac [r=zkrynicki][bug=][author=dannf]"

2948. By Launchpad Translations on behalf of checkbox-dev on 2014-04-26

Launchpad automatic translations update.

2947. By Daniel Manrique on 2014-04-25

"05797dd checkbox-support:parsers:udevadm: Updated heuristic for some VIDEO devices.
5a2fb13 checkbox-support:lib:pci: Added a constant with Intel vendor PCI ID
6ef10c0 checkbox-support:parsers:test_udevadm: Added expectations and data for Vostro 270.

    Usually a video device has a PCI subclass_id of Pci.CLASS_DISPLAY_VGA or
    Pci.CLASS_DISPLAY_3d. However, some manufacturers which hadn't
    previously used a subclass_id of Pci.CLASS_DISPLAY_OTHER (0x80 or
    decimal 128) have begun to do so (as of 2013-2014); and others which
    previously used this class for uninteresting devices are now using it
    for actual video devices.

    Previously we identified AMD as doing this, and added a simple heuristic
    to accomodate that; on a few newer systems, Intel has also started using
    the OTHER display class, but erratically. This code implements a more
    versatile heuristic we can expand if/when we find other vendors/devices
    that break our old expectations.

    A previous commit added a test and data file to confirm that things
    behave as expected, and also that none of the other tests break,
    indicating that this change doesn't cause old systems to "sprout" new
    VIDEO devices on account of faulty heuristics.
 [r=zkrynicki][bug=1312736][author=roadmr]"

2946. By Zygmunt Krynicki on 2014-04-25

"automatic merge by tarmac [r=roadmr][bug=][author=zkrynicki]"

2945. By Zygmunt Krynicki on 2014-04-25

"automatic merge by tarmac [r=brendan-donegan][bug=1309128][author=zkrynicki]"

2944. By Daniel Manrique on 2014-04-25

"Added some firmware-related test definitions and whitelist items to certification lists.

8f54fbe providers:certification-client: added wakealarm and uefirtvariable tests to whitelist
0951d74 providers:checkbox: Added fwts oops test.
307ef99 providers:certification-client: removed s3/s4 log checking jobs.
f021cc7 providers:checkbox: Added cpu cstates job.
6a0fe4c providers:checkbox: Added fwts_desktop_diagnosis job.
ed5cbb2 providers:certification-client: removed fwts_test and log attachment from 14.04 client [r=brendan-donegan][bug=1306468][author=roadmr]"