| Age | Commit message (Collapse) | Author |
|---|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
(pattern)
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Starting at a certain commit in the past, run patchtest until HEAD.
This is useful for scripts requiring pretest execution because
testing patch can be merge in every iteration.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
The oe-core setup is the way is indicated in the subject, so oe-init-build-env
script finds bitbake in the standard place.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Patchtest and its test suite is intended for oe-core so set it as the reference
repo.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
This filter is already done inside the create-summary script so removing it
from post script.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
The series/revision is included by default by patchwork [1], so there is no
need to include it again at the summary.
[1] c1cea7c880fb2babd36c6eaa2204ed92ef5da843 at patchwork repo.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Now the results contain a commit "object" (really a dict) we need to
select what to show. Select the subject since that will contain the
[PATCH x/y] sequence number that may be useful in identifying the patch
in question.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
At the moment we're referring to a module-level variable ("opt") in the
function, and we were passing certain parameters into the rendering that
aren't necessary (template name, output file name). To fix both of
these, pass some of these explicitly to the function and then pass the
parameters to go into the rendering separately.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
In order to place the emphasis on information that is most important to
the submitter, we need to alter the presentation in the default jinja2
template to present the commit first (if applicable), the issue (suffixed
with the test name), the suggested fix (if any), and then any optional
additional fields.
In order to do this and at the same time allow the template a bit more
control over formatting, require tests to send a dict with "id" and
"issue" elements, plus optionally "fix", "commit" and "data". These are
passed verbatim into the jinja2 template when formatting the summary.
(As a general rule though for human readability I believe we should be
expressing additional information as part of the issue / fix text rather
than supplying it in field form via the data parameter.)
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
The script has a --fail option, it seems logical that it should filter
the results passed through it to include only failures rather than
relying on an intermediate grep to do the same.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
Touch up the grammar slightly.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
Importing the functions directly from Python modules such as re and
os is unusual, not to mention that Python has a built-in compile()
function already. Improve readability by importing the modules as
normal and calling functions within them.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
Important for future Python 3 compatibility. Drop one useless print
statement in the process.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
When I ran patchtest I ended up with IDs such as
"tests.test_metadata_src_uri.SrcUri.test_src_uri_checksums_not_changed",
which has one extra item beyond what the code here was expecting. Use
the last three items so that it works in either situation.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
If we point to /usr/bin/python we break the ability to use the scripts
within virtualenv, so use /usr/bin/env python instead.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
* Rewrap lines
* Remove erroneous - after project name
* Fix first sentence to be a bit more grammatically correct
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
kernel and rootfs command line arguments effectively replace the image argument.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
This summary is intended to be sent to Patchwork
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Patch also does minor changes on module header and usage
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
For the moment, results coming from patchtest wont be stored into a upstream
repo, so there is no need to track results with git. Instead, just store
results and log in simple plain texts in the output directory.
Also, store results in JSON, instead of default patchtest output. The main reason
is that results are processed later from another script thus json format is
better suited for post-processing.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Also, make this image default at the pt-cronjob script
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Otherwise, if outrepodir is not a repository, it will never be created.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Populates a folder with patchtest and related repos then launch a qemu machine.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
In case -A is given, it will merge the mbox no matter the result.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
This parameter indicates a starting point in time to poll events. Thus in case
user does not provide series.revisions, a timestamp can be provided or if not
of these are provided, it will rely on the .git-poll file. If the latter is not
found, the system will try to poll ALL events, so next fetch the system is ready.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
This is a much cleaner way to see either pt results and
log.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
|
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com>
|
Leonardo Sandoval
Paul Eggleton