Hello.
On 21.12.19 00:06, Alexander Aring wrote:
Hi Stefan,
On Wed, 18 Sep 2019 at 15:58, Stefan Schmidt <stefan@xxxxxxxxxxxxxxxxxx> wrote:
Hello.
This series does really not have any theme, besides me starting
to look into rpld and playing around with it.
I had to fix meson for my lua detection and make sure the ifdef
for SCOPE_ID is honoured. After getting it all building locally
I added Travis support and submitted builds to Coverity.
Afterwards I did a quick run through the reports from address
sanitizer and coverity scan.
Not sure if you would prefer review for rpld as patches here on
the list or as pull requests on github. Up to you.
I can't deal with the github gui! I click always the wrong buttons
there. Thanks for setting up this travis support, does it work like
wpan-tools to push it into branch "coverity_scan"?
Yes
I would like to review patches via mail, what do you prefer? What is
the procedure when somebody submits pull requests on github then,
somebody of us send it via mail? Sounds more work... Can we do both as
"whatever you like"? Can github somehow send pull request via github
to this mailinglist? Would be nice to have something to connect the
whole github thing to a mailinglist and you can control it with that
as well... at least the commenting system but I think this isn't
possible and the new kids wants all pull request via github gui. We
didn't answered this question for wpan-tools as well and I once
clicked the "merge" button (actually not by accident).
I do not mind any of the two methods. Getting Github and mailing lists
working together is to big of a task on its own for such a small project
to tackle in my opinion.
Given the low volumes of patches we are seeing on wpan-tools and maybe
also rpld I would not mind if we say that we accept both ways. Mailing
patches as well as pull requests. That would ease the contributions for
some people I hope. And while it puts a bit more burden on us it should
not be to hard given the low volume.
Also how we do that with signed off by thing? I can still fix it
somehow what I did there...
Signed off by makes only sense if we have a document descriping what we
mean by it. If we want it and use the one from the kernel we need to
communicate this clearly. If not, the sign offs are off no use.
regards
Stefan Schmidt
