On Thu, 08 Nov 2018 15:38:38 +0100, Jaroslav Kysela wrote:
Dne 8.11.2018 v 15:20 Takashi Sakamoto napsal(a):
Hi Jaroslav,
Thanks for your reply.
On 2018/11/06 18:20, Jaroslav Kysela wrote:
I have two concerns.
- For Ubuntu 14.04 LTS (trusty), End-of-life (EOL) is scheduled April
- A few months remained but it's better to use recent LTSs such as
18.04 (bionic) to reduce future maintenance cost. (I guess packages except on 'main' pocket are not already maintained for security updates.)
I think that we have limited choices:
Oh, indeed. It's reasonable currently, but I continue to take care of travis-ci action against the EOF.
- Message for PR
In my opinion, notification to alsa-devel list is an alternative of cover-letters in the past. But in this time it doesn't includes change summary, like:
Takashi Sakamoto (3): aplay: delete paragraph for obsoleted '--sleep-min' ('-s') option from aplay manual aplay: add a paragraph for '--samples' ('-s') option to aplay manual aplay: improve available conditions for '--samples' and '--duration' options aplay/aplay.1 | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)If it's difficult to include the change summary automatically into the notification, it's worth to discuss that PR senders should include it handy to PR message.
Actually, I can work directly only with the information from the webhooks:
https://developer.github.com/v3/activity/events/types/#pullrequestevent
The webhook code used on my server is in python, so we can probably fetch the patch from github and generate the diffstat from it.
Sounds good, but I don't mind to postpone the idea because this is not so critical issue. Readers of posted issue can see diffstat in github.com.
Well, I have another concern when having conversation with Daniel Baluta[1]. He post his 'Reviewed-by' tag to the github issue. In this case, how can we apply the tag to commit history? I don't know exactly services in github.com have good solution for this issue, or not...
[1] https://github.com/alsa-project/alsa-utils/pull/1#issuecomment-436164565
I believe that the maintainer who signs and pushes the commits should handle those tags, too.
It means that we are not allowed to use the github web merge button! (no signing - seems like on github's issue list - https://github.com/todogroup/gh-issues/issues/50)
OK, let me handle this case as an exercise.
thanks,
Takashi