-
-
Notifications
You must be signed in to change notification settings - Fork 419
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
release 0.11.4 #1738
Comments
DO IT! |
I've been distracted by https://github.com/killerswan/homebrew-pony (for # 1732) this afternoon, but will do this tomorrow, when more people are around. |
@killerswan do you have this or should I do it? i'd rather you did it because if something does go sideways you are in best position to address. |
I'll give it a shot now. 😅 |
Haha, our clever usage of [skip ci] is uncool here. Because what I have to say is that our potentially golden commit, 44ef576 (master), is actually a couple changes since the last master build that was green on Travis (delta from CI: b99d5b6...44ef576, delta from last release to master: 917610c...44ef576).
|
Wait that's fine. Master doesn't deploy anymore, wake up Kevin. But look it was passing tests both on AppVeyor and Travis: |
Why do you say it won't deploy anymore? |
o because 44ef576 should have triggered a travis build (but not an appveyor one), correct? |
O no, that had skip ci in it. I'm not sure why you say master won't deploy. What do you mean? Everything looks fine from here. |
I was looking at Bintray and saying "WHERE ARE MY BINARIES!? 😿 " Only release builds go to Bintray. |
👍 So now with 44ef576 dubbed "golden", I'm running the release script (after creating an "origin" remote, haha)... |
ah yes, was the release doc not updated to reflect that we don't do master binaries? i thought i updated that. |
Just my brain needed a couple cycles there. |
OK, after a bit of tweaking of my local repo (origin, and tracking setup) to make And I've added the new release here: But this change on master is not what I expected: 44ef576...master. @SeanTAllen what gives? |
CI:
|
@killerswan i've never seen it to do that to the CHANGELOG before. Can you roll it back and fix up by hand? Sorry, not rollback nec. just fix up the oops, why did you blow all that stuff away-ness. |
@Theodus the release script did weird things to the changelog. I assume its related to your last change. Can you take a look? |
@killerswan travis release failed on building documentation to push to standard lib update. |
It's hard to build the docs when the Pony build failed. Not sure why this didn't stop the build here:
|
So, that problem is the version of gcc is old. I believe we recently fixed that. |
yeah, that was supposed to be fixed by: 82ac4a3 did that get undone somehow? |
also WEEEEEEEE! computers 😅 |
|
I thought you were in this repo rather than your own when you did everything. The release script and instructions assume you are working in a clone of the ponyc repo. We should update the release instructions to make that clear. You have access to do from a clone of this repo, yes @killerswan ? |
These are the remotes I have set up:
That was why my comment about tracking: git pull is not a command I ever type myself. |
@killerswan write, the release instructions assume you aren't using a fork, but are using upstream: a clone of ponylang/ponyc. |
Forks and clones are the same thing. |
@killerswan i dont think it caused any problems either, i just think someone less experienced than you could have gotten tripped up so i wanted to note that i'd seen and addressed in docs. |
we wanted to update to |
That's also what we have in our Dockerfile. Are those breaking? |
ugh. it appears we never verified this worked on travis. |
i'm about to disappear on you for a couple hours. i'd say, we turn off the tune for now, remove the change log entry for it from release notes and address later. |
((nod)) |
also what's the emoji for BLARG |
My removal of 'tune=intel' is going on master now from #1743. |
To amend this 0.11.4 release:
|
OK, so release is now 0.11.4 at 5618e86, and master's at 8fc1403. Builds happening here:
|
Issues so far:
|
@kulibali do you have any idea why AppVeyor wouldn't run even after I've pushed a few new commits to the release branch (and moved the 0.11.4 tag to its latest)? |
I have a fix for the includePattern (path "does not exist") error in progress here: #1744. |
@killerswan are the commits all for things in the appveyor skip list? |
Ahahaa, possibly! |
@killerswan the previous 0.11.4 should be fine for windows. |
OK, so that leaves the release.sh issue. The changes here between 0.11.3 and now didn't introduce the problem: 917610c...eaf9a86#diff-e55bb7573f1110c8b2a6922fe8cccf48 Have I failed to set some envvar to avoid using macOS's sed? 👻 |
I've filed a new ticket for that, #1745, since someone other than me might be able to fix it faster. |
OK, to really recap, before I go eat food:
Work to fix release.sh should be tested on another repository, and proceed on master, now. I'm now wrapping up:
|
@stefantalpalaru, FYI, we're deploying the 0.11.4 release now! |
@SeanTAllen I've made a post here: https://pony.groups.io/g/user/message/940 Feeling I've not used enough emojis yet, but there you go. |
|
Gentoo ebuild updated. |
Thanks, @stefantalpalaru! 🐴 |
There's a wee bit of stuff in here now that I'd like to confirm also works right on a release branch: release...master
Anything pending that we should wait up for? @sylvanc @jemc
The text was updated successfully, but these errors were encountered: