Smalltalk › Squeak › Squeak VM

[OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

11 messages Options

David T Lewis

[OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

Based on appveyor/ci#1865

[skip appveyor]

You can view, comment on, or merge this pull request online at:

https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/349

Commit Summary

.appveyor.yml test - try to add [skip appveyor]

File Changes

M .appveyor.yml (73)

Patch Links:

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

@akgrant43 pushed 1 commit.

6b29a6a Add [ci skip] as well

—
You are receiving this because you are subscribed to this thread.
View it on GitHub or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

In reply to this post by David T Lewis

Hi @akgrant43,
Have you seen this? https://www.appveyor.com/docs/how-to/filtering-commits/

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

In reply to this post by David T Lewis

Hi Fabio,

My understanding with the default commit filter is that it requires the [skip ci] message in the commit title. Most people don't want the first line of the commit polluted by this type of message, so the idea here is to allow the [skip ci] to be placed anywhere in the commit message (it has it's own drawbacks, see https://help.appveyor.com/discussions/problems/11603-skip-ci-doesnt-work-as-expected for a discussion (thanks to Ben for providing this link).

Cheers,
Alistair

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

In reply to this post by David T Lewis

@akgrant43 pushed 1 commit.

9778817 Restore all builds

—
You are receiving this because you are subscribed to this thread.
View it on GitHub or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

In reply to this post by David T Lewis

@akgrant43 pushed 1 commit.

3e6e557 Remove duplicate key error

—
You are receiving this because you are subscribed to this thread.
View it on GitHub or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

In reply to this post by David T Lewis

Interesting... We could be using something like this

only_commits:
    files:
        - *src/
        - platforms/win32/
        - platforms/Cross/
        - build.win*/
        - third-party/

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

In reply to this post by David T Lewis

Hi Nicolas,

Interesting... We could be using something like this

only_commits:
    files:
        - *src/
        - platforms/win32/
        - platforms/Cross/
        - build.win*/
        - third-party/

The drawback of this approach is that we identify the VM by the commit id, and this will prevent the same commit id being present on all platforms whenever a change is made that doesn't affect Windows.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

In reply to this post by David T Lewis

The drawback of this approach is that we identify the VM by the commit id, and this will prevent the same commit id being present on all platforms whenever a change is made that doesn't affect Windows.

Well shouldn't it work that way? I mean, there are other drawbacks of identifying VMs by commit id (e.g. the need for git hooks). More importantly, why rebuild all Linux/macOS VMs when only Windows code was modified? I have never expected bleeding edge "releases" on Bintray to be complete, no one really should because they are mainly for development and builds happen to fail from time to time. Stable release builds will still be triggered simultaneously on AppVeyor and Travis. So we should be fine, right?

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

In reply to this post by David T Lewis

Well shouldn't it work that way? I mean, there are other drawbacks of identifying VMs by commit id (e.g. the need for git hooks).

I'm not sure. If we introduced a version number or build number there may still be a need to substitute it in to other files, so there may still be a use for the commit hooks. And event with the other identifier, I wouldn't want to get rid of the CommitHash - it makes it easy as a developer to find the code.

More importantly, why rebuild all Linux/macOS VMs when only Windows code was modified?

This is a good point, of course.

I have never expected bleeding edge "releases" on Bintray to be complete, no one really should because they are mainly for development and builds happen to fail from time to time. Stable release builds will still be triggered simultaneously on AppVeyor and Travis. So we should be fine, right?

Implicit in what you're saying is that someone then manually starts a full build for every release candidate.

So, I'm fine with the change you're proposing, it just requires buy-in from the people that are actually responsible for the releases in the various flavours (and thus are responsible for ensuring a full build for the released VM).

It also doesn't completely remove the usefulness of being able to skip building when desired. :-)

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.

David T Lewis

Re: [OpenSmalltalk/opensmalltalk-vm] .appveyor.yml test - try to add [skip appveyor] (#349)

In reply to this post by David T Lewis

Merged #349 into Cog.

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub, or mute the thread.