Signed-off-by: Evan Jarrett evan@evanjarrett.com
+2
-4
appview/db/pipeline.go
+1
-1
appview/pages/templates/repo/compare/compare.html
+1
-1
appview/pages/templates/repo/settings/general.html
+1
appview/pages/templates/user/fragments/editBio.html
-3
appview/pipelines/pipelines.go
-2
appview/pulls/pulls.go
+10
-14
appview/repo/compare.go
+14
-1
appview/repo/repo_util.go
+1
appview/repo/router.go
+1
-1
nix/pkgs/knot-unwrapped.nix
+1
-1
spindle/engines/nixery/engine.go
-73
spindle/engines/nixery/setup_steps.go
+157
spindle/models/clone.go
+364
spindle/models/clone_test.go
History
4 rounds
5 comments
1 commit
expand
collapse
Signed-off-by: Evan Jarrett <evan@evanjarrett.com>
expand 1 comment
2 commits
expand
collapse
Signed-off-by: Evan Jarrett <evan@evanjarrett.com>
Signed-off-by: Evan Jarrett <evan@evanjarrett.com>
expand 2 comments
Not entirely sure why this says there are merge conflicts when i rebased master This is an attempt to go off of your design with returning a whole step struct. I think it turns out essentially the same code... so maybe i'm missing something obvious.
that is a bit strange, i don't see anything off with the logs on the knotserver itself. that aside, some comments on the code itself:
- don't think we need to
cd workspaceDir, we already have this upon contrainer creation (the WorkingDir is set):
resp, err := e.docker.ContainerCreate(ctx, &container.Config{
Image: addl.image,
Cmd: []string{"cat"},
OpenStdin: true, // so cat stays alive :3
Tty: false,
Hostname: "spindle",
WorkingDir: workspaceDir,
- the rest of the clone code itself looks good to me!
- could you squash the two commits into one? the commit message could be
spindle: introduce common clone stepor similar
2 commits
expand
collapse
Signed-off-by: Evan Jarrett <evan@evanjarrett.com>
Signed-off-by: Evan Jarrett <evan@evanjarrett.com>
expand 0 comments
1 commit
expand
collapse
Signed-off-by: Evan Jarrett <evan@evanjarrett.com>
expand 2 comments
thanks for the contribution! couple of nits:
- we already have a top level
workflowpackage that also definesCloneOpts(to unmarshal the workflow yaml file however), it is a bit confusing to introduce anotherworkflowpackage insidespindlewith a similar structCloneOptions(it also shares some of the same fields) - IMO a simpler route here would be to have a generic
CloneStepmethod that is calculated fromtangled.Pipeline_Workflowand returns a struct that follows themodels.Stepinterface
@oppi.li What package would you want this generic cloneStep in?
I think your solution is better, there was no precedent for a custom models.Step outside of nixery, so I was originally trying to avoid that.
I would still eventually want RepoURL and CommitSHA exposed for my usecase. but I might have to just submit a different PR for system-level env vars.
Gave it a once-over; looks good! And thanks for the tests.