Surface Step Results in BuildRun Status #871

shahulsonhal · 2021-09-21T04:01:29Z

Changes

Reference shipwright-io/community#13

Add git commit author result for git source step

Add bundle image digest result for local source code step

Add a function to surface TaksRun results to BuildRun

This PR only includes a subset of the enhancements called for, those are:

Build Results Sources API Proposal (excluded surfacing the http source type results since it is an MVP implementation)
Build Results Output API Proposal

and not including the Strategy Results API Proposal that allows strategy authors to surface their desire result.

Submitter Checklist

Includes tests if functionality changed/was added
Includes docs if changes are user-facing
Set a kind label on this PR
Release notes block has been filled in, or marked NONE

See the contributor guide
for details on coding conventions, github and prow interactions, and the code review process.

Release Notes

You can now see the TaskRun results in BuildRun status

shahulsonhal · 2021-09-21T04:02:55Z

/kind api-change

gabemontero · 2021-09-21T13:12:07Z

hey @shahulsonhal

First, thanks for contributing!

This is an interesting one, both in that I think you are surfacing information users will care about, buin that it surfaces somewhat explicitly TaskRun details in the BuildRun. Admittedly, at least IMO, we have been "walking a tightrope" if you are familiar with the expression, of how much of our Tekton dependency we surface in the API, and how much we do not surface.

I at least need to look at this in more detail, but while the information could be useful, we might need to spend some time on the names of things, minimally. But I should have better input once I can focus on this.

I also think this needs attention from other of our key contributors. I'll assign them as well.

I will also mention we do have an enhancement proposal process (similar to KEPs and TEPs) over at https://github.com/shipwright-io/community ... perhaps we can avoid the need for a SHP for this change, but let's see what kind of consensus we get with the other folks.

/assign @gabemontero
/assign @adambkaplan
/assign @SaschaSchwarze0
/assign @imjasonh
/assign @sbose78

I didn't assign everyone :-) but hopefully that is at least a sufficient meets min (we do have some folks that are out currently)

gabemontero · 2021-09-21T13:43:12Z

ok @adambkaplan just told me there is a SHP :-) .... apologies

I'll refresh my memory with all that ... and the PR for the SHP was mentioned in the description :-( .... sheesh apologies again

adambkaplan · 2021-09-21T13:44:02Z

We do have a relevant SHIP here:

https://github.com/shipwright-io/community/blob/main/ships/0023-surface-results-in-buildruns.md

HeavyWombat

@shahulsonhal Thanks for PR! This looks pretty solid. So far, I only went through the suggested changes by looking at the code. If time permits, I would like to checkout this branch to test it locally. Until then, maybe you can have a look at my comments already. Almost all of them can be considered nitpicking, but I do think some changes could help with regards to readability or compactness.

HeavyWombat · 2021-09-21T18:44:40Z

docs/buildrun.md

+  - git:
+      commitAuthor: xxx xxxxxx
+      commitSha: f25822b85021d02059c9ac8a211ef3804ea8fdde
+    name: default


Nit: Readability.

Suggested change

- git:

commitAuthor: xxx xxxxxx

commitSha: f25822b85021d02059c9ac8a211ef3804ea8fdde

name: default

- name: default

git:

commitAuthor: xxx xxxxxx

commitSha: f25822b85021d02059c9ac8a211ef3804ea8fdde

HeavyWombat · 2021-09-21T18:45:24Z

docs/buildrun.md

+  - bundle:
+      digest: sha256:0f5e2070b534f9b880ed093a537626e3c7fdd28d5328a8d6df8d29cd3da760c7
+    name: default


Nit: Readability.

Suggested change

- bundle:

digest: sha256:0f5e2070b534f9b880ed093a537626e3c7fdd28d5328a8d6df8d29cd3da760c7

name: default

- name: default

bundle:

digest: sha256:0f5e2070b534f9b880ed093a537626e3c7fdd28d5328a8d6df8d29cd3da760c7

HeavyWombat · 2021-09-21T18:48:33Z

pkg/bundle/bundle.go

+func PullAndUnpack(
+	ref name.Reference,
+	targetPath string,
+	options ...remote.Option,
+) (*v1.Image, error) {


Open for debate, but with two arguments and one variadic argument it is still readable in one line.

Suggested change

func PullAndUnpack(

ref name.Reference,

targetPath string,

options ...remote.Option,

) (*v1.Image, error) {

func PullAndUnpack(ref name.Reference, targetPath string, options ...remote.Option) (*v1.Image, error) {

Thanks for the suggestion and I will update it.

TBH, I am not a big fan of long lines (it has 104 chars in a single line). It would be better if we consider having a lint rule for the line length.

I must mention that I am a big fan of long lines. :-)

Long lines allow me to see more logic because more horizontal lines without scrolling. If details matter, then I can look at the details (for example which variables get passed into a function call) of the longer line.

But just my few cents, within any project with more than one contributor, there'll rarely be a common opinion on those aspects. ;-)

HeavyWombat · 2021-09-21T18:55:50Z

pkg/reconciler/buildrun/resources/results_test.go

+			tr.Status.TaskRunResults = append(tr.Status.TaskRunResults, pipelinev1beta1.TaskRunResult{
+				Name:  "shp-image-digest",
+				Value: imageDigest,
+			}, pipelinev1beta1.TaskRunResult{
+				Name:  "shp-image-size",
+				Value: "230",
+			})


Nit: Readability.

Suggested change

tr.Status.TaskRunResults = append(tr.Status.TaskRunResults, pipelinev1beta1.TaskRunResult{

Name: "shp-image-digest",

Value: imageDigest,

}, pipelinev1beta1.TaskRunResult{

Name: "shp-image-size",

Value: "230",

})

tr.Status.TaskRunResults = append(tr.Status.TaskRunResults,

pipelinev1beta1.TaskRunResult{

Name: "shp-image-digest",

Value: imageDigest,

},

pipelinev1beta1.TaskRunResult{

Name: "shp-image-size",

Value: "230",

})

HeavyWombat · 2021-09-21T19:00:20Z

cmd/bundle/main.go

+		if err = ioutil.WriteFile(
+			flagValues.imageDigest, []byte(digest.String()), 0644,
+		); err != nil {


Nit: Short enough to be on a single line.

Suggested change

if err = ioutil.WriteFile(

flagValues.imageDigest, []byte(digest.String()), 0644,

); err != nil {

if err = ioutil.WriteFile(flagValues.imageDigest, []byte(digest.String()), 0644); err != nil {

HeavyWombat · 2021-09-21T19:09:45Z

cmd/git/main_test.go

@@ -331,4 +331,39 @@ var _ = Describe("Git Resource", func() {
 			})
 		})
 	})
+
+	Context("store result", func() {


Suggested change

Context("store result", func() {

Context("store details in result files", func() {

HeavyWombat · 2021-09-21T19:12:34Z

pkg/reconciler/buildrun/resources/results.go

+	"fmt"
+
+	buildv1alpha1 "github.com/shipwright-io/build/pkg/apis/build/v1alpha1"
+	"github.com/tektoncd/pipeline/pkg/apis/pipeline/v1beta1"


I know we have no clear line here so far, but new files give us the opportunity to be more specific with the import names. Without an explicit name, the import is just v1beta1 and this is ambiguous.

Suggested change

"github.com/tektoncd/pipeline/pkg/apis/pipeline/v1beta1"

pipeline "github.com/tektoncd/pipeline/pkg/apis/pipeline/v1beta1"

Also, I would not mind to change the build imports to be build instead of buildv1alpha1, but that is just my personal opinion.

HeavyWombat · 2021-09-21T19:15:03Z

pkg/reconciler/buildrun/resources/results.go

+	sources.defaultSource.Git = &buildv1alpha1.GitSourceResult{}
+	sources.bundleSource.Bundle = &buildv1alpha1.BundleSourceResult{}


Could we move that into the if blocks that follow? Theoretically speaking, we do not need to initialize both as usually only one of them will be used.

I have moved the initialization to the switch case of func updateBuildRunStatus(), so that it will only initialize the source being used.

HeavyWombat · 2021-09-21T19:17:01Z

pkg/reconciler/buildrun/resources/results_test.go

+	. "github.com/onsi/ginkgo"
+	. "github.com/onsi/gomega"
+
+	"github.com/shipwright-io/build/pkg/apis/build/v1alpha1"


Similar to the other comment, just the "version" is ambiguous.

Suggested change

"github.com/shipwright-io/build/pkg/apis/build/v1alpha1"

build "github.com/shipwright-io/build/pkg/apis/build/v1alpha1"

HeavyWombat · 2021-09-21T19:18:29Z

pkg/reconciler/buildrun/resources/sources/bundle.go

+		"--result-file-image-digest",
+		fmt.Sprintf(
+			"$(results.%s-source-%s-bundle-image-digest.path)",
+			prefixParamsResultsVolumes, name,
+		),


Nit: Could fit into one line.

Suggested change

"--result-file-image-digest",

fmt.Sprintf(

"$(results.%s-source-%s-bundle-image-digest.path)",

prefixParamsResultsVolumes, name,

),

"--result-file-image-digest", fmt.Sprintf("$(results.%s-source-%s-bundle-image-digest.path)", prefixParamsResultsVolumes, name),

gabemontero

OK managed to review the EP before looking at this again :-)

Share similar sentiments with @HeavyWombat on the changes here @shahulsonhal - thanks!

That said, a few minor requests for changes below.

And while it may be obvious to the uninitiated, I'm going to explicitly note (as I did not see an explicit note in the PR description) that this PR takes on a subset of the enhancements called for.

Of course that is fine. Just wanting to make sure it is clear.

In particular, https://github.com/shipwright-io/community/blob/main/ships/0023-surface-results-in-buildruns.md#strategy-results-api-proposal is not tackled with this one.

Also, there was a BR result type noted in the EP that I did not see here. See my question below.

Thanks again.

gabemontero · 2021-09-22T12:12:15Z

docs/buildrun.md

@@ -213,6 +215,46 @@ To make it easier for users to understand why did a BuildRun failed, users can i

 In addition, the `Status.Conditions` will host under the `Message` field a compacted message containing the `kubectl` command to trigger, in order to retrieve the logs.

+### Step Results in BuildRun Status
+
+After the successful completion of a `BuildRun`, the `.status` field contains the results (`.status.taskResults`) emitted from the `TaskRun` steps. These results contain valuable metadata for users, like the _image digest_ or the _commit sha_ of the source code used for building.


nit:

emitted from the TaskRun steps. -> emitted from the TaskRun steps generate by the BuildRun controller as part of processing the BuildRun.

gabemontero · 2021-09-22T12:13:44Z

docs/buildstrategies.md

@@ -277,7 +277,7 @@ Contrary to the strategy `spec.parameters`, you can use system parameters and th

 ## System results

-You can optionally store the size and digest of the image your build strategy created to some files. This information will eventually be made available in the status of the BuildRun.
+You can optionally store the size and digest of the image your build strategy created to some files.


some files -> a set of files

gabemontero · 2021-09-22T12:21:23Z

pkg/apis/build/v1alpha1/buildrun_types.go

+
+	// CommitAuthor holds the commit author of a git source
+	CommitAuthor string `json:"commitAuthor,omitempty"`
+}


The EP calls for a HttpSourceResult as well. Was it an explicit choice to not do that type at this time, and follow up later, or was it an oversight?

It was an explicit choice since the http type is an MVP implementation (assuming that everything that comes under .sources is http type) and as of now not emitting any results.

gabemontero · 2021-09-22T12:22:51Z

pkg/bundle/bundle.go

@@ -18,6 +18,7 @@ import (

 	"github.com/go-git/go-git/v5/plumbing/format/gitignore"
 	"github.com/google/go-containerregistry/pkg/name"
+	v1 "github.com/google/go-containerregistry/pkg/v1"


no need for the v1 qualifier with the package name ends in v1

surprised govet or goimports did not catch this

either drop the v1 import qualifier, or change to something more meaniful like containerregv1

or containerreg is fine too, as I just noticed the pattern you employed in the new results.go file

shahulsonhal · 2021-09-22T16:48:01Z

@gabemontero thanks

And while it may be obvious to the uninitiated, I'm going to explicitly note (as I did not see an explicit note in the PR description) that this PR takes on a subset of the enhancements called for.

I have updated the description.

Also, there was a BR result type noted in the EP that I did not see here. See my question below.

I assume you meant the .status.results field. That is not included in this PR.

SaschaSchwarze0

Very nice work. Some suggestions.

cmd/bundle/main.go

docs/buildrun.md

docs/configuration.md

pkg/config/config.go

pkg/reconciler/buildrun/resources/results.go

pkg/reconciler/buildrun/resources/sources/bundle.go

pkg/reconciler/buildrun/resources/sources/git.go

Reference https://github.com/shipwright-io/community/blob/main/ships/0023-surface-results-in-buildruns.md Add git commit author result for git source step Add bunlde image digest result for local source code step Add a function to surface TaksRun results to BuildRun

Source related and type related functionalities were not in corresponding files Add function to append git source results Add function to append bundle source results Add function to append output step results Co-authored-by: Sascha Schwarze <schwarzs@de.ibm.com>

SaschaSchwarze0

A few more items @shahulsonhal.

pkg/apis/build/v1alpha1/buildrun_types.go

pkg/reconciler/buildrun/resources/sources/bundle.go

SaschaSchwarze0 · 2021-09-30T13:57:30Z

pkg/reconciler/buildrun/resources/sources/git.go

+	commitAuthor := findResultValue(results, fmt.Sprintf("%s-source-%s-%s", prefixParamsResultsVolumes, name, commitAuthorResult))
+	commitSha := findResultValue(results, fmt.Sprintf("%s-source-%s-%s", prefixParamsResultsVolumes, name, commitSHAResult))
+
+	buildRun.Status.Sources = append(buildRun.Status.Sources, buildv1alpha1.SourceResult{


Same here, let's add the if not empty check.

test/e2e/e2e_test.go

gabemontero · 2021-09-30T16:50:51Z

I'll add the

/approve

and let @SaschaSchwarze0 and @shahulsonhal sort out the final changes prior to lgtm

openshift-ci · 2021-09-30T16:50:58Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: gabemontero

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

~~OWNERS~~ [gabemontero]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

Extend existing test case to add surfacing the taskRun result to buildRun test case Update output image size type from string to integer

SaschaSchwarze0

Small suggestions, answering your question regarding the empty check.

pkg/reconciler/buildrun/resources/results.go

pkg/reconciler/buildrun/resources/results_test.go

pkg/reconciler/buildrun/resources/sources/bundle.go

SaschaSchwarze0

/lgtm

openshift-ci bot added the release-note Label for when a PR has specified a release note label Sep 21, 2021

openshift-ci bot requested review from imjasonh and sbose78 September 21, 2021 04:01

openshift-ci bot added the kind/api-change Categorizes issue or PR as related to adding, removing, or otherwise changing an API label Sep 21, 2021

shahulsonhal force-pushed the enhance-buildrun-status branch from d0b5062 to cabaa44 Compare September 21, 2021 05:03

openshift-ci bot assigned adambkaplan, gabemontero, imjasonh, SaschaSchwarze0 and sbose78 Sep 21, 2021

HeavyWombat requested changes Sep 21, 2021

View reviewed changes

shahulsonhal requested a review from HeavyWombat September 22, 2021 07:15

shahulsonhal force-pushed the enhance-buildrun-status branch from 43f7310 to 7e81dba Compare September 22, 2021 07:49

gabemontero suggested changes Sep 22, 2021

View reviewed changes

shahulsonhal requested a review from gabemontero September 22, 2021 17:33

shahulsonhal force-pushed the enhance-buildrun-status branch 2 times, most recently from 6df9fd2 to 255182f Compare September 23, 2021 15:23

SaschaSchwarze0 requested changes Sep 27, 2021

View reviewed changes

openshift-ci bot added the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Sep 29, 2021

Shahul added 5 commits September 30, 2021 10:39

Test case for surface step results in BuildRun status

9a9adda

Add documentation for surface step results in BuildRun status

32a9749

Address review comments

9c8bd2d

Resolve review comments

17fe826

shahulsonhal force-pushed the enhance-buildrun-status branch from 255182f to ffbc582 Compare September 30, 2021 05:40

openshift-ci bot removed the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Sep 30, 2021

shahulsonhal force-pushed the enhance-buildrun-status branch from ffbc582 to d9e6e3f Compare September 30, 2021 09:27

shahulsonhal requested a review from SaschaSchwarze0 September 30, 2021 10:01

SaschaSchwarze0 requested changes Sep 30, 2021

View reviewed changes

openshift-ci bot added the approved Indicates a PR has been approved by an approver from all required OWNERS files. label Sep 30, 2021

Address review comments

6635eeb

Extend existing test case to add surfacing the taskRun result to buildRun test case Update output image size type from string to integer

shahulsonhal requested a review from SaschaSchwarze0 October 1, 2021 12:15

SaschaSchwarze0 requested changes Oct 1, 2021

View reviewed changes

pkg/reconciler/buildrun/resources/results.go Outdated Show resolved Hide resolved

pkg/reconciler/buildrun/resources/results_test.go Outdated Show resolved Hide resolved

pkg/reconciler/buildrun/resources/sources/bundle.go Outdated Show resolved Hide resolved

Address review comments

4027ace

shahulsonhal force-pushed the enhance-buildrun-status branch from 27b6030 to 4027ace Compare October 1, 2021 13:27

shahulsonhal requested a review from SaschaSchwarze0 October 1, 2021 15:48

SaschaSchwarze0 approved these changes Oct 4, 2021

View reviewed changes

openshift-ci bot added the lgtm Indicates that a PR is ready to be merged. label Oct 4, 2021

openshift-merge-robot merged commit c6b6a03 into shipwright-io:main Oct 4, 2021

adambkaplan added this to the release-v0.6.0 milestone Oct 6, 2021

SaschaSchwarze0 mentioned this pull request Oct 6, 2021

Add Shahul as reviewer #889

Merged

4 tasks

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Surface Step Results in BuildRun Status #871

Surface Step Results in BuildRun Status #871

shahulsonhal commented Sep 21, 2021 •

edited

Loading

shahulsonhal commented Sep 21, 2021

gabemontero commented Sep 21, 2021

gabemontero commented Sep 21, 2021 •

edited

Loading

adambkaplan commented Sep 21, 2021

HeavyWombat left a comment

HeavyWombat Sep 21, 2021

HeavyWombat Sep 21, 2021

HeavyWombat Sep 21, 2021

shahulsonhal Sep 22, 2021

SaschaSchwarze0 Sep 27, 2021

HeavyWombat Sep 21, 2021

HeavyWombat Sep 21, 2021

HeavyWombat Sep 21, 2021

HeavyWombat Sep 21, 2021

HeavyWombat Sep 21, 2021

HeavyWombat Sep 21, 2021

shahulsonhal Sep 22, 2021

HeavyWombat Sep 21, 2021

HeavyWombat Sep 21, 2021

gabemontero left a comment

gabemontero Sep 22, 2021

gabemontero Sep 22, 2021

gabemontero Sep 22, 2021

shahulsonhal Sep 22, 2021

gabemontero Sep 22, 2021

gabemontero Sep 22, 2021

shahulsonhal commented Sep 22, 2021 •

edited

Loading

SaschaSchwarze0 left a comment

SaschaSchwarze0 left a comment

SaschaSchwarze0 Sep 30, 2021

gabemontero commented Sep 30, 2021

openshift-ci bot commented Sep 30, 2021

SaschaSchwarze0 left a comment

SaschaSchwarze0 left a comment

	Context("store result", func() {
	Context("store details in result files", func() {

	"github.com/tektoncd/pipeline/pkg/apis/pipeline/v1beta1"
	pipeline "github.com/tektoncd/pipeline/pkg/apis/pipeline/v1beta1"

		sources.defaultSource.Git = &buildv1alpha1.GitSourceResult{}
		sources.bundleSource.Bundle = &buildv1alpha1.BundleSourceResult{}

	"github.com/shipwright-io/build/pkg/apis/build/v1alpha1"
	build "github.com/shipwright-io/build/pkg/apis/build/v1alpha1"

Surface Step Results in BuildRun Status #871

Surface Step Results in BuildRun Status #871

Conversation

shahulsonhal commented Sep 21, 2021 • edited Loading

Changes

Submitter Checklist

Release Notes

shahulsonhal commented Sep 21, 2021

gabemontero commented Sep 21, 2021

gabemontero commented Sep 21, 2021 • edited Loading

adambkaplan commented Sep 21, 2021

HeavyWombat left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

gabemontero left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

shahulsonhal commented Sep 22, 2021 • edited Loading

SaschaSchwarze0 left a comment

Choose a reason for hiding this comment

SaschaSchwarze0 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

gabemontero commented Sep 30, 2021

openshift-ci bot commented Sep 30, 2021

SaschaSchwarze0 left a comment

Choose a reason for hiding this comment

SaschaSchwarze0 left a comment

Choose a reason for hiding this comment

shahulsonhal commented Sep 21, 2021 •

edited

Loading

gabemontero commented Sep 21, 2021 •

edited

Loading

shahulsonhal commented Sep 22, 2021 •

edited

Loading