Cache bincapz-samples repository to speed up subsequent tests by egibs · Pull Request #448 · chainguard-dev/malcontent

egibs · 2024-09-12T14:47:54Z

Closes: #445

Repeatedly cloning the bincapz-samples directory (which is >150MB) wastes a lot of time. This PR attempts to cache the output in out/ so that subsequent test runs do not need to re-clone the entire repository.

First run:

$ make test
mkdir -p out
git clone git@github.com:chainguard-dev/bincapz-samples.git out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7
Cloning into 'out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7'...
remote: Enumerating objects: 267, done.
remote: Counting objects: 100% (48/48), done.
remote: Compressing objects: 100% (44/44), done.
remote: Total 267 (delta 5), reused 35 (delta 2), pack-reused 219 (from 1)
Receiving objects: 100% (267/267), 157.86 MiB | 1.15 MiB/s, done.
Resolving deltas: 100% (22/22), done.
pushd out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7; git checkout bdcb8c2e9bf557a0abe3e2b0144f437d456299b7; popd
~/repos/chainguard-dev/bincapz/out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7 ~/repos/chainguard-dev/bincapz
Note: switching to 'bdcb8c2e9bf557a0abe3e2b0144f437d456299b7'.

You are in 'detached HEAD' state. You can look around, make experimental
changes and commit them, and you can discard any commits you make in this
state without impacting any branches by switching back to a branch.

If you want to create a new branch to retain commits you create, you may
do so (now or later) by using -c with the switch command. Example:

  git switch -c <new-branch-name>

Or undo this operation with:

  git switch -

Turn off this advice by setting config variable advice.detachedHead to false

HEAD is now at bdcb8c2 Add samples for: Darkcracks, helmet-validate, noblox, aspdasdksa2 (#4)
~/repos/chainguard-dev/bincapz
rm -rf samples; cp -a out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7 samples/
cp -a test_data/. samples/
for file in caddy.xz chezmoi.xz minio_x86_64.xz mongosh.xz neuvector_agent_aarch64.xz opa.xz ; do \
		tar -xJvf samples/linux/clean/$file -C samples/linux/clean; \
	done
x caddy
x chezmoi
x minio.x86_64
x mongosh
x neuvector-agent.aarch64
x opa
tar -xJvf samples/macOS/clean/bincapz.xz -C samples/macOS/clean
x bincapz
go test github.com/chainguard-dev/bincapz github.com/chainguard-dev/bincapz/out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7/does-nothing github.com/chainguard-dev/bincapz/pkg/action github.com/chainguard-dev/bincapz/pkg/bincapz github.com/chainguard-dev/bincapz/pkg/compile github.com/chainguard-dev/bincapz/pkg/profile github.com/chainguard-dev/bincapz/pkg/render github.com/chainguard-dev/bincapz/pkg/report github.com/chainguard-dev/bincapz/pkg/version github.com/chainguard-dev/bincapz/rules github.com/chainguard-dev/bincapz/samples github.com/chainguard-dev/bincapz/samples/does-nothing github.com/chainguard-dev/bincapz/third_party
?   	github.com/chainguard-dev/bincapz	[no test files]
?   	github.com/chainguard-dev/bincapz/out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7/does-nothing	[no test files]
?   	github.com/chainguard-dev/bincapz/pkg/bincapz	[no test files]
?   	github.com/chainguard-dev/bincapz/pkg/compile	[no test files]
?   	github.com/chainguard-dev/bincapz/pkg/render	[no test files]
?   	github.com/chainguard-dev/bincapz/pkg/version	[no test files]
?   	github.com/chainguard-dev/bincapz/rules	[no test files]
ok  	github.com/chainguard-dev/bincapz/pkg/action	(cached)
ok  	github.com/chainguard-dev/bincapz/pkg/profile	(cached)
ok  	github.com/chainguard-dev/bincapz/pkg/report	(cached)
?   	github.com/chainguard-dev/bincapz/samples/does-nothing	[no test files]
?   	github.com/chainguard-dev/bincapz/third_party	[no test files]
ok  	github.com/chainguard-dev/bincapz/samples	56.764s

Subsequent runs:

$ make test
rm -rf samples; cp -a out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7 samples/
cp -a test_data/. samples/
for file in caddy.xz chezmoi.xz minio_x86_64.xz mongosh.xz neuvector_agent_aarch64.xz opa.xz ; do \
		tar -xJvf samples/linux/clean/$file -C samples/linux/clean; \
	done
x caddy
x chezmoi
x minio.x86_64
x mongosh
x neuvector-agent.aarch64
x opa
tar -xJvf samples/macOS/clean/bincapz.xz -C samples/macOS/clean
x bincapz
go test github.com/chainguard-dev/bincapz github.com/chainguard-dev/bincapz/out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7/does-nothing github.com/chainguard-dev/bincapz/pkg/action github.com/chainguard-dev/bincapz/pkg/bincapz github.com/chainguard-dev/bincapz/pkg/compile github.com/chainguard-dev/bincapz/pkg/profile github.com/chainguard-dev/bincapz/pkg/render github.com/chainguard-dev/bincapz/pkg/report github.com/chainguard-dev/bincapz/pkg/version github.com/chainguard-dev/bincapz/rules github.com/chainguard-dev/bincapz/samples github.com/chainguard-dev/bincapz/samples/does-nothing github.com/chainguard-dev/bincapz/third_party
?   	github.com/chainguard-dev/bincapz	[no test files]
?   	github.com/chainguard-dev/bincapz/out/samples-bdcb8c2e9bf557a0abe3e2b0144f437d456299b7/does-nothing	[no test files]
?   	github.com/chainguard-dev/bincapz/pkg/bincapz	[no test files]
?   	github.com/chainguard-dev/bincapz/pkg/compile	[no test files]
?   	github.com/chainguard-dev/bincapz/pkg/render	[no test files]
?   	github.com/chainguard-dev/bincapz/pkg/version	[no test files]
?   	github.com/chainguard-dev/bincapz/rules	[no test files]
ok  	github.com/chainguard-dev/bincapz/pkg/action	(cached)
ok  	github.com/chainguard-dev/bincapz/pkg/profile	(cached)
ok  	github.com/chainguard-dev/bincapz/pkg/report	(cached)
?   	github.com/chainguard-dev/bincapz/samples/does-nothing	[no test files]
?   	github.com/chainguard-dev/bincapz/third_party	[no test files]
ok  	github.com/chainguard-dev/bincapz/samples	55.816s

Signed-off-by: egibs <20933572+egibs@users.noreply.github.com>

Makefile

tstromberg · 2024-09-12T15:09:46Z

Makefile

+
+update-samples: out/samples-$(SAMPLES_HASH)
+	rm -rf samples; cp -a out/samples-$(SAMPLES_HASH) samples/
+	cp -a test_data/. samples/


Can we leave the test_data where it is? It seems weird to pollute the samples data this way. I'm OK if we leave this for a future PR though as it would involve changing tests around.

I've thought about that and it's definitely messy as-is. I'll keep thinking on it.

Makefile

Signed-off-by: egibs <20933572+egibs@users.noreply.github.com>

tstromberg · 2024-09-12T19:30:18Z

.github/workflows/go-tests.yaml


      - name: Prepare samples
        run: |
-          cp -a ${{ github.workspace }}/test_data/. ${{ github.workspace }}/samples/


Can we run the makefile rule instead of duplicating this logic?

I had some hang up about that mentally but it may just be because I cloned the other repository separately and wasn't sure if our anonymous clone would work.

The make targets should work in CI as well.

Updated in 4ad123f (#448).

tstromberg · 2024-09-12T19:31:28Z

Makefile

+	mkdir -p out
+	git clone https://github.com/chainguard-dev/bincapz-samples.git out/samples-$(SAMPLES_HASH).tmp
+	git -C out/samples-$(SAMPLES_HASH).tmp checkout $(SAMPLES_HASH)
+	for file in caddy.xz chezmoi.xz minio_x86_64.xz mongosh.xz neuvector_agent_aarch64.xz opa.xz ; do \


Could this work to avoid hardcoding filenames?

find <path> -name "*.xz" -exec xz -d {}

Implemented a variant of this in 4ad123f (#448).

Signed-off-by: egibs <20933572+egibs@users.noreply.github.com>

tstromberg · 2024-09-13T12:21:46Z

Makefile

@@ -113,16 +126,7 @@ update-third-party:
 .PHONY: refresh-sample-testdata out/bincapz
 refresh-sample-testdata: clone-samples out/bincapz
 	cp ./test_data/refresh-testdata.sh samples/


note: in a future PR, we can switch to running this script from the test_data directory.

tstromberg · 2024-09-13T12:23:06Z

merging so I can work on bincapz without wifi :)

…uard-dev#448) * Cache bincapz-samples repository to speed up subsequent tests Signed-off-by: egibs <20933572+egibs@users.noreply.github.com> * Update -> prepare Signed-off-by: egibs <20933572+egibs@users.noreply.github.com> * Address PR comments; rework out/samples target Signed-off-by: egibs <20933572+egibs@users.noreply.github.com> * Use Make targets in CI; simplify + clean up targets Signed-off-by: egibs <20933572+egibs@users.noreply.github.com> --------- Signed-off-by: egibs <20933572+egibs@users.noreply.github.com>

Cache bincapz-samples repository to speed up subsequent tests

4767de0

Signed-off-by: egibs <20933572+egibs@users.noreply.github.com>

egibs requested a review from tstromberg September 12, 2024 14:47

Update -> prepare

7054bdd

Signed-off-by: egibs <20933572+egibs@users.noreply.github.com>

tstromberg suggested changes Sep 12, 2024

View reviewed changes

egibs marked this pull request as draft September 12, 2024 18:28

Address PR comments; rework out/samples target

74ebc63

Signed-off-by: egibs <20933572+egibs@users.noreply.github.com>

egibs force-pushed the cache-samples branch from 5a0f132 to 74ebc63 Compare September 12, 2024 18:59

tstromberg reviewed Sep 12, 2024

View reviewed changes

egibs force-pushed the cache-samples branch from ff2bc32 to 04a6d33 Compare September 12, 2024 20:14

Use Make targets in CI; simplify + clean up targets

4ad123f

Signed-off-by: egibs <20933572+egibs@users.noreply.github.com>

egibs force-pushed the cache-samples branch from 04a6d33 to 4ad123f Compare September 12, 2024 20:25

egibs marked this pull request as ready for review September 12, 2024 20:30

egibs requested a review from tstromberg September 12, 2024 20:30

tstromberg reviewed Sep 13, 2024

View reviewed changes

tstromberg approved these changes Sep 13, 2024

View reviewed changes

tstromberg merged commit 9223b64 into chainguard-dev:main Sep 13, 2024

egibs deleted the cache-samples branch September 22, 2024 19:31

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Cache bincapz-samples repository to speed up subsequent tests#448

Cache bincapz-samples repository to speed up subsequent tests#448
tstromberg merged 4 commits intochainguard-dev:mainfrom
egibs:cache-samples

egibs commented Sep 12, 2024

Uh oh!

Uh oh!

Uh oh!

tstromberg Sep 12, 2024

Uh oh!

egibs Sep 12, 2024

Uh oh!

Uh oh!

Uh oh!

tstromberg Sep 12, 2024

Uh oh!

egibs Sep 12, 2024

Uh oh!

egibs Sep 12, 2024

Uh oh!

tstromberg Sep 12, 2024

Uh oh!

egibs Sep 12, 2024

Uh oh!

tstromberg Sep 13, 2024

Uh oh!

tstromberg commented Sep 13, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Conversation

egibs commented Sep 12, 2024

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

tstromberg commented Sep 13, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants