Small fixes after recent PRs #1116
Conversation
We actually want to run the litd itests on the staging branch that has the custom channel tests. We also can speed up the tests by removing the yarn build for the static files, since we don't need those for the tests (an empty index.html just needs to exist for the build to succeed).
Pull Request Test Coverage Report for Build 10715230733Details
💛 - Coveralls |
| uses: actions/checkout@v3 | ||
| with: | ||
| repository: lightninglabs/lightning-terminal | ||
| ref: ${{ env.LITD_ITEST_BRANCH }} |
There was a problem hiding this comment.
not sure if related, but there's a tiny misbehavior according to an itest
https://github.com/lightninglabs/taproot-assets/actions/runs/10715230733/job/29710290030?pr=1116#step:4:52159
we get a "status completed" instead of "proofs transferred"
There was a problem hiding this comment.
Hmm, that seems to be a GitHub only flake... I ran this test a bunch of time locally and it never failed for me...
Going to re-run CI and just hope it goes away.
There was a problem hiding this comment.
Man, this has been driving me bananas! I had something similar way back and never understood why that happens on GitHub.
The weird thing is that in my case when the test ran on Github it expected a different state than what the test was actually supposed to be expecting, and I think that's the case here as well, right?
The test expects SendStateWaitTxConf
taproot-assets/itest/psbt_test.go
Lines 1148 to 1152 in 6f4d65f
But according to the run on GitHub the test expects SendStateTransferProofs which makes no sense whatsoever:
https://github.com/lightninglabs/taproot-assets/actions/runs/10715230733/job/29710290030?pr=1116#step:4:52158
Maybe an enumeration is off by one?
There was a problem hiding this comment.
The enum values passed into AssertSendEvents() is a "from" and "to". So it's expecting all events from SendStateWaitTxConf in order up until SendStateComplete. And SendStateTransferProofs just happens to be an in-between one.
So I think it's a pure timing issue, perhaps because GitHub is slower than the local machine.
Fixes formatting of build tag excluded files the linter didn't pick up on.
And fixes the new itest to actually run against the correct
litdbranch (I missed that during review). At the same time we slightly optimize the litd itest build.