test(e2e): assert the fixture's preconditions, and say where connect stopped - #498
Open
sotashimozono wants to merge 1 commit into
Open
test(e2e): assert the fixture's preconditions, and say where connect stopped#498sotashimozono wants to merge 1 commit into
sotashimozono wants to merge 1 commit into
Conversation
…stopped
Three different faults have been producing one message — "no shadow
vault entry appeared, connect command likely never fired" — and that
collapse cost three rounds of fixing the wrong thing. The message stayed
identical while its cause changed underneath: palette not interactive,
then the command possibly unregistered, then a ConnectModal whose second
pane was never clicked.
Two changes so the next red names itself.
The scaffold now asserts what connect DEPENDS ON, and throws as what
they are — fixture faults, not product failures:
- the test private key exists (without it, auth cannot succeed);
- for the default `rpc` transport, a daemon binary was actually staged
(without it, there is nothing to connect to).
And `connectAndWaitForShadowVault` attaches the evidence that separates
the remaining two: the tail of the plugin's own log (how far connect
got) and the labels of the visible modal buttons (whether the click
landed). Both are best-effort, since a diagnostic that throws would
replace the error it was meant to explain.
Refs #429
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Codecov Report✅ All modified and coverable lines are covered by tests.
Flags with carried forward coverage won't be shown. Click here to find out more. 🚀 New features to boost your workflow:
|
This was referenced Aug 3, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stacked on #495.
The problem this fixes is diagnostic, not behavioural
Three different faults have been producing one message:
That collapse cost three rounds of fixing the wrong thing. The text stayed identical while its cause changed underneath: the palette was not interactive (#492), then the command might not have been registered (#493), then
ConnectModal's second pane was never clicked (#495). Each fix was right; none of them could be known to be right from the message.What now names itself
The scaffold asserts what connect depends on and throws as a fixture fault, not a product failure:
rpctransport, a daemon binary was actually staged — without it there is nothing to connect to.Either now reads
fixture missing: …with the path and the fix, instead of surfacing as "connect is broken" in fifteen specs.connectAndWaitForShadowVaultattaches the evidence that separates the remaining two causes:Both are best-effort: a diagnostic that throws would replace the error it exists to explain.
Why this instead of another guess
The last E2E run still fails connect nearly everywhere, and I have now twice mistaken this message for a cause I could name. The honest next step is to make the run answer the question rather than to patch a fourth hypothesis into it.
Verification
tsc --noEmitgreen locally. No unit-test surface. No auto-merge.Refs #429
🤖 Generated with Claude Code