Deletes the third of three "standing limits" in README.md's End-to-End Tests section — the one claiming src/content/inpage.js rebuilds every rejection as new Error(message), so the page catches an error with no code.
Evidence it is fixed on next (9dcd875, #274): src/content/inpage.js defines ProviderRpcError and toPageError() returns new ProviderRpcError(error.code, message, error.data) whenever a code arrived; tests/e2e/run.jsassertUserRejection() asserts outcome.hasCode, outcome.code === 4001 and outcome.name === "ProviderRpcError" on the page's caught error, alongside the wire assertion.
The two surviving limits (stubbed RPC, and the toolbar-anchored site-connection prompt not being observable in headless Chromium) were checked against the current code and left unchanged.
Docs-only: README.md and TODO.md. make check green (703 tests, 18 verify-build cases, prettier clean).
Closes https://git.eeqj.de/sneak/AutistMask/issues/285.
Deletes the third of three "standing limits" in `README.md`'s End-to-End Tests section — the one claiming `src/content/inpage.js` rebuilds every rejection as `new Error(message)`, so the page catches an error with no `code`.
Evidence it is fixed on `next` (`9dcd875`, https://git.eeqj.de/sneak/AutistMask/issues/274): `src/content/inpage.js` defines `ProviderRpcError` and `toPageError()` returns `new ProviderRpcError(error.code, message, error.data)` whenever a code arrived; `tests/e2e/run.js` `assertUserRejection()` asserts `outcome.hasCode`, `outcome.code === 4001` and `outcome.name === "ProviderRpcError"` on the page's caught error, alongside the wire assertion.
The two surviving limits (stubbed RPC, and the toolbar-anchored site-connection prompt not being observable in headless Chromium) were checked against the current code and left unchanged.
Docs-only: `README.md` and `TODO.md`. `make check` green (703 tests, 18 verify-build cases, prettier clean).
The End-to-End Tests section listed, as the third of three standing limits
of the dApp coverage, the EIP-1193 error code not surviving the last hop
into the page. That stopped being true when `9dcd875` landed
(#274): the provider now rejects
with a `ProviderRpcError` carrying `code`, and the e2e suite asserts the
code on the page as well as on the wire.
The other two limits are still accurate against the current code and are
left alone.
clawbot
self-assigned this 2026-08-17 08:02:20 +02:00
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Closes #285.
Deletes the third of three "standing limits" in
README.md's End-to-End Tests section — the one claimingsrc/content/inpage.jsrebuilds every rejection asnew Error(message), so the page catches an error with nocode.Evidence it is fixed on
next(9dcd875, #274):src/content/inpage.jsdefinesProviderRpcErrorandtoPageError()returnsnew ProviderRpcError(error.code, message, error.data)whenever a code arrived;tests/e2e/run.jsassertUserRejection()assertsoutcome.hasCode,outcome.code === 4001andoutcome.name === "ProviderRpcError"on the page's caught error, alongside the wire assertion.The two surviving limits (stubbed RPC, and the toolbar-anchored site-connection prompt not being observable in headless Chromium) were checked against the current code and left unchanged.
Docs-only:
README.mdandTODO.md.make checkgreen (703 tests, 18 verify-build cases, prettier clean).