Five defects traced to one fact: src/background/index.js read and wrote the
module-level `state` singleton in src/shared/state.js, which the MV3 service
worker never populates and which answered an unpopulated read out of
DEFAULT_STATE in silence. Every previous fix added a loadState() before the
access, and that is what produced the fifth: a load detaches the objects an
in-flight handler is holding.
So the reachability goes rather than a sixth call site.
The background now has its own storage layer, src/background/state.js:
getState() is a detached, normalized per-call read, and updateState() is a
queued read-modify-write whose read is one storage round trip ahead of its
write. Nothing in the background holds an in-memory copy of the profile. The
write is the whole record, and updateState()'s header now names what that costs:
a popup write landing inside that one-round-trip window is reverted.
- Every handler takes one snapshot and answers from it, including the address
it names: activeAddressOf(s) replaced a second, later storage read that
could disagree with the first.
- wallet_switchEthereumChain applies applyChainSwitchFields() (split out of
chainSwitch.js, which keeps the singleton path for the popup) inside
updateState() instead of calling onChainSwitch() on the singleton.
- The remembered site decision is a read-modify-write, not a load-mutate-save
around a prompt the user takes seconds to answer.
- backgroundRefresh() refreshes a private copy of the wallets and applies the
balances that came back by address, so it never publishes an object other
in-flight work holds, and a wallet added or deleted during the round trip
survives its write.
- The transaction attempt takes its chain id and its endpoint from the same
snapshot. They used to come from different moments, so a chain switch
committed in between moved the endpoint under an artifact already verified
against the old chain.
getProvider(rpcUrl, networkId) now REQUIRES the network id and validates it
against networks.js. That closes the cold-worker wrong-chain send at its shape
rather than at one call site: the hint used to default to currentNetwork() off
the unpopulated singleton, so the endpoint was the user's chain and ethers
fixed chainId at 0x1, and the wallet's own verifySignedTx then refused every
non-mainnet dApp send. refreshBalances(), lookupTokenInfo(), scanForAddresses()
and resolveEnsName() carry the id through; balances.js no longer requires
state.js at all.
The prohibition is enforced mechanically, not by review, and it is enforced by
the bundler rather than by a guess at what the bundler does. The table of
modules an entry point's bundle may not contain lives in
script/lib/forbiddenBundleInputs.js — one copy, read by both layers that act on
it — and build.js's assertNoForbiddenInputs() fails the build when esbuild's
metafile reports src/shared/state.js as an input of a background bundle, naming
the import chain from the metafile's own graph. That is the resolution the
shipped bundle was built from, so no specifier syntax, no hop and no resolution
rule can slip past it; Dockerfile:42 runs make build, so it holds in CI.
A background entry point the table does not name fails the build as well. The
five defects were accidents, and so is adding a second worker entry point
without knowing that a table elsewhere needs a line for it: entry points under
src/background/ are prohibited by default and must be listed, rather than
protected only when someone remembers. That prefix is the build's only notion of
"the background", and eslint.config.js scopes the lint rule from the same
constant so the two layers cannot disagree about it.
Every way the table can rot is a failure rather than a quiet pass: a key no
bundled entry point matched, a listed module this build bundled nowhere, and an
entry that lists no modules. The second is what makes a rename of
src/shared/state.js loud instead of silently disarming the check, and it is
stronger than an existsSync() because it also fails when the module is still
there but has dropped out of every bundle. The third is refused at require time,
where the table is defined, because an empty list also empties the lint rule's
forbidden set — one character, and a plain require of the singleton in the
worker was green in make test, make lint and make build alike.
An entry is recorded as checked only once its bundle's inputs are in hand. It
used to be recorded before the output lookup that produces them, so an early
return past that point left both halves of the guarantee satisfied by a bundle
nothing had examined.
What the assertion does NOT cover is a COPY of the singleton at another path: it
is keyed by path, so a copy builds and lints clean. That is stated where the
table lives, with what the residual actually is — a copy carries the singleton's
own guard, so an unloaded read is a loud StateNotLoadedError and defects 1-3
cannot recur silently, but a copy carries loadState() too, so defects 4 and 5
(a stale read several awaits after a load, a load detaching objects an in-flight
handler is mutating) would recur over it in silence.
make check does not run make build, so the assertion is unit tested against
synthetic metafiles in tests/buildForbiddenInputs.test.js: build.js runs its
build() only as a program now and exports the checks. Executing a check in CI
is not testing it — without that file, inverting the condition leaves every
check in this repo green with the singleton back in the worker. Each vacuous
pass above has a case, including the output lookup that finds nothing, the empty
list, the unlisted second entry point, and recordBundledInputs() itself, which
every other case used to hand-seed.
A custom ESLint rule walks the CommonJS require graph from every src/background/
file and reports the same thing in the editor, before a full bundle. It reads
the same table, and it matches specifiers textually, so it is best-effort fast
feedback and not the guarantee — two earlier revisions of it shipped holes (a
template literal, a dynamic import(), a comment inside the call, a directory
resolved through package.json main). Those are covered now and pinned by
tests/backgroundStateLintRule.test.js. Two shapes it does not report are pinned
there as asserted non-reports, so the header's list of its bounds is measured
rather than claimed: a computed specifier (require("../shared/" + "state"),
which esbuild constant-folds into the bundle) and a symlink to the module
(esbuild reports the real path). Each is make lint exit 0 and make build exit 2.
Reading a persisted field of the singleton before any load now throws
StateNotLoadedError instead of serving DEFAULT_STATE.
Test stubs: chrome.storage.local is a serialization boundary, and eight files
stubbed it with an aliasing get, so the object a module held and the object
"storage" held were one object — an assertion could pass on a build that never
wrote anything. Every test that drives real persistence now goes through
tests/support/storageStub.js, which structured-clones in both directions.
closes#320
manifest/chrome.json now carries a fixed public key, so the extension id and the chrome.storage.local partition holding the wallet stay stable across checkout moves and re-clones instead of being derived from the absolute path. A release entrypoint produces a self-contained versioned artifact per browser, including the files that sit at dist/ root outside both browser directories. One version source of truth, enforced: the build fails naming the culprit when the two manifests and package.json disagree, and BUILD_COMMIT now marks a dirty tree as dirty. Firefox ships an unsigned XPI; the README states that release Firefox and ESR refuse it, that Developer Edition or Unbranded is required, and that Remove is irreversible except from the recovery phrase, which is asserted by test.
A failed release build no longer leaves a complete, loadable debug bundle in dist/ whose every wallet uses the publicly committed test recovery phrase. Each step of the release build runs through script/discard-dist-on-failure, which removes dist/ on failure, says on stderr that it did and why, and returns the step's own status. build-debug is deliberately unwrapped. script/verify-build is untouched.
README, the script synopsis, its header paragraph and the check_dist_tree comment now all say the same thing: regular files and symlinks under dist/ are covered; fifos, sockets, device nodes and empty directories are not, and why. No behaviour change — the walk is untouched.
verify-build read its expectation from AUTISTMASK_DEBUG in its own environment
and the Makefile invoked it bare, so an operator with that variable exported
who ran the release target got an INSECURE debug build — every wallet it
creates uses the publicly committed test phrase — verified green, exit 0. It
also had no provenance: a 26-byte file containing the right marker string
passed, the content script and manifest.json were never inspected, and an
entire hand-written dist/ passed.
--expect release|debug and --receipt PATH are now both required, with no
defaults and nothing read from the environment. build.js records every file it
emits with its sha256 and writes the receipt; the Makefile mktemps it outside
the repo per invocation with a trap, and build.js refuses a receipt path inside
dist/. Verification runs three passes in a load-bearing order — receipt shape,
full dist/ walk, then per-file bytes — so an unwalkable subtree cannot make
files look absent. dist/constants-bundles.txt, which was an unsigned trust root
living inside the tree it vouched for, is gone.
What this proves is bounded and stated as such: dist/ is byte-for-byte the
output of the build.js run that just finished, within one make build
invocation. It proves nothing about the honesty of the source tree or build.js,
and nothing to anyone handed a dist/ from elsewhere — that is signing, #310.
The standalone make verify-build target is removed because its only input would
be dist/ itself, i.e. the artifact vouching for itself.
Verified: make check green, test-verify-build 39 cases (was 18), test-e2e 55/55
and test-e2e-firefox 8/8 with make build running uncached inside both images.
All four original bypasses now exit 1. Mutations: digests disabled fails
exactly 4 cases, dropping the dist/ walk fails exactly 8, restoring the ambient
fallback fails exactly 1.
Drives the real popup in a real Firefox with dist/firefox/ installed as an
unpacked MV2 temporary add-on via geckodriver. make test-e2e-firefox, outside
make check like the Chrome suite. Zero npm dependencies: plain fetch and
child_process against geckodriver's HTTP API. Base image, Firefox tarball and
geckodriver are each pinned by digest and verified at build time.
Error capture reads the privileged console service through Marionette's chrome
context, not WebDriver BiDi. BiDi delivers nothing at all for extension pages,
so a BiDi-based harness would observe zero events and report success -- the
vacuous-check shape this repo has shipped twice. Both the driver and the README
say so where someone would be tempted to simplify.
Demonstrated to discriminate: a background page that throws at the top of the
file, a missing import, and an async throw where every UI assertion still
passes each fail the run.
Three limits are measured and documented rather than papered over: capture is
poll-based so an error is attributed to a step, not a moment; the console ring
buffer holds 250 messages and evicts the oldest, measured against a clean-run
peak of 4; and the drained window ends roughly 1.5s after the last step, with
observed jitter rather than a hard boundary. Content-script capture is marked
unverified because --network none leaves no page to inject into, and that same
choice inverts coverage of network-dependent code.
build.js records which emitted bundles contain src/shared/constants.js, and
constants.js carries a marker constant-folded from DEBUG itself. script/verify-build
cross-checks the two and fails on every way of not knowing, so deleting the
__BUILD_DEBUG__ define now breaks the build instead of shipping a live debug branch.
Runs the real popup in a pinned containerized Chrome and fails on any uncaught
page error or console.error. Also fixes the two defects it caught: the missing
showView import in addToken.js and the missing addressDotHtml import in
transactionDetail.js.
closes#150closes#151
