Compare commits

..

1 Commits

Author SHA1 Message Date
84c58a97ab fix: render the view "Back" lands on after the popup is reopened (closes #268)
All checks were successful
check / check (push) Successful in 36s
goBack() only unhid its target. A reopened popup renders the wallet list
and the one view restoreView() lands on, so every other view is still the
blank static template from index.html: pressing Back from Settings onto an
address showed an empty address line and no balances, and the same held for
address-token, receive, confirm-tx and transaction.

The per-view dispatch and its data guards move out of restoreView() into
src/popup/viewRouter.js, and goBack() now routes a popped view through the
same code by way of a renderer index.js registers with setBackRenderer().
A view whose backing state is gone falls back to Home the way the restore
does, rather than showing an empty template.

The Back path renders only a view this page load has not rendered yet.
viewRouter.js keeps a page-load-scoped set of rendered views, written by
showView() — the last thing every render path runs, forward navigation and
the restore alike, so a view added later registers itself rather than
needing to be remembered. Back onto a view already in that set unhides it
and nothing more, exactly as it does for a view outside RESTORABLE_VIEWS:
rendering a second time would re-fetch and overwrite what the view holds,
such as an edit typed into Settings and not yet saved. Home stays the
exception and is re-rendered on every Back, as it was when goBack() called
the renderWalletList() registered through setRenderMain(), so the wallet
list reflects what changed while the user was away from it.

tests/backNavigation.test.js drives the real goBack() over the reproduction
and over each of address-token, receive, confirm-tx, transaction, success-tx
and error-tx, with and without their backing state, and pins that a view
this page load already rendered — a live-session view, or Settings revisited
— is only unhidden, while Home still re-renders. tests/e2e/run.js adds three
cases against the real popup in a real browser, because make check cannot
see a blank view: a real close and reopen then Back onto the address screen
and onto Receive, and an in-session Back onto Settings that must keep
unsaved input. All three were demonstrated failing first.
2026-08-12 11:32:46 +00:00
15 changed files with 86 additions and 823 deletions

View File

@@ -1,49 +0,0 @@
name: e2e
on: [push]
# The browser end-to-end suites, one job per browser, deliberately kept out
# of the check workflow: REPO_POLICIES.md caps make test at 20 seconds and
# script/cibuild is a plain `docker build .` whose Dockerfile runs
# make check, so folding a browser suite into either would blow that cap
# and slow the local fast path. Before this workflow every browser-level
# guarantee in this repo held only when a human remembered to run it.
#
# One job per browser rather than two steps in one job, so a Chrome failure
# does not hide the Firefox result.
#
# Each job is one script and nothing else. Both scripts need docker and
# nothing else — they deliver the repo to the daemon as a build context and
# build the extension inside the pinned image — which is what makes them
# runnable here at all: the runner executes the job in a container against
# the host's docker socket, so a `-v "$PWD:/work"` source path is resolved
# by the host daemon and mounts an empty directory, and the runner image's
# node is too old to install this repo's dependencies.
#
# These jobs REPORT, they do not gate. Whether a check blocks a merge is
# Gitea branch protection, which this repo does not configure, so a failure
# here is a red mark a reviewer has to account for rather than a hard
# block. Making e2e-chrome a required check is blocked on the measured
# flake in the dApp signing wait -- two of six runs of unmutated code on a
# loaded machine -- tracked as
# https://git.eeqj.de/sneak/AutistMask/issues/287. A gate that fails at
# random teaches people to merge past red.
#
# Nothing here may pass vacuously. There is no continue-on-error and no
# `|| true`. Both scripts exit non-zero when docker is missing, when the
# image build fails, and when the browser fails to start; the Chrome
# harness aborts the suite outright if its network interception is not in
# effect.
jobs:
e2e-chrome:
runs-on: ubuntu-latest
steps:
# actions/checkout v4.2.2, 2026-02-22
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- run: script/test-e2e
e2e-firefox:
runs-on: ubuntu-latest
steps:
# actions/checkout v4.2.2, 2026-02-22
- uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683
- run: script/test-e2e-firefox

View File

@@ -83,11 +83,10 @@ provide:
git pre-commit hook
- `script/projectname` — print the project name (used for the Docker image tag)
- `script/test` — run the test suite (jest)
- `script/test-e2e` — run the Chrome browser end-to-end suite (docker is the
only prerequisite: it builds a pinned image that carries the repo and a fresh
extension build, see [End-to-End Tests](#end-to-end-tests))
- `script/test-e2e-firefox` — run the Firefox browser end-to-end suite (same,
against an image with a pinned Firefox and geckodriver, see
- `script/test-e2e` — run the Chrome browser end-to-end suite (docker required;
see [End-to-End Tests](#end-to-end-tests))
- `script/test-e2e-firefox` — run the Firefox browser end-to-end suite (docker
required; builds its own pinned image, see
[End-to-End Tests](#end-to-end-tests))
- `script/lint` — run the linter
- `script/fmt` — format all files (writes)
@@ -137,12 +136,11 @@ are outside `make check`.
`make test-e2e` builds `dist/chrome/` and drives the **real popup in a real
Chrome**, loaded as an unpacked MV3 extension inside a pinned
`mcr.microsoft.com/playwright` container (pinned by digest in
`tests/e2e/Dockerfile`, which is also where the extension is built; docker is
required and the suite fails loudly rather than skipping if it is unavailable).
The suite lives in `tests/e2e/` and is driven by `playwright-core`, whose
version must stay matched to the container's Playwright version — the browsers
ship inside the image.
`mcr.microsoft.com/playwright` container (pinned by digest in `script/test-e2e`;
docker is required and the suite fails loudly rather than skipping if it is
unavailable). The suite lives in `tests/e2e/` and is driven by
`playwright-core`, whose version must stay matched to the container's Playwright
version — the browsers ship inside the image.
It covers popup load, WebAssembly compilation under the shipped CSP (see
[Content Security Policy](#content-security-policy)), wallet creation through
@@ -322,46 +320,9 @@ Two limits are worth knowing, both real differences from the Chrome suite:
Neither `make test-e2e` nor `make test-e2e-firefox` is part of `make check` or
`make test`. `REPO_POLICIES.md` caps `make test` at 20 seconds and a browser
suite does not fit; nothing in `tests/e2e/` is named `*.test.js`, so jest cannot
pick it up either. Run them locally before changing anything under
`src/popup/views/`.
### In CI
`.gitea/workflows/e2e.yml` runs both suites on every push, as two jobs —
`e2e-chrome` and `e2e-firefox` — separate from the `check` workflow, so the
20-second `make test` cap and the local fast path are untouched. Each job is a
checkout and the matching `script/` entrypoint, nothing else.
Docker is the only thing either job needs from the runner, and that is not an
accident. The runner executes a job inside a container against the **host's**
docker daemon, so a `docker run -v "$PWD:/work"` source path is resolved by the
host and mounts an empty directory, and the runner image's node is too old to
install this repo's dependencies. Both suites therefore ship the repo to the
daemon as a build context and build the extension inside the image, which works
identically on a laptop.
The jobs **report, they do not gate.** A failure is a red mark against the
commit that a reviewer has to account for, not a hard block: whether a check
blocks a merge is Gitea branch protection, which this repo does not configure.
That is not only a statement about configuration. The Chrome suite is
**measurably flaky under load** — two of six runs of unmutated code on a busy
machine lost the approval popup out from under the dApp signing wait, always in
the `#183` section, tracked as
[#287](https://git.eeqj.de/sneak/AutistMask/issues/287). So a red `e2e-chrome`
has to be read before it is believed, and that flake is the blocker to ever
making this a required check. Do not answer it with a retry wrapper: a suite
that reruns until it is green stops being evidence.
Nothing in either job can pass vacuously. There is no `continue-on-error` and no
`|| true`; both scripts exit non-zero when docker is missing, when the image
build fails, and when the browser fails to start; the Chrome harness aborts the
suite outright if its network interception is not in effect.
Measured on this repo's runner: `e2e-chrome` about 1m55s cold, almost all of it
the one-time pull of the pinned ~800MB Playwright layer, and well under a minute
once that layer is cached. `e2e-firefox` about 1m05s cold, and it caches its
Firefox and geckodriver downloads the same way.
pick it up either. Neither is wired into the Gitea workflow yet —
docker-in-docker in CI is a separate question. Run them locally before changing
anything under `src/popup/views/`.
## Rationale

53
TODO.md
View File

@@ -33,8 +33,7 @@ The backlog lives on the
authoritative; this file does not duplicate it. Full policy file set present.
Real-browser end-to-end suites (`make test-e2e` for Chrome,
`make test-e2e-firefox` for Firefox) now sit alongside `make check`, which
cannot see a runtime `ReferenceError` in a popup view, and
`.gitea/workflows/e2e.yml` runs both of them on every push.
cannot see a runtime `ReferenceError` in a popup view.
# Next Step
@@ -46,38 +45,6 @@ undefined identifiers, which is how
# Completed Steps
- 2026-08-14: CI runs the browser end-to-end suites. `.gitea/workflows/e2e.yml`
runs `script/test-e2e` and `script/test-e2e-firefox` as two jobs on every
push, separate from `check`, so `make check` and its 20-second `make test` cap
are untouched. Every browser-level guarantee in this repo — the WASM-under-CSP
check, the recovery-phrase and private-key DOM wipes, the ConfirmTx spend
gate, the dApp approval round trips — was enforced only when a human
remembered to run it by hand. The suites could not run on the runner as they
stood: the runner executes a job in a container against the host's docker
daemon, so `docker run -v "$PWD:/work"` mounts an empty directory (measured),
and the runner image's node cannot install this repo's dependencies. Both
suites now ship the repo to the daemon as a build context and build the
extension inside the pinned image, so docker is the only prerequisite on a
runner or a laptop, and both run the image by ID rather than by tag so
concurrent clones cannot swap it. The jobs report rather than gate — this repo
configures no branch protection, and the Chrome suite is measurably flaky
under load, filed as [#287](https://git.eeqj.de/sneak/AutistMask/issues/287)
rather than papered over
([#259](https://git.eeqj.de/sneak/AutistMask/issues/259)).
- 2026-08-12: EIP-1193 error codes now reach the page. `src/content/inpage.js`
rebuilt every failure as `new Error(error.message)`, so the code the
background produced and the content script relayed intact was dropped in the
last hop and a dApp checking `err.code === 4001` saw `undefined` — a wallet
the user deliberately declined was indistinguishable from one that broke. The
provider now rejects with a `ProviderRpcError` carrying `code` and, where the
boundary sent one, `data`, passed through verbatim rather than matched against
a list, so 4001, 4100 and 4902 all arrive and a future code needs no edit
here. An error the background sent with no code stays a plain `Error` with no
`code` property, and `message` is unchanged in every case. All four request
entry points (`request`, `enable`, `send`, `sendAsync`) are covered by
`tests/inpageErrors.test.js`, and the e2e probe that printed the missing code
now requires it on the page's Error as well as on the wire, for all four
rejected flows ([#274](https://git.eeqj.de/sneak/AutistMask/issues/274)).
- 2026-08-12: "Back" now renders the screen it lands on instead of only unhiding
it. A reopened popup renders the wallet list and the one screen it restores
onto, so every screen further down the stack was still the blank template from
@@ -95,20 +62,6 @@ undefined identifiers, which is how
and by three end-to-end cases against the real popup, each demonstrated
failing on the unfixed build
([#268](https://git.eeqj.de/sneak/AutistMask/issues/268)).
- 2026-08-12: `KNOWN_SYMBOLS` now maps a symbol to the set of contract addresses
that bear it, not to one of them. A ticker is not unique: seven of the 512
bundled tokens — `FRAX`, `REUSD`, `TON`, `EURE`, `MSUSD`, `MUSD` and `JPYC`
share a symbol with another bundled entry at a different real contract, and
the table, built from the list first-wins, kept only the earlier one. The
other seven were judged spoofs of their own symbol at their own address and
hidden from the balance list, the history and the send selector, so a holder
could not spend them. Both contracts of each pair come from the same CoinGecko
fetch of 2026-02-27, so neither was stale and neither was dropped.
`isSpoofedSymbol()` asks set membership instead of equality, which does not
loosen the rule — a contract outside the set is still a spoof — and a test now
walks `TOKENS` asserting no bundled token is filtered at its own address,
which is the walk the suite lacked
([#276](https://git.eeqj.de/sneak/AutistMask/issues/276)).
- 2026-08-12: The dApp approval round trips are driven end to end in the
browser. A test page served by the harness speaks EIP-1193 to the real inpage
provider through the real content script, background worker and approval popup
@@ -387,5 +340,9 @@ tracker.
- Pre-1.0 security review of the extension (key handling, DEBUG mode policy, RPC
input validation) before any 1.0rc tag. Individual filed issues are parts of
it, but the review is broader than any of them.
- Decide whether docker-in-docker makes `make test-e2e` and
`make test-e2e-firefox` runnable in the Gitea workflow. Extending the Chrome
suite itself is tracked as
[#183](https://git.eeqj.de/sneak/AutistMask/issues/183).
- Cut 1.0.0 once the milestone is empty, then continue tagging as milestones
land.

View File

@@ -7,29 +7,17 @@
# caps make test at 20 seconds and a browser suite does not fit. Run it
# yourself before touching popup views; it is the only check that can see
# a used-but-not-imported identifier blow up at runtime.
# .gitea/workflows/e2e.yml also runs it on every push, in a job separate
# from check so that cap and the local fast path both stay intact.
#
# Docker is the only prerequisite. The repo reaches the container as a
# build context and the extension is built inside it (see
# tests/e2e/Dockerfile), so nothing here depends on the node, yarn or make
# on the machine that starts the run. That is not a convenience: a bind
# mount cannot work under Gitea Actions, and the runner image's node is too
# old to install this repo's dependencies.
set -eu
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)"
ROOT="$(cd "$SCRIPT_DIR/.." && pwd -P)"
ROOT="$(cd "$(dirname "$0")/.." && pwd -P)"
IMAGE="$("$SCRIPT_DIR/projectname")-e2e-chrome"
IIDFILE=""
cleanup() {
if [ -n "$IIDFILE" ]; then
rm -f "$IIDFILE"
fi
}
# mcr.microsoft.com/playwright:v1.56.0-noble, 2026-08-09
#
# The playwright-core devDependency is pinned to the matching Playwright
# version (1.56.0) and the two must be bumped together: the browsers ship
# inside this image, and playwright-core looks for the exact browser
# revision its own version expects. A mismatch fails at launch.
IMAGE="mcr.microsoft.com/playwright@sha256:35246d87a7c88ea9b771c65d33171b2611b02a8253b4b12ce6f94376c55f99f2"
main() {
cd "$ROOT"
@@ -39,23 +27,14 @@ main() {
exit 1
fi
IIDFILE="$(mktemp)"
trap cleanup EXIT
trap 'cleanup; exit 130' INT TERM
echo "Building the Chrome e2e image (extension included)..."
docker build --iidfile "$IIDFILE" -t "$IMAGE" -f tests/e2e/Dockerfile .
echo "Building extension for e2e..."
yarn run build 2>&1
echo "Running e2e suite in the pinned Playwright container..."
# The image is run by ID, not by tag: where two clones of this repo run
# the suite at once, the other build can move the tag between this
# build and this run, and the suite would then silently test the other
# checkout.
#
# --ipc=host: Chromium's shared-memory needs more than the default
# 64MB /dev/shm or renderers crash.
# HOME=/tmp: the image's root home is not a reliable place for the
# browser profile.
# --user: keep files the suite touches owned by the caller, not root.
# HOME=/tmp: the mapped uid has no home directory in the image.
# PW_EXPERIMENTAL_SERVICE_WORKER_NETWORK_EVENTS=1: without it,
# ctx.route() intercepts page requests only, and every fetch made by
# the MV3 background service worker — including the phishing
@@ -72,10 +51,13 @@ main() {
# on a deliberate bump.
docker run --rm \
--ipc=host \
--user "$(id -u):$(id -g)" \
-e HOME=/tmp \
-e PW_EXPERIMENTAL_SERVICE_WORKER_NETWORK_EVENTS=1 \
-e "E2E_TRACE_NETWORK=${E2E_TRACE_NETWORK:-0}" \
"$(cat "$IIDFILE")" \
-v "$ROOT:/work" \
-w /work \
"$IMAGE" \
node tests/e2e/run.js
}

View File

@@ -5,17 +5,12 @@
#
# Deliberately NOT called by script/check or script/test, for the same
# reason as the Chrome suite: REPO_POLICIES.md caps make test at 20 seconds
# and a browser suite does not fit. .gitea/workflows/e2e.yml also runs it
# on every push, in a job separate from check.
# and a browser suite does not fit.
#
# Unlike script/test-e2e this builds its base image locally, because no
# Unlike script/test-e2e this builds its image locally, because no
# published image carries both a pinned Firefox and a matching geckodriver.
# All three external artifacts are pinned by digest inside the Dockerfile;
# see tests/e2e/firefox/Dockerfile, which also explains why the repo and
# the extension build are baked into the image rather than mounted.
#
# Docker is the only prerequisite: nothing here depends on the node, yarn
# or make on the machine that starts the run.
# see tests/e2e/firefox/Dockerfile.
set -eu
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd -P)"
@@ -23,14 +18,6 @@ ROOT="$(cd "$SCRIPT_DIR/.." && pwd -P)"
IMAGE="$("$SCRIPT_DIR/projectname")-e2e-firefox"
IIDFILE=""
cleanup() {
if [ -n "$IIDFILE" ]; then
rm -f "$IIDFILE"
fi
}
main() {
cd "$ROOT"
@@ -39,20 +26,16 @@ main() {
exit 1
fi
IIDFILE="$(mktemp)"
trap cleanup EXIT
trap 'cleanup; exit 130' INT TERM
echo "Building extension for e2e..."
yarn run build 2>&1
echo "Building the pinned Firefox e2e image (extension included)..."
docker build --iidfile "$IIDFILE" -t "$IMAGE" \
-f tests/e2e/firefox/Dockerfile .
# The build context is tests/e2e/firefox/ and holds nothing but the
# Dockerfile: the harness itself arrives over the bind mount below, so
# editing it never invalidates an image layer.
echo "Building the pinned Firefox e2e image..."
docker build -t "$IMAGE" "$ROOT/tests/e2e/firefox"
echo "Running the Firefox e2e suite..."
# The image is run by ID, not by tag: where two clones of this repo run
# the suite at once, the other build can move the tag between this
# build and this run, and the suite would then silently test the other
# checkout.
#
# --shm-size=1g: Firefox needs more than the default 64MB /dev/shm.
# --network none: the suite stubs nothing, so this is what keeps the
# run offline and deterministic. The extension swallows its own
@@ -60,8 +43,8 @@ main() {
# network note in README.md. Weaker than the Chrome suite's
# fixture interception, and honestly so — it proves no request
# escaped, but it cannot report which ones were attempted.
# HOME=/tmp: the image's root home is not a reliable place for the
# browser profile.
# --user: keep files the suite touches owned by the caller, not root.
# HOME=/tmp: the mapped uid has no home directory in the image.
#
# No --privileged. Firefox's sandbox logs
# "CanCreateUserNamespace() clone() failure: EPERM" on startup here;
@@ -69,8 +52,11 @@ main() {
docker run --rm \
--shm-size=1g \
--network none \
--user "$(id -u):$(id -g)" \
-e HOME=/tmp \
"$(cat "$IIDFILE")" \
-v "$ROOT:/work" \
-w /work \
"$IMAGE" \
node tests/e2e/firefox/run.js dist/firefox
}

View File

@@ -11,39 +11,6 @@
let nextId = 1;
const pending = {};
// EIP-1193 ProviderRpcError: `code`, `message`, optional `data`. A class
// rather than properties bolted onto an Error because this object crosses
// no boundary after construction — it is built in the page's own realm and
// handed straight to the caller's catch — so the prototype survives and
// `error.name` is a stable thing for a dApp to see.
class ProviderRpcError extends Error {
constructor(code, message, data) {
super(message);
this.name = "ProviderRpcError";
this.code = code;
if (data !== undefined) this.data = data;
}
}
// Rebuild a boundary error as the error the page catches, carrying the
// code (and data) the extension reported. Without this a dApp cannot tell
// a user's refusal (4001) from a wallet that broke, and retries or shows
// an error instead of accepting the refusal.
//
// Whatever code arrived is passed through verbatim rather than being
// matched against a list: the extension emits 4001, 4100 and 4902 today,
// and a code this file has never heard of is still the truth about what
// happened. An error reported with no code at all stays a plain Error —
// a ProviderRpcError whose `code` is undefined would advertise a
// conformance it does not have. `message` is untouched in every case.
function toPageError(error) {
const message = (error && error.message) || "Request failed";
if (error && error.code !== undefined && error.code !== null) {
return new ProviderRpcError(error.code, message, error.data);
}
return new Error(message);
}
// Listen for responses from the content script
window.addEventListener("message", function onUuid(event) {
if (event.source !== window) return;
@@ -53,7 +20,7 @@
if (!p) return;
delete pending[id];
if (error) {
p.reject(toPageError(error));
p.reject(new Error(error.message || "Request failed"));
} else {
p.resolve(result);
}

View File

@@ -8,19 +8,12 @@
// either verdict alone, because the balance list is where the user forms
// their belief about what they own (issue #235).
//
// KNOWN_SYMBOLS maps a symbol to the set of lowercased contract addresses
// that may bear it, or to null. Null means the symbol belongs to the native
// asset, which has no contract at all, so no contract may bear it and every
// one that does is a spoof. "ETH" is the only such entry today; the rule is
// KNOWN_SYMBOLS maps a symbol to the lowercased contract address that may
// bear it, or to null. Null means the symbol belongs to the native asset,
// which has no contract at all, so no contract may bear it and every one
// that does is a spoof. "ETH" is the only such entry today; the rule is
// written so that a second one needs no change here or at any call site.
//
// The value is a set because a ticker is not unique: seven symbols in the
// bundled list belong to two real contracts each, and answering with one of
// them hid the other one's holders' money (issue #276). Membership, not
// equality, is therefore the question — but it is the same question, asked of
// a table that can now state the truth. Every address in a set is one the
// wallet ships as a real token; a contract outside the set is still a spoof.
//
// The symbol is attacker-controlled — it is whatever the ERC-20 contract
// returns — so the lookup is done on a normalized form (issue #260): the
// question is whether the symbol reaches the user's eye as a known one,
@@ -100,7 +93,7 @@ function isSpoofedSymbol(symbol, contractAddress) {
if (!KNOWN_SYMBOLS.has(sym)) return false;
const legit = KNOWN_SYMBOLS.get(sym);
if (legit === null) return true;
return !legit.has(contract);
return contract !== normalizeAddress(legit);
}
module.exports = {

View File

@@ -3607,33 +3607,14 @@ for (const t of TOKENS) {
TOKEN_BY_ADDRESS.set(t.address.toLowerCase(), t);
}
// Build a map of symbol (uppercased) -> the set of contract addresses
// (lowercased) that legitimately bear it. Used for spoofed-symbol detection.
// "ETH" maps to null: the native asset has no contract, so no contract may
// bear its symbol.
//
// The value is a set and not a single address because tickers are not unique
// and the list above proves it: seven of these 512 tokens share a symbol with
// another entry — FRAX, REUSD, TON, EURE, MSUSD, MUSD and JPYC — at two
// different real contracts each, all of them from the same source fetch. A
// one-address-per-symbol table can only answer that by picking a winner, and
// the loser is then a token in our own bundled list that the spoof filter
// hides from the balance list, the history and the send selector at its own
// address, so the user cannot spend it (issue #276). Naming every address
// that bears the symbol is the only shape that says what is true; it does not
// loosen the rule, because a contract outside the set is still a spoof.
// Build a map of symbol (uppercased) -> legitimate contract address (lowercased).
// Used for spoofed-symbol detection. "ETH" maps to null (native token).
const KNOWN_SYMBOLS = new Map();
KNOWN_SYMBOLS.set("ETH", null);
for (const t of TOKENS) {
const upper = t.symbol.toUpperCase();
if (!KNOWN_SYMBOLS.has(upper)) {
KNOWN_SYMBOLS.set(upper, new Set());
}
const addresses = KNOWN_SYMBOLS.get(upper);
// A null entry is the native asset and stays null: an ERC-20 that reports
// the native symbol does not thereby become entitled to it.
if (addresses !== null) {
addresses.add(t.address.toLowerCase());
KNOWN_SYMBOLS.set(upper, t.address.toLowerCase());
}
}

View File

@@ -1,34 +0,0 @@
# Chrome end-to-end image: the pinned Playwright image with this repo and a
# freshly built extension inside it, built by script/test-e2e. The suite is
# still started with `docker run`, so every runtime flag the harness needs
# (--ipc=host in particular) applies as before.
#
# The repo is baked in rather than bind-mounted because a bind mount does
# not resolve under Gitea Actions: the runner runs the job in a container
# against the HOST's docker socket, so the source side of a -v is resolved
# by the host daemon while the job's checkout lives on a docker volume that
# is not a host path -- the mount silently succeeds and /work is empty. A
# build context is streamed to the daemon and so works from anywhere.
# Building the extension here too means the machine starting a run needs
# docker and nothing else.
# mcr.microsoft.com/playwright:v1.56.0-noble, 2026-08-09
#
# The playwright-core devDependency is pinned to the matching Playwright
# version (1.56.0) and the two must be bumped together: the browsers ship
# inside this image, and playwright-core looks for the exact browser
# revision its own version expects. A mismatch fails at launch.
FROM mcr.microsoft.com/playwright@sha256:35246d87a7c88ea9b771c65d33171b2611b02a8253b4b12ce6f94376c55f99f2
WORKDIR /work
# Same layering as the root Dockerfile: script/bootstrap installs the
# prerequisites and the dependencies, and the manifests are copied first so
# that layer is cached until they change.
COPY script/ script/
COPY package.json yarn.lock ./
RUN script/bootstrap
COPY . .
RUN make build

View File

@@ -1,24 +1,10 @@
# Firefox end-to-end image: stock Firefox plus geckodriver on a node base,
# with this repo and a freshly built extension inside it, built by
# script/test-e2e-firefox. The harness itself has no dependencies, so
# nothing is installed for it.
# built by script/test-e2e-firefox. The repo is bind-mounted at /work; the
# harness itself has no dependencies, so nothing is installed for it.
#
# The build context is the repo root. The repo is baked in rather than
# bind-mounted because a bind mount does not resolve under Gitea Actions:
# the runner runs the job in a container against the HOST's docker socket,
# so the source side of a -v is resolved by the host daemon while the job's
# checkout lives on a docker volume that is not a host path -- the mount
# silently succeeds and /work is empty. Baking the build in is also the
# only way this suite can have both a built extension and the
# `--network none` it runs under, since a container with no network cannot
# install anything.
#
# All three external artifacts are pinned by digest, and are fetched in
# layers above the repo copy, so editing the harness or any source file
# re-runs only the two cheap layers at the bottom. The Firefox version in
# particular must not float: -remote-allow-system-access is mandatory on
# 153 and was not on 142, so the flag the harness passes is
# version-coupled.
# All three external artifacts are pinned by digest. The Firefox version in
# particular must not float: -remote-allow-system-access is mandatory on 153
# and was not on 142, so the flag the harness passes is version-coupled.
# node:22-bookworm-slim, 2026-08-12
FROM node@sha256:d649c27dae7ba0137b3cef5dd75baa422c08dc3d9e3fc0c23dfb172dc3cc6436
@@ -62,16 +48,4 @@ ENV FIREFOX_BIN=/opt/firefox/firefox
ENV GECKODRIVER=/usr/local/bin/geckodriver
WORKDIR /work
# Same layering as the root Dockerfile: script/bootstrap installs the
# prerequisites and the dependencies, and the manifests are copied first so
# that layer is cached until they change.
COPY script/ script/
COPY package.json yarn.lock ./
RUN script/bootstrap
COPY . .
RUN make build
CMD ["node", "tests/e2e/firefox/run.js", "dist/firefox"]

View File

@@ -86,11 +86,9 @@ const DAPP_URL = DAPP_ORIGIN + "/";
// never drive the popup that has to settle it; start() files the promise
// under a key and settle() collects it once the prompt has been dealt with.
//
// The rejection branch records the whole observable shape of the error as it
// arrives — name, message, and whether a `code` is present at all as distinct
// from its value. EIP-1193 says a user rejection is a ProviderRpcError
// carrying code 4001; what the page can actually see is recorded here rather
// than assumed, and asserted in run.js.
// The rejection branch records `code` as it arrives. EIP-1193 says a user
// rejection is a ProviderRpcError carrying code 4001; what the page can
// actually see is recorded here rather than assumed, and asserted in run.js.
//
// The message log is the page's half of the boundary observation: every
// AUTISTMASK_* message that crosses between this page and the content
@@ -122,7 +120,6 @@ const DAPP_HTML = [
" return {",
" settled: 'rejected',",
" message: String((error && error.message) || error),",
" name: error ? error.name : undefined,",
" hasCode: !!error && 'code' in Object(error),",
" code: error ? error.code : undefined,",
" };",

View File

@@ -1757,14 +1757,15 @@ async function lastResponseError(page) {
}
// A rejected prompt, asserted at both ends: the page's promise rejected
// rather than hanging or resolving, and EIP-1193 code 4001 is present both
// on the wire and on the Error the calling page catches.
// rather than hanging or resolving, and the response that crossed the
// boundary carried EIP-1193 code 4001.
//
// Both ends matter because they used to disagree. The code crossed the
// boundary correctly and src/content/inpage.js then threw it away, rebuilding
// every rejection as `new Error(error.message)` — so a dApp branching on
// `err.code === 4001` saw undefined and could not tell a refusal from a
// failure (#274). Asserting only the wire would leave that gap invisible.
// The code is asserted on the wire because that is the only place it
// survives. src/content/inpage.js rebuilds the rejection as `new
// Error(error.message)`, so the Error the calling page catches carries the
// message and no code. That is reported rather than asserted either way —
// locking in the current behaviour would make the gap permanent, and
// asserting the code on the Error would fail today.
async function assertUserRejection(page, key, label) {
const outcome = await settleRequest(page, key);
assert(
@@ -1786,32 +1787,15 @@ async function assertUserRejection(page, key, label) {
" did not carry EIP-1193 code 4001 across the boundary: " +
JSON.stringify(error),
);
assert(
outcome.hasCode,
label +
" reached the page as an error with no code property at all, so a " +
"dApp cannot tell the user's refusal from a failure: " +
JSON.stringify(outcome),
);
assert(
outcome.code === 4001,
label +
" reached the page with code " +
JSON.stringify(outcome.code) +
" rather than EIP-1193 4001",
);
assert(
outcome.name === "ProviderRpcError",
label +
" reached the page as " +
JSON.stringify(outcome.name) +
" rather than an EIP-1193 ProviderRpcError",
);
console.log(
"# " +
label +
": code 4001 on the wire and on the page's " +
outcome.name,
": boundary code=" +
error.code +
" page Error.code=" +
JSON.stringify(outcome.code) +
" page Error carries a code=" +
outcome.hasCode,
);
return outcome;
}

View File

@@ -1,310 +0,0 @@
// The EIP-1193 error the page actually catches (src/content/inpage.js).
//
// The bug this pins down (issue #274): the provider rebuilt every failure as
// `new Error(error.message)`, so the `code` the background produced and the
// content script relayed intact was thrown away in the last hop. A dApp
// checking `err.code === 4001` — the standard way to tell "the user said no"
// from "the wallet broke" — saw undefined, and well-behaved sites showed an
// error or retried instead of accepting the refusal.
//
// inpage.js is a bare IIFE injected into the page's JS context, not a module:
// it takes no import and exports nothing, and reaches for `window` at load.
// So it is evaluated here the way the browser evaluates it, against a stub
// window, and the provider is collected from `window.ethereum`. The globals it
// touches are passed in as function parameters rather than assigned to
// globalThis: nothing leaks between tests, and the source is compiled in this
// realm, so the errors it constructs are comparable against this file's own
// `Error` — which a second realm's intrinsics would silently defeat.
//
// There is no jsdom in this repo; see tests/txStatus.test.js.
const fs = require("fs");
const path = require("path");
const { webcrypto } = require("crypto");
const SOURCE = fs.readFileSync(
path.join(__dirname, "..", "src", "content", "inpage.js"),
"utf8",
);
const loadInto = new Function(
"window",
"self",
"crypto",
"Event",
"CustomEvent",
SOURCE,
);
class StubEvent {
constructor(type) {
this.type = type;
}
}
class StubCustomEvent extends StubEvent {
constructor(type, init) {
super(type);
this.detail = init && init.detail;
}
}
// Every code the background emits on the RPC path today, read out of
// src/background/index.js. The provider must not know this list — it passes
// through whatever arrived — but the cases below are the real ones.
const REJECTED = 4001; // user rejected the request
const UNAUTHORIZED = 4100; // site not connected / wrong address
const UNRECOGNIZED_CHAIN = 4902; // switch/add to an unsupported chain
// A stub window with the four things inpage.js touches: message listeners,
// postMessage out to the content script, window.ethereum, and dispatchEvent
// for the EIP-6963 announcement.
function loadProvider() {
const messageListeners = [];
const posted = [];
const win = {
addEventListener(type, fn) {
if (type === "message") messageListeners.push(fn);
},
removeEventListener(type, fn) {
const i = messageListeners.indexOf(fn);
if (type === "message" && i !== -1) messageListeners.splice(i, 1);
},
postMessage(data) {
posted.push(data);
},
dispatchEvent() {
return true;
},
};
win.window = win;
loadInto(win, win, webcrypto, StubEvent, StubCustomEvent);
// Deliver the content script's answer to an outstanding request. The id is
// read back off the wire rather than assumed: inpage.js issues its own
// eth_chainId at load, so the first id a test sees is not 1.
function respond(response) {
const request = posted
.filter((m) => m.type === "AUTISTMASK_REQUEST")
.pop();
expect(request).toBeDefined();
const event = {
source: win,
data: { type: "AUTISTMASK_RESPONSE", id: request.id, ...response },
};
for (const fn of messageListeners.slice()) fn(event);
}
return { provider: win.ethereum, posted, respond };
}
// Start a request, answer it with `response`, and hand back the rejection.
// Fails the test if the call resolves instead.
async function rejectionFrom(start, response) {
const { provider, respond } = loadProvider();
const settled = start(provider).then(
(result) => ({ resolved: result }),
(error) => ({ error }),
);
// The provider posts synchronously, so the request is already on the wire.
respond(response);
const outcome = await settled;
expect(outcome).not.toHaveProperty("resolved");
return outcome.error;
}
describe("an EIP-1193 code reaches the page", () => {
test("a user rejection arrives as code 4001", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "eth_requestAccounts" }),
{
error: {
code: REJECTED,
message: "User rejected the request.",
},
},
);
expect(err.code).toBe(REJECTED);
expect(err.message).toBe("User rejected the request.");
});
test("it is a ProviderRpcError, and an Error", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "eth_requestAccounts" }),
{
error: {
code: REJECTED,
message: "User rejected the request.",
},
},
);
expect(err).toBeInstanceOf(Error);
expect(err.name).toBe("ProviderRpcError");
});
test("4100 unauthorized arrives intact", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "personal_sign", params: ["0x00"] }),
{ error: { code: UNAUTHORIZED, message: "Unauthorized" } },
);
expect(err.code).toBe(UNAUTHORIZED);
expect(err.message).toBe("Unauthorized");
});
test("4902 unrecognized chain arrives intact", async () => {
const message =
"AutistMask supports Ethereum Mainnet and Sepolia Testnet only.";
const err = await rejectionFrom(
(p) => p.request({ method: "wallet_switchEthereumChain" }),
{ error: { code: UNRECOGNIZED_CHAIN, message } },
);
expect(err.code).toBe(UNRECOGNIZED_CHAIN);
expect(err.message).toBe(message);
});
// The provider is not allowed to know the list above: a code added to the
// background later must reach the page without this file being edited.
test("a code the provider has never heard of is passed through", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "eth_accounts" }),
{ error: { code: 4900, message: "Disconnected" } },
);
expect(err.code).toBe(4900);
});
test("data is carried when the boundary sent it", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "eth_call" }),
{
error: {
code: -32000,
message: "execution reverted",
data: "0x08c379a0",
},
},
);
expect(err.code).toBe(-32000);
expect(err.data).toBe("0x08c379a0");
});
test("no data property is invented when the boundary sent none", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "eth_requestAccounts" }),
{
error: {
code: REJECTED,
message: "User rejected the request.",
},
},
);
expect("data" in err).toBe(false);
});
});
describe("the message is untouched", () => {
test("a coded error keeps the message byte for byte", async () => {
const message =
"This site asked to sign as an address that is not " +
"the active one.";
const err = await rejectionFrom(
(p) => p.request({ method: "personal_sign" }),
{ error: { code: UNAUTHORIZED, message } },
);
expect(err.message).toBe(message);
});
test("an error the background sent with no code keeps its message", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "eth_sendTransaction" }),
{ error: { message: "No accounts available" } },
);
expect(err.message).toBe("No accounts available");
});
// A ProviderRpcError whose code is undefined would claim a conformance it
// does not have, and `'code' in err` is exactly what a careful dApp asks.
test("an error with no code gets no code property at all", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "eth_sendTransaction" }),
{ error: { message: "No accounts available" } },
);
expect(err).toBeInstanceOf(Error);
expect("code" in err).toBe(false);
});
test("an error with no message keeps the generic fallback", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "eth_sendTransaction" }),
{ error: { code: REJECTED } },
);
expect(err.message).toBe("Request failed");
expect(err.code).toBe(REJECTED);
});
});
// Every entry point the provider exposes, not just eth_requestAccounts. They
// all funnel through the same response listener, and this is what says so.
describe("every request path carries the code", () => {
const rejection = {
error: { code: REJECTED, message: "User rejected the request." },
};
test("request()", async () => {
const err = await rejectionFrom(
(p) => p.request({ method: "eth_requestAccounts" }),
rejection,
);
expect(err.code).toBe(REJECTED);
});
test("enable()", async () => {
const err = await rejectionFrom((p) => p.enable(), rejection);
expect(err.code).toBe(REJECTED);
});
test("send(method, params)", async () => {
const err = await rejectionFrom(
(p) => p.send("eth_requestAccounts", []),
rejection,
);
expect(err.code).toBe(REJECTED);
});
test("send({ method, params })", async () => {
const err = await rejectionFrom(
(p) => p.send({ method: "personal_sign", params: ["0x00"] }),
rejection,
);
expect(err.code).toBe(REJECTED);
});
test("sendAsync() hands the code to its callback", async () => {
const { provider, respond } = loadProvider();
const called = new Promise((resolve) => {
provider.sendAsync({ id: 1, method: "eth_requestAccounts" }, (e) =>
resolve(e),
);
});
respond(rejection);
const err = await called;
expect(err.name).toBe("ProviderRpcError");
expect(err.code).toBe(REJECTED);
expect(err.message).toBe("User rejected the request.");
});
});
describe("the success path is unchanged", () => {
test("a result still resolves", async () => {
const { provider, respond } = loadProvider();
const settled = provider.request({ method: "eth_requestAccounts" });
respond({ result: ["0xb61264DEFB0c4B8afb3D73724be15310036743a5"] });
await expect(settled).resolves.toEqual([
"0xb61264DEFB0c4B8afb3D73724be15310036743a5",
]);
expect(provider.selectedAddress).toBe(
"0xb61264DEFB0c4B8afb3D73724be15310036743a5",
);
});
});

View File

@@ -56,7 +56,7 @@ global.chrome = {
};
const { isSpoofedSymbol } = require("../src/shared/symbolSpoof");
const { TOKENS, KNOWN_SYMBOLS } = require("../src/shared/tokenList");
const { KNOWN_SYMBOLS } = require("../src/shared/tokenList");
const { filterTransactions } = require("../src/shared/transactions");
const {
fetchTokenBalances,
@@ -284,137 +284,11 @@ describe("the shared rule: symbols that render as a known symbol", () => {
// asserts the claim it stands for — `[ -~]` would admit an interior
// space and let a whitespace-bearing entry through the guard.
test("no bundled symbol is touched by the normalization", () => {
for (const [symbol, addresses] of KNOWN_SYMBOLS) {
for (const [symbol, address] of KNOWN_SYMBOLS) {
expect(symbol).toBe(symbol.trim());
expect(symbol).toMatch(/^[!-~]+$/);
if (addresses === null) continue;
for (const address of addresses) {
expect(isSpoofedSymbol(symbol, address)).toBe(false);
}
}
});
});
// Issue #276: the guard that was missing. The suite walked KNOWN_SYMBOLS,
// which is built from TOKENS, so it could only ever assert that the table
// agrees with itself. Seven symbols appear twice in the bundled list at two
// different real contracts, and the table kept whichever came first, so the
// other seven contracts — tokens in our own shipped list, at their own
// addresses — were judged spoofs and hidden from the balance list, the
// history and the send selector. That is the over-filtering direction: it
// hides a holding the user cannot then spend.
//
// This walk is over TOKENS, the data the wallet actually ships, so it fails
// whenever a bundled token would be filtered at its own address no matter
// which side of the table the mistake is on.
describe("the shipped token list", () => {
test("no bundled token is filtered at its own address", () => {
const filtered = TOKENS.filter((t) =>
isSpoofedSymbol(t.symbol, t.address),
).map((t) => t.symbol + " @ " + t.address);
expect(filtered).toEqual([]);
});
// The third failure mode the issue asks about: a symbol whose table entry
// names an address that is in neither the table nor the list would be a
// contract we vouch for and do not ship. There is none, and the table is
// built from the list, so this asserts the derivation has not acquired a
// hand-written entry.
test("every address the table vouches for is a bundled token", () => {
const bundled = new Set(TOKENS.map((t) => t.address.toLowerCase()));
for (const [symbol, addresses] of KNOWN_SYMBOLS) {
if (addresses === null) continue;
expect(addresses.size).toBeGreaterThan(0);
for (const address of addresses) {
expect(address).toBe(address.toLowerCase());
expect(bundled.has(address)).toBe(true);
// And it is the token that actually reports that symbol.
const token = TOKENS.find(
(t) => t.address.toLowerCase() === address,
);
expect(token.symbol.toUpperCase()).toBe(symbol);
}
}
});
// Both contracts behind a shared ticker must pass, from either side: a
// rule that admits only the one the table happens to visit first is the
// bug, not the fix.
test("both contracts behind a shared ticker are admitted", () => {
const bySymbol = new Map();
for (const t of TOKENS) {
const upper = t.symbol.toUpperCase();
if (!bySymbol.has(upper)) bySymbol.set(upper, []);
bySymbol.get(upper).push(t);
}
const shared = [...bySymbol].filter(([, list]) => list.length > 1);
// The shared tickers are a fact about the shipped data; if a future
// list has none, this test would silently assert nothing.
expect(shared.length).toBeGreaterThan(0);
for (const [, list] of shared) {
for (const t of list) {
expect(isSpoofedSymbol(t.symbol, t.address)).toBe(false);
}
}
});
// The seven from issue #276, named so that the reconciliation is a fact
// in the suite: each is two real contracts from the same source fetch,
// and the table now holds both rather than the one that came first.
test("the seven shared tickers each name both bundled contracts", () => {
const expected = {
TON: [
"0x582d872a1b094fc48f5de31d3b73f2d9be47def1", // Toncoin
"0x2be5e8c109e2197d077d13a82daead6a9b3433c5", // Tokamak Network
],
FRAX: [
"0x853d955acef822db058eb8505911ed77f175b99e", // Legacy Frax Dollar
"0x3432b6a60d23ca0dfca7761b7ab56459d9c964d0", // Frax (prev. FXS)
],
REUSD: [
"0x5086bf358635b81d8c47c66d1c8b9e567db70c72", // Re Protocol reUSD
"0x57ab1e0003f623289cd798b1824be09a793e4bec", // Resupply USD
],
EURE: [
"0x39b8b6385416f4ca36a20319f70d28621895279d", // Monerium EUR emoney
"0x3231cb76718cdef2155fc47b5286d82e6eda273f", // Monerium EUR emoney [OLD]
],
MSUSD: [
"0x4ba01f22827018b4772cd326c7627fb4956a7c00", // Main Street USD
"0xab5eb14c09d416f0ac63661e57edb7aecdb9befa", // Metronome Synth USD
],
MUSD: [
"0xaca92e438df0b2401ff60da7e4337b687a2435da", // MetaMask USD
"0xdd468a1ddc392dcdbef6db6e34e89aa338f9f186", // Mezo USD
],
JPYC: [
"0x431d5dff03120afa4bdf332c61a6e1766ef37bdb", // JPY Coin
"0x2370f9d504c7a6e775bf6e14b3f12846b594cd53", // JPY Coin v1
],
};
for (const [symbol, addresses] of Object.entries(expected)) {
expect([...KNOWN_SYMBOLS.get(symbol)].sort()).toEqual(
[...addresses].sort(),
);
for (const address of addresses) {
expect(isSpoofedSymbol(symbol, address)).toBe(false);
}
}
});
// The other direction, on the same symbols: widening the table to hold
// every bundled address for a ticker must not turn it into a pass for
// any other contract.
test("a shared ticker from a third contract is still a spoof", () => {
const bySymbol = new Map();
for (const t of TOKENS) {
const upper = t.symbol.toUpperCase();
if (!bySymbol.has(upper)) bySymbol.set(upper, []);
bySymbol.get(upper).push(t);
}
for (const [symbol, list] of bySymbol) {
if (list.length < 2) continue;
expect(isSpoofedSymbol(symbol, FAKE_ETH_CONTRACT)).toBe(true);
if (address === null) continue;
expect(isSpoofedSymbol(symbol, address)).toBe(false);
}
});
});

View File

@@ -207,8 +207,8 @@ describe("token list assumptions the fixtures rely on", () => {
});
test("USDC and WETH map to their genuine lowercased contracts", () => {
expect([...KNOWN_SYMBOLS.get("USDC")]).toEqual([USDC_CONTRACT]);
expect([...KNOWN_SYMBOLS.get("WETH")]).toEqual([WETH_CONTRACT]);
expect(KNOWN_SYMBOLS.get("USDC")).toBe(USDC_CONTRACT);
expect(KNOWN_SYMBOLS.get("WETH")).toBe(WETH_CONTRACT);
});
test("the spam fixture symbol is not in the known token list", () => {