If the initial network when the extension is started is something other
than Mainnet, the swaps controller will never successfully retrieve
swap quotes. This is because the `ethers` provider used by the swaps
controller doesn't allow network changes by default - it assumes that
the network remains the same as when the provider was initialized.
This was fixed by hard-coding Mainnet as the initial chain ID for this
`ethers` provider used by the swaps controller.
Some adjustments needed to be made to the `provider` stub to allow
setting `1` as the network ID and chain ID in unit tests.
This reduces the footprint of each Node process in an attempt to try
and lower the failure rate.
> Expected behavior with –max-old-space-size within container constraints
>
> By default, Node.js (up to 11.x) uses a maximum heap size of 700MB and 1400MB
> on 32-bit and 64-bit platforms, respectively. For current defaults, see the
> reference mentioned at the end of blog.
[1]:https://developer.ibm.com/languages/node-js/articles/nodejs-memory-management-in-container-environments/
`ganache-core` and `ganache-cli` have been updated to the latest
published versions.
Two Yarn resolutions have been made unnecessary by this update, so they
have been removed. They were added to update dependencies of
`ganache-core` to address security advisories. They have since been
updated in the latest `ganache-core` release.
If the attempt to fetch quotes fails, an error message is now logged to
the console explaining why it failed. This makes it dramatically easier
to understand what's happening when quotes fail.
At some point later on we should also look specifically for errors we
expect to occur, and report to Sentry in unexpected cases. I've added a
TODO comment about this.
* origin/develop: (57 commits)
Remove unused parameter in styles build script (#9710)
Fix `yarn build styles:dev` (#9709)
Update main-quote-summary designs/styles (#9612)
@metamask/test-dapp@3.2.0 (#9707)
Add ses lockdown to build system (#9568)
Robustify waiting logic in e2e test (#9704)
Prevent React error for close
Prevent memory leak from selected account copy tooltip
Lint
Clean up events
Make the dropdown widgets for swaps keyboard accessible
Fix mocha/max-top-level-suites issues (#9699)
Provide image sizing so there's no jump when opening the swaps token search
Bump @metamask/controllers from 3.1.0 to 3.2.0 (#9692)
Fix pull request template location
Bump @metamask/inpage-provider from 6.1.0 to 6.3.0 (#9691)
Fix 9632 - Prevent old fetches from polluting the swap state
Remove broken Storybook stories (#9690)
Add a GitHub Dependabot config (#9664)
Fix PropType error on Awaiting Swap page (#9688)
...
The `devMode` parameter being passed to the `buildScss` function was
not being used. The `buildScss` function was declared _inside_ the
function in which it is invoked, so the `devMode` variable is already
in scope - it doesn't need to be passed in.
The `styles:dev` step of the development build was broken in #9568 when
the `lockdown()` function from SES was introduced. One of the effects
of `lockdown()` is preventing the use of `RegExp.lastMatch`, which is
relied upon by `postcss` for parsing inline sourcemaps.
This problem has been worked around by re-arranging the build pipeline
for styles.
Firstly, the `autoprefixer` stage was moved _before_ sourcemaps were
generated. Its placement _after_ sourcemaps was a mistake in the first
place - it should have always been before they were written, so that
the sourcemaps could keep track of the changes made by `autoprefixer`.
Second, the `-rtl` sourcemap generation was moved to a separate build
pipeline. The `rtl` stage also uses `postcss`, and it must come after
the sourcemaps are written for the non-RTL styles, so splitting it into
a separate build was the only way to avoid passing in inline
sourcemaps. This does make the build slightly slower, but it also
decreases the size of the RTL sourcemaps dramatically.
* Update main-quote-summary designs/styles
* Clean up css: use className instead of element types
* Style fixes to symbol elements in main-quote-view
* Use correct source for token iconUrls passed to main-quote-view
* Improve vertical spacing on view-quote screen and with new main-quote-view designs
* Remove unused classes
* Tweak space around large quote amount text in main-quote-summary
* Add ses lockdown to build system using lavamoat-core
* use proper object.assign version
* disable lint rules for ses lockdown
* deps - update rtlcss
Co-authored-by: kumavis <aaron@kumavis.me>
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
These two stories showcased components that used React context and
Redux state that was not made available in the stories themselves. They
both crashed with an error when rendered in Storybook.
Instead of going through a great deal of effort to setup mocks for the
global Redux state and context, they have been removed. They are
perhaps the wrong layer for us to showcase in Storybook. Storybook is
more well suited for individual components that contain just UI logic.
This change adds a GitHub Dependabot configuration to enable daily checks for dependency updates.
[See the docs for more information.][1]
[1]:https://docs.github.com/en/free-pro-team@latest/github/administering-a-repository/configuration-options-for-dependency-updates
This config enables checking for dependencies in the root `package.json` file (`directory`) every weekday (`schedule.interval`) only for lockfile updates, ignoring any new versions that would require package manifest changes. (if necessary, `versioning-strategy`). This is all restricted to the organization's `@metamask/*` packages.
A PropType error was shown in the console when on the Awaiting Swap
page in certain error scenarios. The `symbol` property was sometimes
not set if `destinationTokenInfo` was missing.
The `symbol` prop has been removed, as the `AwaitingSwap` component
already selected the `fetchParams`, and the `destinationTokenInfo`
object within. The prop was effectively a duplicate.
The e2e test for the contract deposit action was unnecessarily reliant
upon timing. After initiating a deposit, it would grab the first
transaction in the transaction list and assume it was the deposit that
it had just initiated. If it looked prior to the unapproved transaction
being added to the list, it would grab the wrong transaction.
It now looks specifically for _unconfirmed_ transactions, meaning it
will block until the deposit transaction is rendered.
This was discovered in testing a test-dapp PR:
https://github.com/MetaMask/test-dapp/pull/76
Refs #9663
See [`node/no-unsupported-features/node-builtins`][1] for more information.
This change enables `node/no-unsupported-features/node-builtins` and fixes the issue
raised by the rule.
The `engines.node` version is updated to reflect the version specified by the `.nvmrc`
file and the version used by CircleCI:
```bash
$ cat .nvmrc
v10.18.1
```
```bash
$ docker run --rm -it circleci/node@sha256:e16740707de2ebed45c05d507f33ef204902349c7356d720610b5ec6a35d3d88 node --version
v10.18.1
```
Refs #9663
See [`node/no-deprecated-api`][1] for more information.
This change enables `node/no-deprecated-api` and fixes the issues raised by the rule.
[1]:https://github.com/mysticatea/eslint-plugin-node/blob/v11.1.0/docs/rules/no-deprecated-api.md
The change to the way that `punycode` is imported is to address the fact that
third-party module is hidden by the built-in. This is a silly hack but it works.
Our ENS resolver for the browser address bar was incorrectly resolving
addresses that included query strings. We were concatenating the `path`
property with the `search` property, despite the fact that the `path`
property already contains `search`. As a result, `search` was
duplicated in the resolved addresses.
For example, if an IPFS content ID was found for this address, the
resolved address for `metamask.eth/?foo=bar` would have the path
`/?foo=bar?foo=bar`
The original intent was likely to use `pathname` in place of `path`.
The resolver has been updated to use `pathname`, and the query string
now appears only once in the resolved address.
Consolidates the background and UI segment implementations into a shared solution.
This results in the introduction of our first shared module.
Co-authored-by: Erik Marks <25517051+rekmarks@users.noreply.github.com>
