Two tabs have been created on the home screen: 'Assets' and 'History'.
This tabbed view is shown only on small screens (e.g. in the popup).
The fullscreen view is unchanged.
The toggle-able left sidebar no longer exists, so some 'sidebar-left'
specific code and styles have been removed. The button in the menu bar
has been removed as well.
The 'History' title of the transaction history is now redundant when
where are no pending transactions, so it as been conditionally hidden.
A passthrough for `data-testid` has been added to the Tab component for
convenience in e2e tests.
This commit updates the existing _Connected Sites_ section to a modal using
the `Popover` component. This will serve as a base for the new modal design.
The user-specified seed phrase during the first-time-flow import step
required the phrase to be entered in all lowercase. The case does not
add any extra entropy to the seed, so there's no reason to be case
sensitive. Flexibility here will improve the onboarding UX.
This commit makes the entered seed phrase case-insensitive.
Fixes#8171
The "i18n-provider" module has been replaced by a new `i18n.js` module
in the `contexts` directory which provides the `t` function via the new
React Context API.
The legacy context API is still used throughout the codebase, so a
legacy context provider has also been added as a shim until we migrate
away from the old API. The migration does require changing every single
place where the `t` function is used, so it is a non-trivial amount of
work. This shim allows us to tackle it one piece at a time without
breaking anything.
This was placed in a new `contexts` directory because it didn't seem
to belong in any existing categories. It certainly isn't a higher-order
component.
The translation helper function we use everywhere (`t`) would fallback
to the message `[${key}]` for any key not found in the set of localized
messages. Instead it how returns `undefined` in that case.
`[${key}]` isn't something you'd typically want to show to users, so
this fallback wasn't overly useful in practice. At best it served to
hide errors.
The falsey return value in the case where the message is missing makes
it easier to implement a fallback for that case. Such a fallback is
used in the `confirm-transaction-base` component, to restore the
fallback behavior accidentally changed in #8211
These two functions were not especially useful. `tOrDefault` was used
only by `tOrKey`, and `tOrKey` was only used in one place. All it did
was return the given `key` directly if it was falsey, so it was easily
replaced by a condition.
Previously a few mostly-empty `div`s would be shown if a render
happened while the confirm page was loading. Now nothing is shown. This
shouldn't impact users at all, as this condition should only last a
fraction of a second.
The address is blank momentarily when navigating to the confirmation
screen when sending a token. The address is updated in a subsequent
render.
The ENS reverse resolution is now only attempted if the address is
given. It has also been updated to attempt resolution when the address
is finally set, which fixes the reverse resolution for token sends.
I had hoped to get rid of this interim render-without-address, but that
turned out to be a bit more challenging. The problem is that the UI
submits transactions through the provider just as a dapp would, and the
provider doesn't say when the transaction is submitted. The promise
returned doesn't resolve until after confirmation. We would have to
start calling the background methods directly and bypass the provider
to get the feedback we need, and that sounded potentially dangerous.
Definitely a challenge for another day.
The `metamask.send.from` field was assumed by various selectors to be
an object, but instead it was recently set to a string. The selectors
have been updated to assume it's a string, and to fetch the full
account object explicitly.
The selector `getSendFromObject` was repurposed for this, as that's
basically what it already did. The optional address parameter was
removed though, as that was only used to fetch the `from` address in
cases where the `send` state was set without there being a `from`
address set. That case is no longer possible, as the `from` address is
always set upon the initialization of the `send` state.
The `getSendFromObject` selector no longer fetches the 'name' of that
address from the address book state either. This property was not used
in either of the cases this selector was used.
* Update i18n-helper to allow substitutions of react components and wrapping of translation substrings
* Simplify code in i18n-helper.js related to substitutions, including react substitutions.
* Remove wrapper support from i18n in favour of using translations in substitutions.
* Fix i18n-helper substitution logic: ensure correct index of substitution is applied
* Throw error if there are not enough substitutions for a translation phrase
* Adds unit tests for now i18n-helper substitution functionality
* Fix grammar, react element line spacing and test layout+readability in i18n-helper.test.js
The "Transaction View" component has been merged with the Home
component. The division between these two components seemed wrong
because the "Transaction View" contained the menu bar (distinctly a
"home" thing, not a "transaction" thing), and we will be adding more
non-transaction-related components shortly.
This also let us use a single `Media` component instead of two.
The notifications displayed on the home screen were being passed
through the `TransactionView` and `TransactionList` components before
being rendered. This was unnecessary because the notifications are
absolutely positioned.
They are now rendered directly in the home component where they're
defined. A helper function has been written to improve readability.
The styles used for the Home component were in the huge
"newui-sections" SCSS file. Instead they've been moved into an SCSS
module alongside the component, to follow our conventions.
The `main-container` class was left as-is because it is shared between
here and the settings page.
Implement `eth_decrypt` and `eth_getEncryptionPublicKey`. This allows decryption backed by the user's private key. The message decryption uses a confirmation flow similar to the messaging signing flow, where the message to be decrypted is also able to be decrypted inline for the user to read directly before confirming.
ENS currently supports a variety of tlds in addition to `.eth`, and
more will be supported in the future. Rather than hard-code a list of
supported ENS tlds, all valid domain names will now be interpreted as
potential ENS addresses in our address input component.
Closes#7978
The QR scanner component error handling would sometimes redirect the
user to the wrong page. It was also confusingly handled in two places;
the action used to open the QR scanner, and the scanner component.
The error handling has now been corrected, simplified, and integrated
into the QR scanner component itself.
The old way of handling an error within the component was to close the
modal, then call the action to open it all over again. This action took
a route parameter, which instructed the action on which route to open
if the fullscreen UI needed to be opened (as the fullscreen UI is the
only one where the browser will show the camera permission prompt).
This redirection worked just fine for handling the initial opening
of the QR scanner modal. But for any subsequent errors the same action
was used with a _default route_, meaning the user could click "try
again" and find themselves on a completely different screen.
Instead, errors now trigger a state change instead of closing and re-
opening the modal. The permission checks in the action have been
integrated into the component as well.
One functional change is that the scenario where you have an invalid
QR code has been made an error. Previously this just showed the error
message below the video preview, but left the user with no way to try
again. There error page has a "Try again" button, so it seemed better
suited as an error. Also the message literally started with "Error:".
Another functional change is that _all_ errors during initialization
will result in the error UI being shown. Previously there was one error
case that would instead log to the console and leave the user stuck.
* Use @metamask/eslint-config@1.1.0
* Use eslint-plugin-mocha@6.2.2
* Mark root ESLint config as root
* Update Mocha ESLint rules with shared ESLint config
* Various component tests and some conditional statements
Conditional in account-menu in removeAccount when keyring sometimes is not initially provideed
Conditional on unlock-page when there is no target.getBoundingClientRect on the element.
* Update helpers
* Remove component debugging
* Add default params for render helpers
* Remove stubComponent for old Mascot
Changes in https://github.com/MetaMask/metamask-extension/pull/7893 has prevented the need to stub it out.
Change logout to lock in account-menu test
The custom spend limit was previously not validated. It did have a
minimum of zero set, but this didn't have any affect (that minimum is
used for form constraint validation, and this field wasn't in a form).
The field was never checked to ensure the contents didn't exceed the
maximum.
The field is now checked for values that exceed the maximum, and
invalid values in general (including negative values).
The parameters to the `showEditApprovalPermissionModal` were also
alphabetized to make them easier to read. In the course of doing this,
I noticed that the origin was missing from one of the calls. This was
responsible for the modal saying "Spend limit requested by undefined"
when clicking "Edit" under the transaction details. This has been
fixed.
These two functions differ slightly in options, but none of those
options are being used by us, so in these cases they're functionally
equivalent. They're even both descendants of the original `debounce`
function from `underscore`.
This was done to reduce the number of direct dependencies we have. It
should not affect bundle size, as we still depend upon the `debounce`
package transitively.
In the case where the initial spend limit for the `approve` function
was set to the maximum amount, editing this value would result in the
new limit being silently ignored. The transaction would be submitted
with the original max spend limit.
This occurred because function to generate the new custom data would
look for the expected spend limit in the existing data, then bail if
it was not found (and in these cases, it was never found).
The reason the value was not found is that it was erroneously being
converted to a `Number`. A JavaScript `Number` is not precise enough to
represent larger spend limits, so it would give the wrong hex value
(after rounding had taken place in the conversion to a floating-point
number).
The data string is now updated without relying upon the original token
value; the new value is inserted after the `spender` argument instead,
as the remainder of the `data` string is guaranteed to be the original
limit. Additionally, the conversion to a `Number` is now omitted so
that the custom spend limit is encoded correctly.
Fixes#7915
After updating the custom spend limit on the approve screen, the data
for the transaction was not being updated. Instead it showed the
original transaction data. The transaction data was being updated
correctly in the final transaction though.
The approve screen has been updated to ensure changes to the custom
spend limit are reflected correctly in the data shown.
* Update lodash
All versions of the full `lodash` package have been updated to 4.17.15.
The only exception is v4.17.14 which is pinned by `ganache-core`.
* Switch to using `lodash` instead of per-method packages
We have the full lodash package _ten times_ as a production transitive
dependency, so including per-method packages is not saving space (it
might instead result in slightly more space being used).
Any error caught during a React component render or lifecycle method
will now be caught by the top-level error boundary, which shows the
user this new error page. The error page will display a simple error
message, and will show the details of the error in a collapsible
section.
The caught error is also reported to Sentry.
In development the error will be re-thrown to make it easier to see on
the console, but it is not re-thrown in production.
* Remove unnecessary `getEnvironmentType` parameter
The default value of the first parameter is `window.location.href`, so
there is no need to pass it in explicitly.
* Remove junk parameter from `getEnvironmentType` invocation
`getEnvironmentType` doesn't need to be passed any parameter, as the
default value is `window.location.href` which is generally what is
wanted. In this case, the variable `location.href` was always
`undefined` anyway. This particular `location` variable is from React
Router, and does not have an `href` property.
* Fix comment for `getEnvironmentType`
One of the possible return values was referred to by the wrong name.
`withRouter` has been removed from any components that were not using
any of the three props injected by `withRouter`: `history`, `location`,
and `match`.
`compose` is a no-op when called upon a single component, so it has
been removed in all such cases.
The filename is seeded by a simple use of Math.random() pulling from an alphanumeric character bank, as opposed to a more cryptographically random solution. This provides a simple layer of difficulty for bad actors to seek out the recovery phrase file.
The 'can retype the seed phrase' test would fail sometimes when one of
the words in the seed phrase was a subset of another word (e.g. 'issue'
and 'tissue'). This is because the selector used to find the word
looked for the first element that contained the text, rather than an
exact match.
To simplify the selector and make it more reliable, test ids were added
to each seed phrase word. The selector now uses CSS instead of XPath,
and it only finds exact matches.
A test id was also added to the div containing the shuffled seed words
to select from, so that the chosen seed words wouldn't be selected
in place of the real target when the same word appears twice.
The `confirm-seed-phrase` component extends PureComponent, so it
doesn't need a `shouldComponentUpdate` function. The state is
effectively immutable, as all state is either a primitive type or is
updated with new instances rather than mutation.
Removing this function will silence a warning message printed to the
console during e2e tests (React doesn't want you to set this function
on `PureComponent`s).
Removing this function also exposed an unused piece of state, which has
also been removed.
The `withFixtures` helper will instantiate ganache, a web driver, and
a fixture server initialized with the given set of fixtures. It is
meant to facilitating writing small, isolated e2e tests.
The first example test has been added: simple-send. It ensures that the
user can send 1 ETH to another account.
These new e2e tests will run during the normal e2e test run.
Closes#6548
* Remove unused functions from `mapDispatchToProps`
The actions import was also updated to import only the two actions
used, rather than all actions.
* Remove unused container component
Well, technically it was the props injected by this container that were
unused. The container served no purpose, so the component it surrounded
is now used directly instead.
* Remove both unused `getCurrentViewContext` selectors
* Remove unused SHOW_CONFIG_PAGE action
* Remove checks for `currentView` with name `config`
Now that the SHOW_CONFIG_PAGE action has been removed, it's no longer
possible for `currentView.name` to be set to that value.
* Remove unused `wallet-view` container props
* Delete unused SHOW_SEND_PAGE and SHOW_ADD_TOKEN_PAGE actions
* Remove unused `account-menu.container` props
* Remove unused SHOW_INFO_PAGE action
* Remove unused SET_NEW_ACCOUNT_FORM action
These rows on the Advanced Settings page were being looked up in the
e2e tests by the order they appeared in. Instead they're now referenced
by data id, so that we can add new settings and re-arrange them without
breaking the e2e tests.
This component used to persist form contents to LocalStorage. This was
especially useful for the popup UI, as each time the mouse left the
popup, the UI was completely torn down and state was lost.
This component was only being referenced by one form, and it wasn't
even being used there (e.g. no fields were labelled appropriately to
be persisted).
This was a useful component, and it seems this feature was lost
somewhere in the past couple of years. It was tempting to re-instate it
rather than delete it, but I decided not to because I'd likely approach
the problem differently if we wanted to reinstate it again today (maybe
by using a React Hook, or storing the state in Redux and persisting a
subset of the Redux store instead).
These unused props weren't being caught by ESLint because this
component extended another, which I guess made it difficult for the
React plugin to determine what was unused.
The `componentWillUpdate` logic was moved into `componentDidUpdate` so
that it would be picked up by ESLint. Also it seemed like a sensible
place for it to go. Having three redundant gas updates as part of
the same lifecycle function seemed too far, so I ensured it's
only called once.
* Make gas estimate update on debounced token amount change, not just on blur after change
* Updated tests
* Ensure `updateGas` is bound early
Co-authored-by: Mark Stacey <markjstacey@gmail.com>
* Add `react/no-unused-prop-types` rule
All detected unused prop types have been removed. I have attempted to
ensure these props are no longer passed in either.
* Update handling of props to avoid false positive lint errors
These cases were detected by `react/no-unused-prop-types` as being
unused props, even though they were used. These minor adjustments
prevent them from being flagged as errors.
* Update unit tests
Many of these tests were just checking that specific props were passed
from containers or to a child component. These were deleted, as I can't
imagine how they'd be useful.
* Disable `react/no-unused-prop-types` in `componentWillReceiveProps
The rule `react/no-unused-prop-types` doesn't seem to be detecting
props used within `UNSAFE_componentWillReceiveProps`. The two cases
have been disabled temporarily until we can replace these unsafe
lifecycle functions.
* Use arrow property initializer functions
* Use pure components where applicable
* Add UNSAFE_ prefix for deprecated lifecycle hooks
* Add allow UNSAFE_
* Removed unused "Component"
* Replace boron with 'fade-modal'
* Upgrade react/no-deprecated to an error
* Paste react-tooltip-component source directly
* Use arrow functions to bind `this`
* Add UNSAFE_ prefix
* Update react-redux, react-router-dom
* Remove things from inlined 'fade-modal'
* Adjust mountWithRouter to get unit tests passing again
* Remove domkit
* Add Wrapper to render-helpers
* Upgrade @storybook/addon-knobs
The connect route now takes a route parameter: the permissions request
id. This id is set whenever the permissions connect screen is opened,
ensuring that that tab is for that specific request alone.
This makes handling of multiple permissions requests a bit more
intuitive. Previously whenever opening multiple permissions requests,
the first one would be shown on each successive tab, whereas you
would expect each tab to show the request that prompted the tab to
open. Users may now address permissions request in whichever order
they'd like to, rather than being forced to deal with them
chronologically.
The MetaMask logo was throwing an exception because the `history` prop
from React Router wasn't being passed in. However, even if it was
passed in, clicking on the logo wouldn't have done anything because the
home route redirects back to the permission connect page.
Instead the logo now isn't given an `onClick` handler. The
`MetaFoxLogo` component has been updated to only set `cursor: pointer`
if given an `onClick` handler.
