We'd rather 'play it safe' for our selection detection (which avoids
following links if you have some text selected). This commit adds an
additional check of 'more than 1 character' to the 'isSelected' check.
This is based on the fact that it's far quicker to type the character so
why would a user select on character? This therefore caters for any
fairly large (in the scheme of things) incidential mouse moves whilst
clicking a link.
Overall this is probably overkill
1. Re-focus the input element on phrase removal
2. Move all actions to `actions:`
3. Move to a form looking `value` rather than `items`
4. Move placeholder functionalit yinto the component
5. Force DDAU instead of two way binding with `slice` and `onchange`
6. Begin to deprecate the `searchable` interface
1. EventSources now pass themselves thorugh to the run function as a
second argument. This enables the use of arrow functions along with the
EventSources API
`(configuration, source) => source.close()`. Order of arguments could
potentially be switched at a later date.
2. BlockingEventSources now let you pass an 'event' through at
instantation time. If you do this, the event will immediately be dispatched
once the EventSource is opened. The usecase for this is for 'unfreezing' cached
BlockingEvents. This makes it easier to provide a cache for
BlockingEventSources by caching its data rather than the entire
BlockingEventSource itself.
```
new BlockingEventSource(
(config, source) => { /* something */ },
{
cursor: 1024, // this would also come from a cache
currentEvent: getFromSomeSortOfCache(eventSourceId) //this is the new bit
}
);
// more realistically
new BlockingEventSource(
(config, source) => { return data.findSomething(slug, config) },
getFromSomeSortOfCache(eventSourceId)
);
```
Makes listening to multiple events on one target slightly easier.
Adding events can be rolled up into passing through an object, and the
returned remove function removes all of the handlers in the object
For example:
```
//this.listen...
const remove = listeners.add(
{
'message': handler,
'error': handler
}
);
remove(); // removes all listeners in the object
```
The entire API for listeners is now becoming slightly overloaded, so
potentially we'd use this API always and remove the ability to use a
string/function pair.
Throughout the app we mutate the value of the root node classList on
navigation between separate pages (basically on URL change).
Every template has a unique classList for example `template-service
template-show` and `template-service template-list` etc etc.
When navigating between 2 pages, both pages using the same template yet
with different data, previoulsy we would entirely clear out the
`html.classList` and then refill it again with eaxctly the same classes.
This commit moves this to perform a diff previous to mutating the
classList, and then potentially no classList mutating is needed when
moving between 2 pages of the same template.
You can potentially close an EventSource before its first tick by
immediately setting the readyState to a non-open state. Therefore it
never opens.
Calling `open` will then open it.
'Open' fits better than 'reopen' when taking the above into account
Currently these gherkin dictionaries are contained in
`ember-cli-api-double`, which isn't the correct place for them.
They belong with the steps themselves.
This is also less complicated now we can require/import javascript
using a standard approach
Also runs an update of `ember-cli-api-double` which will be the last
installation of version 1 of this module.
Migrate roughly half of the base components into base
Adds a target for easily formatting CSS
Further CSS amends/migration (#5921)
1. tooltips within tables where a slightly bit troublesome due to a mix
of `inline-flex`, `overflow` and the need for truncation. This refineds
tooltips a slight bit more to work 'everywhere'.
2. We also move tooltip to use the correct color and min-width from
structure, but we overwrite the min-width here until we get confirmation
on widths/alignment of text within a tooltip.
3. Tiny fixes for breadcrumbs and toggle-buttons in tabular listings
4. Now we inline-flex our table cells, it means it is impossible to
truncate text without wrapping it in another element. This wraps all
Description like text in `<p>` tags. Generally the first column of text
is already wrapped in an `<a>` tag. Other items such as consul tags and
policy names etc get 'cutoff' rather than truncated.
5. We are now using all the icons from `@hashicorp/structure-icons`
EventSources will wait for 1 tick before 'opening'. There is always the
chance that the EventSource is '.close()'ed before that tick. We
therefore check the 'readyState' before opening the EventSource
* ui: Allow text selection of clickable elements and their contents
This commit disables a click on mousedown be removing the `href`
attribute and moving it to a `data-href` attribute. On mouseup it will
only move it back if there is no selection. This means that an anchor
will only be followed on click _if_ there is no selection.
This fixes the fact that whenever you select some copy within a
clickable element it immediately throws you into the linked page when
you release your mouse.
Further notes:
We use the `isCollapsed` property here which 'seems' to be classed as
'experimental' in one place where I researched it:
https://developer.mozilla.org/en-US/docs/Web/API/Selection/isCollapsed
Although in others it makes no mention of this 'experimental' e.g:
- https://webplatform.github.io/docs/dom/Selection/isCollapsed/
- https://w3c.github.io/selection-api/#dom-selection-iscollapsed
I may have gone a little overboard in feature detection for this, but I
conscious of that fact that if `isCollapsed` doesn't exist at some point
in the future (something that seems unlikely). The code here will have
no effect on the UI. But I'd specifically like a second pair of eyes on
that.
* ui: Don't break right click, detects a secondary click on mousedown
* ui: Put anchor selection capability behind an ENV var
* ui: Reconciliate ember-data store when records are deleted via blocking
Currently we are barely using the ember-data store/cache, but it will
still cache records in the store even though technically we aren't using
it.
This adds a SyncTime to every record that uses blocking queries so we
can delete older records from the ember-data cache to prevent them
building up
* ui: Add basic timestamp method we can access from tests, fixup tests
Adds a timestamp method that we can access from within tests so we can
test that the SyncTime is being set.
There is probably a better way to do this, but this is also probably the
simplest approach - we are also likely to revisit this at a later date
1. Adds a Listeners class, which lets us...
2. Add Listeners recursively. So you can
createListeners().add(createListeners())
3. Also add the ability to `.add` as a plain function
This moves the entire idea more towards a generic teardown utility
* Store primaries root in secondary after intermediate signature
This ensures that the intermediate exists within the CA root stored in raft and not just in the CA provider state. This has the very nice benefit of actually outputting the intermediate cert within the ca roots HTTP/RPC endpoints.
This change means that if signing the intermediate fails it will not set the root within raft. So far I have not come up with a reason why that is bad. The secondary CA roots watch will pull the root again and go through all the motions. So as soon as getting an intermediate CA works the root will get set.
* Make TestAgentAntiEntropy_Check_DeferSync less flaky
I am not sure this is the full fix but it seems to help for me.
When this test flakes sometimes this happens:
--- FAIL: TestCoordinate_Node (1.69s)
panic: interface conversion: interface {} is nil, not structs.Coordinates [recovered]
FAIL github.com/hashicorp/consul/agent 19.999s
Exit code: 1
panic: interface conversion: interface {} is nil, not structs.Coordinates [recovered]
panic: interface conversion: interface {} is nil, not structs.Coordinates
There is definitely a bug lurking, but the code seems to imply this can
only return nil on 404. The tests previously were not checking the
status code.
The underlying cause of the flake is unknown, but this should turn the
failure into a more normal test failure.
When there is an node name conflicts, such messages are displayed within Consul:
`consul.fsm: EnsureRegistration failed: failed inserting node: Error while renaming Node ID: "e1d456bc-f72d-98e5-ebb3-26ae80d785cf": Node name node001 is reserved by node 05f10209-1b9c-b90c-e3e2-059e64556d4a with name node001`
While it is easy to find the node that has reserved the name, it is hard to find
the node trying to aquire the name since it is not registered, because it
is not part of `consul members` output
This PR will display the IP of the offender and solve far more easily those issues.
Since FUNCNAME is not defined when running outside a function,
trap does not work and display wrong error message.
Example from https://circleci.com/gh/hashicorp/consul/69506 :
```
⨯ FAIL
/home/circleci/project/test/integration/connect/envoy/run-tests.sh: line 1: FUNCNAME[0]: unbound variable
make: *** [GNUmakefile:363: test-envoy-integ] Error 1
```
This fix will avoid this error message and display the real cause.
The embedded `Server` field on a `DNSServer` is only set inside of the
`ListenAndServe` method. If that method fails for reasons like the
address being in use and is not bindable, then the `Server` field will
not be set and the overall `Agent.Start()` will fail.
This will trigger the inner loop of `TestAgent.Start()` to invoke
`ShutdownEndpoints` which will attempt to pretty print the DNS servers
using fields on that inner `Server` field. Because it was never set,
this causes a nil pointer dereference and crashes the test.
* website: Update middleman-hashicorp container and Gemfile.lock
Time marches on, and so do security vulnerabilities in Nokogiri. So it's time
for a new container.
As with last time, here's a reminder for the next person who needs to update
this:
- You shouldn't just update the dependency in Gemfile.lock, because your build
times will go to heck as you compile Nokogiri from source on every run. So you
need an updated container with all the dependencies.
- To update the container, you need to push a new tag to the middleman-hashicorp
repo. Teamcity does the rest, and will ship a new container to Docker Hub
(unless its credentials are out of date, in which case go ask team-eng-serv.)
- Once that's pushed:
- Update Makefile
- Update the Gemfile
- Delete Gemfile.lock
- `make website` until it comes up, then ctrl-C
- Commit the changes
* website: Specify a different json version in Gemfile.lock
The Consul website uses different containers for preview and deploy, and this
oddball JSON version was causing issues. This commit sacrifices a little bit
of preview startup speed for (hopefully) working deploys.
Previously `verify_incoming` was required when turning on `auto_encrypt.allow_tls`, but that doesn't work together with HTTPS UI in some scenarios. Adding `verify_incoming_rpc` to the allowed configurations.
John Cowen
Sarah Adams
Pierre Souchay
Freddy
Aestek
Matt Keeler
R.B. Boyer
Alvin Huang
tradel
Hans Hasselberg
Nick Fagerlund
Mike Morris