Merge main #1
Merged
Merge main #1
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
The prose says we are changing the `presign_url` function, but really it means to to `presign_upload`. This becomes evident when reading through the whole context, but can be very confusing if you don't. I suspect the function was originally called `presign_url` but got refactored to `presign_upload` and the prose got missed. Co-authored-by: Benjamin Porter <[email protected]>
…erty (phoenixframework#3213) If there is no scroll container, but the body or html element have the overflow-y property, we listened for scroll events on those. Instead we must still listen for scroll events on the document itself.
…3207) * Warn on version mismatch between client and server * define LV_VSN in js tests
* Add cheatsheet for special phx-* HTML attributes * Update guides/cheatsheets/html-attrs.cheatmd Co-authored-by: José Valim <[email protected]> * Update guides/cheatsheets/html-attrs.cheatmd --------- Co-authored-by: José Valim <[email protected]> Co-authored-by: Chris McCord <[email protected]>
In circumstances where `assign_async/4` is being used to async assign multiple results, it is possible that some of the results should persist during loading, and others should not. In this case, a list of keys can be passed to the `reset` option, and only those keys will be reset during the async operation.
…et-async-list Allow partially resetting an async result when using multiple assign
* reject empty string in class_names
…ring (phoenixframework#3090) Co-authored-by: José Valim <[email protected]>
* merge test coverages * update deprecated actions, fix artifact paths + update playwright to 1.41.2 * allow coverage >= 85% * fix merge command * update README * use mix test * apply suggestions from review * add default jest reporter back * update playwright to 1.43.1 * fix port in playwright teardown script * use liveview esm script for correct coverage reporting * import isUsedInput from module
Closes phoenixframework#3306 While we use refs (vector clocks) and ack's to prevent races like double form submissions and duplicate interactions for latent clients, there existed two bugs – one on the server, and one on client, that could allow for an ack to be processed too early in some conditions involving LiveComponents with live navigation. On the server, a LiveComponent would issue a nav like push_patch, and we we would ack the LiveComponent event before the parent's handle_params diff event. This caused a race where the LiveComponent is ack'd and pending states undone too early on the client, before the parent's handle_params diff has been sent. This allows scenarios where the diff from the parent LiveView handle_params removes the LiveComponent from the page, but there is a window where the LC is interactive since we ack'd it early. The fix on the server is to still send the LiveComponent diff on redirect, but send it out of band, and ack the original LiveComponent event within the parent's handle_params diff that will necessarily be triggered. On the client, a bug existed for phx-remove. Elements are delayed from being removed from the DOM to allow a short transition to occur, such as a fadeout of 200-300ms. Within this window, the elements remained interactive, which allowed for an "unlocked" element like a form button to be clicked after it was processed and ack'd by the server during the phx-remove transition. The fix for the client race is to prevent event propagation on the removed elements during the phx-remove transition.
Bumps [ws](https://github.com/websockets/ws) from 8.16.0 to 8.17.1. - [Release notes](https://github.com/websockets/ws/releases) - [Commits](websockets/ws@8.16.0...8.17.1) --- updated-dependencies: - dependency-name: ws dependency-type: indirect ... Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
In phoenixframework#3201 we changed the way form recovery works by using a document fragment to dispatch the form recovery events. When traversing the DOM to check if a LC is in the same LiveView, this broke the check, because when iterating over node.parentNode, we would never arrive at the fragment itself. This commit fixes this by checking by using the framgnet's first child instead, which is the actual LV element. Fixes phoenixframework#3314.
Co-authored-by: Steffen Deusch <[email protected]>
* Improve doc of put_flash * Update lib/phoenix_live_view.ex --------- Co-authored-by: Steffen Deusch <[email protected]>
* mix format * support noformat modifier on sigil_H It's not officially documented for now. * check if formatted in CI * raise on invalid sigil_H modifiers
slvust
approved these changes
Jul 17, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.