Skip to content

Commit

Permalink
chore: add 2020-09-21 API WG meeting notes (#342)
Browse files Browse the repository at this point in the history
  • Loading branch information
jkleinsc authored Sep 24, 2020
1 parent 342e5d6 commit 82277c7
Showing 1 changed file with 65 additions and 0 deletions.
65 changes: 65 additions & 0 deletions wg-api/meeting-notes/2020-09-21.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,65 @@
# API Working Group

## Date Sept 21 2020

### Attendees
* @nornagon
* @zcbenz
* @codebytere
* @jkleinsc
* @MarshallOfSound
* @samuelmaddock

### Agenda

* [API Review Requested PRS](https://github.com/electron/electron/pulls?q=is%3Apr+label%3A%22api-review%2Frequested+%F0%9F%97%B3%22+is%3Aopen+-label%3Aapi-review%2Freviewed+sort%3Acreated-asc):

* [[WIP] feat: Expose renderer spellcheck API](https://github.com/electron/electron/pull/25060)
* No longer WIP
* There’s no Web API for this
* (Sam:) Not convinced this should be on webFrame
* [ACTION: Sam to discuss this. Where should this go?]
* [feat: add support for will-navigate and will-fail-load custom error pages](https://github.com/electron/electron/pull/25128)
* Weird returnValue thing?
* Prevents white page.
* Where does the error code go?
* Netlog
* Devtools console
* event.failWith(...)?
* This doesn’t trigger for `fetch`
* Could we expose a NavigationThrottle API directly…?
* Should be marked as experimental, using the flag feature
* [feat: implement allowFileAccess loadExtension option](https://github.com/electron/electron/pull/25198)
* This is ~= the “Allow access to file urls” in the extensions preferences in chrome
* Should we expose ExtensionPrefs? (For remembering whether allowFileAccess: true across reloads) Not for now.
* (SamM) Followup to previous responses in PR. Avoid ExtensionPrefs.
* [feat: add BrowserWindow.isTabletMode API](https://github.com/electron/electron/pull/25209)
* Seems fine
* [feat: add serial api support](https://github.com/electron/electron/pull/25237)
* We should make sure we expose the same surfaces that Chrome does in terms of UI
* When do the added/removed events start getting emitted…?
* Let’s not back ourselves into a hole w.r.t. WebUSB which uses similar APIs
* re-review
* [feat: [net] add "credentials" option to net.request](https://github.com/electron/electron/pull/25284)
* Gives the ability to opt out of sending session auth.
* More closely matches fetch().
* :check:
* [feat: add macOS fast user switching evt](https://github.com/electron/electron/pull/25321)
* Should this be on powerMonitor…?
* It’s next to lock-screen and unlock-screen
* Maybe we could have a new “system” module…? [for later followup]
* :check:
* [feat(extensions): expose ExtensionRegistryObserver events in Session](https://github.com/electron/electron/pull/25385)
* Should this maybe be on session.extensions.on(...)?
* meh
* NB the pdf extension always emits the “ready” event
* :check:
* [feat: custom toast xml and failure reporting for notifications](https://github.com/electron/electron/pull/25401)
* Can’t be called “error” because that would be a breaking change.
* :check:
* [feat: allow setting SSL config](https://github.com/electron/electron/pull/25461)
* Should mention what the defaults are in the docs.
* :check:
* [[wip] feat: add webFrameMain API to the main process](https://github.com/electron/electron/pull/25464)
* SamM: Figure out lifetime of WebFrame instances (lookup vs hold references).
* Finalize API in followup next meeting.

0 comments on commit 82277c7

Please sign in to comment.