| Summary: | File upload cancels before files can be uploaded. | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Megan Gardner <megan_gardner> | ||||||
| Component: | New Bugs | Assignee: | Megan Gardner <megan_gardner> | ||||||
| Status: | RESOLVED FIXED | ||||||||
| Severity: | Normal | CC: | bdakin, commit-queue, thorton, wenson_hsieh | ||||||
| Priority: | P2 | Keywords: | InRadar | ||||||
| Version: | WebKit Nightly Build | ||||||||
| Hardware: | Unspecified | ||||||||
| OS: | Unspecified | ||||||||
| Attachments: |
|
||||||||
|
Description
Megan Gardner
2020-03-09 19:34:20 PDT
Created attachment 393117 [details]
Patch
Comment on attachment 393117 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=393117&action=review > Source/WebKit/ChangeLog:11 > + With the new UIContextMenus, we needed to cancel in order for everything to be > + reset correclty in WebCore. But this can only be done in the animation completetion > + handler, but we only want to cancel if we aren't presenting a different picker. Spelling: "correctly" Spelling: "completetion" We should figure out how to test this (though I have a feeling any way we test this would probably bypass exactly the parts that are a problem here). > Source/WebKit/UIProcess/ios/forms/WKFileUploadPanel.mm:442 > + self->_isPresentingSubMenu = YES; Should this be held weakly? Is it possible for WKFileUploadPanel to go away before the handler is called? > Source/WebKit/UIProcess/ios/forms/WKFileUploadPanel.mm:473 > + [self _cancel]; Who cancels after the submenu is dismissed? If it's really necessary, it seems like a problem that it doesn't happen. Comment on attachment 393117 [details] Patch View in context: https://bugs.webkit.org/attachment.cgi?id=393117&action=review >> Source/WebKit/UIProcess/ios/forms/WKFileUploadPanel.mm:473 >> + [self _cancel]; > > Who cancels after the submenu is dismissed? If it's really necessary, it seems like a problem that it doesn't happen. Megan notes that the submenus themselves already call _cancel, so this should be OK Created attachment 393168 [details]
Patch for landing
Comment on attachment 393168 [details] Patch for landing Clearing flags on attachment: 393168 Committed r258225: <https://trac.webkit.org/changeset/258225> All reviewed patches have been landed. Closing bug. |