| Summary: | On iOS, getData can't get text set by setData during copy event | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | WebKit | Reporter: | Ryosuke Niwa <rniwa> | ||||||||
| Component: | HTML Editing | Assignee: | Ryosuke Niwa <rniwa> | ||||||||
| Status: | RESOLVED FIXED | ||||||||||
| Severity: | Normal | CC: | cdumez, commit-queue, darin, kaspern, thorton, webkit-bug-importer, wenson_hsieh | ||||||||
| Priority: | P2 | Keywords: | InRadar | ||||||||
| Version: | WebKit Nightly Build | ||||||||||
| Hardware: | Unspecified | ||||||||||
| OS: | Unspecified | ||||||||||
| Attachments: |
|
||||||||||
Created attachment 320872 [details]
WIP
Let me fix this myself. I need to apply one more fix to dataTransfer.types, which always returns the list of all supported types at the moment :( Created attachment 321183 [details]
Fixes the bug by refactoring
Comment on attachment 321183 [details] Fixes the bug by refactoring View in context: https://bugs.webkit.org/attachment.cgi?id=321183&action=review > Source/WebCore/platform/Pasteboard.h:189 > + virtual void commitToPasteboard(Pasteboard&) { ASSERT_NOT_REACHED(); } Can we make this pure virtual instead? I’m sure there is some reason we can’t, but failing at runtime instead of compile time is not a good pattern. (In reply to Darin Adler from comment #5) > Comment on attachment 321183 [details] > Fixes the bug by refactoring > > View in context: > https://bugs.webkit.org/attachment.cgi?id=321183&action=review > > > Source/WebCore/platform/Pasteboard.h:189 > > + virtual void commitToPasteboard(Pasteboard&) { ASSERT_NOT_REACHED(); } > > Can we make this pure virtual instead? I’m sure there is some reason we > can’t, but failing at runtime instead of compile time is not a good pattern. So Pasteboard is the "real" pasteboard we use to write to pasteboard so we can't quite do that. I think the cleaner approach is to only add this to StaticPasteboard and just dynamic_cast dataTransfer->pasteboard(). I was debating that and just adding a virtual function but I'd just add a dynamic_cast instead. Well, an even cleaner approach is if there was a local variable referencing to a StaticPasteboard in dispatchClipboardEvent itself but that's a bit tricky because StaticPasteboard is not ref-counted (uses unique_ptr). Created attachment 321232 [details]
Patch for landing
Comment on attachment 321232 [details]
Patch for landing
Wait for EWS.
Comment on attachment 321232 [details] Patch for landing Clearing flags on attachment: 321232 Committed r222228: <http://trac.webkit.org/changeset/222228> All reviewed patches have been landed. Closing bug. *** Bug 179439 has been marked as a duplicate of this bug. *** |
On iOS, calling getData('text/plain') after calling setData('text/plain', 'foo') doesn't return 'foo'.