Summary: | WebsiteDataStore operations need to grab background assertions | ||||||
---|---|---|---|---|---|---|---|
Product: | WebKit | Reporter: | Anders Carlsson <andersca> | ||||
Component: | New Bugs | Assignee: | Anders Carlsson <andersca> | ||||
Status: | RESOLVED FIXED | ||||||
Severity: | Normal | CC: | mitz | ||||
Priority: | P2 | ||||||
Version: | 528+ (Nightly build) | ||||||
Hardware: | Unspecified | ||||||
OS: | Unspecified | ||||||
Attachments: |
|
Description
Anders Carlsson
2015-07-16 14:12:13 PDT
Created attachment 256924 [details]
Patch
Comment on attachment 256924 [details]
Patch
We normally have the callers create the callbacks using the create function that takes an activity token. That’s why we added that parameter. Can you follow that established pattern in this case?
(In reply to comment #2) > Comment on attachment 256924 [details] > Patch > > We normally have the callers create the callbacks using the create function > that takes an activity token. That’s why we added that parameter. Can you > follow that established pattern in this case? I forgot to mention that we normally do this using CallbackMap. (In reply to comment #3) > (In reply to comment #2) > > Comment on attachment 256924 [details] > > Patch > > > > We normally have the callers create the callbacks using the create function > > that takes an activity token. That’s why we added that parameter. Can you > > follow that established pattern in this case? > > I forgot to mention that we normally do this using CallbackMap. I'm not going to make such an invasive change at this point. Comment on attachment 256924 [details]
Patch
Seems worth adding FIXMEs about using CallbackMap in the future.
Committed r187023: <http://trac.webkit.org/changeset/187023> (In reply to comment #5) > Comment on attachment 256924 [details] > Patch > > Seems worth adding FIXMEs about using CallbackMap in the future. The ability to do async messages with replies should be built into IPC::Connection, not layered on top of it. |