WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED WONTFIX
113605
Cancel pending spellcheck requests to avoid lagging markers with slow spellcheck
https://bugs.webkit.org/show_bug.cgi?id=113605
Summary
Cancel pending spellcheck requests to avoid lagging markers with slow spellcheck
Rouslan Solomakhin
Reported
2013-03-29 13:37:53 PDT
Cancel pending spellcheck requests to avoid lagging markers with slow spellcheck
Attachments
Patch
(4.63 KB, patch)
2013-03-29 13:48 PDT
,
Rouslan Solomakhin
no flags
Details
Formatted Diff
Diff
View All
Add attachment
proposed patch, testcase, etc.
Rouslan Solomakhin
Comment 1
2013-03-29 13:48:34 PDT
Created
attachment 195791
[details]
Patch
Rouslan Solomakhin
Comment 2
2013-03-29 13:50:35 PDT
Haven't added any tests yet.
Hajime Morrita
Comment 3
2013-03-31 19:45:16 PDT
Doesn't this cancel requests that we might want? For example, If we type "Helllo.", then type "How are you?", A request for "Helllo." seems to be cancelled. Is this right?
Rouslan Solomakhin
Comment 4
2013-04-01 09:01:21 PDT
(In reply to
comment #3
)
> Doesn't this cancel requests that we might want? > For example, If we type "Helllo.", then type "How are you?", > A request for "Helllo." seems to be cancelled. Is this right?
Spell Check requests incorporate whole paragraphs. If you type "Helllo. How are you?" then you should be fine. However, if you type "Helllo How are you?" then the request for the first paragraph might be cancelled. I see your point. Do you have a suggestion on how to better avoid lagging markers with slow spellcheck? Here is the Chromium bug-report with more details that you might find useful:
https://code.google.com/p/chromium/issues/detail?id=179639
Hajime Morrita
Comment 5
2013-04-01 21:22:22 PDT
> > How are you?" then the request for the first paragraph might be cancelled. I see your point. Do you have a suggestion on how to better avoid lagging markers with slow spellcheck? Here is the Chromium bug-report with more details that you might find useful:
https://code.google.com/p/chromium/issues/detail?id=179639
Thanks for the explanation. We could compare the position of Nodes using Node::compareDocumentPosition and doing cancellation for requests which overlap, or are located after the newly-requested range. Also, i noticed that we have kinda similar cancellation logic in SpellChecker::enqueueRequest() where we discard newly-requested check when there is pending request for same node. We could put the cancellation logic in this patch there.
Rouslan Solomakhin
Comment 6
2013-04-02 09:42:14 PDT
Let me rethink this patch and get back to this later.
Levi Weintraub
Comment 7
2013-04-08 10:52:58 PDT
Comment on
attachment 195791
[details]
Patch Clearing flags.
Note
You need to
log in
before you can comment on or make changes to this bug.
Top of Page
Format For Printing
XML
Clone This Bug