WebKit Bugzilla
New
Browse
Log In
×
Sign in with GitHub
or
Remember my login
Create Account
·
Forgot Password
Forgotten password account recovery
RESOLVED FIXED
210369
Move more from live range to SimpleRange: callers of absoluteTextRects
https://bugs.webkit.org/show_bug.cgi?id=210369
Summary
Move more from live range to SimpleRange: callers of absoluteTextRects
Darin Adler
Reported
2020-04-10 19:43:08 PDT
Move more from live range to SimpleRange: callers of absoluteTextRects
Attachments
Patch
(36.58 KB, patch)
2020-04-11 08:39 PDT
,
Darin Adler
no flags
Details
Formatted Diff
Diff
Patch
(36.81 KB, patch)
2020-04-11 10:44 PDT
,
Darin Adler
andersca
: review+
Details
Formatted Diff
Diff
Show Obsolete
(1)
View All
Add attachment
proposed patch, testcase, etc.
Darin Adler
Comment 1
2020-04-11 08:39:35 PDT
Created
attachment 396164
[details]
Patch
Anders Carlsson
Comment 2
2020-04-11 08:48:52 PDT
Comment on
attachment 396164
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=396164&action=review
> Source/WebCore/dom/Range.cpp:-1183 > -void Range::absoluteTextRects(Vector<IntRect>& rects, bool useSelectionHeight, RangeInFixedPosition* inFixed, OptionSet<BoundingRectBehavior> rectOptions) const
I would change this to return the rects instead of using an out parameter.
> Source/WebCore/html/HTMLTextFormControlElement.cpp:463 > + return { { { *innerText, 0 }, { *innerText, 0 } } };
Maybe specify SimpleRange here to cut down on some of the curly braces.
> Source/WebCore/html/HTMLTextFormControlElement.cpp:488 > + return { { { *startNode, start }, { *endNode, end } } };
Maybe specify SimpleRange here to cut down on some of the curly braces.
> Source/WebCore/rendering/RenderObject.cpp:1904 > + else if (is<RenderText>(renderer)) {
I assume is<RenderText> will return false here if renderer is null?
> Source/WebKitLegacy/mac/WebView/WebFrame.mm:1114 > + return range ? [range textRects] : @[];
range.textRects?
Darin Adler
Comment 3
2020-04-11 10:01:26 PDT
Comment on
attachment 396164
[details]
Patch View in context:
https://bugs.webkit.org/attachment.cgi?id=396164&action=review
Looks like build failures are because WebFrame.mm needs to include DOM.h.
>> Source/WebCore/dom/Range.cpp:-1183 >> -void Range::absoluteTextRects(Vector<IntRect>& rects, bool useSelectionHeight, RangeInFixedPosition* inFixed, OptionSet<BoundingRectBehavior> rectOptions) const > > I would change this to return the rects instead of using an out parameter.
Yes, agreed. I would do that if I was planning to keep this. But I am planning to delete it. I only removed the argument as a prelude to removing the entire function soon.
>> Source/WebCore/html/HTMLTextFormControlElement.cpp:463 >> + return { { { *innerText, 0 }, { *innerText, 0 } } }; > > Maybe specify SimpleRange here to cut down on some of the curly braces.
I’d be willing to do that. Not sure it’s better. If you are sure, I will do it. It allows me to remove the outer set.
>> Source/WebCore/html/HTMLTextFormControlElement.cpp:488 >> + return { { { *startNode, start }, { *endNode, end } } }; > > Maybe specify SimpleRange here to cut down on some of the curly braces.
Ditto.
>> Source/WebCore/rendering/RenderObject.cpp:1904 >> + else if (is<RenderText>(renderer)) { > > I assume is<RenderText> will return false here if renderer is null?
Yes.
>> Source/WebKitLegacy/mac/WebView/WebFrame.mm:1114 >> + return range ? [range textRects] : @[]; > > range.textRects?
Will do.
Darin Adler
Comment 4
2020-04-11 10:44:38 PDT
Created
attachment 396174
[details]
Patch
Darin Adler
Comment 5
2020-04-11 10:44:54 PDT
Anders, you didn’t set review+. Did you mean to?
Anders Carlsson
Comment 6
2020-04-11 12:12:23 PDT
(In reply to Darin Adler from
comment #5
)
> Anders, you didn’t set review+. Did you mean to?
I wanted to wait until everything was green!
Darin Adler
Comment 7
2020-04-11 12:36:50 PDT
Committed
r259933
: <
https://trac.webkit.org/changeset/259933
>
Radar WebKit Bug Importer
Comment 8
2020-04-11 12:37:16 PDT
<
rdar://problem/61644860
>
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