Closed
Bug 251775
Opened 21 years ago
Closed 21 years ago
[FIX]Remove direct drag-drop knowledge from nsFrame::HandlePress
Categories
(Core :: Layout, defect, P1)
Tracking
()
RESOLVED
FIXED
mozilla1.8alpha3
People
(Reporter: bzbarsky, Assigned: bzbarsky)
References
Details
Attachments
(1 file, 1 obsolete file)
32.72 KB,
patch
|
Details | Diff | Splinter Review |
![]() |
Assignee | |
Comment 1•21 years ago
|
||
![]() |
Assignee | |
Comment 2•21 years ago
|
||
Comment on attachment 153466 [details] [diff] [review]
Proposed patch
biesi, jst, what do you think?
Attachment #153466 -
Flags: superreview?(jst)
Attachment #153466 -
Flags: review?(cbiesinger)
![]() |
Assignee | |
Comment 3•21 years ago
|
||
That patch fixes bug 244859 too.
Blocks: 244859
Priority: -- → P1
Summary: Remove direct drag-drop knowledge from nsFrame::HandlePress → [FIX]Remove direct drag-drop knowledge from nsFrame::HandlePress
Target Milestone: --- → mozilla1.8beta
Comment 4•21 years ago
|
||
Comment on attachment 153466 [details] [diff] [review]
Proposed patch
thanks!
Index: content/base/public/nsContentUtils.h
+ * @return the nsIImage corresponding to the first frame of the image
+ */
+ static nsresult GetImageFromContent(nsIImageLoadingContent* aContent,
+ nsIImage** aImage);
+ * @return whether it's draggable
+ */
+ static PRBool ContentIsDraggable(nsIContent* aContent) {
This is a bit inconsistent. the @return refers to two different things here...
why not use something like:
@param [out]
in the first case?
![]() |
Assignee | |
Comment 5•21 years ago
|
||
> This is a bit inconsistent. the @return refers to two different things here...
That seems to be the style of the file, basically...
Comment 6•21 years ago
|
||
Comment on attachment 153466 [details] [diff] [review]
Proposed patch
content/base/src/nsContentUtils.cpp
+ return CallGetInterface(ir.get(), aImage);
is the .get() needed?
+ nsCOMPtr<nsIURI> absURI;
what's this COMptr for?
content/base/src/nsContentAreaDragDrop.cpp
+ NS_PRECONDITION(outParent, "Null out param?");
I kinda find these useless, but ok.
void
nsTransferableFactory::FindParentLinkNode(nsIDOMNode* inNode,
nsIDOMNode** outParent)
maybe this should return an already_AddRefed, instead of using an out param
Attachment #153466 -
Flags: review?(cbiesinger) → review+
![]() |
Assignee | |
Comment 7•21 years ago
|
||
> is the .get() needed?
Yes, if you want it to compile.
> + nsCOMPtr<nsIURI> absURI;
> what's this COMptr for?
Oops. That's from an earlier patch iteration. Fixed.
> maybe this should return an already_AddRefed, instead of using an out param
Hmm.. yeah, I could do that. There're just a _lot_ of callsites...
Comment 8•21 years ago
|
||
Comment on attachment 153466 [details] [diff] [review]
Proposed patch
- In content/base/public/nsContentUtils.h:
+ * Method to get an nsIImage from an image loading content
+ *
+ * @param aContent The image loading content. Must not be null.
+ * @return the nsIImage corresponding to the first frame of the image
+ */
+ static nsresult GetImageFromContent(nsIImageLoadingContent* aContent,
+ nsIImage** aImage);
Seems like this could return already_AddRefed().
sr=jst with that and the other comments addressed.
Attachment #153466 -
Flags: superreview?(jst) → superreview+
![]() |
Assignee | |
Comment 9•21 years ago
|
||
![]() |
Assignee | |
Updated•21 years ago
|
Attachment #153466 -
Attachment is obsolete: true
![]() |
Assignee | |
Comment 10•21 years ago
|
||
Fixed.
Status: NEW → RESOLVED
Closed: 21 years ago
Resolution: --- → FIXED
![]() |
Assignee | |
Updated•21 years ago
|
Target Milestone: mozilla1.8beta → mozilla1.8alpha3
You need to log in
before you can comment on or make changes to this bug.
Description
•