Closed
Bug 192320
Opened 22 years ago
Closed 21 years ago
Cannot down arrow in file
Categories
(Core :: DOM: Selection, defect)
Core
DOM: Selection
Tracking
()
VERIFIED
FIXED
People
(Reporter: aaronlev, Assigned: KaiE)
References
Details
(Keywords: access, testcase, topembed+, Whiteboard: EDITORBASE+ edt_x3)
Attachments
(2 files)
545 bytes,
text/html
|
Details | |
573 bytes,
patch
|
mjudge
:
review+
kinmoz
:
superreview+
|
Details | Diff | Splinter Review |
When I'm on line 1 of this document, I cannot down arrow.
Reporter | ||
Updated•22 years ago
|
Component: Editor: Core → Selection
Reporter | ||
Comment 1•22 years ago
|
||
Comment 3•22 years ago
|
||
NS7.01 from 11/20/02 does not have this bug. My tip mozilla build from a week ago or so show it, though.
Comment 4•22 years ago
|
||
This source has a p inside a font tag. The "block in inline" quirks code may be involved.
Assignee | ||
Comment 7•21 years ago
|
||
I think this looks like a nice bug to do my first steps with the editor code. Can I take it?
Assignee | ||
Comment 8•21 years ago
|
||
-> me I stepped through the code with the debugger. Although the display only shows 2 lines, the internal data structures counts the content as 5 (logical) lines. When having selected something on visible line 1, the code operates on data structure line 0. When doing shift-cursor-down, the code goes to data structure line 1, and calls nsLineIterator::FindFrameAt() This data structure line 1 has a position of 0/555, but a size of 0/0. I believe the failure happens when doing the check for aX >= line->mBounds.XMost() is reached, which results in signaling success to the caller and that "aXIsAfterLastFrame". This prevents the caller code at nsFrame::GetNextPrevLineFromBlockFrame to not test later lines and fails somehow. I believe the fix is to test for the special case where width is zero and fail in FindFrameAt. This will cause the caller to continue searching in the next line.
Assignee: mjudge → kaie
Assignee | ||
Comment 9•21 years ago
|
||
This patch appears to fix the problem.
Assignee | ||
Comment 10•21 years ago
|
||
jfrancis: Although you said you don't see the problem with NS 7.01, I do see it with 7.01
Assignee | ||
Updated•21 years ago
|
Attachment #117721 -
Flags: review?(mjudge)
Comment 11•21 years ago
|
||
I can repro this one using the current trunk build on win32 (2003031804)
Comment 12•21 years ago
|
||
Comment on attachment 117721 [details] [diff] [review] Patch v1 as long as you have tested it reasonably well
Attachment #117721 -
Flags: review?(mjudge) → review+
Assignee | ||
Comment 13•21 years ago
|
||
Comment on attachment 117721 [details] [diff] [review] Patch v1 Kin, can you please sr?
Attachment #117721 -
Flags: superreview?(kin)
Comment 14•21 years ago
|
||
Comment on attachment 117721 [details] [diff] [review] Patch v1 sr=kin@netscape.com
Attachment #117721 -
Flags: superreview?(kin) → superreview+
Comment 15•21 years ago
|
||
Comment on attachment 117721 [details] [diff] [review] Patch v1 I think I may have hastily sr'd this patch ... kai please test the case where you arrow though blank lines ... I believe blank lines (in composer) only have brs and brs don't have any dimensions so I'm not sure if this "blank" line will be skipped by your patch. Also, does this code get hit from selection via mouse?
Assignee | ||
Comment 16•21 years ago
|
||
It seems to work correctly. I have tested the attached testcase on Windows and Linux. In addition to the testcase I have used another testcase, where instead of a single <br> the file contains <br><br><br> With and without the patch, you have to press cursor down once for each <br> that is present. That behaviour is not changed by the patch. Note there is no visual feedback when selecting the empty lines. I've also tested what happens when using the mouse to select. Again no visual feedback for selecting the empty lines, but it still works. Another test I did was to select all and paste it into an ascii editor. With and without the patch the same amount of whitespace arrives at the destination. I'm not sure whether that code change influences the mouse selection. Should I trace whether that code is getting hit?
Assignee | ||
Comment 17•21 years ago
|
||
checked in, fixed
Status: NEW → RESOLVED
Closed: 21 years ago
Resolution: --- → FIXED
Comment 18•21 years ago
|
||
I have filed bug 200416, bug 200417, and bug 200418 which may be related to the fix for this bug.
Assignee | ||
Comment 19•21 years ago
|
||
I have tested all the three bugs you mentioned. I can reproduce all bugs with earlier versions of Mozilla, the checkin to this bug did not change the behaviour for me.
Comment 20•21 years ago
|
||
reopening; this is a duplicate of 192320 which Kaie just fixed
Status: RESOLVED → REOPENED
Resolution: FIXED → ---
Comment 21•21 years ago
|
||
oops; wrong bug!
Status: REOPENED → RESOLVED
Closed: 21 years ago → 21 years ago
Resolution: --- → FIXED
Comment 22•21 years ago
|
||
verifying with my mac debug build from today carryover keywords from dupe
Comment 23•21 years ago
|
||
*** Bug 192856 has been marked as a duplicate of this bug. ***
Comment 24•21 years ago
|
||
*** Bug 194637 has been marked as a duplicate of this bug. ***
Comment 25•21 years ago
|
||
*** Bug 183137 has been marked as a duplicate of this bug. ***
Comment 26•21 years ago
|
||
This fix was long needed and makes for a much more pleasant experience (for me anyway). I can now down arrow on any page....except for pages with frames. Why can't the main part of any page with frames be selected by default so that the arrow keys work? Right now, using the arrow key will do nothing at all.
You need to log in
before you can comment on or make changes to this bug.
Description
•