647471 - "ASSERTION: We should never hit a case where our height is non-zero but smaller than the caret baseline"

Status: RESOLVED WORKSFORME

(Core :: DOM: Editor, defect)

Description

[Jesse Ruderman]

Attachment: testcase

###!!! ASSERTION: We should never hit a case where our height is non-zero but smaller than the caret baseline...: 'baseline <= mRect.height', file layout/generic/nsInlineFrame.cpp, line 936

This assertion is part of nsInlineFrame::GetCaretBaseline, which was added recently:

changeset:   db8e95fbcd39
user:        Ehsan Akhgari
date:        Mon Mar 28 20:17:16 2011 -0400
summary:     Bug 644428 - Position the caret correctly for empty inline frames; r=roc

Comment 1

[Jesse Ruderman]

Attachment: stack trace

Comment 2

[(no longer active)]

I don't get the assertion on trunk, but I do get this error:

JavaScript error: , line 0: uncaught exception: [Exception... "Component returned failure code: 0x80004003 (NS_ERROR_INVALID_POINTER) [nsIDOMNSHTMLDocument.execCommand]"  nsresult: "0x80004003 (NS_ERROR_INVALID_POINTER)"  location: "JS frame :: https://bug647471.bugzilla.mozilla.org/attachment.cgi?id=523806 :: boom :: line 12"  data: no]

This is caused by nsEditor::GetChildOffset using nsIContent's instead of nsINodes, which cause it to fail if aParent is a document (which happens in this test case).

Comment 3

[(no longer active)]

As with everything in the editor, fixing the js error in comment 2 is not trivial...  I tried a couple of things, and they both failed.  I'm going to attach both of my failed attempts here, for my own future reference.

But this bug in its original form is WFM, since the assertion was removed in <http://hg.mozilla.org/mozilla-central/rev/3c21288157a9>

Comment 4

[(no longer active)]

Attachment: Failed attempt 1

Comment 5

[(no longer active)]

Attachment: Failed attempt 2