Closed Bug 639695 Opened 14 years ago Closed 14 years ago

Caret fails to show on FF4.0 with Orion Editor

Categories

(Core :: DOM: Selection, defect)

defect
Not set
normal

Tracking

()

RESOLVED FIXED
mozilla5

People

(Reporter: johnjbarton, Assigned: ehsan.akhgari)

References

Details

Attachments

(1 file)

1. Install orion http://wiki.eclipse.org/Orion/How_Tos/Install_Orion_on_Localhost use the Lastest releease 0.2 Stream Stable Build 0.2M5 after step 7 of the instructions (you login) you should see a couple of folders. Pick any file in there to bring up the editor. But the editor has no cursor. Works on FF3.6.
Hmm, I gave this a quick try on Mac and couldn't reproduce. I opened a bunch of files in the test server you sent to me by email, and the cursor showed up at the beginning of all of the files that I tried. Does this happen to you on a clean profile, every time? What platform have you tried it under? Which nightly did you test this on?
I just repeated my test using http://localhost:8080/coding.html#http://localhost:8080/file/org.eclipse.orion.client.editor/web/index.html Here is about:support: Application Basics Name Firefox Version 4.0b13pre User Agent Mozilla/5.0 (Windows NT 6.1; rv:2.0b13pre) Gecko/20110303 Firefox/4.0b13pre Profile Directory Open Containing Folder Enabled Plugins about:plugins Build Configuration about:buildconfig Extensions Name Version Enabled ID Modified Preferences Name Value accessibility.typeaheadfind.flashBar 0 browser.places.importBookmarksHTML false browser.places.smartBookmarksVersion 2 browser.startup.homepage_override.buildID 20110303122446 browser.startup.homepage_override.mstone rv:2.0b13pre browser.tabs.warnOnClose false extensions.lastAppVersion 4.0b13pre network.cookie.prefsMigrated true places.history.expiration.transient_current_max_pages 104660 privacy.sanitize.migrateFx3Prefs true Graphics Adapter Description ATI Radeon HD 4300 Series Vendor ID 1002 Device ID 954f Adapter RAM 512 Adapter Drivers atiumdag atidxx32 atidxx64 atiumdva atiumd64 atiumd6a atitmm64 Driver Version 8.632.1.2000 Driver Date 8-17-2009 Direct2D Enabled Blocked on your graphics driver. Try updating your graphics driver to version 10.6 or newer. DirectWrite Enabled false (6.1.7600.16385, font cache n/a) WebGL Renderer (WebGL unavailable) GPU Accelerated Windows 0/2
I can reproduce it on Linux. I see the caret when a file is opened for the first time, but a Reload makes the bug occur - no caret even though the document is still editable. Maybe the same underlying issue as in bug 467333 / bug 407127?
Summary: Cursor fails to show on FF4.0 with Orion Editor → Caret fails to show on FF4.0 with Orion Editor
The corresponding bug on the Eclipse side is here: https://bugs.eclipse.org/bugs/show_bug.cgi?id=335505
This problem can be reproduced by hidden and showing a DIV that has contentEditable set (style.display="none"). See https://bugs.eclipse.org/bugs/show_bug.cgi?id=334587
Here is a sample that can reproduce the problem without the orion editor. https://bugs.eclipse.org/bugs/show_bug.cgi?id=334587#c3
Forgot to mentioned I am running Mozilla/5.0 (Windows NT 5.1; rv:2.0b12) Gecko/20100101 Firefox/4.0b12
My patches in bug 407127 help here to some extent, but there is more work to be done here.
Assignee: nobody → ehsan
Depends on: 407127
Whiteboard: [post-2.0]
Just to let you know, we have a workaround on our side now (that we're not happy with but it does make the Orion editor not lose the caret on Firefox 4).
This is the reincarnation of bug 284245 for contenteditable documents. What happens is that when the presshell changes for an iframe containing a contenteditable document, the editor loses its connection to the selection objects, and won't gain it back until nsEditor::InitializeSelection is called for some reason (most common: the editable area losing and regaining focus). I have a patch to be applied on top of the patches in bug 407127 which fixes this bug.
Attached patch Patch (v1)Splinter Review
Attachment #518858 - Flags: review?(roc)
The testcase from comment 6 fails in Firefox 3.6 for me, but comment 0 says that the problem does not occur in Firefox 3.6. Are we looking at two different bugs here?
(In reply to comment #12) > The testcase from comment 6 fails in Firefox 3.6 for me, but comment 0 says > that the problem does not occur in Firefox 3.6. Are we looking at two different > bugs here? No. The Orion editor is really complex, so whether or not that would fail depends on a lot of factors, but the bug has been there since forever, and it's been something that I wanted to fix for a long time. Note that anything which can trigger an InitializeSelection call would "work around" this issue, but that doesn't mean that the bug doesn't exist.
Depends on: post2.0
Status: NEW → RESOLVED
Closed: 14 years ago
No longer depends on: post2.0
Flags: in-testsuite+
Resolution: --- → FIXED
Whiteboard: [post-2.0]
Target Milestone: --- → mozilla2.2
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: