After opening another message in an existing message window (pref), `Next unread` (N) fails with `TypeError: aParent is null` while executing the button_next command
Categories
(Thunderbird :: Toolbars and Tabs, defect)
Tracking
(Not tracked)
People
(Reporter: v+mozbug, Assigned: mkmelin)
Details
(Whiteboard: [STR comment 4])
Attachments
(1 file)
User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:107.0) Gecko/20100101 Firefox/107.0
Steps to reproduce:
Added "next unread" button to message window toolbar.
Clicked the button while viewing a subfolder of Local Folders with many unread messages.
Actual results:
Doesn't move to next unread message in the current folder.
Expected results:
Should work every time (when there are more unread messages, at least).
Reporter | ||
Comment 1•1 year ago
|
||
related Bug 1792609
I didn't realize my symptom applied to Local Folders, that bug got closed before I could figure out the missing detail.
Comment 2•1 year ago
|
||
Hey Glenn, I am also failing to reproduce this for 1st-level and 2nd-level subfolders of Local Folders node on TB 102.5.1 (64-bit), Win10.
I marked a couple of messages unread, and clicking the Next button in the message reader satellite window will iterate exactly those.
This looks straightforward, so unless you provide more information which will allow us to reproduce, there's nothing we can do here.
We can reopen this bug if you have more information, check the box under comment input [x] Request information from
and add me (:thomas8).
Reporter | ||
Comment 3•1 year ago
|
||
It is very simple to reproduce... I received 3 emails today that went to my Mozilla folder, all from bugzilla.
I click to open the first one in the message window list window, it opens a new window, and marks itself read. It was comment 4 on bug 1792609.
Because the next unread button hasn't worked for me for a long time, I have the habit of clicking next unread in the message list window, so I did that next, and hit enter, which caused the next message to open in the same message window created by the prior operation. It was the comment 2 on bug 1804000.
After reading that you couldn't reproduce it, I tried clicking the next unread button in the new window. It doesn't work - nothing visible happens, except the next unread button itself animates as a result of being clicked.. several times as I click several times. The third message is still unread at this point in time.
What more information should I provide, and what process should I use to provide it? This happens in all my local folders, 1st and 2nd level, all the time.
Is there a setting (possibly a very old setting, as I've been using Thunderbird for years) that is preventing it from working? What setting would that be? What logic is used to react to the click on the message window's next unread button? What could make that process terminate rather than loading the next unread message?
Should I make a screen shot video showing that it doesn't work? Then you could mark the bug "Doesn't work for Glenn" instead of "works for me"?
I think the next obvious step in analyzing the problem is to debug it on my machine, but I have no idea how to do that. Is there logging or tracing or error reporting that can be turned on to analyze the logic flow in that area?
Comment 4•1 year ago
|
||
Hey Glenn, thanks for following up on this and cooperatively providing more information! Sorry for closing this a bit too fast.
Confirming for TB 102.5.1, Win10, based on the new information hidden in comment 3. Already seen on TB 91.13.1 (32-bit).
To those obscure bugs out there, do you really think you can all hide forever? Take that! :-)
New STR
- Set pref
≡ > Settings > General > Reading & Display > Open new messages in: An existing message window
(important!) - Have some unread messages in a mail folder, let's say message 4 and 6.
- Select and open message 1 with
Enter
=> message 1 will open in a new message window W1. - Select and open another message 2 with Enter (important!)
=> message 2 will supersede message 1 in message window W1 - Now press
N
on keyboard to go toNext unread message
(you could also add theNext Unread
button to Mail Toolbar and click that). - Check error console (Ctrl+Shift+J, or Developer Tools via Ctrl+Shift+I, confirm hidden prompt with OK, select
Console
tab; clear that before testing).
Actual result
- Nothing. Message 2 still showing in the satellite window.
- Error in console (see below)
Expected result
- Next unread message (message 4) should be shown in the satellite window W1.
Looks like there's a number of similar bugs on record.
Console error
Uncaught TypeError: aParent is null
FindNextChildFolder chrome://messenger/content/msgViewNavigation.js:34
FindNextFolder chrome://messenger/content/msgViewNavigation.js:77
CrossFolderNavigation chrome://messenger/content/msgViewNavigation.js:162
performNavigation chrome://messenger/content/messageWindow.js:1559
doCommand chrome://messenger/content/messageWindow.js:1474
goDoCommand chrome://global/content/globalOverlay.js:128
oncommand chrome://messenger/content/messageWindow.xhtml:1
msgViewNavigation.js:34:7
An error occurred executing the button_next command: [Exception... "[JavaScript Error: "aParent is null" {file: "chrome://messenger/content/msgViewNavigation.js" line: 34}]'[JavaScript Error: "aParent is null" {file: "chrome://messenger/content/msgViewNavigation.js" line: 34}]' when calling method: [nsIController::doCommand]" nsresult: "0x80570021 (NS_ERROR_XPC_JAVASCRIPT_ERROR_WITH_DETAILS)" location: "JS frame :: chrome://global/content/globalOverlay.js :: goDoCommand :: line 128" data: yes] globalOverlay.js:131
Updated•1 year ago
|
Reporter | ||
Comment 5•1 year ago
|
||
Hah hah, those steps you marked as "important!" just seemed so naturally obvious to do to me, and I leave a message window open most of the time, so it seemed like it never worked. I'll play around, and see if I can get it to work at all....maybe only when I first start Thunderbird, if I follow the right sequence? But as soon as I switch to another folder and open a message there, it'll quit?
Glad you found a specific case to cause the failure, I didn't know what I was doing was so unusual, but it must be different than what any of the current tests do!
Assignee | ||
Comment 6•1 year ago
|
||
Updated•1 year ago
|
Description
•