User-Agent: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.4a) Gecko/20030401 Build Identifier: Mozilla/5.0 (Windows; U; Win98; en-US; rv:1.4a) Gecko/20030401 The line wiggles so fast I can't read the text of either the message or the who sent what. It seems the highlited or message the mouse "blues up" moves also. Reproducible: Always Steps to Reproduce: 1.It just started happening all the time. 2. Even after restart. 3. Actual Results: Nothing unusual. Expected Results: I think your product is great so far. This is my first time doing this. I loaded you guys up just a couple days ago. I love the look and the feel, but I can't use the Mozilla mail. I can't read it its so fast, just a blur.

Not an accessibility bug... I seem to recall this being a bug in trees, but only mail being able to trigger it. Over to mail in the hopes that someone there will recall the bug number....

*** Bug 201856 has been marked as a duplicate of this bug. ***

Confirming; giving precedence to this bug because it's older than the dupe (Bug 201856); however, the description in the dupe is more detailed. I have not managed to duplicate this symptom, myself. Updating summary; original: In Mozilla, while looking at the inbox, the line between the text contained in the message and the messages wiggles or vibrates up and down. New: Horizontal splitter bar between message list and message pane vibrates (flickers,wiggles, jumps) Making this dependent on Bug 201460, which was originally also a dupe but in later discussion found a similar symptom in the browser's Page Info dialog.

*** Bug 203321 has been marked as a duplicate of this bug. ***

*** Bug 204194 has been marked as a duplicate of this bug. ***

If anyone is interested in this bug, I seem to be able to reproduce it as often as like on my computer with 1.4 alpha (I describe it in my dupe, but I'll put it here) I'm using a vanilla windows 2000 install here set screen res to 1024*768 and font size to small install moz 1.4 alpha Set up mail Get exactly 8 mails in a folder (say by making 8 drafts) look in folder See the bounce go! (hopefully!) Note it seems after you have resized the window even once it's not possible to get back to this state.. I'd be interested to know if anyone "in the know" has managed to duplicate this, else I will investigate more trying to get it to happen (because I seem to be able to as often as I like)

There is a patch proposed in bug 201460 since 2003-04-22

adt: nsbeta1-

*** Bug 207929 has been marked as a duplicate of this bug. ***

This bug has followed me throughout the 1.4 builds. Happens very frequently (I tend to not have too many messages in my inbox). Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.4b) Gecko/20030516 Mozilla Firebird/0.6

*** Bug 209829 has been marked as a duplicate of this bug. ***

Would you please read Bug 210184, is this the same Bug? Thanks.

*** Bug 210184 has been marked as a duplicate of this bug. ***

*** Bug 204605 has been marked as a duplicate of this bug. ***

Sorry, but I'm quite new to this BugZilla thing. I'm trying to help here, so I'm very sorry again. This comment has been recopied from Bug 210184 that I have filed and marked duplicate. Basically the thing is, the Email preview pane and the Email Listing Pane in the 3-pane email window flickers / shakes / vibrates up & down when I received a new email. This will happen, let's say there are seven (7) previous / old email. Make sure you delete all newly received email (the one in Bold Typeface). Compose new email (complete with the subject and msg area or email body at least filled-up with some text), send it to yourself. click the send button and send. then click the get Msgs button to retrieve your email. After you have received the new email you've created and send to yourself just now, the all two(2) pane on the right-hand-side on the 3-pane window (I really hope you know what I meant) will flicker up and down sometime slowly then keep pacing up till one point it flicker too fast you can't even read your email. Maybe it may eat up your memory or processor time, coz I experience a slight slowdown in my PC. This can be reproduced on My PC: Intel Pentium 3 - 450 MHZ 320 MB SDRAM on Asus P3B-F MoBos, Enlight ATX Casing, Asus AGPV3800 Ultra 32MB Graphic Card, on 1024X768 32 Bit screen resolution using Sony Trinitron MultiScan100ES Monitor running WinXP Pro Version 2002 SP1 with these history of Windows update Successful 14 June 2003 820291: Recommended Update (Windows XP) Web site Successful 07 June 2003 818529: June 2003, Cumulative Patch for Internet Explorer 6 Service Pack 1 Automatic update Successful 29 May 2003 Q815021 XP: Security Update Web site Successful 29 May 2003 811493: Security Update (Windows XP) Web site Successful 29 May 2003 Q811114: Security Update (Windows XP or Windows XP Service Pack 1) Web site Successful 27 May 2003 Windows MovieMaker 2 Read more... Web site Successful 27 May 2003 Microsoft Windows Journal Viewer (Windows XP) Read more... Web site Successful 27 May 2003 Windows Messenger 4.7 Web site Successful 27 May 2003 Q810243 Update: Watch television shows recorded by Media Center PCs on other Microsoft Windows XP PCs Web site Successful 27 May 2003 Q322011: Recommended Update Read more... Web site Successful 27 May 2003 Microsoft .NET Framework version 1.1 Read more... Web site Successful 27 May 2003 DirectX 9.0a End-User Runtime Read more... Web site Successful 27 May 2003 815485: Recommended Update Web site Successful 27 May 2003 811493: Security Update (Windows XP) Web site Failed 27 May 2003 Windows MovieMaker 2 Read more... Web site Failed 27 May 2003 Microsoft Windows Journal Viewer (Windows XP) Read more... Web site Failed 27 May 2003 Windows Messenger 4.7 Web site Failed 27 May 2003 Q810243 Update: Watch television shows recorded by Media Center PCs on other Microsoft Windows XP PCs Web site Failed 27 May 2003 Q322011: Recommended Update Read more... Web site Failed 27 May 2003 Microsoft .NET Framework version 1.1 Read more... Web site Failed 27 May 2003 DirectX 9.0a End-User Runtime Read more... Web site Failed 27 May 2003 815485: Recommended Update Web site Failed 27 May 2003 811493: Security Update (Windows XP) Web site Failed 27 May 2003 Windows MovieMaker 2 Read more... Web site Failed 27 May 2003 Microsoft Windows Journal Viewer (Windows XP) Read more... Web site Failed 27 May 2003 Windows Messenger 4.7 Web site Failed 27 May 2003 Q810243 Update: Watch television shows recorded by Media Center PCs on other Microsoft Windows XP PCs Web site Failed 27 May 2003 Q322011: Recommended Update Read more... Web site Failed 27 May 2003 Microsoft .NET Framework version 1.1 Read more... Web site Failed 27 May 2003 DirectX 9.0a End-User Runtime Read more... Web site Failed 27 May 2003 815485: Recommended Update Web site Failed 27 May 2003 811493: Security Update (Windows XP) Web site Successful 27 May 2003 Q817287: Critical Update (Catalog Database Corruption in Microsoft Windows XP) Automatic update Successful 27 May 2003 331953: Security Update (Windows XP) Web site Successful 27 May 2003 Q817287: Critical Update (Catalog Database Corruption in Microsoft Windows XP) Web site Successful 27 May 2003 330994: April 2003, Security Update for Outlook Express 6 SP1 Web site Successful 27 May 2003 813489: April 2003, Cumulative Patch for Internet Explorer 6 Service Pack 1 Web site Successful 24 March 2003 Q327405: Recommended Update (Windows XP Professional) Read more... Web site Successful 24 March 2003 813951: Update for Internet Explorer 6 SP1 Web site Successful 24 March 2003 327979: Recommended Update Web site Successful 24 March 2003 814995: Recommended Update Web site Successful 24 March 2003 329170: Security Update Web site Successful 24 March 2003 Q329048: Security Update Read more... Web site Successful 24 March 2003 Q323255: Security Update (Windows XP) Read more... Web site Successful 24 March 2003 Q327696: Internet Information Services Security Roll-up Package Read more... Web site Successful 24 March 2003 814078: Security Update (Microsoft Jscript version 5.6, Windows 2000, Windows XP) Web site Successful 24 March 2003 810577: Security Update Web site Successful 24 March 2003 810833: Security Update (Windows XP) Web site Successful 24 March 2003 328310: Security Update Web site Successful 24 March 2003 Q329390: Security Update Web site Successful 24 March 2003 Q329834: Security Update (Windows XP) Read more... Web site Successful 24 March 2003 810847: February 2003, Cumulative Patch for Internet Explorer 6 Service Pack 1 Web site Successful 23 March 2003 Euro Conversion Tool Read more... Web site Successful 23 March 2003 Windows Media Player 9 Series (Windows XP) Web site Successful 23 March 2003 811630: Critical Update (Windows XP) Read more... Web site Successful 23 March 2003 810565: Critical Update Web site Successful 23 March 2003 Q329115: Security Update (Windows XP) Web site Successful 23 March 2003 814033: Critical Update Web site Successful 23 March 2003 Q329441: Critical Update Web site Successful 23 March 2003 Creative Media Driver Version 5.1.2603.0 Web site Failed 23 March 2003 Creative Media Driver Version 5.1.2603.0 Web site Failed 23 March 2003 Creative Media Driver Version 5.1.2603.0 Web site Successful 23 March 2003 3COM Net Driver Version 04.31.0000.0 Web site This may also be reproduced on my dad PC: Dell OptiPlex GX110 with Dell Monitor Model No E551 running on Intel 82810E Graphics Controller at 800X600 truecolor 24 bit screen resolution with this spec of Intel Processor x86 Family 6 Model 8 Stepping 6 with 129,260 KB RAM running Win2K 5.00.2195 Service Pack 3. I've traced this thing happens back since Mozilla 1.4b (Build ID: 2003050714) on my dad PC, on my PC since Mozilla 1.3.1 Mozilla 1.4 RC1 Mozilla 1.4Beta and Mozilla 1.4RC2 I'm currently on. I've tested all these flavour of Mozilla for this Bug and it happens on all of these flavour. I tried to include my screen capture but do not know how to submit it together with this Bug Report thing. maybe i missed something, should someone show me how to do it? Thanks Very Much To All Anyway and keep up the really nice, interesting and a very good work. Reproducible: Always Steps to Reproduce: Too long to be typed, kindly please refer to the details provided. Thank You Very Much Actual Results: Too long to be typed, kindly please refer to the details provided. Thank You Very Much Expected Results: I don't know, to be looked and consider to be resolved, I supposed. Thanks Very Much To All Anyway and keep up the really nice, interesting and a very good work.

from azmibinsalim - this is the screen capture and some details on which part that flickers / shakes / vibrates. Thanks Very Much.

Thanks Mike Cowperthwaite for the information on the duplicate thing. It is very much appreciated.

I am now 600 miles from the machine on which I observed this bug, however I will confirm it's existence. Win98, PII, ATi Video, I think.

does not block development

It seems that or I do think that this bug 210184 has been resolved in Mozilla 1.4RC3

*** Bug 210736 has been marked as a duplicate of this bug. ***

I also have this problem where in Mozilla 1.4 RC3 the inbox will vibrate rapidly up and down within approximately a quarter inch region of the screen. (About 2/3 cm). Unlike the previous poster however, I can detect no direct pattern to this fault. It is impossible for me to accurately reproduce the bug since it appears to happen purely randomly. Further I can confirm it is a bug over several different platforms. I have a number of systems running WinXP Professional and 1 running W2K03 Enterprise Server and a home based sytem running windows XP media centre edition. This problem occurs in all cases and accross all platforms. It is a minor, but nonetheless irritating bug, which I think deserves your urgent attention before there is any possibility of releasing Mozilla 1.4 to the general public. I trust this information has proved helpful to you. Should you require any further comments, please feel free to let me know. Q

See bug 212130 for a nearly-identical-but-not-quite issue.

*** Bug 212806 has been marked as a duplicate of this bug. ***

Similar but not quite the same bug: bug 131733 . It appears to be much the same problem, but with the separator between the mailbox list and the other two panes.

*** Bug 214194 has been marked as a duplicate of this bug. ***

*** Bug 214704 has been marked as a duplicate of this bug. ***

*** Bug 214642 has been marked as a duplicate of this bug. ***

*** Bug 215161 has been marked as a duplicate of this bug. ***

*** Bug 216575 has been marked as a duplicate of this bug. ***

Mozilla 1.4 (release) on windows 98 (fresh install)

*** Bug 217020 has been marked as a duplicate of this bug. ***

I have the same problem with Mozilla 1.5alpha and 1.5beta with Win XP Pro SP1. When the screen starts to shake and I double click on a mail to read it, the body stays blank. The error is very hard to reproduce. And comes up at almost on random but in general it occurs with long subject lines. For me when it starts and I resize the header pane the jitter stays away until I start again. If someone is interested I can attach some messages which caused the jitter.

Just to add more fuel to the fire...this happened to me yesterday on 1.4 release (classic theme, standard install, no tweaks of any sort), when on WinXP. There were 8 messages in the INBOX. Resizing the window made it better for a bit, but then the shaking returned. resizing the splitter bar between the message list and the message text made the problem go away.

Duplicated with Mozilla 1.4 release on Windows 2003 Server [Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.4) Gecko/20030624]. This server is set up as a multi-user terminal server, and one of our users complained about the problem depicted in the earlier movie attachment. Since the problem did not show up when I ran Messenger on the box, I decided whatever problem exists must be saved in the user's profile somehow. I removed and recreated that user's profile (through Profile Manager) and the problem went away, at least temporarily. The next day the issue returned.

I think I have a reasonable idea what's going on here. The Windows native theme code gives a minimum size for the scrollbar parts, which is also used as the preferred size. This causes the preferred size for the treerows and tree to increase as well (trees by default have a near-zero min size unless rows= is given). Now, here's where the problem happens. When the vbox that contains the thread pane and the message pane lays out, it first gives all the children their preferred size. It then distributes the extra space, according to the flex attributes on the children. So, the size that the thread pane receives is its preferred size + 1/4 of the extra space (the message pane has flex=3). Suppose that the extra space allocated to the thread pane is N times the height of a row. If the number of rows scrolled out of view is <= N, then the thread pane will collapse its scrollbar. Once that happens, the thread pane's size allocation changes -- it's still getting 1/4 of the extra space, but the scrollbar's preferred size is not there in addition to it. The total height is not enough to show all of the rows, so it uncollapses the scrollbar. Repeat ad nauseam. I think this mainly happens with Windows/Classic due to the min sizes the theme code hands back The GTK native theme code does not have this problem, Mac uses native scrollbars, and for non-classic skins we'll use the CSS sizes. The CSS for e.g. Modern _does_ specify min sizes on some elements, but apparently not large enough to cause this problem to appear. I think the kicker is the scrollbar track being given a min height of 34px by the native theme code. We need to ensure that uncollapsing the scrollbar doesn't cause the tree to grow enough to want to uncollapse the scrollbar. One idea is to always make the tree's pref height match what the scrollbar's pref height would be, so that it doesn't change when you uncollapse. Alternately, we could just remove some min heights on the scrollbar parts and hope the problem goes away.

Jan, any thoughts on my last comment?

*** Bug 205464 has been marked as a duplicate of this bug. ***

another way to reproduce this: click on the last "View" link on http://www.xulplanet.com/tutorials/xultu/trees.html, when using the classic skin on win32.

Note that the link that Seth Spitzer gave in the last comment demonstrates the wiggling without any scroll bars being involved. It's a good example because that View link creates a window of the proper size for wiggling without it being dependent on number of messages or a user's screen or window size setting. So there's more to the problem than just Bryan Ryner's excellent analysis about scrollbars in comment #36.

I think the "View" link is the same thing, but for a horizontal scrollbar.

The "Source" link next to the "View" link shows the XUL source code for the window. I don't see where there is a scrollbar, and I don't see any scrollbar appear when I resize the window. It seems to have more to do with the splitter. As the window is resized, the wiggling starts when the splitter crosses over the text line. Am I overlooking a scrollbar somewhere?

Minor correction to my comment: We need to ensure that uncollapsing the scrollbar doesn't cause the tree to grow enough to want to uncollapse the scrollbar. should read: We need to ensure that uncollapsing the scrollbar doesn't cause the tree to grow enough to want to _collapse_ the scrollbar.

> Am I overlooking a scrollbar somewhere? sorry, my mistake, not a horizontal scrollbar, a *vertical* scrollbar. but just for that one row, which seems odd. but it looks like the same overflow / underflow bug to me.

So why can't we allow scrollbar parts to become smaller? Windows can...

The comment in the patch explains fairly well what I'm doing here, but just to summarize, I'm making the tree body have a min-height that's at least the height of a scrollbar. A few concerns I had: - Is globally caching this height correct? (i.e., could some tree have differently sized scrollbars?) I think we're ok here. - Are there other cases in which the cached height needs to be invalidated? - Is it important to be able to make a tree smaller than this? I'd vote no, since it's really a pretty small height, and any tree can potentially overflow.

Comment on attachment 130811 [details] [diff] [review] better version of the above patch Jan, what do you think?

re-assign to bryner, as this fix is all him.

that last patch (plus a missing change to nsTreeBodyFrame.h) also works for me on win2k, classic skin.

Sorry, but autocomplete needs to be able to create 1-line-high trees :-(

That's unfortunate. As far as I know, autocomplete will expand that tree to several lines before it's allowed to overflow, so at least it's not going to shove a scrollbar into 1 line. So, one idea is that we could have a concept of a non-overflowing tree, and not enforce the scrollbar-sized min-height on that (and never uncollapse its scrollbar). I'd like to maintain the idea of nsITheme being able to set a minimum size for the scrollbar, and I'd still like to avoid having a tree's min-height change when the scrollbar is uncollapsed. However, at this point, I'm wondering if we shouldn't come up with a 1.5 fix that simply reduces the min-height that nsNativeThemeWin reports. I suspect removing the min-size on the track would be sufficient - gtk just has a min size for the scrollbar buttons and thumb and does not have this problem.

I haven't yet tested this (in Linux at the moment), but I'm hopeful this will take care of it.

Comment on attachment 130897 [details] [diff] [review] band-aid for 1.5 looks reasonable for 1.5

that last patch works for me on win32. is that patch meant for the trunk, or for just the 1.4.1 / 1.5 branch?

Comment on attachment 130897 [details] [diff] [review] band-aid for 1.5 a=sspitzer for 1.4.1 and 1.5 can you added comment to the code referring back to this bug? also, let's open a new bug about fixing this the right way, so we don't forget. 1.5 definitely, so that it gets picked up for tbird 0.3 and fbird 0.7

*** Bug 218364 has been marked as a duplicate of this bug. ***

*** Bug 214713 has been marked as a duplicate of this bug. ***

Comment on attachment 130897 [details] [diff] [review] band-aid for 1.5 sr=mscott assuming you add some comments that link back to this bug and file a new bug to fix this for real.

Checked in on the trunk and 1.4 branch. Spun off bug 218453 for the XUL issue.

*** Bug 218435 has been marked as a duplicate of this bug. ***

*** Bug 218478 has been marked as a duplicate of this bug. ***

*** Bug 207508 has been marked as a duplicate of this bug. ***

*** Bug 212249 has been marked as a duplicate of this bug. ***

*** Bug 221722 has been marked as a duplicate of this bug. ***

Comment on attachment 130811 [details] [diff] [review] better version of the above patch Can't you just call sbBox->GetMinSize(aBoxLayoutState, aSize) ?

Comment on attachment 130811 [details] [diff] [review] better version of the above patch I still don't like this because it will force the autocomplete tree to be several lines high.

*** Bug 209320 has been marked as a duplicate of this bug. ***

*** Bug 211471 has been marked as a duplicate of this bug. ***

*** Bug 214705 has been marked as a duplicate of this bug. ***

*** Bug 214817 has been marked as a duplicate of this bug. ***

*** Bug 215377 has been marked as a duplicate of this bug. ***

*** Bug 219779 has been marked as a duplicate of this bug. ***

*** Bug 219338 has been marked as a duplicate of this bug. ***

*** Bug 217085 has been marked as a duplicate of this bug. ***

Can I implement the proposed patch on my mozilla 1.4? If yes, how do I go about it? I have the jitter problem for 135 users. I'm using citrix/nt 4.0 thanks in advance

Comment on attachment 130811 [details] [diff] [review] better version of the above patch clearing old request

WORKSFORME because of XXX HACK comment in http://mxr.mozilla.org/mozilla/source/widget/src/windows/nsNativeThemeWin.cpp#2220