Closed Bug 312852 Opened 19 years ago Closed 16 years ago

Tooltip should not appear during a drag (e.g., dragging favicon results in tooltip, blocking part of the bookmarks toolbar)

Categories

(Core :: XUL, defect, P1)

defect

Tracking

()

VERIFIED FIXED
mozilla1.9.2a1

People

(Reporter: andy.ozment, Assigned: mstange)

References

(Blocks 1 open bug)

Details

(Keywords: regression, verified1.9.0.7, verified1.9.1, Whiteboard: [needs test] read comment 172 BEFORE commenting (see comment 129 for STR))

Attachments

(2 files, 4 obsolete files)

User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8b5) Gecko/20051006 Firefox/1.4.1 Build Identifier: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8b5) Gecko/20051006 Firefox/1.4.1 Normal behaviour: drag a URL from the navigation bar to the bookmark toolbar; when you release it, it appears as a bookmark. However, with the latest two Mac alphas (perhaps older as well), this is partially broken. If you drag the URL to the portion of the bookmark toolbar immediately below the URL space on thenavigation bar, a tooltip appears that states: "Drag and drop this icon to create a link to this page" That tooltip blocks the creation of a bookmark in that location. If you release the URL, it is _not_ added as a bookmark. This problem is particularly frustrating if you are attempting to add a bookmark to a folder located in that spot on the bookmark toolbar. Reproducible: Always Steps to Reproduce: 1. Drag URL from navigation bar to bookmark toolbar 2. Drag that URL onto the location of the bookmark toolbar that is immediately below the URL location on the navigation toolbar. 3. Wait briefly for the tooltip "Drag and drop this icon to create a link to this page" to appear. 4. Release the URL while your cursor is over the tooltip, so that a bookmark should be created right under the tooltip. Actual Results: No bookmark is created. Expected Results: A bookmark should be created or the tooltip should move out of the way. Comment 2 in bug 296057 mentions this problem as well: <https://bugzilla.mozilla.org/show_bug.cgi?id=296057#c2> However, bug 296057 describes a different problem and is not a dupe of this bug.
Duplicate of/related to Core bug 133064?
Marcia Knous and I see this bug too. I'm using Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8b5) Gecko/20051024 Firefox/1.5 Marcia says this doesn't happen in Firefox 1.0.x, so I doubt it's a dup of bug 133064.
URL: any
Status: UNCONFIRMED → NEW
Ever confirmed: true
Keywords: regression
Summary: Dragging a URL to the bookmarks toolbar immediately below the navigation bar URL is blocked by the appearance of a tooltip → Dragging proxy icon results in tooltip, blocking part of the bookmarks toolbar
This regressed on trunk between a 2005-06-20 build and a 2005-06-22 build.
I don't see any checkins during that period that reference tooltips or the proxy icon or dragging, but there were some event-handling changes, including: * Bug 282940, Switch Mozilla/Firefox to CFRunLoopSource-based plevent handling. * Bug 296704, Trusted event reinitialization allows data-theft.
As Jesse mentioned, I don't see this happening in 1.0.7. I think this is an annoying enough polish issue that it should be addressed for 1.5.
*** Bug 329981 has been marked as a duplicate of this bug. ***
Just a reminder that this problem is still happening in version 1.5.0.3, and another vote for getting it fixed sooner rather than later.
*** Bug 318772 has been marked as a duplicate of this bug. ***
Josh, is there a simple fix for this regression? I'm guessing that I'll be self-minusing this nomination in my next triage pass, but it's an annoying enough regression that we might want to fix it.
Flags: blocking-firefox2?
Keywords: polish
Target Milestone: --- → Firefox 2 beta2
As predicted, we're not going to block on this, but if there's a small patch that can restore the Firefox 1.0 functionality, we'd love to take it.
Flags: blocking-firefox2? → blocking-firefox2-
Target Milestone: Firefox 2 beta2 → Firefox 3 alpha1
*** Bug 345976 has been marked as a duplicate of this bug. ***
I bet this is caused by not sending NS_MOUSE_EXIT to the last-pointed widget when beginning a drag. That would be a pretty easy fix, but I'd want to verify the behavior against the win32 implementation. I predict we'll be able to fix this regression for 2b2.
Assignee: nobody → mark
Target Milestone: Firefox 3 alpha1 → Firefox 2 beta2
Yeah, looks like that's what's going on.
but it's not
*** Bug 351718 has been marked as a duplicate of this bug. ***
*** Bug 351791 has been marked as a duplicate of this bug. ***
This is really bad for first-time users of Firefox 2 (it seems to work on trunk). I think it should block Ff2.
also bad for users of Firefox 1.5.x....
I agree with Håkan Waara: since this is the "first impression" when new user attempt to set up their links, it does not sit well for recommending to others.
*** Bug 352076 has been marked as a duplicate of this bug. ***
*** Bug 354640 has been marked as a duplicate of this bug. ***
mark, fwiw, the trunk doesn't seem to have this problem
I can confirm that this bug still exists in RC2 and it is very annoying. The only work around is to drag the bookmark, but keep holding the mouse button down for 5 seconds until the tooltip goes away. Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.1) Gecko/20061003 Firefox/2.0
I agree with earlier sentiments about this being rather annoying - still seen in Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.1) Gecko/20061010 Firefox/2.0. Would like to see this addressed in one of the early follow up releases after we ship.
I added my vote and remembered why I switched to Camino for several months. For people who have elaborate bookmark bars with multiple folders, this is extremely annoying. If the help balloon popped up above the address area, the problem would be solved. I'm using a MacBook running OS 10.4.8 and am using Mozilla 2.0
Clarifications: That would be Mozilla Firefox 2.0 (Gecko/20061010 Firefox/2.0). And when I said "over" the address field, I meant above the address field.
A way to prevent this bug was even in the "top 8" list for firefox tweaks at this page: http://www.lifehacker.com/software/firefox-2/geek-to-live--top-firefox-2-config-tweaks-209941.php I'd strongly suggest nominating this for a minor firefox 2-release.
*** Bug 358708 has been marked as a duplicate of this bug. ***
> I'd strongly suggest nominating this for a minor firefox 2-release. Strongly agree. Also note: in firefox 3.0.1a (latest trunk) there's a bug that won't let you drop proxy icons on the toolbar at all.
Flags: blocking1.8.1.1?
Keywords: polish
OS: Mac OS X 10.2 → Mac OS X 10.4
> Also note: in firefox 3.0.1a (latest trunk) there's a bug that > won't let you drop proxy icons on the toolbar at all. Bug 354958?
Yes, it sucks, but until we have an understanding of it, its not a blocker.
Flags: wanted1.8.1.x+
Flags: blocking1.8.1.1?
Flags: blocking1.8.1.1-
One possible way of getting this fixed would be not to not send out tooltip events during a drag. I don't think that is normal on mac anyways.
*** Bug 364959 has been marked as a duplicate of this bug. ***
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9a3pre) Gecko/20070209 Minefield/3.0a3pre I can see this on Windows too with some effort. Weird, for all duplicates are Macs.
(Mac)The tooltip appears on mouse-over of the favicon, just as a tooltip appears on mouse-over of any of the toolbar buttons. The tooltip appears a set number of pixels below the pointer. Is this coded the same for all the toolbar items, or can the tooltip for the favicon location be changed? Over the location bar (and I mean covering the URL as opposed to above the field in the title bar; comment #26) is better than covering the bookmarks toolbar. I've been programmed to just wait out the tooltip. Setting browser.chrome.toolbar_tips = false is also a workaround. nominating for 3.0 in case it doesn't make a 2.0.0.x release.
Flags: blocking-firefox3?
Flags: blocking-firefox3? → blocking-firefox3+
I suspect sspitzer's right, and this doesn't exist on the trunk, making it hard for it to block Fx3, but with bug 358446 not letting the drag actually work, it's hard to be certain.
Depends on: 358446
You're right, this doesn't exist on trunk. When the favicon is clicked for dragging, the tooltip goes away. Thus allowing d-n-d at the users leisure. Bug 358446 doesn't block creating a bookmark this way. d-n-d of the favicon to bookmarks toolbar works fine on trunk. On branch, one must wait out the tooltip if they want to drop the drag in tooltip covered area of the toolbar. I'd back out my 3.0 blocking request but don't have the power since it's been approved.
Bah, you'd think while I was making sure the tooltip still didn't get in the way, I'd actually drop one, and see that it does now work. beltzner: wanna rescind that blocking-firefox3+, so we don't have to mark this branch-only bug wfm for its trunk status?
No longer depends on: 358446
Target Milestone: Firefox 2 beta2 → ---
Version: unspecified → 2.0 Branch
Clearing blocking flag per prior comments.
Flags: blocking-firefox3+
Target Milestone: --- → Firefox 3 beta1
Summary: Dragging proxy icon results in tooltip, blocking part of the bookmarks toolbar → [1.8 branch] Dragging proxy icon results in tooltip, blocking part of the bookmarks toolbar
Target Milestone: Firefox 3 beta1 → ---
Flags: in-testsuite?
Flags: blocking-firefox3+
Flags: blocking-firefox3+
The only work around for this is to not drop the link until the tool tip goes away. When the tool tip disappears, then the link can be dropped. It's a "real" annoying problem, but that appears to be the only way around it. It would be nice if the tooltip could be moved to a different location instead of right over the bookmark toolbar.
Awesome: after all my protestations that it was fixed by Cocoa widgets, apparently I meant "fixed" - looks like prior to bug 395397 we were ignoring the event triggering the tooltip, rather than properly squelching it, so in 20070919 Firefox or Seamonkey, you can drag the page proxy icon, or selected text from the search bar, without a tooltip, but in 20070920 of both, we're back to having tooltips in the way of drags. In fact, they are worse than I remember the Widget: Mac tooltips being, because they don't just go away - not only do they not time out the way they did in 1.8, they don't disappear until another event pushes them out of the way, so if you drag the page proxy for a cached page (a bug page won't work, but an mxr one will), get the tooltip, get tired of waiting, and just drop in the content window, you get a sticky tooltip until the next event you trigger.
Assignee: mark → joshmoz
Component: Bookmarks → Widget: Cocoa
Flags: blocking-firefox2-
Product: Firefox → Core
QA Contact: bookmarks → cocoa
Summary: [1.8 branch] Dragging proxy icon results in tooltip, blocking part of the bookmarks toolbar → Dragging proxy icon results in tooltip, blocking part of the bookmarks toolbar
Version: 2.0 Branch → Trunk
Flags: blocking1.9?
No longer blocks: 354641
Blocks: 399944
I don't see this problem. Does it still exist for others? Please renominate for blocking1.9 status after confirming that the bug exists and posting up-to-date repro steps.
Flags: blocking1.9? → blocking1.9-
Still same repro steps for Firefox 2.0.0.9 on Mac OS X 10.4.10
WFM on trunk.
Flags: blocking1.8.1.10?
I still see it with: Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.4; en-US; rv:1.9a9pre) Gecko/2007110604 Minefield/3.0a9pre STR: 1) Hover over location bar favicon. Notice tooltip covering a portion of bookmarks toolbar 2) click (tooltip goes away) and drag the favicon over the bookmarks toolbar. Notice tooltip covering a portion of bookmarks toolbar The reappearing tooltip is the bug. If you want to dnd to a place that is covered by that tooltip, then you don't really know where you're dropping. Only show the tooltip on hover over the favicon. This has changed a bit, it seems. Sometimes the tooltip will appear just low enough to expose some of the toolbar. Then you can see the dnd feed back, or if over a folder, it expands.
Flags: blocking1.9- → blocking1.9?
I still see it also on Mac OS X 10.4.10 using Firefox version 2.0.0.9.
-'ing for 1.9. Marking as wanted. This has been around for a while.
Flags: blocking1.9? → blocking1.9-
Flags: wanted1.9+
So, Jesse, Josh: are you not-seeing it on 10.5, meaning that the second regression, from bug 395397, is 10.4-only?
I'm still using 10.4.
I can confirm this bug still exists for Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.1.11) Gecko/20071127 Firefox/2.0.0.11 Mac OS X is 10.4.11 The obvious solution would be to make the tool tip disappear once the cursor begins moving on drag-and-drop - after all, this is how the TITLE attribute "tool tip" behaves.
Or perhaps just delaying the pop up tip to 3 or 5 seconds, instead of 1 second.
I was just going to report this problem but seems like it's been there forever. if tool tip could show up below the bookmark menu bar it should solve the problem? seems like the tool tip is covering / blocking URL to be bookmarked. you can, however, drag it to either left or right of the tool tip to bookmark it.. well the problem is when you have a folder. it becomes two step process. You have to add URL to the bookmark menu bar and then drag that into a folder.. when it is ideal to just be able to drag directly into a folder..
Don't see that anything has changed that makes this now a blocker for 2.0.0.12, but it's still "wanted" on the branch should anyone write a patch.
Flags: blocking1.8.1.12? → blocking1.8.1.12-
This Bug still exists in 3.0b2 (MacOSX 10.5.1) and it's still quite annoying. Please, someone simply remove the ToolTip from the Proxy Icon, delay it's dispay for some secs or hide tooltip at least as soon as the mouse starts moving after being inert for some secs.
This doesn't only happen under OS X. I wanted to file this issue for Windows a long time ago but forgot it. Now when adding a bookmark to the toolbar I saw it again with Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9b4pre) Gecko/2008030206 Minefield/3.0b4pre ID:2008030206. You have to use the tiny space above or under the tooltip to place the bookmark into this folder.
Assignee: joshmoz → nobody
Component: Widget: Cocoa → Widget
OS: Mac OS X → All
QA Contact: cocoa → general
Hardware: Macintosh → All
Still happening in FF3b3 -- this really should be a release blocker: it's pretty embarrassing. Perhaps just removing the tooltip until a better resolution can be found would be a good idea. All the tooltip does at present is further annoy people who are already unable to drag&drop new bookmarks.
Second the removal of the tooltip. What good does it do if you can't drop the bookmark.
Blocks: 301482
Attached patch patch (obsolete) — Splinter Review
This fixes it. There is only one very small issue with the patch. When you drag something and you drop it on the same spot as where you started dragging the, the tooltip doesn't appear when mousemoving in that element. It does work again when mousing out and mousing over again. I guess this could be fixed by adding a draggesture event handler for xultooltiplistener, but I think that's probably overkill for such a minor bug.
Attachment #310472 - Flags: review?(neil)
I was working on a browser-chrome test. It works most of the time when you click the "Run All Tests" button quickly enough. But it doesn't work if you wait a little while. No idea why this happens. It doesn't work at all with --autorun. I did some digging, it seems like the mousemove event isn't fired at all. It seems like some code in viewmanager.cpp is not allowing this event to fire: http://lxr.mozilla.org/seamonkey/source/view/src/nsViewManager.cpp#1232 Perhaps this is caused by the patch for bug 248226? Also, it seems like there is no way to bring an unfocused window to the front, which is also annoying.
Attachment #310474 - Attachment is patch: true
Attachment #310474 - Attachment mime type: application/octet-stream → text/plain
Assignee: nobody → martijn.martijn
Component: Widget → XP Toolkit/Widgets: XUL
QA Contact: general → xptoolkit.xul
Comment on attachment 310472 [details] [diff] [review] patch I don't think this is going to work: we only listen to mouse down events after showing the tooltip. (Neil Deakin could confirm this for you.)
Attachment #310472 - Flags: review?(neil)
Yes, I thought that too, when looking at the code, but that doesn't seem to be the case in practice, fwiw.
I think the bug is as Tracy Walker said in comment #50; the bug is that the tool tip reappears during d-n-d when the mouse is over the fav-icons bar. It's the reappearance that's the problem. The tool tip appears properly when the mouse hovers over the URL icon. Then when the button down event occurs, the tool tip correctly disappears, but when the dragging starts and the mouse moves over the fav-icons bar, the tool tip reappears. This reappearance is the bug. The Mac and Windows versions appear similar except for the Mac's re-appearance of the tool tip when the mouse is dragging the URL over the fav-icons bar. I wonder if there is a tool tip trigger event on the bar or some layer under the bar that causes that toop tip to appear again. The tool tip should not be appearing again.
I think comment #69 is right on the money. There is some urgency to this bug. It is not just an annoyance; it is a dealbreaker. I'm not a programmer (except bits of php) but am an early adopter. I used an alternate browser for at least a year, maybe two, because of this bug. I finally figured out that if I held my cursor below the tooltip until it disappeared again, I could move my bookmark into the bookmark folder that had been obscured. Only then did I start using Firefox and its cousins again. How many non-tekkie users would go to that trouble?
Dumb question: is the particular toolbar in question here (over the favicon, which currently says "Verified by: Equifax" or "This web site provides no identity information") even useful, given that clicking the favicon will provide the same information?
(In reply to comment #75) > Dumb question: is the particular toolbar in question here (over the favicon, > which currently says "Verified by: Equifax" or "This web site provides no > identity information") even useful, given that clicking the favicon will > provide the same information? s/toolbar/tooltip?
Attached patch working browserchrome test (obsolete) — Splinter Review
This is a browserchrome test that works with the attached patch (and fails in various cases without it). This only tested with the --autorun method. Although, it doesn't seem necessary, I still would very much like to know a way of bringing an unfocused window to the front.
Attachment #318659 - Attachment is patch: true
Attachment #318659 - Attachment mime type: application/octet-stream → text/plain
Assignee: martijn.martijn → nobody
I'm not sure Bug 434502 is a duplicate of this. It doesn't really have anything to do with the bookmarks toolbar or with dragging URLs. Also, the proposed fixed in this bug here appears to be to display the tooltip in a different position. I am extremely surprised at that. Why do you want the tooltip to appear at all while you're in the middle of dragging an element? Surely you want the tooltip to appear only if the mouse button is not pressed?
(In reply to comment #81) Well, the general problem is the same: that the tooltip is appearing during a drag-and-drop. The correct solution is to correct the general tooltip behavior, which would solve both the general problem you described and the specific subcase that this bug was originally about.
(In reply to comment #82) Come on, devs! This bug has been around since 2005-10-18, and there are several proposed patches and fixes right here in the comments! Comment #82 looks like a good general solution (although I am not a FF hacker). Point is, can we just get this fixed? It can't be that hard!
Martijn, is your patch somehow ready to send out for review? Further updating summary to make this bug better to find.
Summary: Dragging proxy icon results in tooltip, blocking part of the bookmarks toolbar → Dragging favicon results in tooltip, blocking part of the bookmarks toolbar
Updating summary because the favicon tooltip problem is only one case of a general problem. (Martijn's patch addresses the general problem and is not specific to the favicon tooltip)
Summary: Dragging favicon results in tooltip, blocking part of the bookmarks toolbar → Tooltip should not appear during a drag (e.g., dragging favicon results in tooltip, blocking part of the bookmarks toolbar)
Has anyone noticed that if you just wait a couple seconds, the tool tip disappears?
Brad Williamson said: Has anyone noticed that if you just wait a couple seconds, the tool tip disappears?" Yes. I alluded to that in my 2008-04-12 post. I just now counted the seconds, and it's a 4-5 second wait for me.
> Has anyone noticed that if you just wait a couple seconds, the tool tip > disappears? How is that at all relevant to this bug?
Flags: wanted1.9.0.x+
Flags: blocking1.9.0.1?
Yeesh. I would think that the code would be "easy" (semi-psuedocode): if(click_event == URL button) { while(mouse_button1_down == TRUE) // not a click event but dragging event { verified_tooltip = FALSE; // turn off tooltip & other dialogs** move_icon_drag_n_drop(); // or some such thing for mouse movement } } verified_tooltip = TRUE; //note: I haven't written for windows yet so this is all guessing about 'code'. ** or flag for all pop-up texts like tooltip_text or verified_tooltip or whatever That is, the "tooltip" dialog in Firefox 3.0 should not show up while in drag and drop "mode". I'd swear this is a simple code modification / addition to do. It applies to all tooltip like pop-up stuff except of course the X'd out one (invalid drag location) and similar -icon- not dialog. I know the dialog box / text box disappears after 4-5 seconds but it shouldn't appear until AFTER the button_down stops. I'd read the code but I'm 1st semester (and it's been a while) C++ and 4 years C long past.
(In reply to comment #93) > I would think that the code would be "easy" > And it is. Martijn's patch is a testament to the simplicity of the solution. The problem is basically comment 66 and comment 67. So we could either... 1) Adopt the patch in comment 64 and either ignore comment 66 or otherwise find out why comment 64 works despite the fact that reading the code strongly suggests that it shouldn't work 2) Go with a different approach, such as listening for draggesture and dragend as suggested in comment 64 or listening to drag (which would an even "easier" 4-line patch and wouldn't suffer from the problem of the tooltip not showing if the drag ends at the same spot as the start), but both of these methods involve adding a new listener, and Martijn expressed reservation over that in comment 64
There is a new behavior with FF3 that trips this again. I went to drag a icon from the left-hand side of the location bar (which is possible, yeah!) and the darn 'This web site does not display identity information' tooltip (which BTW pops up way too often when a blank icon is enough) gets right in the way of my drag. Arg.
Johnath: would you be OK removing that tooltip? I think I might be ... if so, file that patch and ask for approval. This doesn't block release. We've shipped many, many versions with this bug, I don't think it's all of a sudden increased in priority.
Flags: blocking1.9.0.1? → blocking1.9.0.1-
OS field should be changed to MAC only. I can't confirm this bug on Linux (Maybe doesn't affect Windows/Vista either).
(In reply to comment #99) > Johnath: would you be OK removing that tooltip? I think I might be ... if so, > file that patch and ask for approval. Removing the tooltip on the site button doesn't solve the general case, which Martijn's patch does. I would prefer to find Neil or whomever we need to chime in on the questions in comment 66, and land the patch (with tests!) that already exists. Is there a reason not to do that? > This doesn't block release. We've shipped many, many versions with this bug, I > don't think it's all of a sudden increased in priority. I agree - it should be fixed, but given that it isn't a release blocker, I think it's worth fixing it properly.
I can confirm comment 98 on Windows XP too. Should I open another bug?
(In reply to comment #102) > I can confirm comment 98 on Windows XP too. > Should I open another bug? No, this bug should cover all platforms and should resolve this issue, once the right course of action is decided, and the appropriate changes are reviewed and landed.
(In reply to comment #103) > (In reply to comment #102) > > I can confirm comment 98 on Windows XP too. > > Should I open another bug? > > No, this bug should cover all platforms and should resolve this issue, once the > right course of action is decided, and the appropriate changes are reviewed and > landed. Thanks for your explanation. I understand.
I think this bug is falling into the realm of 'management decision'. I'm going to investigate the 'command structure' (sorry, Star Trek fan) of Mozilla to see where I can get involved as I believe the company to be of the community orientation (having committees of unpaid public to decide things) rather than the pointy haired boss type of normal corporation and the requisite flaws as such. There are many choices of how to handle this bug. There is also a newsgroup that may be discussing it and I hope to discuss it there as people here should as well (and may be). I used to be SQA and know the concepts but can't get too involved yet. Just because of a problem like this that is popular unfortunately doesn't mean it will be fixed as that is a slippery slope and we would wind up with a "Homermobile" from the Simpsons TV show if you know what I mean! :-)
Never saw this tooltip "This web site provides no identity information" in ff2 and it does indeed annoy me when I drag the icon for bookmarking to my toolbar's folder in ff3. What's the so called tip for anyway..
I just realized that using about:config to disable browser.chrome.toolbar_tips also disables tooltips on images with the "title" attribute. So, even with a workaround problems persist.
Oh for God's sake!! [head hits table] This takes care of it. No wonder there was never a patch as there is a setting flag for it! It's not user changeable but it does fix it. Note when you select that value the right click and select 'toggle' to change it's value. Well, "case closed" although it should be user controlled but if you did that with all the features the options section (tabs and all) would be in the thousands overloading the user. This after all needs to be user friendly. I still say it needs that functional change that when the mouse location goes into 'drag mode' the tool tip should disappear and not on a timer. But that can be debatable. *sigh* Thanks so much John for letting us know about it including about:config which I never knew about!
The bottom line, though, is that functionality is broken, and unless the fix is the default state of the setting, things are still broken "out of the box", and that looks bad to new users. Of course, if you do want the tooltips for some reason, then that feature is still broken, even if the default setting is changed, so obviously something still has to happen here (which seems to be where things are heading... eventually).
Oh wow, I had no idea you guys didn't know about that setting. I read about it on a blog (http://phalkunz.com/2006/10/30/disable-annoying-firefox-tooltips/) and submitted a bug report requesting that this setting be modified (https://bugzilla.mozilla.org/show_bug.cgi?id=440890) But anyway, as I said in my last post, disabling this setting also disables tooltips for images (and presumably other elements) that use the "title" attribute. For example, try hovering your mouse on any xkcd comic. A tooltip should appear with words of wisdom, but does not. So even with after modifying this setting in about:config, you will run into some annoying problems.
This behaveour is still observed in FF3 release on an XP machine... Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9) Gecko/2008052906 Firefox/3.0 Clicking the icon already shows the security information, I don´t believe there are a lot of users that would want to see this information as well as a tooltip while dragging to the bookmarks toolbar... Apart from stating the obvious (annoying!) I am unpleasantly surprised that something seemingly simple to fix has been around for so long without any action being taken apart from a patch that will never be found by the average mainstream user. Devs, get your heads together and fix this!!!
I become used to bookmark with d'n'd of the favicon in the personal bookmark tab on ff2 and worked flawlessly. This simple thing on ff3 it's uneasy to obtain, I'm going back and use Ctrl-d most of the time. This change of behavior is the most annoying for the experienced user and the most confusing for newcomers. Either this bug is underrated or the functionality is overestimated, because you can have a very similar tooltip over the lock icon on the lower-right corner. I would want to ask to disable the favicon tooltip asap and re-enabling it when it is ready, to make user experience more satisfactory.
Component: XP Toolkit/Widgets: XUL → XUL
QA Contact: xptoolkit.xul → xptoolkit.widgets
This has affected Windows also since FF3 betas. Very annoying; really should be a release blocker
This is present and highly obstructive in FF 3.0.1 and above. Attempting to drag the current URL opens the info box over the personal bookmarks toolbar, obscuring the intended folders and usually resulting in failed drops, or, dropping directly into the toolbar vs. the intended folder. Quite simple options for fixes.. removing the infobox on drag, making it more transparent, or, making it pop-up -above- the pointer rather than below.
I no longer see the tooltip interfering with the latest trunk build. As soon as you 'mouse-down' to start the drag on the FavIcon the tooltip goes away, no longer hiding the Bookmarks Toolbar. Not sure when this was fixed/corrected... Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.1b1pre) Gecko/20080904035000 Minefield/3.1b1pre Firefox/3.0 ID:20080904035000
This is not fixed yet. Still there with Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.1b1pre) Gecko/20080904035000 Minefield/3.1b1pre ID:20080904035000
Still seeing this in the latest trunk build on Windows XP.
Still not fixed yet, very annoying bug, glad I found out about browser.chrome.toolbar_tips. Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1a2) Gecko/20080829065003 Shiretoko/3.1a2
STR that reproduces this bug: 1. mouse move on say, Larry. 2. *before* the tooltips shows up, mouse down and star dragging (-> Tooltip shows up.) STR that does not reproduce this bug: 1. drag something on larry 2. wait (-> no tooltip)
Whiteboard: see commment 129
There was also a regression on 23-24 Feb 2005, possibly caused by Bug 125386, where the favicon tooltip didn't disappear anymore on mouseout. - Mousedown on favicon, drag a few pixels, the tooltip (re)appears, then as soon as you leave the image, the tooltip should disappear when the mouse enters a location where you can't drop it. Bug 125386 broke this, although I don't know if it was the last regression because I did only random checks.
> Comment #125 From Ria Klaassen (not reading all bugmail) > 2008-09-08 13:00:54 PDT > *** Bug 454150 has been marked as a duplicate of this bug. *** I think that bug #454150 is NOT a duplicate of this bug, as 454150 specifically notes the website identity tooltip whereas this bug is about user info tips like "drag and drop this icon..." please handle the WEBSITE IDENTITY TOOLTIP vs dragging to the bookmarks toolbar!
(In reply to comment #131) > I think that bug #454150 is NOT a duplicate of this bug, as 454150 specifically > notes the website identity tooltip whereas this bug is about user info tips > like "drag and drop this icon..." It's the identity tooltip and not the identity panel! These are different things. The latter one will be opened by a mouse-up event. So it will not be affected by d&d. Resolving the bug as dupe was correct.
Assignee: nobody → mstange
Status: NEW → ASSIGNED
Attached patch v2 (obsolete) — Splinter Review
I unbitrotted Martjin's patch and simplified the test. Martijn's patch works as-is - the answer to comment 66 is: nsXULTooltipListener::AddTooltipSupport adds a listener for the "mouseout" event, and since nsXULTooltipListener implements nsIDOMMouseListener, it gets MouseDown and MouseUp called, too.
Attachment #310472 - Attachment is obsolete: true
Attachment #310474 - Attachment is obsolete: true
Attachment #318659 - Attachment is obsolete: true
Attachment #344084 - Flags: review?(Olli.Pettay)
Comment on attachment 344084 [details] [diff] [review] v2 > nsXULTooltipListener::MouseMove(nsIDOMEvent* aMouseEvent) > { >- if (!sShowTooltips) >+ if (!sShowTooltips || mIsMouseDown) > return NS_OK; Would it be enough to check here that nsIDragService::getCurrentSession doesn't return any session?
I just tried it, but it didn't work. Maybe mouse move events aren't fired anymore once the drag session has started? I think the current patch works because there's a threshold before a drag session is started, during which the crucial mouse move event fires.
i'm asking blocking because actually we have tooltips on Places menus, when a tooltip appear while you are dragging, as soon as you touch the tooltip with the mouse the menus are dismissed due to dragleave. So dragging in menus has become really hard. If we can't fix this we should disable tooltips in Places menus again.
Flags: blocking1.9.1?
Flags: blocking1.9.1? → wanted1.9.1+
Blocks: 465363
Just to note that this bug is still there, other than the tooltip disappearing after like 5 seconds, which is too much, you still get the yellow tooltip over the bookmark bar and can't place it on folders/bar.
Priority: -- → P1
I am also having this issue, and it is driving me up the walls. I hope this gets fixed soon, because the workaround of disabling all tooltips is not an option for me. Too many sites I visit daily use tooltips to carry information I need.
Roc: while a minor nuisance, the use case from bug 467239 comment 0 is indeed quite irksome and a regression of sorts through the combination of this bug and the site identity panel. Not sure if that affects your prioritization.
Attached patch v3Splinter Review
This works.
Attachment #344084 - Attachment is obsolete: true
Attachment #350959 - Flags: review?(Olli.Pettay)
Attachment #344084 - Flags: review?(Olli.Pettay)
I'd advocate for getting this into 1.9.1 if at all possible, thanks for the patch Markus.
On linux dragging something from outside to a browser window doesn't fire 'dragstart' (and I think that is ok per html 5). 'dragenter' and 'dragover' seem to work ok in that case. Would handling either of those be enough to fix this bug?
(In reply to comment #146) > On linux dragging something from outside to a browser window doesn't > fire 'dragstart' (and I think that is ok per html 5). Is that a problem? As far as I can tell, the tooltip only appears if you start your drag gesture inside the browser window.
(In reply to comment #147) > Is that a problem? It *might* be a problem. Tooltip appears after a timeout, so in some very unlikely case user might have been able to start dragging before tooltip is shown. I have no idea how to reproduce that though. On linux window focus handling depends on user's setting - whether it follows mouse or follows mouse after timeout or needs mouse click or whatever, so the problem might be more likely there than elsewhere. But MouseOut should handle at least most of the problematic cases.
Comment on attachment 350959 [details] [diff] [review] v3 So r=me. Would be great to change C-style casting to C++ static_cast<>.
Attachment #350959 - Flags: review?(Olli.Pettay) → review+
Attachment #350959 - Flags: superreview?(roc)
I'll replace the casts on checkin. There are 16 of them in that file. (In reply to comment #148) > (In reply to comment #147) > > Is that a problem? > It *might* be a problem. Tooltip appears after a timeout I'm not sure I understand. The timeout is started on mouse move. But during a drag no mouse move events are fired. I think this is what happens (with the patch) when the user moves his mouse to the button, presses the mouse button and drags away: 1. User moves mouse over button -> Timeout is started. 2. User presses the mouse button -> Timeout is killed. If the timeout has already fired the tooltip is killed, too. 3. User starts dragging -> A mouse move event fires, timeout is started. 4. User crosses dragstart threshold -> Dragstart event fires, timeout and tooltip are killed. 5. User drags away -> No further mouse move events are sent, so no new timeout is started and no tooltip appears.
Attachment #350959 - Flags: superreview?(roc) → superreview+
Status: ASSIGNED → RESOLVED
Closed: 16 years ago
Resolution: --- → FIXED
Target Milestone: --- → mozilla1.9.2a1
Attachment #350959 - Flags: approval1.9.1?
Is this something we can land on 1.9.1 branch as well?
I think so, the patch should be fairly safe. But it won't hurt to let it bake for a while.
Attachment #350959 - Flags: approval1.9.1? → approval1.9.1+
Verified with Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2a1pre) Gecko/20081212 Minefield/3.2a1pre ID:20081212040742 and appropriate build on OS X. Shouldn't the patch have tests to get allowed for check-in into 1.9.1 branch?
Status: RESOLVED → VERIFIED
Blocks: 469464
No longer blocks: 469464
pushed to mozilla-1.9.1: http://hg.mozilla.org/releases/mozilla-1.9.1/rev/0dc120ad2314 My first attempt at writing a test for this had failed. I guess I'll try again.
Keywords: fixed1.9.1
Verified on 1.9.1 with: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1b3pre) Gecko/20081224 Shiretoko/3.1b3pre ID:20081224042714 Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.1b3pre) Gecko/20081225 Shiretoko/3.1b3pre ID:20081225020447
Fwiw, I would have preferred the first patch as that would allow fixing bug 301482.
This bug has been flagged as wanted1.9.0.x. Markus, can we ask for approval on your latest patch?
Target Milestone: mozilla1.9.2a1 → mozilla1.9.1b3
Attachment #350959 - Flags: approval1.9.0.7?
Can we get an automated test for this? Markus?
Whiteboard: see commment 129 → [needs test] see comment 129
This bug is not fixed, http://i43.tinypic.com/eugb43.jpg Mozilla/5.0 (Windows; U; Windows NT 5.2; en-US; rv:1.9.0.5) Gecko/2008120122 Firefox/3.0.5 Windows XP Pro x64 Edition SP2 / Windows Vista x64 Home Premium Edtition Work around is here: browser.chrome.toolbar_tips = false
It's not checked in the 1.9.0.x tree and that probably will not happen, because this is not a security problem or a stability problem.
So is the bug fixed or not?
Yes, the bug is fixed.
(In reply to comment #166) > So is the bug fixed or not? To be a bit more specific. This bug will be fixed with the upcoming Firefox 3.1b3 and for Firefox 3.2.
still not fixed in 3.05
(In reply to comment #170) > still not fixed in 3.05 Yes, that's known. The fix is only checked into the Firefox3.1 tree and current trunk.
(In reply to comment #170) > still not fixed in 3.05 See comment 165. This bug is fixed on trunk (for Firefox 3.2) and the 1.9.1 branch (for Firefox 3.1) but will likely *not* be fixed in Firefox 3.0.x any time soon, if at all. Please do not comment in this bug with reports that it's not fixed in 3.0.x. We know.
Whiteboard: [needs test] see comment 129 → [needs test] read comment 172 BEFORE commenting (see comment 129 for STR)
(In reply to comment #169) > *** Bug 474311 has been marked as a duplicate of this bug. *** Please see my additional comments regarding bug 474311, that were really separate from the drag and drop issues.
Comment on attachment 350959 [details] [diff] [review] v3 Approved for 1.9.0.7, a=dveditz for release-drivers.
Attachment #350959 - Flags: approval1.9.0.7? → approval1.9.0.7+
Checking in layout/xul/base/src/nsXULTooltipListener.cpp; /cvsroot/mozilla/layout/xul/base/src/nsXULTooltipListener.cpp,v <-- nsXULTooltipListener.cpp new revision: 1.72; previous revision: 1.71 done
Keywords: fixed1.9.0.7
Target Milestone: mozilla1.9.1b3 → mozilla1.9.2a1
Verified fixed with Mozilla/5.0 (Macintosh; U; Intel Mac OS X 10.5; en-US; rv:1.9.0.7pre) Gecko/2009021704 GranParadiso/3.0.7pre.
Bug still exists in Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2a1pre) Gecko/20090322 Minefield/3.6a1pre
(In reply to comment #184) > Bug still exists in Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2a1pre) > Gecko/20090322 Minefield/3.6a1pre Lets discuss the Linux part over in bug 482070.
Does the first patch in this bug help for linux?
(In reply to comment #187) > Does the first patch in this bug help for linux? We should move the discussion over to bug 482070. But as a final statement I don't know because I don't build on Linux. Probably worth to create a try server build.
This bug was not fixed for Linux, please reopen it (see Bug 506783).
(In reply to comment #191) > This bug was not fixed for Linux, please reopen it (see Bug 506783). No, this bug is verified fixed, you should go to bug 482070 for Linux.
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: