Steps to reproduce: 1. Over a selected toolbar button (back, forward, etc.), push the left mouse button down. 2. Move the mouse outside of the button (drag). 3. While still holding the mouse button down, alt-tab to another app, then alt- tab back. Result: toolbar button is still "depressed" (a pixel or a few lower and to the right)
I can't duplicate this for the Back, Forward buttons on the left side, but I can see it if I click-hold the Search <titledbutton> on the right, then drag off the button, and then ALT-TAB -- yep, the Search button stays depressed (i.e., in its :active style). [Note: there is a bug for <button> and :active in the trunk builds that prevents one from seeing the Back,Forward buttons depress at all]. At any rate, this is related to bug #30384 "Dragging out of toolbar button while clicking leaves button depressed".
reassigning to evaughan for testing in his latest bits.
Looks like you territory
Even the search button is getting restored properly when you move the mouse outside the button while holding down the mouse button. I'm using today's win32 build. Marking worksforme.
Verified Win98 2000092808
Still broken. 1. Mousedown on reload 2. Alt-tab 3. Mouseup 4. Switch back 5. Move mouse cursor back over reload Result: reload in "depressed" state when the cursor is over it Expected: reload in "hover" state when the cursor is over it Clicking in an empty area of the chrome (taskbar, navigation toolbar, etc) makes it start acting correctly again, but clicking in the content area doesn't.
This bug has been marked "future" because the original netscape engineer working on this is over-burdened. If you feel this is an error, that you or another known resource will be working on this bug,or if it blocks your work in some way -- please attach your concern to the bug for reconsideration.
Houston, we have a problem...it seems that Windows doesn't notify us of an WM_LBUTTONUP if you mouseup after alt+tabbing out of the document. We could fix this in a probably safe but very, very sucky way by resetting the hover/active content states when the window is activated (for a reason still unknown to me, doing so on deactivation doesn't work, even though it gets hit). Dan, Chris, what do you guys think?
*** This bug has been marked as a duplicate of 29405 ***