29050 2009-09-08 14:03:09 -0700 PageUp in at top of multi-line text box scrolls page 2017-07-18 08:29:38 -0700 1 1 1 Unclassified WebKit New Bugs 528+ (Nightly build) PC Windows XP UNCONFIRMED http://gmail.com P2 Normal --- 0 jasneet webkit-unassigned jasneet mrowe oldest_to_newest 145525 0 jasneet 2009-09-08 14:03:09 -0700 I Steps: 1. Login to gmail.com 2. reply to a message with a long thread 3. in the compose box, hit PageDown to page down a few times 4. hit PageUp a few times to go back to the top II Issue: Page up the top of the text box, and then when it's at the top, the next PageUp scrolls the window. Expected : Page up to the top of the text box, then stop. III Other Browsers: IE7: ok FF3: ok IV Nightly tested : 48037 Bug in Chromium : http://code.google.com/p/chromium/issues/detail?id=20549 145542 1 mrowe 2009-09-08 14:39:04 -0700 I believe that this is the expected behavior. 145573 2 ojan 2009-09-08 15:34:00 -0700 pageup/pagedown should have the same expected behavior as up/down, no? Why does it make sense for pageup to scroll the page, but no the up arrow? Is this platform specific somehow? FWIW, Firefox on the Mac does not scroll the page on Mac or Windows. 145631 3 mrowe 2009-09-08 19:36:44 -0700 (In reply to comment #2) > pageup/pagedown should have the same expected behavior as up/down, no? On Mac OS X, up/down move the caret while pageup/pagedown do not. These are quite different behaviors. Pageup/pagedown are more equivalent to scrolling via the scrollwheel, which will scroll the area under the cursor until it reaches the top/bottom and then scroll the containing area. > Why does it make sense for pageup to scroll the page, but no the up arrow? The caret clearly cannot move beyond the start of the editable region. Pageup doesn't move the caret. > Is this platform specific somehow? FWIW, Firefox on the Mac does not scroll the > page on Mac or Windows. It may be. I don't know what the Windows behavior is. 145647 4 ojan 2009-09-08 21:59:28 -0700 (In reply to comment #3) > (In reply to comment #2) > > pageup/pagedown should have the same expected behavior as up/down, no? > > On Mac OS X, up/down move the caret while pageup/pagedown do not. These are > quite different behaviors. Pageup/pagedown are more equivalent to scrolling > via the scrollwheel, which will scroll the area under the cursor until it > reaches the top/bottom and then scroll the containing area. Ah. Right. I forgot about that. I buy the argument that it's more like the scroll wheel on Mac. On Win/Linux pageup/pagedown do (should!) move the cursor, so it seems like it should match up/down. Seem reasonable?