Re: Cursor movement bug in math insets with split views
On 7/04/2014 2:21 p.m., Richard Heck wrote: On 04/06/2014 08:55 PM, aparsloe wrote: On 7/04/2014 12:47 p.m., aparsloe wrote: In a new document open a math inset and enter e=mc^2. Split the view into left and right views. In the right view ensure the cursor lies *outside* the math inset. Now in the left view try to put the cursor *with the mouse* between, say, m and c. I find the cursor remains stuck at the start of the inset (before the e). It does respond to the arrow keys. However, if in the right view I place the cursor *inside* the math inset, then on the left the cursor once more responds to the mouse. The problem is present for me in both 2.1.0beta2 and 2.0.7 on Windows 7, but it has been there for a year or more, including on a Windows Vista computer (with a different mouse) and earlier members of the 2.0.x series. Andrew I should have added that the left and right views behave differently and can't be interchanged to show the bug. If the cursor is placed outside the inset in the left view, it will respond to the mouse in the inset on the right. When the screen is split into upper and lower halves, the top half behaves like the left view and the bottom half like the right view. Can't reproduce on Fedora 17 (a bit out of date, but Qt 4.8.4). Richard I've just checked on my older computer (Vista) which has rc1 installed. The problem is still there. Also, in a long document with a split view, if the formula in the right view scrolls off screen (it seems to need to be about two lines off screen), mouse-driven cursor movement in the formula in the left view becomes available again. (I've also put this on the user's list to see if someone else with a Windows machine can reproduce it.) Andrew
Re: Cursor movement bug in math insets with split views
On 7/04/2014 2:21 p.m., Richard Heck wrote: On 04/06/2014 08:55 PM, aparsloe wrote: On 7/04/2014 12:47 p.m., aparsloe wrote: In a new document open a math inset and enter e=mc^2. Split the view into left and right views. In the right view ensure the cursor lies *outside* the math inset. Now in the left view try to put the cursor *with the mouse* between, say, m and c. I find the cursor remains stuck at the start of the inset (before the e). It does respond to the arrow keys. However, if in the right view I place the cursor *inside* the math inset, then on the left the cursor once more responds to the mouse. The problem is present for me in both 2.1.0beta2 and 2.0.7 on Windows 7, but it has been there for a year or more, including on a Windows Vista computer (with a different mouse) and earlier members of the 2.0.x series. Andrew I should have added that the left and right views behave differently and can't be interchanged to show the bug. If the cursor is placed outside the inset in the left view, it will respond to the mouse in the inset on the right. When the screen is split into upper and lower halves, the top half behaves like the left view and the bottom half like the right view. Can't reproduce on Fedora 17 (a bit out of date, but Qt 4.8.4). Richard I've just checked on my older computer (Vista) which has rc1 installed. The problem is still there. Also, in a long document with a split view, if the formula in the right view scrolls off screen (it seems to need to be about two lines off screen), mouse-driven cursor movement in the formula in the left view becomes available again. (I've also put this on the user's list to see if someone else with a Windows machine can reproduce it.) Andrew
Re: Cursor movement bug in math insets with split views
On 7/04/2014 2:21 p.m., Richard Heck wrote: On 04/06/2014 08:55 PM, aparsloe wrote: On 7/04/2014 12:47 p.m., aparsloe wrote: In a new document open a math inset and enter e=mc^2. Split the view into left and right views. In the right view ensure the cursor lies *outside* the math inset. Now in the left view try to put the cursor *with the mouse* between, say, "m" and "c". I find the cursor remains stuck at the start of the inset (before the "e"). It does respond to the arrow keys. However, if in the right view I place the cursor *inside* the math inset, then on the left the cursor once more responds to the mouse. The problem is present for me in both 2.1.0beta2 and 2.0.7 on Windows 7, but it has been there for a year or more, including on a Windows Vista computer (with a different mouse) and earlier members of the 2.0.x series. Andrew I should have added that the left and right views behave differently and can't be interchanged to show the bug. If the cursor is placed outside the inset in the left view, it will respond to the mouse in the inset on the right. When the screen is split into upper and lower halves, the top half behaves like the left view and the bottom half like the right view. Can't reproduce on Fedora 17 (a bit out of date, but Qt 4.8.4). Richard I've just checked on my older computer (Vista) which has rc1 installed. The problem is still there. Also, in a long document with a split view, if the formula in the right view scrolls off screen (it seems to need to be about two lines off screen), mouse-driven cursor movement in the formula in the left view becomes available again. (I've also put this on the user's list to see if someone else with a Windows machine can reproduce it.) Andrew
