From: Roger H. <rog...@nt...> - 2003-07-01 20:30:22
|
On 30 Jun 2003, David Richards wrote: > Backspace in a DEC keyboard is a DELETE on all other platforms. Hrm, it's very important to get our terms right here. The LK201 keyboard has no key labelled "delete". There's the pad-Remove key (on the editing keypad) which generates some escape sequence that causes VMS to delete the character *following* the cursor. There's CTRL-H, of course, which generates BS, which causes VMS to move the cursor back one space and deletes nothing. There's the key at the upper right of the main key cluster (where the "backspace" key is on a PC-10x board) which is marked with a single glyph looking something like "<X|", generates DEL (0x177, Delete/Rubout), and which causes VMS to delete the character under the cursor and then move the cursor back one space. There's F12, also labelled "(BS)", which some software treats as some sort of backspace, but I never used it enough to learn precisely how it functions; it generates the (DEC-)standard F12 escape sequence. There are layers of stuff going on here, and you have to keep them straight or you'll get a muddle. (I should know: I have an xmodmap that attempts to make a PC104 keyboard act like an LK201, and while I have gotten used to its quirks it would really confuse anybody who has used *either* of those keyboards straight. :-) > Try using control-h for a backspace. Assuming that BS does what you want done. See the difficulty? --------------------------------------------- Yes, the key I am talking about is the "<X|" one. I have got the keymapping file altered so I am returning a PC backspace scancode on that key. I will try Control-H too, I did not think of that. However, the problem is only notepad - in word97, excel, wordpad and the cmd window all operate correctly - the key deletes the character to the left of the cursor, and the cursor moves left. In notepad the character is removed but all the following characters stay there, overwritten - until the display is refreshed. I think the problem might be something to do with the RECT order (looking at order.c) With debug on, I can see that the rect after the key is pressed has reasonable x,y starting, then very large following cx, cy values. Of course, the RECT orders are working correctly in most cases - this is the only display discrepancy I have found so far. I have not yet determined whether its absolute or relative coordinates for this cx, cy, but will try a bit more debug. Roger Haxby |