apm388p@vaxc.cc.monash.edu.au (11/07/90)
Hi There, [I'm not sure if this made it out of here on the first go, so this is a repost] I am looking for a simple utility which will enable me to *clear the screen* (by sending ANSI <esc>[2J ?) by issuing a keystroke, for example while I am in graphics mode of NCSA Telnet or Kermit. (Perhaps this last requirement is irrelevant because presumably the keystroke would be intercepted by the PC before it is sent off.) Is such a thing possible or available anywhere? I looked through SIMTELs screen directory but nothing seemed appropriate. I'm not after a command-line clear as I want it to operate while I'm running a program. Thanks in advance, Michael Page, Mathematics Department, Monash Uni, Melbourne, Oz apm388p%monu1.cc.monash.edu.AU@uunet.uu.NET
roelofs@nas.nasa.gov (Cave Newt) (11/12/90)
Michael Page, apm388p@vaxc.cc.monash.edu.au, writes: >I am looking for a simple utility which will enable me to *clear the screen* >(by sending ANSI <esc>[2J ?) by issuing a keystroke, for example while I am >in graphics mode of NCSA Telnet or Kermit. (Perhaps this last requirement >is irrelevant because presumably the keystroke would be intercepted by the >PC before it is sent off.) It's not clear to me whether you want something to send a clear-screen signal to the host computer or simply a local, PC-based screen-clearer. If the latter, here's a small (and pretty stupid) TSR to do it for you. All this does is check the current video mode and then do a reset to that mode (so if you're in graphics mode, you'll stay that way). It doesn't do anything beyond that, however, so if you're in a 35-line editor mode, you'll lose the bottom 10 lines (and maybe the cursor, too) when the mode gets reset. But hey--that's what caveat emptor is all about. :) Appended is a uuencoded .com file, then source (so you can compare with a debug dump and convince yourself there're no trojans lurking about), and follow-ups to alt.sources.d. Greg roelofs@amelia.nas.nasa.gov begin 600 hotclear.com MZS.0`````/M0Y&`\+G40N$``'H[8H!<`'R0//`IT!E@N_RX#`;`@YB!3M`_- J$#+DS1!;6,^X"37-(8D>`P&,!@4!M"6Z!P'-(:$L`([`M$G-(;HU`<TG ` end ;------------------------------------------------------------------------------ ; hotclear.asm G. Roelofs, 11 Nov 90 ; ; TSR to clear screen when hotkey is pressed. Not very bright: clears screen ; by resetting video mode, which will nuke colors, fonts, etc. (in particular, ; 35-, 43-, 50- and 60-line modes get changed back to 25-line mode). In addi- ; tion, hotclear doesn't bother to check and see if it's already resident, nor ; does it have any mechanism for turning itself off (both are pretty trivial ; to add, however). Default hotkey is ALT-LEFTSHIFT-C, but it may be changed. ;------------------------------------------------------------------------------ BIOS_DATA segment at 40h ; BIOS data area org 17h kb_flag db ? ; keyboard status (shift) BIOS_DATA ends CSEG segment assume cs:CSEG, ds:nothing, es:nothing org 2Ch envir_seg dw ? ; segment addr of environment org 100h ; .com program begin: jmp initialize ; code to make TSR ; Equates, data ; ------------- EOI equ 20h ; end-of-interrupt signal INT_CTRL equ 20h ; 8259 port address KB_DATA equ 60h ; keyboard data port KB_CTRL equ 61h ; keyboard control port KB_STAT equ 64h ; keyboard status port SHIFT_MASK equ 00001111b ; overall shift mask ALT_MASK equ 00001000b ; alt shift key is down CTRL_MASK equ 00000100b ; ctrl shift key is down LEFT_MASK equ 00000010b ; left shift key is down RIGHT_MASK equ 00000001b ; right shift key is down HOTSHIFT equ ALT_MASK+LEFT_MASK ; (or just ALT_MASK, etc.) HOTKEY equ 2Eh ; "c" scan code oldint9h label dword oldint9h_addr dw 2 dup (?) ; old keyboard handler vector ; New interrupt 9h handler: if not hotkey, let old one do the work ; ----------------------------------------------------------------- newint9h proc near sti ; set interrupt enable flag push ax ; save AX in al, KB_DATA ; get scan code from keyboard cmp al, HOTKEY ; our hotkey? jne outta_here ; nope: we're outta_here mov ax, BIOS_DATA ; yep: now check for our shift push ds ; save DS mov ds, ax ; now pointing at BIOS data area assume ds:BIOS_DATA ; so assembler knows what's up mov al, kb_flag ; get keyboard status pop ds ; restore DS assume ds:nothing ; oh, nothing... and al, SHIFT_MASK ; only alt, ctrl, normal shifts cmp al, HOTSHIFT ; see if ours is (are) down je our_thing ; yep: do our_thing outta_here: pop ax ; nope: we're outta_here jmp oldint9h ; let old handler handle it ; The people have spoken: do a hard reset of the video mode to ; whatever it is now. Then leave. our_thing: mov al, EOI ; end-of-interrupt signal (else out INT_CTRL, al ; system locks up) push bx ; gets modified by video call mov ah, 0Fh ; current-video-state function int 10h ; video interrupt xor ah, ah ; set-video-mode function int 10h ; banzai pop bx ; restore regs pop ax ; iret ; hot damn: all done newint9h endp ; Reset interrupt vector, free environment, and become resident ; ------------------------------------------------------------- initialize proc near assume ds:CSEG mov ax, 3509h ; get current keyboard handler int 21h ; vector mov oldint9h_addr, bx ; save it for later use mov oldint9h_addr[2], es ; mov ah, 25h ; set vector to our routine mov dx, OFFSET newint9h ; int 21h ; mov ax, envir_seg ; get environment segment mov es, ax ; mov ah, 49h ; free allocated memory: int 21h ; if fails...too bad mov dx, OFFSET initialize ; end of resident code int 27h ; become TSR and return to DOS initialize endp CSEG ends end begin ; non-TSR entry point