|
|
Does Emacs exist on this system, and if so, where is it?
19 responses total.
We have Jove, which is kind of similar... Do jove at a shell prompt to start it up, or teachjove for a tutorial...
thanks
I played around with emacs on another system today. I feel like throwing out at least half the commands and re-writing the key bindings so I don't have to remember so darn many keys (and maybe make them a little more mnemonic). It'd be great for me if the keys could be made to do the same sort of functions as Qedit...
Kentn: This doesn't help you with Emacs, but if you want to run Jove on here, grab ~power/.joverc... It's set up for an Apple - arrow keys work, and everything! The term prog you're using might make a difference (do you use ProTerm?), but it's pretty obvious what you have to change, if you edit it.....
Thanks, power. I'll give it a look. Sometimes I do log in on Grex with my Apple //e and ProTerm.
<<and does Johnathon object to you swiping his machine??>
It would work with another terminal also, as long as it was vt100, and .... no wait.. I take it back, it should work with any other setup that you can handle all the screen manipulation with.... Hmmmm... (I think).... N-E-way, just go try it, it's nice :)... The arrow keys and the delete key work right and everything! (even if delete is your interrupt)...
<<Jonathan doesn't understand terminal programs and modems, and
even so, he has a II+ to play with, which he hardly does. The
//e was my first computer and will always be so...>>
<<ditto - and thankxx for the spelling correction>>
Since this item is about emacs, and appears stale, I would like to commandeer it to discuss a problem I am having recently in jove. I have been using jove for almost 3 months, and I'm quite happy with it, or I was until it started misbehaving. This began a week or so ago. The problem is that when I have a screen with some text on it and I place the cursor in the middle somewhere and type <cr> to insert a return and thus split the line, Jove does this correctly inside its buffer but fails to move the lines below the insertion point down a line on the screen. ^l (redraw) isn't adequate to fix the screen, because the lines have differing lengths I have to type ESC ^l (clear&redraw) to get my screen to be sane again. It seems to me this used to work fine. Did we get an up(down) grade? I use it from tcsh, with a .joverc which turns on auto-indent. Thanks to anyone who has the clue - it's annoying.
One suggestion: can you emulate another terminal type & try it with that (log on again & specify the new type)? I have problems that sound a little like this in vi and elm, and they're definitely related to my emulation. (But of course, it may be a bug in jove as such, too.)
Thank you davel. You caused me to change my mindset. As soon as I realized that it might be the terminal emulation and not Jove, I also realized that I had changed my terminal emulator a few weeks ago also. I had been using Kermit 0.9(40) which works fine. I switched to MacKermit 0.99(185) which has his problem. I will now report it to the MacKermit developers at Columbia. Thanks again.
It might help them if you were to send a copy of the termcap entry (or better yet, that plus some sort of completely unfiltered copy of the offending data stream produced by Jove).
Good point dave, but I'm at my knowledge limit wrt unix. I do not know what a termcap entry is. I use stty to set my terminal to emulate vt100 when I log in, is that what you're getting at? and...I need a good way to capture the offending data stream. Hmmm. I guess that means running a different comms, because Kermit only does one thing...vt100. Is there a clever way to capture the stream on grex? We may not need this, as the emulation failure is probebly obvious once it's called to their attention. Still..you're right I should find out exactly what sequence is failing.
You may be able to pipe Jove's output to a file (or even through tee); I don't know. If you're setting your term type through the normal login question, you should be able to find out your termcap settings by doing env (or you may want to do env | grep TERMCAP). (Um. That's sh; I'm not sure what the equivalent is for csh. And for vt100 the settings I get run to about 250 characters at least.) If you try piping output through tee or something, I'd keep your session as short as possible - and lots of luck.
I piped my output through tee. I should have thought of that. It worked flawlessly, producing a teed file which captured the effects of my editing operations. When I cat teed while running old MacKermit, I see what the editing session should have looked like. When I cat teed through the new MacKermit, the flaw is visible. I can cat -vt teed to see what the characters are that it loses on. Thanks for the help.
For vt100 phreaks, the command vt100 code jove was issuing appears to be ESC [ <m> ; <n> r where m and n are 8 bit numbers. I suspect the r is the function that says to scroll lines m through n down 1 on the screen, and this is what is failing. I will pass this on to MacKermit developers at Columbia.
"ESC [ <m> ; <n> r" is the "set scrolling region" command -- it is supposed to set lines m through n as the scrollable section of the screen, locking all other lines in place. So e.g. if the cursor is on line n and you send a linefeed character, only lines m through n scroll. I'm sure Jove tries to use this command to implement scrolling in windows.
Thanks, John. I would have sounded more knowledgable when I mailed my bug report to Columbia if I had known that. I think they can dope it out with what I've told them, though.
Response not possible - You must register and login before posting.
|
|
- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss