cotner@ronzoni.berkeley.edu (Carl F. Cotner) (02/19/91)
I am wondering why there is no available msdos port of GNU emacs. The obvious answer seems to be emacs' size. If this is indeed the reason, how much memory would emacs need? Considering that there is an Atari ST port which I assume is smaller than 1 meg, it would seem that one should be able to port it to msdos if you used a dos extender. Has anyone done this? It would sure be nice to have emacs at home. Carl Cotner cotner@math.berkeley.edu
ktw@cbnewsb.ATT.COM (Ken Wolman) (02/19/91)
>>>>> On 19 Feb 91 02:51:23 GMT, cotner@ronzoni.berkeley.edu (Carl F. Cotner) said:
Carl> I am wondering why there is no available msdos port of GNU emacs. The obvious
Carl> answer seems to be emacs' size. If this is indeed the reason, how much memory
Carl> would emacs need? Considering that there is an Atari ST port which I assume is
Carl> smaller than 1 meg, it would seem that one should be able to port it to msdos
Carl> if you used a dos extender. Has anyone done this? It would sure be nice to
Carl> have emacs at home.
This probably won't answer your questions, but what the hell?... I've
tried a variety of MS-DOS Emacs packages--MicroGnuEmacs, Freemacs,
MicroEmacs, and Jove--and the closest to the "real thing" that I've
found in terms of command structure and handling is Freemacs. It is
available via anonymous ftp from grape.ecs.clarkson.edu: the last time
I looked it was in /pub/Freemacs, but don't quote me. It's one
serious flaw is its ability to handle large documents, i.e., anything
over 64K: it just won't do it.
--
Ken Wolman |"At age 35, Paul Gauguin still worked
AT&T Bell Laboratories | as a teller in a Paris bank."
Short Hills, NJ | "At age 35, Mozart was dead."
(201)564-2866--hlwpk!ktw| --NYC subway graffiti, late 1960shalpern@casbah.acns.nwu.edu (David Halpern) (02/19/91)
In article <KTW.91Feb19072720@cbnewsb.ATT.COM> ktw@cbnewsb.ATT.COM (Ken Wolman) writes: > >This probably won't answer your questions, but what the hell?... I've >tried a variety of MS-DOS Emacs packages--MicroGnuEmacs, Freemacs, >MicroEmacs, and Jove--and the closest to the "real thing" that I've >found in terms of command structure and handling is Freemacs. It is >available via anonymous ftp from grape.ecs.clarkson.edu: the last time >I looked it was in /pub/Freemacs, but don't quote me. It's one >serious flaw is its ability to handle large documents, i.e., anything >over 64K: it just won't do it. > I've tried freemacs(from the Clarkson site) on my Zenith 386 pc and had lots of problems. My computer kept on crashing for no good reason. I wonder whether anyone else has experienced the same type of problems. May be there are other flwas besides the 64k barrier that I'm not aware of. Does anyone know whne the next version of freemacs coming out? For those involved in the development of GNU Emacs, it would be great to have a version for pcs. I'm sure that there are lots of people interested. David Halpern halpern@casbah.acns.nwu.edu Northwestern University Evanston Illinois
nelson@sun.soe.clarkson.edu (Russ Nelson) (02/20/91)
In article <3710@casbah.acns.nwu.edu> halpern@casbah.acns.nwu.edu (David Halpern) writes:
I've tried freemacs(from the Clarkson site) on my Zenith 386 pc and had
lots of problems. My computer kept on crashing for no good reason.
I wonder whether anyone else has experienced the same type of problems.
Interesting. I run Freemacs on a Z-386 all the time, no problem.
Perhaps you have a TSR that interferes with it in some way? Could you
send me an email message detailing your problems? Ditto for anyone
else having problems with Freemacs...
May be there are other flwas besides the 64k barrier that I'm not aware
of. Does anyone know whne the next version of freemacs coming out?
I haven't done anything with Freemacs for quite a while. The packet
drivers (a set of mostly Ethernet drivers that can be used by [most]
TCP/IP packages and Novell at the same time) have been sucking up most
of my free software time. I have been struggling with learning C++ so
that I can write Freemacs 2, which will have undo and large buffers.
I can't give you any date on when it will be ready.
--
--russ <nelson@clutx.clarkson.edu> I'm proud to be a humble Quaker.
It's better to get mugged than to live a life of fear -- Freeman Dyson
I joined the League for Programming Freedom, and I hope you'll join too.cbrown@eeserv1.ic.sunysb.edu (Charles T Brown) (02/20/91)
In article <3710@casbah.acns.nwu.edu> halpern@casbah.acns.nwu.edu (David Halpern) writes: >In article <KTW.91Feb19072720@cbnewsb.ATT.COM> ktw@cbnewsb.ATT.COM (Ken Wolman) writes: >> >>This probably won't answer your questions, but what the hell?... I've >>tried a variety of MS-DOS Emacs packages--MicroGnuEmacs, Freemacs, >>MicroEmacs, and Jove--and the closest to the "real thing" that I've >>found in terms of command structure and handling is Freemacs. It is >>available via anonymous ftp from grape.ecs.clarkson.edu: the last time >>I looked it was in /pub/Freemacs, but don't quote me. It's one >>serious flaw is its ability to handle large documents, i.e., anything >>over 64K: it just won't do it. >> >I've tried freemacs(from the Clarkson site) on my Zenith 386 pc and had >lots of problems. My computer kept on crashing for no good reason. >I wonder whether anyone else has experienced the same type of problems. >May be there are other flwas besides the 64k barrier that I'm not aware >of. Does anyone know whne the next version of freemacs coming out? > > >For those involved in the development of GNU Emacs, it would be great to >have a version for pcs. I'm sure that there are lots of people >interested. Umm. A friend of mine, and I, are considering porting it to OS/2 ASAP (as soon as I get an OS/2-running machine! :-); but to the question of porting it to DOS, it just really isn't possible. I mean, what _I_ like about GNU Emacs is the ability to run a sub-shell in it, and feed it input; you want to try writing that in DOS?! Freemacs is really the best thing out so far; but the problem is, it is incredibly slow... If anyone is interested in porting it to DOS, give me a mail; I'd be interested in seeing how they'd do it... :-) > >David Halpern halpern@casbah.acns.nwu.edu >Northwestern University >Evanston Illinois -- "Never put off until tomorrow, that which can be done the day after tomorrow" -- C. Titus Brown, anonymous student, brown@max.physics.sunysb.edu UNIX is good, you say? Which UNIX, say I!
wicklund@intellistor.com (Tom Wicklund) (02/20/91)
In <1991Feb19.025123.12089@agate.berkeley.edu> cotner@ronzoni.berkeley.edu (Carl F. Cotner) writes: >I am wondering why there is no available msdos port of GNU emacs. The obvious >answer seems to be emacs' size. If this is indeed the reason, how much memory >would emacs need? Considering that there is an Atari ST port which I assume is >smaller than 1 meg, it would seem that one should be able to port it to msdos >if you used a dos extender. Has anyone done this? It would sure be nice to >have emacs at home. I've looked a bit at the source to GNU emacs. The primary problems I see to an MSDOS port are: 1. Emacs reads the full source file into memory and manipulates it there. This results in very large memory use. Emacs on a UNIX system can grab several megabytes during routine editing of relatively small files. 2. GNU emacs is not truely portable. Rather it appears to be written to run on 32 bit machines. A 32 bit int is built into the code (or at least, 32 bit minimum int size). 3. The use of 24 bit pointers complicates MSDOS segment:offset addressing. A practical MSDOS port of GNU emacs would probably require replacing the source file handling with a scheme which swaps text to disk along with a similar ability to swap elisp code back to disk. However, this just handles the memory size issue, it doesn't deal with 32 bit ints.
cjeffery@cs.arizona.edu (Clinton Jeffery) (02/20/91)
From article <1991Feb19.204358.16742@sbcs.sunysb.edu>, by cbrown@eeserv1.ic.sunysb.edu (Charles T Brown): > > Freemacs is really the best thing out so far; but the problem is, it > is incredibly slow... Gee, Freemacs is hand-coded in 8086 assembly language! In what way is it slow? Russ Nelson's program is (in my opinion) a high-quality piece of art; I only tossed it out because I wasn't willing to learn its (weird) extension language... -- | Clint Jeffery, U. of Arizona Dept. of Computer Science | cjeffery@cs.arizona.edu -or- {noao allegra}!arizona!cjeffery --
elliss@kira.uucp (Stew Ellis) (02/20/91)
cbrown@eeserv1.ic.sunysb.edu (Charles T Brown) writes: [stuff] >Umm. A friend of mine, and I, are considering porting it to OS/2 ASAP >(as soon as I get an OS/2-running machine! :-); but to the question I doubt it can be ported to OS2 either, because it depends on 32-bit ints I have heard. Several people looked into porting it to 16-bit UNIX/Xenix and gave up. Depending on the OS for virtual memory instead of handling its own swapping will also be a big loss in DOS or OS/2. >of porting it to DOS, it just really isn't possible. I mean, what >_I_ like about GNU Emacs is the ability to run a sub-shell in it, >and feed it input; you want to try writing that in DOS?! Microemacs, jove and I believe freemacs(I don't use freemacs because I can't live with the 64k file limit) allow you to send a file or buffer or just a region out for various kinds of dos services such as compiling and parsing errors, sorting a list of names or whatever you could do at the dos prompt that is not specifically interactive. I routinely compile C code from inside uemacs or jove and use the editor to step through the errors with the editor. I and several other people wrote macros for this for uemacs, but the code is built into jove. At one time Dan Lawrence included a macro for uemacs that allowed you to run an interactive shell in one editor window. The problem with all of these was that they were synchronous because of the lack of multitasking. Jove is able to run interactive shells asynchronously on even the stupidest versions of UNIX. If you do not mind spending a little bit of money (about $120), you have not lived until you have seen Epsilon running an asynchronous subshell in the other window while you continue to edit a file in the current window, all on vanilla DOS. >Freemacs is really the best thing out so far; but the problem is, it >is incredibly slow... >If anyone is interested in porting it to DOS, give me a mail; I'd >be interested in seeing how they'd do it... :-) >> >>David Halpern halpern@casbah.acns.nwu.edu >>Northwestern University >>Evanston Illinois >-- >"Never put off until tomorrow, that which can be done the day after tomorrow" > -- C. Titus Brown, anonymous student, brown@max.physics.sunysb.edu > UNIX is good, you say? Which UNIX, say I! R. Stewart (Stew) Ellis GMI Engineering & Management Institute Flint, MI
cox@sun.software.org (Guy Cox) (02/21/91)
In article <906@caslon.cs.arizona.edu>, cjeffery@cs.arizona.edu (Clinton Jeffery) writes: > From article <1991Feb19.204358.16742@sbcs.sunysb.edu>, by cbrown@eeserv1.ic.sunysb.edu (Charles T Brown): > > > > Freemacs is really the best thing out so far; but the problem is, it > > is incredibly slow... > > Gee, Freemacs is hand-coded in 8086 assembly language! In what way > is it slow? Russ Nelson's program is (in my opinion) a high-quality > piece of art; I only tossed it out because I wasn't willing to learn > its (weird) extension language... > -- > | Clint Jeffery, U. of Arizona Dept. of Computer Science > | cjeffery@cs.arizona.edu -or- {noao allegra}!arizona!cjeffery > -- Wrong-OH on the slow! The only thing slow about freemacs is loading it off the disk and I'm running a 10MhZ Xt. It is the only editor that has prevented me from becoming a full blown schizophrenic (I can use essentially the same editor at home as at work). It's the only editor that I've used since SPF under IBM's MVS ( when people ask me "what's up" I still respond "PF7") that I've felt had all the features that I've needed in an editor and if freemacs doesn't have the features that I need I can add them. True freemacs can pose problems in the compile edit cycle unless you have the latest compile.min .... // //Remember; Tuesday is Soylent green day! // Guy O. Cox, Jr. Software Productivity Consortium. 2214 RockHill Rd Herndon, VA 22090 703-742-7219 cox@software.org
bill@mwca.UUCP (Bill Sheppard) (02/21/91)
In article <1991Feb19.204358.16742@sbcs.sunysb.edu> cbrown@eeserv1.ic.sunysb.edu (Charles T Brown) writes: [various discussion of availability of *emacs for MS-DOS deleted] My distribution of MicroEmacs V3.10 has #ifdefs for MS-DOS, it should compile right up. I received it for OS-9 compilation, but it looks like the "standard" distribution, though I'm not sure where that should come from (I don't have FTP access). >Umm. A friend of mine, and I, are considering porting it to OS/2 ASAP >(as soon as I get an OS/2-running machine! :-); but to the question >of porting it to DOS, it just really isn't possible. I mean, what >_I_ like about GNU Emacs is the ability to run a sub-shell in it, >and feed it input; you want to try writing that in DOS?! One of the _very_nifty_ command files included splits the screen into two windows, one which is MicroEmacs, the other DOS (all being handled by an Emacs macro piping commands to and from DOS). You certainly couldn't run Lotus in the window, but you can do most file operations, and I suspect if you ran a program things would drop back to MicroEmacs after execution (I haven't played with it too much). (Please don't email me requesting the source - I have modified it for my own OS-9 needs, and didn't isolate my changes from the other included source. Also, I'm not well-equipped site-wise to handle many file requests. Thanks!) -- ############################################################################## # Bill Sheppard -- bills@microware.com -- {uunet,sun}!mcrware!mwca!bill # # Microware Systems Corporation --- OS-9: Seven generations beyond OS/2!! # ######Opinions expressed are my own, though you'd be wise to adopt them!######
tree@newton.uvm.edu (Tom Emerson) (02/21/91)
>>>>> On 19 Feb 91 13:27:20 GMT, ktw@cbnewsb.ATT.COM (Ken Wolman) said: Ken> Originator: ktw@cbnewsb.cb.att.com >>>>> On 19 Feb 91 02:51:23 GMT, cotner@ronzoni.berkeley.edu (Carl F. Cotner) said: Carl> I am wondering why there is no available msdos port of GNU emacs. The obvious Carl> answer seems to be emacs' size. If this is indeed the reason, how much memory Carl> would emacs need? Considering that there is an Atari ST port which I assume is Carl> smaller than 1 meg, it would seem that one should be able to port it to msdos Carl> if you used a dos extender. Has anyone done this? It would sure be nice to Carl> have emacs at home. Ken> This probably won't answer your questions, but what the hell?... I've Ken> tried a variety of MS-DOS Emacs packages--MicroGnuEmacs, Freemacs, Ken> MicroEmacs, and Jove--and the closest to the "real thing" that I've Ken> found in terms of command structure and handling is Freemacs. It is Ken> available via anonymous ftp from grape.ecs.clarkson.edu: the last time Ken> I looked it was in /pub/Freemacs, but don't quote me. It's one Ken> serious flaw is its ability to handle large documents, i.e., anything Ken> over 64K: it just won't do it. Ken> -- Ken> Ken Wolman |"At age 35, Paul Gauguin still worked Ken> AT&T Bell Laboratories | as a teller in a Paris bank." Ken> Short Hills, NJ | "At age 35, Mozart was dead." Ken> (201)564-2866--hlwpk!ktw| --NYC subway graffiti, late 1960s
16012_3045@uwovax.uwo.ca (Paul Gomme) (02/22/91)
In article <1991Feb19.025123.12089@agate.berkeley.edu>, cotner@ronzoni.berkeley.edu (Carl F. Cotner) writes: > I am wondering why there is no available msdos port of GNU emacs. Recently, I've been using Mg which is "broadly similar to GNU emacs". It appears to be a strict subset of GNU emacs, but has most of the stuff I want. The code is available on snow.white.utoronto.edu; for the MS-DOS port, you need to remember to pick up ms_dos.shar as well. I've compiled the code under Turbo C 2.0 with minimal changes. -------------------------------------------------------------------------- Bitnet: gomme@uwovax.bitnet gomme@uwovax.uwo.ca Internet: gomme@uwo.ca