johns@cs.umr.edu (John Stone) (04/15/91)
Does anyone know where I could get unix utilities like
vi, rm, gcc, etc.. for OS/2
are they at any ftp sites? Has anyone ported gcc to OS/2
or g++ for that matter? I'm also interested in the
pbmplus utilities, have they been ported to OS/2 ?
Thanks,
John Stone
johns@cs.umr.edu
--
----------------------------------------------------------------------------
| Remember, No matter where you go, there you are.... |
| John E. Stone E-mail: johns@cs.umr.edu |
----------------------------------------------------------------------------brian@king.csd.mot.com (04/15/91)
johns@cs.umr.edu (John Stone) writes: > Does anyone know where I could get unix utilities like >vi, rm, gcc, etc.. for OS/2 The MKS Toolkit has vi, rm, and /etc :-) , but no gcc. There's an OS/2 version. I've used the MSDOS version for a long time and love it. Get info from toolkit@mks.com. Not free, but not really expensive either. -- -Brian Smithson Motorola Inc., Computer Group, Commercial Systems Division 10700 N. De Anza Boulevard, Cupertino, CA 95014 USA, (408)366-4104 brian@csd.mot.com, {apple | pyramid}!motcsd!brian
steveha@microsoft.UUCP (Steve HASTINGS) (04/23/91)
johns@cs.umr.edu (John Stone) writes: > Does anyone know where I could get unix utilities like >vi, rm, gcc, etc.. for OS/2 MKS utilities Thompson Toolkit, from Thompson Automation 503/224-1639 Hamilton C Shell Of the three, I recommend Thompson Toolkit. $180 for the utilities, both real mode (DOS) and protect mode (OS/2 native). I use it. No vi, though. Hamilton C Shell is a professional product specifically written to exploit OS/2 to the fullest. Protect-mode only, $350 list price. No vi. MKS utilities treats "c:foo" as exactly identical to "c:/foo", which makes it, IMHO, unusable. The MKS vi is a very good product, and available separately from the rest of the MKS Toolkit. -- Steve "I don't speak for Microsoft" Hastings ===^=== ::::: uunet!microsoft!steveha steveha@microsoft.uucp ` \\==|
gerry@dialogic.com (Gerry Lachac) (04/30/91)
In article <72009@microsoft.UUCP> steveha@microsoft.UUCP (Steve Hastings) writes: > >MKS utilities treats "c:foo" as exactly identical to "c:/foo", which makes >it, IMHO, unusable. The MKS vi is a very good product, and available >separately from the rest of the MKS Toolkit. Just out of curiousity, what's wrong with that? C: is device C, just as SYS: is device SYS on Novell. In a Unix-type shell, if I did a "cp C:" I would want it to go to the root of the C device, not some arbitrary directory that C: had been cd'ed to. (ie if I had done a "cd C:/foo/devs" and then did a "cp C:" and everything went to "C:/foo/devs" I would scream.) MKS does a pretty good job of imitating a UNIX ksh. -gerry -- uunet!dialogic!gerry | "Even a dead plant turns | Dialogic Corporation OR | over a new leaf | 300 Littleton Rd gerry@dialogic.UUCP | when the wind blows." | Parsippany, NJ 07054 | | (201)334-8450
resnicks@netcom.COM (Steve Resnick) (05/01/91)
In article <1991Apr30.133157.6453@dialogic.com> gerry@dialogic.com (Gerry Lachac) writes: >In article <72009@microsoft.UUCP> steveha@microsoft.UUCP (Steve Hastings) writes: >> >>MKS utilities treats "c:foo" as exactly identical to "c:/foo", which makes >>it, IMHO, unusable. The MKS vi is a very good product, and available >>separately from the rest of the MKS Toolkit. > >Just out of curiousity, what's wrong with that? C: is device C, just >as SYS: is device SYS on Novell. In a Unix-type shell, if I did a "cp >C:" I would want it to go to the root of the C device, not some >arbitrary directory that C: had been cd'ed to. (ie if I had done a >"cd C:/foo/devs" and then did a "cp C:" and everything went to >"C:/foo/devs" I would scream.) MKS does a pretty good job of imitating >a UNIX ksh. I never have known UNIX to refer to drives at all. The problem with C: refering to C:/ is that it's inconsitant with the DOS/OS2 file system. If, in a program (not a shell) I open "C:FOO.BAR" and the CWD on drive C is \FILES\FOAD then the file handle returned will refer to the explicit path C:\FILES\FOAD\FOO.BAR. What UNIX are you using that uses drive identifiers as pathname components? Not SysV, Xenix, or Sun OS ... Cheers! Steve -- ------------------------------------------------------------------------------- resnicks@netcom.com, steve@camphq, IFNA: 1:143/105.0, resnicks@192.100.81.100 Real life: Steve Resnick. Chief Software Architect, Process Scientific, Inc Flames, grammar and spelling errors >/dev/null 0x2b |~ 0x2b, THAT is the question. The Asylum OS/2 BBS - (408)263-8017 12/2400,8,1 - Running Maximus CBCS 1.2 -------------------------------------------------------------------------------
brian@king.csd.mot.com (05/01/91)
steveha@microsoft.UUCP (Steve HASTINGS) writes: >MKS utilities treats "c:foo" as exactly identical to "c:/foo", which makes >it, IMHO, unusable. The MKS vi is a very good product, and available >separately from the rest of the MKS Toolkit. Are you sure about that? In the MSDOS version, it treats c:foo the same as MSDOS does -- look on C: for whatever your working directory is/was on that drive and then look for FOO there. That is *not* equivalent to c:/foo. In any case, why do you consider the toolkit unusable on that basis, IYHO?
gerry@dialogic.com (Gerry Lachac) (05/02/91)
In article <1991Apr30.234748.10836@netcom.COM> resnicks@netcom.COM (Steve Resnick) writes: > >I never have known UNIX to refer to drives at all. The problem with C: >refering to C:/ is that it's inconsitant with the DOS/OS2 file system. Ok, I'll agree with the fact that it is inconsistant with the way DOS does things. >What UNIX are you using that uses drive identifiers as pathname components? >Not SysV, Xenix, or Sun OS ... Your right, no UNIX shell that I know of lets me do that. Call me an Amiga-weenie if you will, but I can do it that way under AmigaDOS, which is something I'm used to doing at home. Also Novell (arguablly a DOS file-system) lets you refer to network drives by drive identifier. I can "DIR SYS:" or "DIR SYS:/PUBLIC", but I can just "SYS:" and get to the "SYS:" drive. (Actually you are really supposed to map the SYS: drive to a letter device like L: so you can do the standard DOS file-system things) And in addition, VAX VMS uses DEVICE: as a device (with other arcane syntax, it's been 8 years since I used a VAX). I have alway been uncomfortable refering to a device (such as C:) that can change on the fly. I like to think of the letter/name to the left of the colon to be a physical or logical device that doesn't jump around, so if I delete C:*, I know exactly WHAT I am deletely. To each there own... :-) -gerry -- uunet!dialogic!gerry | "Even a dead plant turns | Dialogic Corporation OR | over a new leaf | 300 Littleton Rd gerry@dialogic.UUCP | when the wind blows." | Parsippany, NJ 07054 | | (201)334-8450
lenox@media-lab.media.mit.edu.MEDIA.MIT.EDU (Lenox H. Brassell) (05/02/91)
I find it indefensible that the MKS Korn Shell doesn't even use the "intuitive" definition of "C:.", the current directory on drive C! Instead, they insist on interpreting it as "the root directory on drive C:! It isn't unreasonable to want to, say, copy a bunch of files to "the current directory on drive C:", but there's no easy way to do this using the MKS Korn Shell. Less defensible still, is the fact that the MKS utilities for OS/2 _still_ do not support long file names for OS/2 HPFS files. Yet they continue to advertise their wonderful support for OS/2 in all of their magazine ads. I called them a couple of weeks ago to ask about this feature, and the salesman said that their "might" be an update for the OS/2 Toolkit this fall (as in SIX MONTHS FROM NOW) and that there "might" be support for long filenames in that release, if it came. It's not as if long filenames are a new feature, either. I'm not sending these guys any more of _my_ money, and I'm not reccommending their products to anyone I know, either. Regards, --Lenox [lenox@media-lab.mit.edu]
lenox@media-lab.media.mit.edu.MEDIA.MIT.EDU (Lenox H. Brassell) (05/02/91)
In article <5774@media-lab.media.mit.edu.MEDIA.MIT.EDU>, lenox@media-lab.media.mit.edu.MEDIA.MIT.EDU (Lenox H. Brassell) writes: > > It isn't unreasonable to want to, say, copy a bunch of files to "the > current directory on drive C:", but there's no easy way to do this > using the MKS Korn Shell. > I haven't bothered using the MKS Korn Shell for a long time, and my memory failed me. "cp *.c C:." actually does do "the right thing". But when you ask the Korn Shell to "CD C:.", it puts you into the root of drive "C". This is not an unreasonable thing to want to do, either. This, despite the fact that the underlying operating system kept track of what "the current directory on drive C:" really meant. It was particularly, uhm, "interesting" to me that all of the Toolkit .EXE utilities interpreted "C:." to mean what I wanted it to mean, but that the Korn Shell had a different interpretation. --Lenox [lenox@media-lab.mit.edu]
roy@mas1.UUCP (Roy Ho) (05/06/91)
In article <5774@media-lab.media.mit.edu.MEDIA.MIT.EDU> lenox@media-lab.media.mit.edu.MEDIA.MIT.EDU (Lenox H. Brassell) writes: >Less defensible still, is the fact that the MKS utilities for OS/2 >_still_ do not support long file names for OS/2 HPFS files. When I was at the Windows expo a couple of months ago, the salescritter at the MKS booth claimed that HPFS support was going to be released Real Soon Now. Also, has anyone noticed that the MKS vi editor is a memory hog? Editting an empty file sucks up 1100K of memory on my OS/2 1.3 system. -- Roy Ho Measurex Automation Systems ...!apple!mas1!roy
feustel@netcom.COM (David Feustel) (05/07/91)
Argosoft is advertising a BSD 4.3 set of tools that runs on OS/2 and supports HPFS. I meant to call them today but forgot. The package runs ca. $400-500. -- David Feustel, 1930 Curdes Ave, Fort Wayne, IN 46805, (219) 482-9631 EMAIL: netcom.com