From: UnixOS2 Archive To: "UnixOS2 Archive" Date: Tue, 19 Feb 2002 04:15:49 EST-10EDT,10,-1,0,7200,3,-1,0,7200,3600 Subject: [UnixOS2_Archive] No. 140 ************************************************** Monday 18 February 2002 Number 140 ************************************************** Subjects for today 1 Re: OSFree... : Henry Sobotka 2 Re: total.cmd : Charles R. Hunter" 3 Re: OSFree... : John Poltorak 4 Re: LESS v370 : John Poltorak 5 Re: termcap.dat : James Cannon 6 Re: LIBEMX suggestion : John Poltorak 7 Re: LIBEMX suggestion : John Poltorak 8 Re: OSFree... : Stepan Kazakov 9 MAN pages : John Poltorak 10 Re: OSFree... : Adrian Gschwend" 11 Re: termcap.dat : Charles R. Hunter" 12 Re: LIBEMX suggestion : Holger Veit 13 Re: MAN pages : Holger Veit 14 Re: termcap.dat : Charles R. Hunter" 15 Re: termcap.dat : Thomas Dickey 16 Re: OSFree... : Dave and Natalie" 17 EMACS build error : John Poltorak 18 Re: termcap.dat : Thomas Dickey 19 libtermcap : John Poltorak 20 Re: EMACS build error : John Poltorak 21 Re: total.cmd : John Poltorak 22 Re: EMACS build error : Holger Veit 23 termcap.dat : John Poltorak 24 Re: OSFree... : Kris Steenhaut 25 Re: OSFree... : Adrian Gschwend" 26 Re: termcap.dat : Thomas Dickey 27 Re: Termcap and VIO (was: termcap.dat) : James Cannon 28 Re: termcap.dat : John Poltorak 29 Re: LESS v370 : Akira Hatakeyama 30 Re: Re: EMACS build error : John Poltorak 31 Re: termcap.dat : John Poltorak 32 Re: termcap.dat : Thomas Dickey 33 Re: Newbie Q re gcc 3.0.2 : John Poltorak 34 Re: Termcap and VIO (was: termcap.dat) : Thomas Dickey 35 Re: Newbie Q re gcc 3.0.2 : Anders Jarnberg" 36 Re: MAN pages : John Poltorak 37 Re: Termcap and VIO (was: termcap.dat) : Thomas Dickey 38 Re: termcap.dat : John Poltorak 39 termcap.c : John Poltorak 40 Re: termcap.dat : John Poltorak 41 Re: Termcap and VIO (was: termcap.dat) : James Cannon 42 Re: OSFree... : Kris Steenhaut 43 Re: LIBEMX suggestion : Stefan Neis 44 Re: LIBEMX suggestion : Stefan Neis 45 Re: OSFree... : Lyn St George" 46 Re: Sleepycat Software licence : Stefan Neis 47 Re: OSFree... : Stefan Neis 48 Re: termcap.dat : Stefan Neis 49 Re: total.cmd : Andreas Buening 50 Re: EMACS build error : Andreas Buening 51 Re: OSFree... : Stefan Neis **= Email 1 ==========================** Date: Tue, 19 Feb 2002 01:01:25 -0500 From: Henry Sobotka Subject: Re: OSFree... Stepan Kazakov wrote: > > do you know the author of OSFree ? it is interesting to contact him.. The hibernation component has me thinking it might be Andrew Belov . h~ **= Email 2 ==========================** Date: Tue, 19 Feb 2002 07:37:26 -0500 From: "Charles R. Hunter" Subject: Re: total.cmd On Mon, Feb 18, 2002 at 10:55:48AM +0000, John Poltorak (jp at eyup.org) wrote: > > I've written this simple AWK script called total.cmd which totals up the > values in a file:- > > [SNIP] > I don't see why every line is getting printed. How do I surpress it? > what version of awk are you using? Your program works properly for me. ( on FreeBSD ) Charles > -- > John > -- Charles R. Hunter Director, Physics Computer Network Purdue University crh at physics.purdue.edu **= Email 3 ==========================** Date: Tue, 19 Feb 2002 08:34:19 +0000 From: John Poltorak Subject: Re: OSFree... On Mon, Feb 18, 2002 at 11:27:09PM +0100, Adrian Gschwend wrote: > Something interesting poped up: > > http://hobbes.nmsu.edu/pub/incoming/osf_en.zip > > Noncommercial OS/2-compatible operating system > I guess this thingy is based on some "leaked" sources from IBM. And I > don't think IBM will like this... It worries me that IBM could get Hobbes shut down through this... Following on from various discussions about licensing etc. I'm trying to understand the status of this:- Terms and conditions -------------------- Downloading this product indicates your acceptance of the following terms and conditions: 1. You may not use the product for any business, government or institutional activities. To my mind it is completely void. There is no copyright and no attribution and I don't see how it can have any meaning. > cu > > Adrian > > > -- > Adrian Gschwend > at OS/2 Netlabs > > ICQ: 22419590 > ktk at netlabs.org > ------- > The OS/2 OpenSource Project: > http://www.netlabs.org > -- John **= Email 4 ==========================** Date: Tue, 19 Feb 2002 08:59:57 +0000 From: John Poltorak Subject: Re: LESS v370 On Tue, Feb 19, 2002 at 03:36:57PM +0900, Akira Hatakeyama wrote: > On Mon, Feb 18, 2002 at 05:40:11PM +0000, John Poltorak wrote: > > There is an OS/2 port of LESS v370 on Hobbes which is maintained by > > Kyosuke Tokoro . Do any of our Japanese friends know > > of him? I'd like to try and get his patches and rebuild LESS if possible, > > but have been unable to get any reply from him. > > Seems no update since 2002/01/01. But as his 'Whats New' section > (in japanese), his patches was inlcuded in original less version. > You can get it from > > http://www.greenwoodsoftware.com/less/ Thanks for pointing this out. Indeed the latest source for v373 does contain an OS/2 makefile, but you wouldn't realise that OS/2 is officially supported by looking at the home page which says:- 3. OS/2 Binaries OS/2 binaries are packaged in a zip archive. Binaries for OS/2 are maintained by Kai Ewe Rommel (rommel at ars.muc.de). OS/2 binaries are not currently available. If you have contact with the porter, can you find out if he would have any objection to the less homepage having a link for the OS/2 version pointing to? :- ftp://unixos2.com/pub/unixos2/unixos2-current/unixos2/a1/less.zip > -- > Akira Hatakeyama E-Mail: akira at sra.co.jp > http://www.sra.co.jp/people/akira/index.html > chigasaki-minami, tsuzuki ward, yokohama, japan -- John **= Email 5 ==========================** Date: Tue, 19 Feb 2002 09:07:13 -0800 (PST) From: James Cannon Subject: Re: termcap.dat --- John Poltorak wrote: > On Tue, Feb 19, 2002 at 09:56:10AM -0500, Charles R. > Hunter wrote: > > On Tue, Feb 19, 2002 at 02:02:04PM +0000, John > Poltorak (jp at eyup.org) wrote: > > > > > > Does the file termcap.dat exist on Unix systems? On HP-UX (10.20), termcap is located in: /usr/share/lib/termcap Using head -20 reveals the following: # at (#) $Revision: 72.3 $ # Important: # This file is here only for compatibility purposes. # It is not referenced by the curses(3X) library routines, # the termcap(3X) emulation routines, or the terminfo routines. # Instead of /usr/share/lib/termcap these routines use the terminfo(5) # terminal data base under /usr/share/lib/terminfo. su|dumb|un|unknown:co#80:os:am: sp|plugboard:co#80:os:am: se|ethernet|network:co#80:os:am: sd|du|dialup:co#80:os:am: sb|bussiplexer:co#80:os:am: sa|arpanet|network:co#80:os:am: # # HEWLETT-PACKARD TERMINALS # # hpsub: > > /etc/termcap [SNIP] > > Charles > > > > > > > > > > > -- > > > John > > > > > > > -- > > Charles R. Hunter > > Director, Physics Computer Network > > Purdue University crh at physics.purdue.edu > > > -- > John > > ===== Sincerely, James Cannon Using OS/2 Warp in the beautiful Wine Country of Northen California! __________________________________________________ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.yahoo.com **= Email 6 ==========================** Date: Tue, 19 Feb 2002 09:11:37 +0000 From: John Poltorak Subject: Re: LIBEMX suggestion On Mon, Feb 18, 2002 at 03:25:43PM -0500, Henry Sobotka wrote: > John Poltorak wrote: > > > > It is Copyright Sleepycat Software, but I've never heard of this outfit. > > Keith Bostic's firm. He's one of the original authors of BDB. Can't say I'd come across the name before, but I guess he must be famous... > > One thing I noticed in the archive is the presence of both db.(a,lib) and > > libdb.(a,lib). We only should have one, and my preference would be libdb.*. > > > > Anyone (dis)agree? > > Problem is that'll break every makefile containing -ldb for anyone using > a version of gcc that doesn't prepend 'lib'. AIUI, this is a constraint imposed by LD from gcc 2.8.1, but in the context of UnixOS/2, any GCC package assembled will contain a newer version of which can handle a 'lib' prefix. I'm not an expert in this area, so please correct me if I have this wrong. > h~ -- John **= Email 7 ==========================** Date: Tue, 19 Feb 2002 09:17:07 +0000 From: John Poltorak Subject: Re: LIBEMX suggestion On Mon, Feb 18, 2002 at 10:25:10PM +0000, Lyn St George wrote: > > Backward compatibility - just looked for that and the only thing I can > see is in configure: > Build DB 1.85 compatibility API > which I did not enable. If you think it should be enabled, then I'll > build a new one and test it, but it will take a couple of days or so. Any chance of you providing some build instructions? > > Cheers > Lyn St George > +--------------------------------------------------------------------------------- > + http://www.zolotek.net .. eCommerce hosting, consulting > + http://www.os2docs.org .. some 'How To' stuff ... > +---------------------------------------------------------------------------------- > -- John **= Email 8 ==========================** Date: Tue, 19 Feb 2002 09:26:03 -0500 From: Stepan Kazakov Subject: Re: OSFree... Adrian Gschwend wrote: > Something interesting poped up: > http://hobbes.nmsu.edu/pub/incoming/osf_en.zip > Noncommercial OS/2-compatible operating system > Long Description: osFree, a noncommercial OS/2-compatible system, sees > the light of day. Englishand Russian versions are available at Hobbes. > The project invites new participants... btw i just done some PM* DLLs for embedded OS/2 ;) with Prf* and some others funcs implementations. now weasel, ftpser, telnetd ... etc working in OS/2 textmode env ;) do you know the author of OSFree ? it is interesting to contact him.. -- madded. [Red Hot Chili Hackers] **= Email 9 ==========================** Date: Tue, 19 Feb 2002 09:36:07 +0000 From: John Poltorak Subject: MAN pages I've recompiled Jun SAWATAISHI's port of MAN 1.5g, basically because I didn't like the default location of man.conf being in \os2\etc. It should shortly appear here:- ftp://unixos2.com/pub/unixos2/unixos2-current/unixos2/ap1/ To actually view MAN pages properly you need to get MAN, GROFF, LESS and TERMCAP to work in harmony. This works seemlessly on Unix, but is somewhat contorted on OS/2, but I'd like it to work just as easily on OS/2. Can anyone explain how it manages to work seemlessly on Unix? I guess that is due to the correct default compile options or use of established environment variables... On OS/2 I use this MAN.CMD as a wrapper for man.exe :- at echo off set GROFF_TMAC_PATH=c:/usr/lib/groff/share/groff/tmac set GROFF_FONT_PATH=c:/usr/lib/groff/share/groff/font set GROFF_TMPDIR=c:/tmp set GROFF_TYPESETTER=ascii set MANPATH=c:/usr/man; set path=c:\usr\lib\groff\bin;%path% SET TERMCAP=c:/etc/termcap.dat SET TERM=ansi c:\usr\bin\man %1 If possible, I'd like to get rid of it. -- John **= Email 10 ==========================** Date: Tue, 19 Feb 2002 09:51:25 +0100 (CET) From: "Adrian Gschwend" Subject: Re: OSFree... On Tue, 19 Feb 2002 09:26:03 -0500, Stepan Kazakov wrote: >btw i just done some PM* DLLs for embedded OS/2 ;) >with Prf* and some others funcs implementations. >now weasel, ftpser, telnetd ... etc working in OS/2 textmode env ;) nice :-) >do you know the author of OSFree ? it is interesting to contact him.. no I don't know him. For me the whole thing is very exiting but also very dangerous. I don't know how IBM will react on this but I doubt they will like it unfortunately... I also don't know how they want to release the source because it's almost for sure not rewriten but recompiled (of the original source obviously). cu Adrian -- Adrian Gschwend at OS/2 Netlabs ICQ: 22419590 ktk at netlabs.org ------- The OS/2 OpenSource Project: http://www.netlabs.org **= Email 11 ==========================** Date: Tue, 19 Feb 2002 09:56:10 -0500 From: "Charles R. Hunter" Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 02:02:04PM +0000, John Poltorak (jp at eyup.org) wrote: > > Does the file termcap.dat exist on Unix systems? /etc/termcap > > Where would I expect to find it, and is it likely to be usable on OS/2? yes and no... the VIO session underneath is not a tty as UNIX programs expect. You can receive raw keyboard codes instead of terminal codes ( like ANSI ) if you want. This work has already been done for you with EMX's libtermcap.a Charles > > > -- > John > -- Charles R. Hunter Director, Physics Computer Network Purdue University crh at physics.purdue.edu **= Email 12 ==========================** Date: Tue, 19 Feb 2002 10:33:58 +0100 From: Holger Veit Subject: Re: LIBEMX suggestion On Tue, Feb 19, 2002 at 09:11:37AM +0000, John Poltorak wrote: > On Mon, Feb 18, 2002 at 03:25:43PM -0500, Henry Sobotka wrote: > > John Poltorak wrote: > > > > > > It is Copyright Sleepycat Software, but I've never heard of this outfit. > > > > Keith Bostic's firm. He's one of the original authors of BDB. > > Can't say I'd come across the name before, but I guess he must be famous... Check the BSD (m)annuals. > > > One thing I noticed in the archive is the presence of both db.(a,lib) and > > > libdb.(a,lib). We only should have one, and my preference would be libdb.*. > > > > > > Anyone (dis)agree? > > > > Problem is that'll break every makefile containing -ldb for anyone using > > a version of gcc that doesn't prepend 'lib'. > > AIUI, this is a constraint imposed by LD from gcc 2.8.1, but in the > context of UnixOS/2, any GCC package assembled will contain a newer > version of which can handle a 'lib' prefix. Right. Holger -- Please update your tables to my new e-mail address: holger.veit$ais.fhg.de (replace the '$' with ' at ' -- spam-protection) **= Email 13 ==========================** Date: Tue, 19 Feb 2002 10:55:19 +0100 From: Holger Veit Subject: Re: MAN pages On Tue, Feb 19, 2002 at 09:36:07AM +0000, John Poltorak wrote: > > I've recompiled Jun SAWATAISHI's port of MAN 1.5g, basically because I > didn't like the default location of man.conf being in \os2\etc. > It should shortly appear here:- > > ftp://unixos2.com/pub/unixos2/unixos2-current/unixos2/ap1/ > > > To actually view MAN pages properly you need to get MAN, GROFF, LESS and > TERMCAP to work in harmony. This works seemlessly on Unix, but is somewhat > contorted on OS/2, but I'd like it to work just as easily on OS/2. > > Can anyone explain how it manages to work seemlessly on Unix? I guess that > is due to the correct default compile options or use of established > environment variables... Preformat them. See the "catman" command. Holger -- Please update your tables to my new e-mail address: holger.veit$ais.fhg.de (replace the '$' with ' at ' -- spam-protection) **= Email 14 ==========================** Date: Tue, 19 Feb 2002 11:01:05 -0500 From: "Charles R. Hunter" Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 03:28:08PM +0000, John Poltorak (jp at eyup.org) wrote: > On Tue, Feb 19, 2002 at 09:56:10AM -0500, Charles R. Hunter wrote: > > On Tue, Feb 19, 2002 at 02:02:04PM +0000, John Poltorak (jp at eyup.org) wrote: > > > > > > Does the file termcap.dat exist on Unix systems? > > > > /etc/termcap > > I think we should use the same name on UnixOS/2... agreed. I don't know Why Eberhardt Mattes used termcap.dat in the first place... maybe for early compatibility with djgpp or something. ( since EMX was originally dual mode DOS and OS/2 ) > > I did find such a file on my FreeBSD CD but it was zero length. > > > > > > > Where would I expect to find it, and is it likely to be usable on OS/2? > > > > yes and no... the VIO session underneath is not a tty as UNIX > > programs expect. You can receive raw keyboard codes instead of > > terminal codes ( like ANSI ) if you want. > > > > This work has already been done for you with EMX's libtermcap.a > > Where exactly? > > The components of termcap.a are termcap.h, termcap.c, tparam.c, version.c > and termcap.dat. > > I'm just trying to establish the source of all these files. > > There seem to be a number of different termcap.h files around, and I'd > like to know how EMX's termcap differs from the one in BSD, but I can't > find the source for that one... all the source code for EMX should be here: http://hobbes.nmsu.edu/pub/os2/dev/emx/v0.9d/*src*.zip specifically termcap here: http://hobbes.nmsu.edu/pub/os2/dev/emx/v0.9d/gnusrc.zip charles > > > Charles > > > > > > > > > > > -- > > > John > > > > > > > -- > > Charles R. Hunter > > Director, Physics Computer Network > > Purdue University crh at physics.purdue.edu > > > -- > John > -- Charles R. Hunter Director, Physics Computer Network Purdue University crh at physics.purdue.edu **= Email 15 ==========================** Date: Tue, 19 Feb 2002 11:21:28 -0500 From: Thomas Dickey Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 11:01:05AM -0500, Charles R. Hunter wrote: > > There seem to be a number of different termcap.h files around, and I'd > > like to know how EMX's termcap differs from the one in BSD, but I can't > > find the source for that one... is he looking for the "the" termcap in BSD, or the source for EMX. It sounds like the former. Actually, EMX's termcap wasn't based on that. > all the source code for EMX should be here: > > http://hobbes.nmsu.edu/pub/os2/dev/emx/v0.9d/*src*.zip > > specifically termcap here: > > http://hobbes.nmsu.edu/pub/os2/dev/emx/v0.9d/gnusrc.zip > > charles -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net **= Email 16 ==========================** Date: Tue, 19 Feb 2002 11:50:54 +0000 From: "Dave and Natalie" Subject: Re: OSFree... On Tue, 19 Feb 2002 14:17:53 +0100, Kris Steenhaut wrote: > >I don't think patents would be an issue. Afaik the only patented systems >in OS/2 are HPFS and hpfs386, and it's M$ owning the patent. > >That's is, if M$ had renewed the patent end last year. Dunno if they have >done so. intersesting reading at http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=/netahtml/search-bool.html&r=2&f=G&l=50&co1=AND&d=ft95&s1=hpfs&s2=microsoft&OS=hpfs+AND+microsoft&RS=hpfs+AND+microsoft I like these quotes However, the present invention is particularly adapted for use with the OS/2 operating system developed by Microsoft. .... The FAT file system provides excellent performance with volumes which are less than 1 Mb. However, as volumes increase in size over 1 Mb, the performance of the FAT file system quickly degrades. This has become an increasingly severe problem as the size of readily available hard disks is rapidly increasing. Dave **= Email 17 ==========================** Date: Tue, 19 Feb 2002 11:53:19 +0000 From: John Poltorak Subject: EMACS build error I get the following error when attempting to compile emacs 20.7 :- make[1]: Entering directory `/eval/emacs/emacs-20.7/lib-src' gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O -o test-distrib test-distrib.c test-distrib testfile gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O make-docfile.c -lc -o make-docfile gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O profile.c -lc -o profile gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O digest-doc.c -lc -o digest-doc gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O sorted-doc.c -lc -o sorted-doc gcc -Zexe -DOS2 -c -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O -Demacs -DMAIL_USE_POP movemail.c gcc -Zexe -DOS2 -c -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O -DMAIL_USE_POP pop.c gcc -Zexe -DOS2 -c -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O getopt.c gcc -Zexe -DOS2 -c -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O getopt1.c gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O -DMAIL_USE_POP movemail.o pop.o get opt.o getopt1.o -lc -lsocket -o movemail movemail.c:739 (movemail.o): Undefined symbol _fchown referenced from text segment make[1]: *** [movemail] Error 1 make[1]: Leaving directory `/eval/emacs/emacs-20.7/lib-src' make: *** [lib-src] Error 2 Where should _fchown be defined ? -- John **= Email 18 ==========================** Date: Tue, 19 Feb 2002 12:04:49 -0500 From: Thomas Dickey Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 04:34:07PM +0000, John Poltorak wrote: > On Tue, Feb 19, 2002 at 11:21:28AM -0500, Thomas Dickey wrote: > > On Tue, Feb 19, 2002 at 11:01:05AM -0500, Charles R. Hunter wrote: > > > > There seem to be a number of different termcap.h files around, and I'd > > > > like to know how EMX's termcap differs from the one in BSD, but I can't > > > > find the source for that one... > > > > is he looking for the "the" termcap in BSD, > > Yes... "the" termcap doesn't exist (the 3 BSD's have three views of this - some are based on ncurses, and the ones that are not are conceded to be poorly maintained) > > or the source for EMX. It > > sounds like the former. Actually, EMX's termcap wasn't based on that. > > It looks to me as though EMX's termcap is based on GLIBC, although I'm not > sure of the significance of having both btermcap and termcap. glibc doesn't distribute termcap. termcap is a separate item. (Actually I've seen more than one termcap for EMX, and incorporated much of the more recent one into ncurses as emx.src). None of the versions that I've seen are based on the BSD termcap, but instead are a mix of generic stuff (such as vt100 and xterm) and OS/2-specific stuff that isn't found in that form elsewhere. > I'd like to know if a BSD based termcap could be created but don't know > where to look. ftp://invisible-island.net/ncurses/termcap.src.gz -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net **= Email 19 ==========================** Date: Tue, 19 Feb 2002 12:09:01 +0000 From: John Poltorak Subject: libtermcap Where do I get the source for building a libtermcap ? -- John **= Email 20 ==========================** Date: Tue, 19 Feb 2002 12:50:41 +0000 From: John Poltorak Subject: Re: EMACS build error On Tue, Feb 19, 2002 at 01:30:27PM +0100, Holger Veit wrote: > Better enjoy with the emacs we now have for now; Emacs is an order > of magnitude too tricky for someone who has already a problem with fchown > (well-meant advice). I'm not trying to do anything clever, merely trying to apply an OS/2 patch kit. There were some instructions on the list a couple of months ago about how to build emacs v20.7 provided by Masaru Nomiya. I'll repost them here for anyone who missed them the first time:- These files are required. ftp://ftp.mirror.ac.uk/sites/ftp.gnu.org/pub/gnu/emacs/emacs-20.7.tar.gz ftp://ftp.mirror.ac.uk/sites/ftp.gnu.org/pub/gnu/emacs/leim-20.7.tar.gz http://www.asahi-net.or.jp/~ik3a-nsmr/emacs-20.7-KIT1.9.tgz To build Emacs-20.7, 1. tar zxf SOMEWHERE\emacs-20.7.tar.gz 2. tar zxf SOMEWHERE\leim-20.7.tar.gz 3. cd emacs-20.7 4. chmod +rw -R* 5. untgz SOMEWHERE\emacs-20.7-KIT1.9.tgz 6. configure.cmd --with-bsd-socket --with-xf86sup-pty Execution of configure.cmd ask you twice, just press Enter key 8. edit src\puresize.h as follows (line 36-43); #ifndef SYSTEM_PURESIZE_EXTRA /* #define SYSTEM_PURESIZE_EXTRA 0 */ #define SYSTEM_PURESIZE_EXTRA 200000 $B"+(B this number #endif #ifndef SITELOAD_PURESIZE_EXTRA #define SITELOAD_PURESIZE_EXTRA 200000 $B"+(B this nimber #endif Moreover, I have edited src\config.h as follows (line 246-248); #define EMACS_CONFIGURATION "AthlonXP OS/2_Warp_V4.51 emx09d_fix04" #define EMACS_CONFIG_OPTIONS "--emacs-configuration=AthlonXP OS/2_Warp_V4.51 emx09d_fix04 --with-bsd-socket=yes --with-xf86sup-pty=yes" 7. SET INCLUDE=x:/emx/include SET NAME=John Poltorak $B"+(B ADD this in CONDIF.SYS 8. make 9. del src\emacs.exe 10. make 11. add emacs-20.7\src to SET PATH in the CONFIG.SYS In my case, SET PATH=....;\g:\usr\local\emacs-20.7\src; And add these 3 lines SET EMACSROOT=g:/usr/local/emacs-20.7 SET EMACSLOADPATH=g:/usr/local/emacs-20.7/lisp;g:/usr/local/emacs-20.7/site- lisp;g:/usr/local/emacs-20.7/leim SET NAME=John Poltrak (Yes, I executed make under the g:\usr\local directory.) 12. make emacs.cmd like this g:\usr\local\emacs-20.7\src\emacs.exe then relate with a nice icon (emacs.ico) which is in the ..\emacs-20.7\os2. 13. Reboot That's all. > Holger > > -- > Please update your tables to my new e-mail address: > holger.veit$ais.fhg.de (replace the '$' with ' at ' -- spam-protection) > -- John **= Email 21 ==========================** Date: Tue, 19 Feb 2002 13:02:55 +0000 From: John Poltorak Subject: Re: total.cmd On Tue, Feb 19, 2002 at 07:37:26AM -0500, Charles R. Hunter wrote: > On Mon, Feb 18, 2002 at 10:55:48AM +0000, John Poltorak (jp at eyup.org) wrote: > > > > I've written this simple AWK script called total.cmd which totals up the > > values in a file:- > > > > > > [SNIP] > > I don't see why every line is getting printed. How do I surpress it? > > > > what version of awk are you using? C:\>awk --version GNU Awk 3.1.0 Copyright (C) 1989, 1991-2001 Free Software Foundation. Actually, I just tried it with an older port and it worked there too, so it looks as though it is a problem in the port of v3.1.0... > Charles > > -- > Charles R. Hunter > Director, Physics Computer Network > Purdue University crh at physics.purdue.edu -- John **= Email 22 ==========================** Date: Tue, 19 Feb 2002 13:30:27 +0100 From: Holger Veit Subject: Re: EMACS build error On Tue, Feb 19, 2002 at 11:53:19AM +0000, John Poltorak wrote: > > I get the following error when attempting to compile emacs 20.7 :- > > make[1]: Entering directory `/eval/emacs/emacs-20.7/lib-src' > gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O -o test-distrib test-distrib.c test-distrib testfile > gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O make-docfile.c -lc -o make-docfile > gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O profile.c -lc -o profile > gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O digest-doc.c -lc -o digest-doc > gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O sorted-doc.c -lc -o sorted-doc > gcc -Zexe -DOS2 -c -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O -Demacs -DMAIL_USE_POP movemail.c > gcc -Zexe -DOS2 -c -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O -DMAIL_USE_POP pop.c > gcc -Zexe -DOS2 -c -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O getopt.c > gcc -Zexe -DOS2 -c -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O getopt1.c > gcc -Zexe -DOS2 -D_BSD_SOURCE -DHAVE_CONFIG_H -I. -I../src -g -O -DMAIL_USE_POP movemail.o pop.o get > opt.o getopt1.o -lc -lsocket -o movemail > movemail.c:739 (movemail.o): Undefined symbol _fchown referenced from text segment You could ignore this (or rather exclude it in the corresponding source file); it is for security that OS/2 does not yet have. Other things in emacs are also not too difficult; encounter link(), getuid(), setpgrp() and other functions that EMX doesn't have. They are not the real problems. You rather have a more difficult problem because emacs will at some stage build a temacs.exe which will attempt to modify itself, by "dumping" some lisp source code into its own data segment. Here the stuff will become a challenge. It was already a challenge with Xemacs for me; but still this dumps core depending on the phase of the moon - haven't got enough time to debug the problem behind that. Better enjoy with the emacs we now have for now; Emacs is an order of magnitude too tricky for someone who has already a problem with fchown (well-meant advice). > Where should _fchown be defined ? In emxlibc :-) It isn't... Holger -- Please update your tables to my new e-mail address: holger.veit$ais.fhg.de (replace the '$' with ' at ' -- spam-protection) **= Email 23 ==========================** Date: Tue, 19 Feb 2002 14:02:04 +0000 From: John Poltorak Subject: termcap.dat Does the file termcap.dat exist on Unix systems? Where would I expect to find it, and is it likely to be usable on OS/2? -- John **= Email 24 ==========================** Date: Tue, 19 Feb 2002 14:17:53 +0100 From: Kris Steenhaut Subject: Re: OSFree... Adrian Gschwend schreef: > On Tue, 19 Feb 2002 09:26:03 -0500, Stepan Kazakov wrote: > > >btw i just done some PM* DLLs for embedded OS/2 ;) > >with Prf* and some others funcs implementations. > >now weasel, ftpser, telnetd ... etc working in OS/2 textmode env ;) > > nice :-) > > >do you know the author of OSFree ? it is interesting to contact him.. > > no I don't know him. For me the whole thing is very exiting but also > very dangerous. I don't know how IBM will react on this but I doubt > they will like it unfortunately... > Big Blue already has pulled the plug out of MCP, so there isn't much left for them to dislike. > > I also don't know how they want to release the source because it's > almost for sure not rewriten but recompiled (of the original source > obviously). Would be hard to prove either way, isn't it? Th only problem I'm seeing now is they've forgotten to mention the trade marking. You don't say " Compatible with OS/2 Warp 4" but you have to say: " Compatible with OS/2 Warp 4 -Tm- (*)" (*) OS/2 and Warp are trade marked products by International Business Machines. (Or something like that). I don't think patents would be an issue. Afaik the only patented systems in OS/2 are HPFS and hpfs386, and it's M$ owning the patent. That's is, if M$ had renewed the patent end last year. Dunno if they have done so. -- Groeten uit Gent, Kris **= Email 25 ==========================** Date: Tue, 19 Feb 2002 14:56:31 +0100 (CET) From: "Adrian Gschwend" Subject: Re: OSFree... On Tue, 19 Feb 2002 14:17:53 +0100, Kris Steenhaut wrote: >Big Blue already has pulled the plug out of MCP, so there isn't much left >for them to dislike. that doesn't change anything. I agree that the only solution for OS/2 is a free OS based on OS/2 ideas (and maybe parts of it if we could get that from IBM officialy). But stolen source does not help at all, in no way. It will just make IBM even more closed against such ideas. >Would be hard to prove either way, isn't it? well the build level is from the leaked sources so not that hard. For sure the prove would be to see it but it's quite obvious. cu Adrian -- Adrian Gschwend at OS/2 Netlabs ICQ: 22419590 ktk at netlabs.org ------- The OS/2 OpenSource Project: http://www.netlabs.org **= Email 26 ==========================** Date: Tue, 19 Feb 2002 15:00:43 -0500 From: Thomas Dickey Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 07:40:04PM +0000, John Poltorak wrote: > > ftp://invisible-island.net/ncurses/termcap.src.gz > > I was actually looking for source to create a libtermcap, but this file is > interesting in itself... There is mention of emx entries - it says > 'distributed with EMX 0.9b'. I don't have this version available, but the > emx section doesn't bear much of a resemblance to termcap.dat which I have in > \emx\etc from 0.9d. > > Is there any indication of exactly where these entries came from or who > put them there? no - that file's EMX entries are actually older than the emx.src which I put in the misc directory of ncurses' source. I probably should merge them, but there are some conflicts with the emx.src vs the other stuff, so I left them separate - ymmv). -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net **= Email 27 ==========================** Date: Tue, 19 Feb 2002 15:28:05 -0800 (PST) From: James Cannon Subject: Re: Termcap and VIO (was: termcap.dat) --- Ælfred_þe_Lef wrote: > On Tue, 19 Feb 2002, Charles R. Hunter wrote: > > > yes and no... the VIO session underneath is not a > tty as UNIX > > programs expect. You can receive raw keyboard > codes instead of > > terminal codes ( like ANSI ) if you want. To the best of my knowledge, if using ANSI, you must set the OS/2 Command Session to ANSI using: ANSI ON (or at ANSI ON). This is similiar to using a terminal emulation program that is set to ansi. Also, one can do this in a situation, where the Unix host does not recognize your terminal, by using the, "set TERM=ansi;export $TERM" or "export $TERM=ansi". I don't know how relevant this is to the discussion, but I thought I would add my $ .02 > Incidentally, can anyone suggest a good > termcap/terminfo entry for a VIO > window? I know they are supposedly ANSI terminals > but none of the entries > distributed with EMX or ncurses (including the "os2" > terminfo entry) seem > to work quite right for me. I managed to compile > Pico/Pine using the > termcap and terminfo code instead of the old VIO > code, and it works great > in an xterm but has trouble in a VIO window with > clearing the screen and > using the cursor keys. > > Nik S. ===== Sincerely, James Cannon Using OS/2 Warp in the beautiful Wine Country of Northen California! __________________________________________________ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.yahoo.com **= Email 28 ==========================** Date: Tue, 19 Feb 2002 15:28:08 +0000 From: John Poltorak Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 09:56:10AM -0500, Charles R. Hunter wrote: > On Tue, Feb 19, 2002 at 02:02:04PM +0000, John Poltorak (jp at eyup.org) wrote: > > > > Does the file termcap.dat exist on Unix systems? > > /etc/termcap I think we should use the same name on UnixOS/2... I did find such a file on my FreeBSD CD but it was zero length. > > > > Where would I expect to find it, and is it likely to be usable on OS/2? > > yes and no... the VIO session underneath is not a tty as UNIX > programs expect. You can receive raw keyboard codes instead of > terminal codes ( like ANSI ) if you want. > > This work has already been done for you with EMX's libtermcap.a Where exactly? The components of termcap.a are termcap.h, termcap.c, tparam.c, version.c and termcap.dat. I'm just trying to establish the source of all these files. There seem to be a number of different termcap.h files around, and I'd like to know how EMX's termcap differs from the one in BSD, but I can't find the source for that one... > Charles > > > > > > > -- > > John > > > > -- > Charles R. Hunter > Director, Physics Computer Network > Purdue University crh at physics.purdue.edu -- John **= Email 29 ==========================** Date: Tue, 19 Feb 2002 15:36:57 +0900 From: Akira Hatakeyama Subject: Re: LESS v370 On Mon, Feb 18, 2002 at 05:40:11PM +0000, John Poltorak wrote: > There is an OS/2 port of LESS v370 on Hobbes which is maintained by > Kyosuke Tokoro . Do any of our Japanese friends know > of him? I'd like to try and get his patches and rebuild LESS if possible, > but have been unable to get any reply from him. Seems no update since 2002/01/01. But as his 'Whats New' section (in japanese), his patches was inlcuded in original less version. You can get it from http://www.greenwoodsoftware.com/less/ -- Akira Hatakeyama E-Mail: akira at sra.co.jp http://www.sra.co.jp/people/akira/index.html chigasaki-minami, tsuzuki ward, yokohama, japan **= Email 30 ==========================** Date: Tue, 19 Feb 2002 16:08:41 +0000 From: John Poltorak Subject: Re: Re: EMACS build error On Wed, Feb 20, 2002 at 01:01:50AM +0900, Masaru Nomiya wrote: > Hello, > > My mail is empty! What's the matter, I wonder? > > I post again. It may be a problem with Majordomo... if there is a Cc: in the msg, a blank msg goes to that recipient :-(... -- John **= Email 31 ==========================** Date: Tue, 19 Feb 2002 16:34:07 +0000 From: John Poltorak Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 11:21:28AM -0500, Thomas Dickey wrote: > On Tue, Feb 19, 2002 at 11:01:05AM -0500, Charles R. Hunter wrote: > > > There seem to be a number of different termcap.h files around, and I'd > > > like to know how EMX's termcap differs from the one in BSD, but I can't > > > find the source for that one... > > is he looking for the "the" termcap in BSD, Yes... > or the source for EMX. It > sounds like the former. Actually, EMX's termcap wasn't based on that. It looks to me as though EMX's termcap is based on GLIBC, although I'm not sure of the significance of having both btermcap and termcap. I'd like to know if a BSD based termcap could be created but don't know where to look. > > charles > > -- > Thomas E. Dickey > http://invisible-island.net > ftp://invisible-island.net -- John **= Email 32 ==========================** Date: Tue, 19 Feb 2002 16:34:53 -0500 From: Thomas Dickey Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 08:52:41PM +0000, John Poltorak wrote: > Now, according to man termcap:- > > DESCRIPTION > These functions extract and use capabilities from a terminal capability > data base, usually /usr/share/misc/termcap > > > I guess every flavour of Unix puts this file in a different place, using > a different format, and a different name.... not exactly - it used to be "always" /etc/termcap, but was moved as part of the rearrangement of filesystems in the past several years. -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net **= Email 33 ==========================** Date: Tue, 19 Feb 2002 17:32:00 +0000 From: John Poltorak Subject: Re: Newbie Q re gcc 3.0.2 On Tue, Feb 19, 2002 at 06:24:09PM +0100, Anders Jarnberg wrote: > On Mon, 18 Feb 2002 09:06:03 +0000, John Poltorak wrote: > >> Yep, still dumps core. > >Does v2.8.1 work OK? > > Yes, 2.8.1 works ok and can compile code. > > > >> >Sounds like something wrong in the environment. > >> Probably, but I'm having trouble finding it. Did I mention I'm > >> running on eCS 1.0 ? > >I can't see how this would be a factor. > > I've now done the same installation > on a box with Warp 4 & FP10 and I can get 3.0.2 working > without any problems there. The same installation on > eCS creates a core when compiling. > > Has anyone gotten 3.0.2 to work on eComStation 1.0 ? ISTR seeing some GNU stuff such as SED on eCS. Maybe there are also some DLLs conflicting with gcc 3.0.2. Check the libpath. > > -- > Anders Jarnberg > in Stockholm, Sweden > running eComStation & Mandrake > > -- John **= Email 34 ==========================** Date: Tue, 19 Feb 2002 17:50:20 -0500 From: Thomas Dickey Subject: Re: Termcap and VIO (was: termcap.dat) On Wed, Feb 20, 2002 at 09:01:04AM +1100, ?lfred ?e Lef wrote: > On Tue, 19 Feb 2002, Charles R. Hunter wrote: > > > yes and no... the VIO session underneath is not a tty as UNIX > > programs expect. You can receive raw keyboard codes instead of > > terminal codes ( like ANSI ) if you want. > > Incidentally, can anyone suggest a good termcap/terminfo entry for a VIO > window? I know they are supposedly ANSI terminals but none of the entries > distributed with EMX or ncurses (including the "os2" terminfo entry) seem > to work quite right for me. I managed to compile Pico/Pine using the > termcap and terminfo code instead of the old VIO code, and it works great > in an xterm but has trouble in a VIO window with clearing the screen and > using the cursor keys. I generally use the 'ansi' one in ncurses' emx.src (but that wouldn't be applicable to xterm). That's a little different from 'os2'. -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net **= Email 35 ==========================** Date: Tue, 19 Feb 2002 18:24:09 +0100 (CET) From: "Anders Jarnberg" Subject: Re: Newbie Q re gcc 3.0.2 On Mon, 18 Feb 2002 09:06:03 +0000, John Poltorak wrote: >> Yep, still dumps core. >Does v2.8.1 work OK? Yes, 2.8.1 works ok and can compile code. >> >Sounds like something wrong in the environment. >> Probably, but I'm having trouble finding it. Did I mention I'm >> running on eCS 1.0 ? >I can't see how this would be a factor. I've now done the same installation on a box with Warp 4 & FP10 and I can get 3.0.2 working without any problems there. The same installation on eCS creates a core when compiling. Has anyone gotten 3.0.2 to work on eComStation 1.0 ? -- Anders Jarnberg in Stockholm, Sweden running eComStation & Mandrake **= Email 36 ==========================** Date: Tue, 19 Feb 2002 19:05:47 +0000 From: John Poltorak Subject: Re: MAN pages On Wed, Feb 20, 2002 at 12:53:43AM +0900, Jun Sawataishi wrote: > If anyone wants to use newly ported GNU groff or man, please mail me. > In a month or in a week, I will put them in my web page and hobbes. I'm interested in getting uptodate versions of everything... > # OS/2 is not a question, it's a solution. > # SAWATAISHI Jun > # http://www2s.biglobe.ne.jp/~vtgf3mpr/ > -- John **= Email 37 ==========================** Date: Tue, 19 Feb 2002 19:19:50 -0500 From: Thomas Dickey Subject: Re: Termcap and VIO (was: termcap.dat) On Wed, Feb 20, 2002 at 11:02:28AM +1100, Nichals Sheppard wrote: > On Tue, 19 Feb 2002, James Cannon wrote: > > > To the best of my knowledge, if using ANSI, you must > > set the OS/2 Command Session to ANSI using: > > > > ANSI ON (or at ANSI ON). > > I have ANSI on; if you don't you just get a whole lot of escape codes > appearing on the screen. The problem is that the ansi definition supplied > by EMX and the version of ncurses I have don't have the escape codes for > the extended keys in them, and some don't have "clear to end of display" > which Pico uses to clear the screen. I'll try Thomas Dickey's suggestion > and see how it goes. iirc, EMX doesn't implement the clear to end of display (nor scrolling). Though that might simply be that I overlooked it (being more intent on verifying the correctness of a particular terminfo entry than finding omissions). But the extended keys are present - I tested that, in particular because I was curious about the embedded nulls in some of the strings. -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net **= Email 38 ==========================** Date: Tue, 19 Feb 2002 19:40:04 +0000 From: John Poltorak Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 12:04:49PM -0500, Thomas Dickey wrote: > On Tue, Feb 19, 2002 at 04:34:07PM +0000, John Poltorak wrote: > "the" termcap doesn't exist > > (the 3 BSD's have three views of this - some are based on ncurses, and the > ones that are not are conceded to be poorly maintained) > > > or the source for EMX. It > > > sounds like the former. Actually, EMX's termcap wasn't based on that. > > > > It looks to me as though EMX's termcap is based on GLIBC, although I'm not > > sure of the significance of having both btermcap and termcap. > > glibc doesn't distribute termcap. termcap is a separate item. (Actually > I've seen more than one termcap for EMX, and incorporated much of the > more recent one into ncurses as emx.src). None of the versions that I've > seen are based on the BSD termcap, but instead are a mix of generic stuff > (such as vt100 and xterm) and OS/2-specific stuff that isn't found in that > form elsewhere. > > > I'd like to know if a BSD based termcap could be created but don't know > > where to look. > > ftp://invisible-island.net/ncurses/termcap.src.gz I was actually looking for source to create a libtermcap, but this file is interesting in itself... There is mention of emx entries - it says 'distributed with EMX 0.9b'. I don't have this version available, but the emx section doesn't bear much of a resemblance to termcap.dat which I have in \emx\etc from 0.9d. Is there any indication of exactly where these entries came from or who put them there? > -- > Thomas E. Dickey > http://invisible-island.net > ftp://invisible-island.net -- John **= Email 39 ==========================** Date: Tue, 19 Feb 2002 20:47:59 +0000 From: John Poltorak Subject: termcap.c After searching round the Internet for a while, I finally discovered *two* versions of termcap.c right under my nose in my emx distribution:- C:\emx\bsd\libterm\termcap.c C:\emx\gnu\termcap\termcap.c Both directories have a makefile which creates termcap.a... The former seems to get renamed as btermcap.a though. I can find the source of the GNU file, but the BSD version is still proving difficult to locate. So, can anyone explain why we have two such files? Has anyone come across any makefiles which need a -lbtermcap ? -- John **= Email 40 ==========================** Date: Tue, 19 Feb 2002 20:52:41 +0000 From: John Poltorak Subject: Re: termcap.dat On Tue, Feb 19, 2002 at 11:01:05AM -0500, Charles R. Hunter wrote: > On Tue, Feb 19, 2002 at 03:28:08PM +0000, John Poltorak (jp at eyup.org) wrote: > > On Tue, Feb 19, 2002 at 09:56:10AM -0500, Charles R. Hunter wrote: > > > On Tue, Feb 19, 2002 at 02:02:04PM +0000, John Poltorak (jp at eyup.org) wrote: > > > > > > > > Does the file termcap.dat exist on Unix systems? > > > > > > /etc/termcap > > > > I think we should use the same name on UnixOS/2... > agreed. Now, according to man termcap:- DESCRIPTION These functions extract and use capabilities from a terminal capability data base, usually /usr/share/misc/termcap I guess every flavour of Unix puts this file in a different place, using a different format, and a different name.... > charles > > -- > Charles R. Hunter > Director, Physics Computer Network > Purdue University crh at physics.purdue.edu -- John **= Email 41 ==========================** Date: Tue, 19 Feb 2002 21:24:00 -0800 (PST) From: James Cannon Subject: Re: Termcap and VIO (was: termcap.dat) Have you tried VioGetAnsi? this call returns the current ANSI status On/Off state. #define INCL_VIO #include USHORT VioGetAnsi (PUSHORT pfAnsi, HVIO hvio); pfAnsi (PUSHORT) - output Address of the current ANSI status. A value of indicates ANSI is active, and a value of 0 indicates ANSI is not active. hvio (HVIO) - input This must be zero unless the caller is a Presentation Manager application, in which case it must be the value returned by VioGetPs. Return value 0 NO_ERROR 436 ERROR_VIO_INVALID_HANDLE 465 ERROR_VIO_DETACHED --- Thomas Dickey wrote: > On Wed, Feb 20, 2002 at 11:02:28AM +1100, Nichals > Sheppard wrote: > > On Tue, 19 Feb 2002, James Cannon wrote: > > > > > To the best of my knowledge, if using ANSI, you > must > > > set the OS/2 Command Session to ANSI using: > > > > > > ANSI ON (or at ANSI ON). > > > > I have ANSI on; if you don't you just get a whole > lot of escape codes > > appearing on the screen. The problem is that the > ansi definition supplied > > by EMX and the version of ncurses I have don't > have the escape codes for > > the extended keys in them, and some don't have > "clear to end of display" > > which Pico uses to clear the screen. I'll try > Thomas Dickey's suggestion > > and see how it goes. > > iirc, EMX doesn't implement the clear to end of > display (nor scrolling). > Though that might simply be that I overlooked it > (being more intent on > verifying the correctness of a particular terminfo > entry than finding > omissions). But the extended keys are present - I > tested that, in particular > because I was curious about the embedded nulls in > some of the strings. > > -- > Thomas E. Dickey > http://invisible-island.net > ftp://invisible-island.net > ===== Sincerely, James Cannon Using OS/2 Warp in the beautiful Wine Country of Northen California! __________________________________________________ Do You Yahoo!? Yahoo! Sports - Coverage of the 2002 Olympic Games http://sports.yahoo.com **= Email 42 ==========================** Date: Tue, 19 Feb 2002 22:25:19 +0100 From: Kris Steenhaut Subject: Re: OSFree... Adrian Gschwend schreef: > On Tue, 19 Feb 2002 14:17:53 +0100, Kris Steenhaut wrote: > > >Big Blue already has pulled the plug out of MCP, so there isn't much left > >for them to dislike. > > that doesn't change anything. I agree that the only solution for OS/2 > is a free OS based on OS/2 ideas What's an "OS/2" idea? > (and maybe parts of it if we could get > that from IBM officialy). But stolen source does not help at all, in no > way. Stolen? Be happy I'm not concerned. Otherwise I could/would sue you. As a starter, on what evidence is your statement based there has been stolen anything? Maybe it's better to think twice before making such serious allegations. > It will just make IBM even more closed against such ideas. > By pulling the plug out of MCP Big Blue finally tries to perform the (OS/2) kill by itself. So, their opinion doesn't matter much anymore. > > >Would be hard to prove either way, isn't it? > > well the build level is from the leaked sources so not that hard. For > sure the prove would be to see it but it's quite obvious. > Could only be done in court. -- Groeten uit Gent, Kris **= Email 43 ==========================** Date: Tue, 19 Feb 2002 22:30:56 +0100 (CET) From: Stefan Neis Subject: Re: LIBEMX suggestion On Mon, 18 Feb 2002, John Poltorak wrote: > > Omflibs.cmd is quite useful, even if this new distro > > shouldn't need it. > > > omflibs only seems to be a wrapper for emxomf and looks to be specific to > the .a libs supplied with EMX/GCC. But it's a nice example on how to use emxomf. To me, it was more helpful to see a real example than to read the manual page. > I was thinking of running omflibs as part of the installation routine > instead of including .lib files, but I'll probably leave that until > another time. That would be nice though. A huge decrease in download size (which means download time which means money over here) for a very small penalty in computing time is something I always do like. BTW, if you're removing this BSD curses stuff, I think it would be nice to include ncurses headers and lib, so whatever software using anything like curses (less?) can easily be reconfigured and rebuilt. Regards, Stefan -- Micro$oft is not an answer. It is a question. The answer is 'no'. **= Email 44 ==========================** Date: Tue, 19 Feb 2002 22:32:41 +0100 (CET) From: Stefan Neis Subject: Re: LIBEMX suggestion On Mon, 18 Feb 2002, John Poltorak wrote: > include/bsd/* only contains headers that can be superceeded by ncurses, > but I'm probably mistaken about deleting bsd.a. Oh yes, right. You are. It contains useful stuff like (s)random which is frequently needed. Regards, Stefan -- Micro$oft is not an answer. It is a question. The answer is 'no'. **= Email 45 ==========================** Date: Tue, 19 Feb 2002 22:40:45 +0000 From: "Lyn St George" Subject: Re: OSFree... On Tue, 19 Feb 2002 23:07:27 +0100 (CET), Stefan Neis wrote: >On Tue, 19 Feb 2002, Adrian Gschwend wrote: > >> well the build level is from the leaked sources so not that hard. For >> sure the prove would be to see it but it's quite obvious. > >Well, cloning the kernel's build level information might be very helpful >in making the rest of the system work, even if the kernel is based on >a completely independent source code. So I think one shouldn't jump to >conclusions just from the build level which the thing displays... > >And I wonder how the kernel sources could possibly have leaked... >Bad publicity for IBM, if they have to admit _that_. That could >really cause some harm to them - some independent kernel for some >OS which the don't even market seems to be by far the smaller >problem. ;-) The sources /did/ leak some time ago - this happened just after Stardock's attempt to become a reseller of an OEM version of OS/2 failed. It was apparently everything except the WPS which "leaked" out. There was a lot of heat and bluster about at the time - "IBM announces death of OS/2" on Slashdot - and it seems that certain people within IBM let the hot blood get to their heads. I had always been under the impression though that the sources were pretty much locked up at Netlabs and had never been made widely available - though this could well be way off. > Regards, > Stefan >-- >Micro$oft is not an answer. It is a question. The answer is 'no'. > > Cheers Lyn St George +--------------------------------------------------------------------------------- + http://www.zolotek.net .. eCommerce hosting, consulting + http://www.os2docs.org .. some 'How To' stuff ... +---------------------------------------------------------------------------------- **= Email 46 ==========================** Date: Tue, 19 Feb 2002 22:55:44 +0100 (CET) From: Stefan Neis Subject: Re: Sleepycat Software licence On Mon, 18 Feb 2002, John Poltorak wrote: > * 3. Redistributions in any form must be accompanied by information on > * how to obtain complete source code for the DB software and any ^^^^^^^ > * accompanying software that uses the DB software. ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ I don't like that kind of stuff for my own software, but then I'm not using db.a anyway. Still, with that kind of licence change, it would be nice to keep the old version available somewhere, even if it is only an incomplete replacement. And put a prominent reminder into documentation that "(lib)db.a MAY BE USED IN OPEN SOURCE SOFTWARE ONLY!!!". Regards, Stefan -- Micro$oft is not an answer. It is a question. The answer is 'no'. **= Email 47 ==========================** Date: Tue, 19 Feb 2002 23:07:27 +0100 (CET) From: Stefan Neis Subject: Re: OSFree... On Tue, 19 Feb 2002, Adrian Gschwend wrote: > well the build level is from the leaked sources so not that hard. For > sure the prove would be to see it but it's quite obvious. Well, cloning the kernel's build level information might be very helpful in making the rest of the system work, even if the kernel is based on a completely independent source code. So I think one shouldn't jump to conclusions just from the build level which the thing displays... And I wonder how the kernel sources could possibly have leaked... Bad publicity for IBM, if they have to admit _that_. That could really cause some harm to them - some independent kernel for some OS which the don't even market seems to be by far the smaller problem. ;-) Regards, Stefan -- Micro$oft is not an answer. It is a question. The answer is 'no'. **= Email 48 ==========================** Date: Tue, 19 Feb 2002 23:18:31 +0100 (CET) From: Stefan Neis Subject: Re: termcap.dat On Tue, 19 Feb 2002, Charles R. Hunter wrote: > agreed. I don't know Why Eberhardt Mattes used termcap.dat in > the first place... maybe for early compatibility with djgpp or > something. ( since EMX was originally dual mode DOS and OS/2 ) You might be surprised to hear that emx-0.9d still is dual mode. Actually, more like triple mode, using RSX and or RSXNT DLL's generated executables still work even on the various windows flavours. Not that anyone seems to be actively using this object. For some unknown reason Cygwin (which still seems to be a rather poor alternative when compared to RSXNT) gets all the hype. Cygnus way to do things reminds me a bit of MS's marketing. :-( > > I did find such a file on my FreeBSD CD but it was zero length. There seems to be a tendency to replace termcap by terminfo which means that termcap file is unused (thus empty in newer distributions) and termcap functions just call their terminfo counterparts. I'm not sure yet of the pattern, I though it was the usual SysV vs. BSD kind of stuff but with FreeBSD using terminfo as well, things get confusing. Maybe just newer vs. older releases? Regards, Stefan -- Micro$oft is not an answer. It is a question. The answer is 'no'. **= Email 49 ==========================** Date: Tue, 19 Feb 2002 23:43:24 +0100 From: Andreas Buening Subject: Re: total.cmd John Poltorak wrote: > > On Tue, Feb 19, 2002 at 07:37:26AM -0500, Charles R. Hunter wrote: > > On Mon, Feb 18, 2002 at 10:55:48AM +0000, John Poltorak (jp at eyup.org) wrote: > > > > > > I've written this simple AWK script called total.cmd which totals up the > > > values in a file:- > > > > > > > > > > [SNIP] > > > I don't see why every line is getting printed. How do I surpress it? > > > > > > > what version of awk are you using? > > C:\>awk --version > GNU Awk 3.1.0 > Copyright (C) 1989, 1991-2001 Free Software Foundation. > > Actually, I just tried it with an older port and it worked there too, so > it looks as though it is a problem in the port of v3.1.0... Yes, I can reproduce this behaviour with cmd and gawk 3.1.0. However, this bug seems to be fixed in gawk 3.1.30 (not officially released). bye, Andreas -- One OS to rule them all, One OS to find them, One OS to bring them all and in the darkness bind them In the Land of Redmond where the Shadows lie. **= Email 50 ==========================** Date: Tue, 19 Feb 2002 23:44:12 +0100 From: Andreas Buening Subject: Re: EMACS build error Holger Veit wrote: > On Tue, Feb 19, 2002 at 11:53:19AM +0000, John Poltorak wrote: [compiling emacs] > > Where should _fchown be defined ? > > In emxlibc :-) It isn't... Only an idea: Would it be possible to define these function (fchown(), link(), ) somewhere in the EMX headers? I'm thinking about something like #ifdef __DUMMY_FUNCTIONS__ int fchown(...) { errno = WHATEVER; return -1; } int link(...) { errno = WHATEVER; return -1; } ... #endif This could help in porting applications that use those functions but don't really depend on them. (at least every Unix program _should_ be able to handle a failure of those functions because the user really might have no permission to execute them). Using -D__DUMMY_FUNCTIONS__ as compiler flag would be much easier than patching every source file. And, the emx* libraries wouldn't need a recompilation. It's just an idea. bye, Andreas -- One OS to rule them all, One OS to find them, One OS to bring them all and in the darkness bind them In the Land of Redmond where the Shadows lie. **= Email 51 ==========================** Date: Tue, 19 Feb 2002 23:59:21 +0100 (CET) From: Stefan Neis Subject: Re: OSFree... On Tue, 19 Feb 2002, Lyn St George wrote: > The sources /did/ leak some time ago - this happened just after > Stardock's attempt to become a reseller of an OEM version of > OS/2 failed. It was apparently everything except the WPS which > "leaked" out. I see. And I suppose that OSFree happens to show the build level of exactly that leaked version? I have to admit that's at least a strange "coincidence". And extremely "stupid" to not just change it at least a bit, if it really is the leaked code just in recompiled form. Thanks for clarification, Stefan -- Micro$oft is not an answer. It is a question. The answer is 'no'.