From: UnixOS2 Archive To: "UnixOS2 Archive" Date: Thu, 18 Apr 2002 04:23:48 EST-10EDT,10,-1,0,7200,3,-1,0,7200,3600 Subject: [UnixOS2_Archive] No. 197 ************************************************** Wednesday 17 April 2002 Number 197 ************************************************** Subjects for today 1 Re: CDDB Server : Thomas E. Dickey" 2 Re: CDDB Server : Thomas E. Dickey" 3 Re: CDDB Server : Thomas E. Dickey" 4 Re: CDDB Server : John Poltorak 5 Re: Building GTAK : Ken Ames 6 Re: CDDB Server : John Poltorak 7 "Re: Problem with cmd that calls Perl" : Edwin =?iso-8859-1?Q?G=FCnthner?= 8 Re: CDDB Server : Holger Veit 9 Re: compiling c++ program under gcc 3.0.3 : Andrew Zabolotny" 10 Re: CDDB Server : Thomas Dickey 11 Building GTAK : John Poltorak 12 Re: Building GTAK : John Poltorak 13 Re: "Re: Problem with cmd that calls Perl" : Henry Sobotka **= Email 1 ==========================** Date: Thu, 18 Apr 2002 05:46:01 -0400 (EDT) From: "Thomas E. Dickey" Subject: Re: CDDB Server On Thu, 18 Apr 2002, John Poltorak wrote: > > This is a typedef in some sys header file. It is quite standard. > > If you don't have it, simply add a > > typedef int socklen_t; > > somewhere. > > 'somewhere' is a little vague... Isn't there a correct place for this to > be defined? Doesn't ANSI have any mention of it? > > > Holger ANSI doesn't define anything related to networking. (POSIX ignores most of it as well fwiw). -- T.E.Dickey http://invisible-island.net ftp://invisible-island.net **= Email 2 ==========================** Date: Thu, 18 Apr 2002 06:11:05 -0400 (EDT) From: "Thomas E. Dickey" Subject: Re: CDDB Server On Thu, 18 Apr 2002, John Poltorak wrote: > On Thu, Apr 18, 2002 at 05:46:01AM -0400, Thomas E. Dickey wrote: > > On Thu, 18 Apr 2002, John Poltorak wrote: > > > > > > This is a typedef in some sys header file. It is quite standard. > > > > If you don't have it, simply add a > > > > typedef int socklen_t; > > > > somewhere. > > > > > > 'somewhere' is a little vague... Isn't there a correct place for this to > > > be defined? Doesn't ANSI have any mention of it? > > > > > > > Holger > > > > ANSI doesn't define anything related to networking. (POSIX ignores most > > of it as well fwiw). > > Well if there isn't a standard place to define socklen_t where is it > defined in *BSD? I don't see it defined in SunOS 4.x or Solaris. The former is BSD system (perhaps the code you're looking at is a later version). -- T.E.Dickey http://invisible-island.net ftp://invisible-island.net **= Email 3 ==========================** Date: Thu, 18 Apr 2002 06:14:48 -0400 (EDT) From: "Thomas E. Dickey" Subject: Re: CDDB Server On Thu, 18 Apr 2002, John Poltorak wrote: > > ANSI doesn't define anything related to networking. (POSIX ignores most > > of it as well fwiw). > > Well if there isn't a standard place to define socklen_t where is it > defined in *BSD? for instance http://sprg.ssl.berkeley.edu/~fastops/ITOS/portable/socklen$_t.html (since the mindset in *BSD is not-invented-here, it's hard to relate it to "standards"). -- T.E.Dickey http://invisible-island.net ftp://invisible-island.net **= Email 4 ==========================** Date: Thu, 18 Apr 2002 08:46:32 +0100 From: John Poltorak Subject: Re: CDDB Server On Wed, Apr 17, 2002 at 11:17:51PM +0200, Holger Veit wrote: > On Tue, Apr 16, 2002 at 08:16:55PM -0800, Dave and Natalie wrote: > > On Tue, 16 Apr 2002 19:28:33 -0800, Dave and Natalie wrote: > > > > >On Tue, 16 Apr 2002 19:31:38 +0100, John Poltorak wrote: > > > > > >> > > >>Has anyone tried building a CDDB server? > > >> > > >>It's available here:- > > >> > > >>ftp://ftp.freedb.org/pub/freedb/cddbd-1.4.tar.gz > > >> > > >> > > >>I tried but got a number of compile errors:- > > >> > > >>gcc -O -c inet.c > > >>inet.c: In function `get_rmt_hostname': > > >>inet.c:132: `socklen_t' undeclared (first use in this function) > > > > This function isn't implemented in emx, maybe in posix/2? > > This is a typedef in some sys header file. It is quite standard. > If you don't have it, simply add a > typedef int socklen_t; > somewhere. 'somewhere' is a little vague... Isn't there a correct place for this to be defined? Doesn't ANSI have any mention of it? > Holger -- John **= Email 5 ==========================** Date: Thu, 18 Apr 2002 10:01:29 -0400 (EDT) From: Ken Ames Subject: Re: Building GTAK On Thu, 18 Apr 2002 16:14:58 +0100, John Poltorak wrote: > >As it stands, GTAK258 will not build out of the box, but someone has >unearthed a very old copy here:- > >ftp://ftp.lair.co.za/pub/hobbes.cd1/all/diskutil/GTAK100S.ZIP > >Does anyone know if this can be rebuilt? Looks like Zortec C may do the trick. > >It actually came out before a 32-bit OS/2 so there may be some problems... > >Apparently it won't build with gcc, and I don't know what could be used >instead. > > >-- >John > Ken **= Email 6 ==========================** Date: Thu, 18 Apr 2002 10:56:03 +0100 From: John Poltorak Subject: Re: CDDB Server On Thu, Apr 18, 2002 at 05:46:01AM -0400, Thomas E. Dickey wrote: > On Thu, 18 Apr 2002, John Poltorak wrote: > > > > This is a typedef in some sys header file. It is quite standard. > > > If you don't have it, simply add a > > > typedef int socklen_t; > > > somewhere. > > > > 'somewhere' is a little vague... Isn't there a correct place for this to > > be defined? Doesn't ANSI have any mention of it? > > > > > Holger > > ANSI doesn't define anything related to networking. (POSIX ignores most > of it as well fwiw). Well if there isn't a standard place to define socklen_t where is it defined in *BSD? > -- > T.E.Dickey -- John **= Email 7 ==========================** Date: Thu, 18 Apr 2002 11:13:15 +0200 From: Edwin =?iso-8859-1?Q?G=FCnthner?= Subject: "Re: Problem with cmd that calls Perl" Hi there, i now had the chance to look into my popups2.log: It says: 04-18-2002 09:45:59 SYS3175 PID 0049 TID 0001 Slot 0073 F:\PERL\BIN\PERL_.EXE c0000005 1e507e7c P1=00000001 P2=00000000 P3=XXXXXXXX P4=XXXXXXXX EAX=00000000 EBX=00000000 ECX=00000000 EDX=001cad01 ESI=00000000 EDI=0019d140 DS=0053 DSACC=d0f3 DSLIM=3fffffff ES=0053 ESACC=d0f3 ESLIM=3fffffff FS=150b FSACC=00f3 FSLIM=00000030 GS=0000 GSACC=**** GSLIM=******** CS:EIP=005b:1e507e7c CSACC=d0df CSLIM=3fffffff SS:ESP=0053:030efa98 SSACC=d0f3 SSLIM=3fffffff EBP=00000000 FLG=00012246 PHCMIB.DLL 0001:00007e7c where phcmib.dll is the one that belongs to my self-compiled module ... Does this help any further? **= Email 8 ==========================** Date: Thu, 18 Apr 2002 12:29:08 +0200 From: Holger Veit Subject: Re: CDDB Server On Thu, Apr 18, 2002 at 06:11:05AM -0400, Thomas E. Dickey wrote: > On Thu, 18 Apr 2002, John Poltorak wrote: > > > On Thu, Apr 18, 2002 at 05:46:01AM -0400, Thomas E. Dickey wrote: > > > On Thu, 18 Apr 2002, John Poltorak wrote: > > > > > > > > This is a typedef in some sys header file. It is quite standard. > > > > > If you don't have it, simply add a > > > > > typedef int socklen_t; > > > > > somewhere. > > > > > > > > 'somewhere' is a little vague... Isn't there a correct place for this to > > > > be defined? Doesn't ANSI have any mention of it? > > > > > > > > > Holger > > > > > > ANSI doesn't define anything related to networking. (POSIX ignores most > > > of it as well fwiw). > > > > Well if there isn't a standard place to define socklen_t where is it > > defined in *BSD? > > I don't see it defined in SunOS 4.x or Solaris. The former is BSD system > (perhaps the code you're looking at is a later version). It is in for Solaris 5.7: #ifndef _SOCKLEN_T #define _SOCKLEN_T #if defined(_XPG4_2) && !defined(_XPG5) && !defined(_LP64) typedef size_t socklen_t; #else typedef uint32_t socklen_t; #endif /* defined(_XPG4_2) && !defined(_XPG5) && !defined(_LP64) */ #if defined(_XPG4_2) typedef socklen_t *Psocklen_t; #else typedef void *Psocklen_t; #endif /* defined(_XPG4_2) */ #endif /* _SOCKLEN_T */ This means it is sort of "defined". Holger -- Please update your tables to my new e-mail address: holger.veit$ais.fhg.de (replace the '$' with ' at ' -- spam-protection) **= Email 9 ==========================** Date: Thu, 18 Apr 2002 14:09:34 +0400 From: "Andrew Zabolotny" Subject: Re: compiling c++ program under gcc 3.0.3 On Wed, 17 Apr 2002 09:03:12 -0400, Arnstein.Prytz at jcu.edu.au wrote: >I had done all the things mentioned in the installation file, as >was evidenced by all sorts of .lib files appearing with the new >(yesterday's) date. I found, however, that I have to keep > >LIBRARY_PATH=e:/usr/local/emx/lib > >otherwise that particular program was unable to find crt0.o. >Any ideas what that is about? I cannot imagine that this >directory should not be included because a whole bunch of .lib >files were updated there as a result of the make operation. If the compiler is installed in the e:/usr/local/emx/bin directory, the above environment variable should not be needed. The OS/2 port of gcc (since 2.95.x) finds the path of executable (e.g. e:/usr/local/emx/bin/gcc.exe) and appends ".. /lib" to it and adds the resulting path to LIBRARY_PATH. Same about include files: it appends "../include;../lib/gcc-lib/3.0.1/include" to the source path and then everything to C_INCLUDE_PATH. I'm not sure why it doesn't work for you. Greetings, _\ndy **= Email 10 ==========================** Date: Thu, 18 Apr 2002 15:43:07 -0400 From: Thomas Dickey Subject: Re: CDDB Server On Thu, Apr 18, 2002 at 12:29:08PM +0200, Holger Veit wrote: > > > Well if there isn't a standard place to define socklen_t where is it > > > defined in *BSD? > > > > I don't see it defined in SunOS 4.x or Solaris. The former is BSD system > > (perhaps the code you're looking at is a later version). > > It is in for Solaris 5.7: I was looking at Solaris 2.5.1 (getting old, but he did ask about BSD - past that Sun pushed the BSD stuff even farther into the attic). > #if defined(_XPG4_2) > typedef socklen_t *Psocklen_t; > #else > typedef void *Psocklen_t; > #endif /* defined(_XPG4_2) */ > > #endif /* _SOCKLEN_T */ > > This means it is sort of "defined". sort of. The ifdef for XPG4_2 is promising. -- Thomas E. Dickey http://invisible-island.net ftp://invisible-island.net **= Email 11 ==========================** Date: Thu, 18 Apr 2002 16:14:58 +0100 From: John Poltorak Subject: Building GTAK As it stands, GTAK258 will not build out of the box, but someone has unearthed a very old copy here:- ftp://ftp.lair.co.za/pub/hobbes.cd1/all/diskutil/GTAK100S.ZIP Does anyone know if this can be rebuilt? It actually came out before a 32-bit OS/2 so there may be some problems... Apparently it won't build with gcc, and I don't know what could be used instead. -- John **= Email 12 ==========================** Date: Thu, 18 Apr 2002 17:07:16 +0100 From: John Poltorak Subject: Re: Building GTAK On Fri, Apr 19, 2002 at 02:02:06AM +1000, IanM wrote: > >As it stands, GTAK258 will not build out of the box, but someone has > >unearthed a very old copy here:- > > http://www.os2site.com/sw/util/backup/gtak258.zip > > :-) I'm obviously missing something here... Is this any different to the one on Hobbes? > Cheers > IanM -- John **= Email 13 ==========================** Date: Thu, 18 Apr 2002 19:33:04 -0400 From: Henry Sobotka Subject: Re: "Re: Problem with cmd that calls Perl" Edwin Günthner wrote: > > PHCMIB.DLL 0001:00007e7c > > where phcmib.dll is the one that belongs to my > self-compiled module ... > > Does this help any further? You'll have to look at what's happening at that address. Since your script works when launched directly, it may just be that running inside the REXX interpreter triggers bad chemistry. h~