Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.os2.programmer.porting > #219 > unrolled thread

Problems building unzip 6.10b

Back to article view | Back to comp.os.os2.programmer.porting

Contents

  Problems building unzip 6.10b Lewis Rosenthal <lgrosenthal@2-de-sp-am-2rosenthals.com> - 2013-06-01 14:07 -0400
    Re: Problems building unzip 6.10b Dave Yeo <dave.r.yeo@gmail.com> - 2013-06-01 12:07 -0700
    Re: Problems building unzip 6.10b "Andy" <abwillis1@yahoo.com> - 2013-06-01 23:30 +0000
      Re: Problems building unzip 6.10b Lewis Rosenthal <lgrosenthal@2-de-sp-am-2rosenthals.com> - 2013-06-01 22:12 -0400
        Re: Problems building unzip 6.10b Lewis Rosenthal <lgrosenthal@2-de-sp-am-2rosenthals.com> - 2013-06-01 22:41 -0400
          Re: Problems building unzip 6.10b Dave Yeo <dave.r.yeo@gmail.com> - 2013-06-01 20:12 -0700
    Re: Problems building unzip 6.10b Dave Yeo <dave.r.yeo@gmail.com> - 2013-06-01 20:10 -0700
      Re: Problems building unzip 6.10b Lewis Rosenthal <lgrosenthal@2-de-sp-am-2rosenthals.com> - 2013-06-03 22:28 -0400
      Re: Problems building unzip 6.10b Lewis Rosenthal <lgrosenthal@2-de-sp-am-2rosenthals.com> - 2013-06-04 00:28 -0400
        Re: Problems building unzip 6.10b Dave Yeo <dave.r.yeo@gmail.com> - 2013-06-03 23:28 -0700
        Re: Problems building unzip 6.10b Dave Yeo <dave.r.yeo@gmail.com> - 2013-06-03 23:50 -0700
          Re: Problems building unzip 6.10b Lewis Rosenthal <lgrosenthal@2-de-sp-am-2rosenthals.com> - 2013-06-04 13:21 -0400
            Re: Problems building unzip 6.10b Dave Yeo <dave.r.yeo@gmail.com> - 2013-06-04 19:33 -0700
              Re: Problems building unzip 6.10b Lewis Rosenthal <lgrosenthal@2-de-sp-am-2rosenthals.com> - 2013-06-05 01:18 -0400
      Re: Problems building unzip 6.10b "Andy" <abwillis1@yahoo.com> - 2013-06-04 17:14 +0000
        Re: Problems building unzip 6.10b Lewis Rosenthal <lgrosenthal@2-de-sp-am-2rosenthals.com> - 2013-06-04 13:23 -0400
          Re: Problems building unzip 6.10b "Andy" <abwillis1@yahoo.com> - 2013-06-04 20:50 +0000
            Re: Problems building unzip 6.10b Lewis Rosenthal <lgrosenthal@2-de-sp-am-2rosenthals.com> - 2013-06-04 17:12 -0400

#219 — Problems building unzip 6.10b

Hi, guys...

Prompted by Dave's mention in eCS-Technical (see
http://tech.groups.yahoo.com/group/eCS-Technical/message/17592 et seq)
and by http://mantis.smedley.id.au/view.php?id=549 , I decided to try to
give building the latest Unzip a shot...

I'm not making much headway. If I set aside all of the options I'd like
to add, and from bash, just run:

[Apollo|J:/devel/unzip610b]make -f os2/makefile.os2 gcc
make -f os2/makefile.os2 all \
        CC="gcc -Zomf -O -I." \
        CFLAGS="-Wall -DOS2 -DASM_CRC " \
        NFLAGS="" \
        DLLFLAG="" \
        AS="gcc -Zomf" \
        ASFLAGS="-Di386" \
        LDFLAGS="-o ./" \
        LDFLAGS2="-Zsmall-conv -Zstack 320 -Zsys -s" \
        OUT="-o" \
        OBJ=".obj" \
        CRCA_O="crc_gcc.obj" \
        DEF="os2/unzip.def"
make.exe[1]: Entering directory `J:/devel/unzip610b'
gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC   unzip.c
In file included from unzpriv.h:637,
                 from unzip.h:734,
                 from unzip.c:69:
/@unixroot/usr/include/ctype.h:42: error: expected declaration
specifiers or '..
.' before '(' token
/@unixroot/usr/include/ctype.h:45: error: expected declaration
specifiers or '..
.' before '(' token
make.exe[1]: *** [unzip.obj] Error 1
make.exe[1]: Leaving directory `J:/devel/unzip610b'
make: *** [gcc] Error 2

The INSTALL file specifically states that for OS/2, makefile.os2 should
remain in the os2 directory and be called with -f as I have above.

I've actually had some recent successes in building things with gcc
4.4.6, so I think that my build system is relatively intact.

I understand (I think) that the "expected declaration" errors shown are
displayed as a result of -Wall, but looking at ctype.h, I don't
understand what's so different about those two particular lines than the
rest:

[32]	__BEGIN_DECLS
[33]	int     isalnum(int);
[34]	int     isalpha(int);
[35]	int     iscntrl(int);
[36]	int     isdigit(int);
[37]	int     isgraph(int);
[38]	int     islower(int);
[39]	int     isprint(int);
[40]	int     ispunct(int);
[41]	int     isspace(int);
[42]	int     isupper(int);
[43]	int     isxdigit(int);
[44]	int     toupper(int);
[45]	int     tolower(int);

So, what's it complaining about? Am I missing a locale specifier somewhere?

TIA

-- 
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC                www.2rosenthals.com
Need a managed Wi-Fi hotspot?                www.hautspot.com
visit my IT blog                www.2rosenthals.net/wordpress
-------------------------------------------------------------

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [next] | [standalone]

#220

Lewis Rosenthal wrote:
> Hi, guys...
>
> Prompted by Dave's mention in eCS-Technical (see
> http://tech.groups.yahoo.com/group/eCS-Technical/message/17592 et seq)
> and by http://mantis.smedley.id.au/view.php?id=549 , I decided to try to
> give building the latest Unzip a shot...
>
> I'm not making much headway. If I set aside all of the options I'd like
> to add, and from bash, just run:
>
> [Apollo|J:/devel/unzip610b]make -f os2/makefile.os2 gcc
> make -f os2/makefile.os2 all \
>          CC="gcc -Zomf -O -I." \
>          CFLAGS="-Wall -DOS2 -DASM_CRC " \
>          NFLAGS="" \
>          DLLFLAG="" \
>          AS="gcc -Zomf" \
>          ASFLAGS="-Di386" \
>          LDFLAGS="-o ./" \
>          LDFLAGS2="-Zsmall-conv -Zstack 320 -Zsys -s" \

This stack declaration is way to small and possibly what is causing the 
crash. The default of 1MB is probably sufficient. Actually just comment 
out the whole line as the rest is for EMX except -s, you can strip it 
with lxlite later.


>          OUT="-o" \
>          OBJ=".obj" \
>          CRCA_O="crc_gcc.obj" \
>          DEF="os2/unzip.def"
> make.exe[1]: Entering directory `J:/devel/unzip610b'
> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC   unzip.c
> In file included from unzpriv.h:637,
>                   from unzip.h:734,
>                   from unzip.c:69:
> /@unixroot/usr/include/ctype.h:42: error: expected declaration
> specifiers or '..
> .' before '(' token
> /@unixroot/usr/include/ctype.h:45: error: expected declaration
> specifiers or '..
> .' before '(' token
> make.exe[1]: *** [unzip.obj] Error 1
[...]

>
> So, what's it complaining about? Am I missing a locale specifier somewhere?

It's probably a totally unrelated error. Try commenting out the #include 
<ctype.h> line to see if you get a more meaningful error or add -E to 
the CFLAGS to get the preprocessor output and examine unzip.o (or unzip.obj)
Dave

[toc] | [prev] | [next] | [standalone]

#221

On Sat, 1 Jun 2013 18:07:55 UTC, Lewis Rosenthal 
<lgrosenthal@2-de-sp-am-2rosenthals.com> wrote:

<snip>
> make.exe[1]: Entering directory `J:/devel/unzip610b'
> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC   unzip.c
> In file included from unzpriv.h:637,
>                  from unzip.h:734,
>                  from unzip.c:69:
> /@unixroot/usr/include/ctype.h:42: error: expected declaration
> specifiers or '..
> .' before '(' token
> /@unixroot/usr/include/ctype.h:45: error: expected declaration
> specifiers or '..
> .' before '(' token
> make.exe[1]: *** [unzip.obj] Error 1
> make.exe[1]: Leaving directory `J:/devel/unzip610b'
> make: *** [gcc] Error 2
> 
<snip>
I just reproduced it with my environment I use for building Mozilla.  
I also decided to try OpenWatcom and it gave slightly different error 
about expecing a ) and finding a ( but also in ctype.h but I haven't 
found anything else so far.
Andy
-- 

[toc] | [prev] | [next] | [standalone]

#222

On 06/01/13 07:30 pm, Andy thus wrote :
> On Sat, 1 Jun 2013 18:07:55 UTC, Lewis Rosenthal 
> <lgrosenthal@2-de-sp-am-2rosenthals.com> wrote:
> 
> <snip>
>> make.exe[1]: Entering directory `J:/devel/unzip610b'
>> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC   unzip.c
>> In file included from unzpriv.h:637,
>>                  from unzip.h:734,
>>                  from unzip.c:69:
>> /@unixroot/usr/include/ctype.h:42: error: expected declaration
>> specifiers or '..
>> .' before '(' token
>> /@unixroot/usr/include/ctype.h:45: error: expected declaration
>> specifiers or '..
>> .' before '(' token
>> make.exe[1]: *** [unzip.obj] Error 1
>> make.exe[1]: Leaving directory `J:/devel/unzip610b'
>> make: *** [gcc] Error 2
>>
> <snip>
> I just reproduced it with my environment I use for building Mozilla.  
> I also decided to try OpenWatcom and it gave slightly different error 
> about expecing a ) and finding a ( but also in ctype.h but I haven't 
> found anything else so far.
> 
Thanks, guys. I'll give this another go this evening. I'm glad I'm not
the only one seeing weirdness, here, at least.

-- 
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC                www.2rosenthals.com
Need a managed Wi-Fi hotspot?                www.hautspot.com
visit my IT blog                www.2rosenthals.net/wordpress
-------------------------------------------------------------

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [prev] | [next] | [standalone]

#223

On 06/01/13 10:12 pm, Lewis Rosenthal thus wrote :
> On 06/01/13 07:30 pm, Andy thus wrote :
>> On Sat, 1 Jun 2013 18:07:55 UTC, Lewis Rosenthal 
>> <lgrosenthal@2-de-sp-am-2rosenthals.com> wrote:
>>
>> <snip>
>>> make.exe[1]: Entering directory `J:/devel/unzip610b'
>>> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC   unzip.c
>>> In file included from unzpriv.h:637,
>>>                  from unzip.h:734,
>>>                  from unzip.c:69:
>>> /@unixroot/usr/include/ctype.h:42: error: expected declaration
>>> specifiers or '..
>>> .' before '(' token
>>> /@unixroot/usr/include/ctype.h:45: error: expected declaration
>>> specifiers or '..
>>> .' before '(' token
>>> make.exe[1]: *** [unzip.obj] Error 1
>>> make.exe[1]: Leaving directory `J:/devel/unzip610b'
>>> make: *** [gcc] Error 2
>>>
>> <snip>
>> I just reproduced it with my environment I use for building Mozilla.  
>> I also decided to try OpenWatcom and it gave slightly different error 
>> about expecing a ) and finding a ( but also in ctype.h but I haven't 
>> found anything else so far.
>>
> Thanks, guys. I'll give this another go this evening. I'm glad I'm not
> the only one seeing weirdness, here, at least.
> 
Ugh...

After hacking a little bit, (following Dave's advice, I removed the:

LDFLAGS2="-Zsmall-conv -Zstack 320 -Zsys -s" \

line completely, but there's still something which is passing these
params, as I keep bumping into stuff mentioning that string.

Further, I get a mess of stuff happening in os2acls.c (which is included
from os2.c, and deals with handling LAN Server ACLs). My head's not in
it, at least not until I pull 6.0 and give that a go to try to determine
whether there's just something with 6.10 which is causing the confusion.

Thanks again for the help, and for confirming that I'm not alone in what
I'm seeing. ;-)

-- 
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC                www.2rosenthals.com
Need a managed Wi-Fi hotspot?                www.hautspot.com
visit my IT blog                www.2rosenthals.net/wordpress
-------------------------------------------------------------

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [prev] | [next] | [standalone]

#225

Lewis Rosenthal wrote:
> Thanks again for the help, and for confirming that I'm not alone in what
> I'm seeing.

Of course I replied while you were replying :) See other post for fixes
Dave

[toc] | [prev] | [next] | [standalone]

#224

Lewis Rosenthal wrote:
> Hi, guys...
>
> Prompted by Dave's mention in eCS-Technical (see
> http://tech.groups.yahoo.com/group/eCS-Technical/message/17592 et seq)
> and by http://mantis.smedley.id.au/view.php?id=549 , I decided to try to
> give building the latest Unzip a shot...
>
> I'm not making much headway. If I set aside all of the options I'd like
> to add, and from bash, just run:
>
> [Apollo|J:/devel/unzip610b]make -f os2/makefile.os2 gcc
> make -f os2/makefile.os2 all \
>          CC="gcc -Zomf -O -I." \
>          CFLAGS="-Wall -DOS2 -DASM_CRC " \
>          NFLAGS="" \
>          DLLFLAG="" \
>          AS="gcc -Zomf" \
>          ASFLAGS="-Di386" \
>          LDFLAGS="-o ./" \
>          LDFLAGS2="-Zsmall-conv -Zstack 320 -Zsys -s" \
>          OUT="-o" \
>          OBJ=".obj" \
>          CRCA_O="crc_gcc.obj" \
>          DEF="os2/unzip.def"
> make.exe[1]: Entering directory `J:/devel/unzip610b'
> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC   unzip.c
> In file included from unzpriv.h:637,
>                   from unzip.h:734,
>                   from unzip.c:69:
> /@unixroot/usr/include/ctype.h:42: error: expected declaration
> specifiers or '..
> .' before '(' token
> /@unixroot/usr/include/ctype.h:45: error: expected declaration
> specifiers or '..
> .' before '(' token
> make.exe[1]: *** [unzip.obj] Error 1
> make.exe[1]: Leaving directory `J:/devel/unzip610b'
> make: *** [gcc] Error 2
>
> The INSTALL file specifically states that for OS/2, makefile.os2 should
> remain in the os2 directory and be called with -f as I have above.
>
> I've actually had some recent successes in building things with gcc
> 4.4.6, so I think that my build system is relatively intact.
>
> I understand (I think) that the "expected declaration" errors shown are
> displayed as a result of -Wall, but looking at ctype.h, I don't
> understand what's so different about those two particular lines than the
> rest:
>
> [32]	__BEGIN_DECLS
> [33]	int     isalnum(int);
> [34]	int     isalpha(int);
> [35]	int     iscntrl(int);
> [36]	int     isdigit(int);
> [37]	int     isgraph(int);
> [38]	int     islower(int);
> [39]	int     isprint(int);
> [40]	int     ispunct(int);
> [41]	int     isspace(int);
> [42]	int     isupper(int);
> [43]	int     isxdigit(int);
> [44]	int     toupper(int);
> [45]	int     tolower(int);
>
> So, what's it complaining about? Am I missing a locale specifier somewhere?
>

Probably missing a locale, os2cfg.h redefines toupper and tolower 
anyways so just wrap the #include <ctype.h> with #ifndef __EMX__ #endif.
I also did,
--- makefile.os2.orig   2013-06-01 19:55:04.000000000 -0700
+++ makefile.os2        2013-06-01 19:56:16.000000000 -0700
@@ -394,9 +394,9 @@
         AS="gcc -Zomf" \
         ASFLAGS="-Di386" \
         LDFLAGS="-o ./" \
-       LDFLAGS2="-Zsmall-conv -Zstack 320 -Zsys -s" \
+       LDFLAGS2="-s" \
         OUT="-o" \
-       OBJ=".obj" \
+       OBJ=".o" \
         CRCA_O="crc_gcc.obj" \
         DEF="os2/unzip.def"

and

--- os2acl.c.orig       2013-06-01 19:57:08.000000000 -0700
+++ os2acl.c    2013-06-01 19:59:42.000000000 -0700
@@ -71,7 +71,7 @@
  typedef ULONG U_INT;
  #ifdef __EMX__
  #define PSTR16 _far16ptr
-#define PTR16(x) _emx_32to16(x)
+#define PTR16(x) _libc_32to16(x)
  #else /* other 32-bit */
  #define PSTR16 PCHAR16
  #define PTR16(x) ((PCHAR16)(x))
@@ -123,7 +123,7 @@
             _THUNK_FLAT (pbBuffer);
             _THUNK_SHORT (cbBuffer);
             _THUNK_FLAT (pcbTotalAvail);
-           _THUNK_CALLI (_emx_32to16(_NetAccessGetInfo)));
+           _THUNK_CALLI (_libc_32to16(_NetAccessGetInfo)));
  }

  USHORT NetAccessSetInfo(PSZ pszServer, PSZ pszResource, USHORT sLevel,
@@ -137,7 +137,7 @@
             _THUNK_FLAT (pbBuffer);
             _THUNK_SHORT (cbBuffer);
             _THUNK_SHORT (sParmNum);
-           _THUNK_CALLI (_emx_32to16(_NetAccessSetInfo)));
+           _THUNK_CALLI (_libc_32to16(_NetAccessSetInfo)));
  }

  USHORT NetAccessAdd(PSZ pszServer, USHORT sLevel,
@@ -149,7 +149,7 @@
             _THUNK_SHORT (sLevel);
             _THUNK_FLAT (pbBuffer);
             _THUNK_SHORT (cbBuffer);
-           _THUNK_CALLI (_emx_32to16(_NetAccessAdd)));
+           _THUNK_CALLI (_libc_32to16(_NetAccessAdd)));
  }

  #else /* other 32-bit */

Watch line wrap.
This succeeded compiling and quickly testing seems to work. Zip needs 
similar patches and there are still warnings. Really the port needs to 
be done over including adding bzip2 support, making sure large file 
support is working and so on.
Dave

[toc] | [prev] | [next] | [standalone]

#226

On 06/01/13 11:10 pm, Dave Yeo thus wrote :
> Lewis Rosenthal wrote:
>> Hi, guys...
>>
>> Prompted by Dave's mention in eCS-Technical (see
>> http://tech.groups.yahoo.com/group/eCS-Technical/message/17592 et seq)
>> and by http://mantis.smedley.id.au/view.php?id=549 , I decided to try to
>> give building the latest Unzip a shot...
>>
>> I'm not making much headway. If I set aside all of the options I'd like
>> to add, and from bash, just run:
>>
>> [Apollo|J:/devel/unzip610b]make -f os2/makefile.os2 gcc
>> make -f os2/makefile.os2 all \
>>          CC="gcc -Zomf -O -I." \
>>          CFLAGS="-Wall -DOS2 -DASM_CRC " \
>>          NFLAGS="" \
>>          DLLFLAG="" \
>>          AS="gcc -Zomf" \
>>          ASFLAGS="-Di386" \
>>          LDFLAGS="-o ./" \
>>          LDFLAGS2="-Zsmall-conv -Zstack 320 -Zsys -s" \
>>          OUT="-o" \
>>          OBJ=".obj" \
>>          CRCA_O="crc_gcc.obj" \
>>          DEF="os2/unzip.def"
>> make.exe[1]: Entering directory `J:/devel/unzip610b'
>> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC   unzip.c
>> In file included from unzpriv.h:637,
>>                   from unzip.h:734,
>>                   from unzip.c:69:
>> /@unixroot/usr/include/ctype.h:42: error: expected declaration
>> specifiers or '..
>> .' before '(' token
>> /@unixroot/usr/include/ctype.h:45: error: expected declaration
>> specifiers or '..
>> .' before '(' token
>> make.exe[1]: *** [unzip.obj] Error 1
>> make.exe[1]: Leaving directory `J:/devel/unzip610b'
>> make: *** [gcc] Error 2
>>
>> The INSTALL file specifically states that for OS/2, makefile.os2 should
>> remain in the os2 directory and be called with -f as I have above.
>>
>> I've actually had some recent successes in building things with gcc
>> 4.4.6, so I think that my build system is relatively intact.
>>
>> I understand (I think) that the "expected declaration" errors shown are
>> displayed as a result of -Wall, but looking at ctype.h, I don't
>> understand what's so different about those two particular lines than the
>> rest:
>>
>> [32]    __BEGIN_DECLS
>> [33]    int     isalnum(int);
>> [34]    int     isalpha(int);
>> [35]    int     iscntrl(int);
>> [36]    int     isdigit(int);
>> [37]    int     isgraph(int);
>> [38]    int     islower(int);
>> [39]    int     isprint(int);
>> [40]    int     ispunct(int);
>> [41]    int     isspace(int);
>> [42]    int     isupper(int);
>> [43]    int     isxdigit(int);
>> [44]    int     toupper(int);
>> [45]    int     tolower(int);
>>
>> So, what's it complaining about? Am I missing a locale specifier
>> somewhere?
>>
> 
> Probably missing a locale, os2cfg.h redefines toupper and tolower
> anyways so just wrap the #include <ctype.h> with #ifndef __EMX__ #endif.
> I also did,
> --- makefile.os2.orig   2013-06-01 19:55:04.000000000 -0700
> +++ makefile.os2        2013-06-01 19:56:16.000000000 -0700
> @@ -394,9 +394,9 @@
>         AS="gcc -Zomf" \
>         ASFLAGS="-Di386" \
>         LDFLAGS="-o ./" \
> -       LDFLAGS2="-Zsmall-conv -Zstack 320 -Zsys -s" \
> +       LDFLAGS2="-s" \
>         OUT="-o" \
> -       OBJ=".obj" \
> +       OBJ=".o" \
>         CRCA_O="crc_gcc.obj" \
>         DEF="os2/unzip.def"
> 
> and
> 
> --- os2acl.c.orig       2013-06-01 19:57:08.000000000 -0700
> +++ os2acl.c    2013-06-01 19:59:42.000000000 -0700
> @@ -71,7 +71,7 @@
>  typedef ULONG U_INT;
>  #ifdef __EMX__
>  #define PSTR16 _far16ptr
> -#define PTR16(x) _emx_32to16(x)
> +#define PTR16(x) _libc_32to16(x)
>  #else /* other 32-bit */
>  #define PSTR16 PCHAR16
>  #define PTR16(x) ((PCHAR16)(x))
> @@ -123,7 +123,7 @@
>             _THUNK_FLAT (pbBuffer);
>             _THUNK_SHORT (cbBuffer);
>             _THUNK_FLAT (pcbTotalAvail);
> -           _THUNK_CALLI (_emx_32to16(_NetAccessGetInfo)));
> +           _THUNK_CALLI (_libc_32to16(_NetAccessGetInfo)));
>  }
> 
>  USHORT NetAccessSetInfo(PSZ pszServer, PSZ pszResource, USHORT sLevel,
> @@ -137,7 +137,7 @@
>             _THUNK_FLAT (pbBuffer);
>             _THUNK_SHORT (cbBuffer);
>             _THUNK_SHORT (sParmNum);
> -           _THUNK_CALLI (_emx_32to16(_NetAccessSetInfo)));
> +           _THUNK_CALLI (_libc_32to16(_NetAccessSetInfo)));
>  }
> 
>  USHORT NetAccessAdd(PSZ pszServer, USHORT sLevel,
> @@ -149,7 +149,7 @@
>             _THUNK_SHORT (sLevel);
>             _THUNK_FLAT (pbBuffer);
>             _THUNK_SHORT (cbBuffer);
> -           _THUNK_CALLI (_emx_32to16(_NetAccessAdd)));
> +           _THUNK_CALLI (_libc_32to16(_NetAccessAdd)));
>  }
> 
>  #else /* other 32-bit */
> 
> Watch line wrap.
> This succeeded compiling and quickly testing seems to work. Zip needs
> similar patches and there are still warnings. Really the port needs to
> be done over including adding bzip2 support, making sure large file
> support is working and so on.
> Dave
> 
> 
Wow... Thanks, Dave! I'll giv ethis another shot (so much for getting to
bed early tonight, I see). ;-)

Much appreciated.

-- 
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC                www.2rosenthals.com
Need a managed Wi-Fi hotspot?                www.hautspot.com
visit my IT blog                www.2rosenthals.net/wordpress
-------------------------------------------------------------

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [prev] | [next] | [standalone]

#227

On 06/01/13 11:10 pm, Dave Yeo thus wrote :
> Lewis Rosenthal wrote:
>> Hi, guys...
>>
>> Prompted by Dave's mention in eCS-Technical (see
>> http://tech.groups.yahoo.com/group/eCS-Technical/message/17592 et seq)
>> and by http://mantis.smedley.id.au/view.php?id=549 , I decided to try to
>> give building the latest Unzip a shot...
>>
>> I'm not making much headway. If I set aside all of the options I'd like
>> to add, and from bash, just run:
>>
>> [Apollo|J:/devel/unzip610b]make -f os2/makefile.os2 gcc
>> make -f os2/makefile.os2 all \
>>          CC="gcc -Zomf -O -I." \
>>          CFLAGS="-Wall -DOS2 -DASM_CRC " \
>>          NFLAGS="" \
>>          DLLFLAG="" \
>>          AS="gcc -Zomf" \
>>          ASFLAGS="-Di386" \
>>          LDFLAGS="-o ./" \
>>          LDFLAGS2="-Zsmall-conv -Zstack 320 -Zsys -s" \
>>          OUT="-o" \
>>          OBJ=".obj" \
>>          CRCA_O="crc_gcc.obj" \
>>          DEF="os2/unzip.def"
>> make.exe[1]: Entering directory `J:/devel/unzip610b'
>> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC   unzip.c
>> In file included from unzpriv.h:637,
>>                   from unzip.h:734,
>>                   from unzip.c:69:
>> /@unixroot/usr/include/ctype.h:42: error: expected declaration
>> specifiers or '..
>> .' before '(' token
>> /@unixroot/usr/include/ctype.h:45: error: expected declaration
>> specifiers or '..
>> .' before '(' token
>> make.exe[1]: *** [unzip.obj] Error 1
>> make.exe[1]: Leaving directory `J:/devel/unzip610b'
>> make: *** [gcc] Error 2
>>
>> The INSTALL file specifically states that for OS/2, makefile.os2 should
>> remain in the os2 directory and be called with -f as I have above.
>>
>> I've actually had some recent successes in building things with gcc
>> 4.4.6, so I think that my build system is relatively intact.
>>
>> I understand (I think) that the "expected declaration" errors shown are
>> displayed as a result of -Wall, but looking at ctype.h, I don't
>> understand what's so different about those two particular lines than the
>> rest:
>>
>> [32]    __BEGIN_DECLS
>> [33]    int     isalnum(int);
>> [34]    int     isalpha(int);
>> [35]    int     iscntrl(int);
>> [36]    int     isdigit(int);
>> [37]    int     isgraph(int);
>> [38]    int     islower(int);
>> [39]    int     isprint(int);
>> [40]    int     ispunct(int);
>> [41]    int     isspace(int);
>> [42]    int     isupper(int);
>> [43]    int     isxdigit(int);
>> [44]    int     toupper(int);
>> [45]    int     tolower(int);
>>
>> So, what's it complaining about? Am I missing a locale specifier
>> somewhere?
>>
> 
> Probably missing a locale, os2cfg.h redefines toupper and tolower
> anyways so just wrap the #include <ctype.h> with #ifndef __EMX__ #endif.
> I also did,
> --- makefile.os2.orig   2013-06-01 19:55:04.000000000 -0700
> +++ makefile.os2        2013-06-01 19:56:16.000000000 -0700
> @@ -394,9 +394,9 @@
>         AS="gcc -Zomf" \
>         ASFLAGS="-Di386" \
>         LDFLAGS="-o ./" \
> -       LDFLAGS2="-Zsmall-conv -Zstack 320 -Zsys -s" \
> +       LDFLAGS2="-s" \
>         OUT="-o" \
> -       OBJ=".obj" \
> +       OBJ=".o" \
>         CRCA_O="crc_gcc.obj" \
>         DEF="os2/unzip.def"
> 
> and
> 
> --- os2acl.c.orig       2013-06-01 19:57:08.000000000 -0700
> +++ os2acl.c    2013-06-01 19:59:42.000000000 -0700
> @@ -71,7 +71,7 @@
>  typedef ULONG U_INT;
>  #ifdef __EMX__
>  #define PSTR16 _far16ptr
> -#define PTR16(x) _emx_32to16(x)
> +#define PTR16(x) _libc_32to16(x)
>  #else /* other 32-bit */
>  #define PSTR16 PCHAR16
>  #define PTR16(x) ((PCHAR16)(x))
> @@ -123,7 +123,7 @@
>             _THUNK_FLAT (pbBuffer);
>             _THUNK_SHORT (cbBuffer);
>             _THUNK_FLAT (pcbTotalAvail);
> -           _THUNK_CALLI (_emx_32to16(_NetAccessGetInfo)));
> +           _THUNK_CALLI (_libc_32to16(_NetAccessGetInfo)));
>  }
> 
>  USHORT NetAccessSetInfo(PSZ pszServer, PSZ pszResource, USHORT sLevel,
> @@ -137,7 +137,7 @@
>             _THUNK_FLAT (pbBuffer);
>             _THUNK_SHORT (cbBuffer);
>             _THUNK_SHORT (sParmNum);
> -           _THUNK_CALLI (_emx_32to16(_NetAccessSetInfo)));
> +           _THUNK_CALLI (_libc_32to16(_NetAccessSetInfo)));
>  }
> 
>  USHORT NetAccessAdd(PSZ pszServer, USHORT sLevel,
> @@ -149,7 +149,7 @@
>             _THUNK_SHORT (sLevel);
>             _THUNK_FLAT (pbBuffer);
>             _THUNK_SHORT (cbBuffer);
> -           _THUNK_CALLI (_emx_32to16(_NetAccessAdd)));
> +           _THUNK_CALLI (_libc_32to16(_NetAccessAdd)));
>  }
> 
>  #else /* other 32-bit */
> 
> Watch line wrap.
> This succeeded compiling and quickly testing seems to work. Zip needs
> similar patches and there are still warnings. Really the port needs to
> be done over including adding bzip2 support, making sure large file
> support is working and so on.
> 
Okay, so here's my progress:

I was able to build it - once - (thanks again, so much, Dave).

However, after cleaning everything out and starting over, I get (watch
the line breaks - I've added "\" where I saw them):

# make -f os2/makefile.os2 D_USE_BZ2=-DUSE_BZIP2 L_BZ2=-lbz2 \
LIBBZ2=bzip2/libbz2.a gcc
make -f os2/makefile.os2 all \
        CC="gcc -Zomf -O -I." \
        CFLAGS="-Wall -DOS2 -DASM_CRC -DWILD_STOP_AT_DIR -DOS2_EAS \
-DASM_CRC -DUSE_BZIP2 -DBZIP2_SFX -DUNICODE_SUPPORT" \
        NFLAGS="" \
        DLLFLAG="" \
        AS="gcc -Zomf" \
        ASFLAGS="-Di386" \
        LDFLAGS="-o ./" \
        LDFLAGS2="-s" \
        OUT="-o" \
        OBJ=".o" \
        CRCA_O="crc_gcc.obj" \
        DEF="os2/unzip.def"
make.exe[1]: Entering directory `J:/devel/unzip610b'
gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC -DWILD_STOP_AT_DIR \ -DOS2_EAS
-DASM_CRC -DUSE_BZIP2 -DBZIP2_SFX -DUNICODE_SUPPORT  unzip.c
unzip.c: In function 'get_shortopt':
unzip.c:3986: warning: implicit declaration of function 'isdigit'
gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC -DWILD_STOP_AT_DIR \ -DOS2_EAS
-DASM_CRC -DUSE_BZIP2 -DBZIP2_SFX -DUNICODE_SUPPORT  crc32.c
make.exe[1]: *** No rule to make target `crc_gcc.obj', needed by \
`unzip.exe'.  Stop.
make.exe[1]: Leaving directory `J:/devel/unzip610b'
make: *** [gcc] Error 2

I'm not sure how I got it to work before, but it's possible that
crc_gcc.obj was generated before I applied your changes, and it was a
leftover.

Sorry to be such a pain (and a newbie). Thanks again for all of the help
and guidance.

Cheers/2

-- 
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC                www.2rosenthals.com
Need a managed Wi-Fi hotspot?                www.hautspot.com
visit my IT blog                www.2rosenthals.net/wordpress
-------------------------------------------------------------

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [prev] | [next] | [standalone]

#228

Lewis Rosenthal wrote:
> make.exe[1]: *** No rule to make target `crc_gcc.obj', needed by \
> `unzip.exe'.  Stop.
> make.exe[1]: Leaving directory `J:/devel/unzip610b'
> make: *** [gcc] Error 2
>
> I'm not sure how I got it to work before, but it's possible that
> crc_gcc.obj was generated before I applied your changes, and it was a
> leftover.
>
> Sorry to be such a pain (and a newbie). Thanks again for all of the help
> and guidance.

Your missing
-       OBJ=".obj" \
+       OBJ=".o" \

somewhere in the make file.
Under EMX, -Zomf under CFLAGS caused the default suffix to be .obj which 
is standard for OS/2 etc whereas under kLIBC the default is .o
I'll try your command line
Dave

[toc] | [prev] | [next] | [standalone]

#230

Lewis Rosenthal wrote:
> Okay, so here's my progress:
>
> I was able to build it - once - (thanks again, so much, Dave).
>
> However, after cleaning everything out and starting over, I get (watch
> the line breaks - I've added "\" where I saw them):
>
> # make -f os2/makefile.os2 D_USE_BZ2=-DUSE_BZIP2 L_BZ2=-lbz2 \
> LIBBZ2=bzip2/libbz2.a gcc
> make -f os2/makefile.os2 all \
>          CC="gcc -Zomf -O -I." \
>          CFLAGS="-Wall -DOS2 -DASM_CRC -DWILD_STOP_AT_DIR -DOS2_EAS \
> -DASM_CRC -DUSE_BZIP2 -DBZIP2_SFX -DUNICODE_SUPPORT" \
>          NFLAGS="" \
>          DLLFLAG="" \
>          AS="gcc -Zomf" \
>          ASFLAGS="-Di386" \
>          LDFLAGS="-o ./" \
>          LDFLAGS2="-s" \
>          OUT="-o" \
>          OBJ=".o" \
>          CRCA_O="crc_gcc.obj" \
>          DEF="os2/unzip.def"
> make.exe[1]: Entering directory `J:/devel/unzip610b'
> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC -DWILD_STOP_AT_DIR \ -DOS2_EAS
> -DASM_CRC -DUSE_BZIP2 -DBZIP2_SFX -DUNICODE_SUPPORT  unzip.c
> unzip.c: In function 'get_shortopt':
> unzip.c:3986: warning: implicit declaration of function 'isdigit'
> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC -DWILD_STOP_AT_DIR \ -DOS2_EAS
> -DASM_CRC -DUSE_BZIP2 -DBZIP2_SFX -DUNICODE_SUPPORT  crc32.c
> make.exe[1]: *** No rule to make target `crc_gcc.obj', needed by \
> `unzip.exe'.  Stop.
> make.exe[1]: Leaving directory `J:/devel/unzip610b'
> make: *** [gcc] Error 2
>
> I'm not sure how I got it to work before, but it's possible that
> crc_gcc.obj was generated before I applied your changes, and it was a
> leftover.
>
> Sorry to be such a pain (and a newbie). Thanks again for all of the help
> and guidance.

Using your command line except I used bz2_s.a instead of libbz2.a to 
build static seemed to have worked fine here. Bedtime so all I did was 
test that it didn't crash. For distributing a static linked bz2 seems 
better
Dave

[toc] | [prev] | [next] | [standalone]

#234

On 06/04/13 02:50 am, Dave Yeo thus wrote :
> Lewis Rosenthal wrote:
>> Okay, so here's my progress:
>>
>> I was able to build it - once - (thanks again, so much, Dave).
>>
>> However, after cleaning everything out and starting over, I get (watch
>> the line breaks - I've added "\" where I saw them):
>>
>> # make -f os2/makefile.os2 D_USE_BZ2=-DUSE_BZIP2 L_BZ2=-lbz2 \
>> LIBBZ2=bzip2/libbz2.a gcc
>> make -f os2/makefile.os2 all \
>>          CC="gcc -Zomf -O -I." \
>>          CFLAGS="-Wall -DOS2 -DASM_CRC -DWILD_STOP_AT_DIR -DOS2_EAS \
>> -DASM_CRC -DUSE_BZIP2 -DBZIP2_SFX -DUNICODE_SUPPORT" \
>>          NFLAGS="" \
>>          DLLFLAG="" \
>>          AS="gcc -Zomf" \
>>          ASFLAGS="-Di386" \
>>          LDFLAGS="-o ./" \
>>          LDFLAGS2="-s" \
>>          OUT="-o" \
>>          OBJ=".o" \
>>          CRCA_O="crc_gcc.obj" \
>>          DEF="os2/unzip.def"
>> make.exe[1]: Entering directory `J:/devel/unzip610b'
>> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC -DWILD_STOP_AT_DIR \ -DOS2_EAS
>> -DASM_CRC -DUSE_BZIP2 -DBZIP2_SFX -DUNICODE_SUPPORT  unzip.c
>> unzip.c: In function 'get_shortopt':
>> unzip.c:3986: warning: implicit declaration of function 'isdigit'
>> gcc -Zomf -O -I. -c -Wall -DOS2 -DASM_CRC -DWILD_STOP_AT_DIR \ -DOS2_EAS
>> -DASM_CRC -DUSE_BZIP2 -DBZIP2_SFX -DUNICODE_SUPPORT  crc32.c
>> make.exe[1]: *** No rule to make target `crc_gcc.obj', needed by \
>> `unzip.exe'.  Stop.
>> make.exe[1]: Leaving directory `J:/devel/unzip610b'
>> make: *** [gcc] Error 2
>>
>> I'm not sure how I got it to work before, but it's possible that
>> crc_gcc.obj was generated before I applied your changes, and it was a
>> leftover.
>>
>> Sorry to be such a pain (and a newbie). Thanks again for all of the help
>> and guidance.
> 
> Using your command line except I used bz2_s.a instead of libbz2.a to
> build static seemed to have worked fine here. Bedtime so all I did was
> test that it didn't crash. For distributing a static linked bz2 seems
> better
>
I agree, and a static build is what I'm looking to do.

So, then, just so that I can follow this, how would I know that bz2_s.a
is the static lib? I thought that all I needed to do was state that I
wanted to statically link libbz2.a.

Insofar as missing that .obj -> .o change... I've got:

# emx, gcc, OMF format, statically linked C runtime
gcc:
	$(MAKE) -f os2/makefile.os2 all \
	CC="gcc -Zomf -O -I." \
	CFLAGS="-Wall -DOS2 -DASM_CRC $(LOCAL_UNZIP)" \
	NFLAGS="" \
	DLLFLAG="" \
	AS="gcc -Zomf" \
	ASFLAGS="-Di386" \
	LDFLAGS="-o ./" \
	LDFLAGS2="-s" \
	OUT="-o" \
	OBJ=".o" \
	CRCA_O="crc_gcc.obj" \
	DEF="os2/unzip.def"

and I am specifying gcc as the target. Is it the CRCA_O="crc_gcc.obj"
which is causing a conflict and I need to change that to crc_gcc.o?
Ugh... Never mind; that was it (forest & trees issue).

Getting back to the bz2_s.a thing: Cold you please enlighten me as to
where you got that? My understanding was that all I needed to do WRT
bzip2 was to "dump" the sources into bzip2\ and specify the option and
modify the make line, referencing the library. Clearly, I'm missing
something else.

Thanks again, Dave!!

-- 
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC                www.2rosenthals.com
Need a managed Wi-Fi hotspot?                www.hautspot.com
visit my IT blog                www.2rosenthals.net/wordpress
-------------------------------------------------------------

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [prev] | [next] | [standalone]

#240

Lewis Rosenthal wrote:
> Getting back to the bz2_s.a thing: Cold you please enlighten me as to
> where you got that? My understanding was that all I needed to do WRT
> bzip2 was to "dump" the sources into bzip2\ and specify the option and
> modify the make line, referencing the library. Clearly, I'm missing
> something else.
>
> Thanks again, Dave!!

I built it sometime ago and just copied the file to bzip2\ from lib.
Basic library stuff on OS/2. .a suffix can be either static or import 
library, usually aout with .lib being OMF. _s.a is used for static and 
_dll.a for import library with lib once again used for OMF. The linker 
is smart enough usually to convert aout to OMF using emxomf if there is 
a mismatch, you can't go the other way. Some versions of libtool such as 
the default ones will only create one or another, other ports will 
create both static and import with the different suffixes or somewhere 
in between.
The linker prefers lib as the prefix but the library doesn't need it, 
this is for 8.3 name compatibility.
If you just dumped the sources in bzip2/, it probably built a static 
lib. Static lib is just basically all the objects files concated 
together with a table of contents.
Yuri's patches etc should be in
http://rpm.netlabs.org/release/00/i386/SRPMS/unzip-6.0-5.oc00.src.rpm
http://rpm.netlabs.org/release/00/i386/SRPMS/zip-3.0-5.oc00.src.rpm
Dave

[toc] | [prev] | [next] | [standalone]

#245

On 06/04/13 10:33 pm, Dave Yeo thus wrote :
> Lewis Rosenthal wrote:
>> Getting back to the bz2_s.a thing: Cold you please enlighten me as to
>> where you got that? My understanding was that all I needed to do WRT
>> bzip2 was to "dump" the sources into bzip2\ and specify the option and
>> modify the make line, referencing the library. Clearly, I'm missing
>> something else.
>>
>> Thanks again, Dave!!
> 
> I built it sometime ago and just copied the file to bzip2\ from lib.
> Basic library stuff on OS/2. .a suffix can be either static or import
> library, usually aout with .lib being OMF. _s.a is used for static and
> _dll.a for import library with lib once again used for OMF. The linker
> is smart enough usually to convert aout to OMF using emxomf if there is
> a mismatch, you can't go the other way. Some versions of libtool such as
> the default ones will only create one or another, other ports will
> create both static and import with the different suffixes or somewhere
> in between.
> The linker prefers lib as the prefix but the library doesn't need it,
> this is for 8.3 name compatibility.
> If you just dumped the sources in bzip2/, it probably built a static
> lib. Static lib is just basically all the objects files concated
> together with a table of contents.
> Yuri's patches etc should be in
> http://rpm.netlabs.org/release/00/i386/SRPMS/unzip-6.0-5.oc00.src.rpm
> http://rpm.netlabs.org/release/00/i386/SRPMS/zip-3.0-5.oc00.src.rpm
> 
Ugh... I mirror the whole d-mn repo; now why didn't I see that? I'll
close my ticket. Thanks for pointing that out!!!!

Thanks for the great explanation of this stuff, too. My looking glass is
getting clearer by the minute!

Cheers/2

-- 
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC                www.2rosenthals.com
Need a managed Wi-Fi hotspot?                www.hautspot.com
visit my IT blog                www.2rosenthals.net/wordpress
-------------------------------------------------------------

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [prev] | [next] | [standalone]

#233

On Sun, 2 Jun 2013 03:10:28 UTC, Dave Yeo <dave.r.yeo@gmail.com> 
wrote:

<snip>
> Probably missing a locale, os2cfg.h redefines toupper and tolower 
> anyways so just wrap the #include <ctype.h> with #ifndef __EMX__ #endif.
I ended up including ctype.h in the os2cfg.h file so that both Watcom 
and GCC were happy.
Andy

[toc] | [prev] | [next] | [standalone]

#235

On 06/04/13 01:14 pm, Andy thus wrote :
> On Sun, 2 Jun 2013 03:10:28 UTC, Dave Yeo <dave.r.yeo@gmail.com> 
> wrote:
> 
> <snip>
>> Probably missing a locale, os2cfg.h redefines toupper and tolower 
>> anyways so just wrap the #include <ctype.h> with #ifndef __EMX__ #endif.
> I ended up including ctype.h in the os2cfg.h file so that both Watcom 
> and GCC were happy.
> 
So you got it building? If you set UNZIPOPT=-C, does it complain?

-- 
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC                www.2rosenthals.com
Need a managed Wi-Fi hotspot?                www.hautspot.com
visit my IT blog                www.2rosenthals.net/wordpress
-------------------------------------------------------------

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [prev] | [next] | [standalone]

#237

On Tue, 4 Jun 2013 17:23:36 UTC, Lewis Rosenthal 
<lgrosenthal@2-de-sp-am-2rosenthals.com> wrote:

> On 06/04/13 01:14 pm, Andy thus wrote :
> > On Sun, 2 Jun 2013 03:10:28 UTC, Dave Yeo <dave.r.yeo@gmail.com> 
> > wrote:
> > 
> > <snip>
> >> Probably missing a locale, os2cfg.h redefines toupper and tolower 
> >> anyways so just wrap the #include <ctype.h> with #ifndef __EMX__ #endif.
> > I ended up including ctype.h in the os2cfg.h file so that both Watcom 
> > and GCC were happy.
> > 
> So you got it building? If you set UNZIPOPT=-C, does it complain?
> 

Yes, but I think I already had it set... by clearing it then it no 
longer complains.
Andy
-- 

[toc] | [prev] | [next] | [standalone]

#238

On 06/04/13 04:50 pm, Andy thus wrote :
> On Tue, 4 Jun 2013 17:23:36 UTC, Lewis Rosenthal 
> <lgrosenthal@2-de-sp-am-2rosenthals.com> wrote:
> 
>> On 06/04/13 01:14 pm, Andy thus wrote :
>>> On Sun, 2 Jun 2013 03:10:28 UTC, Dave Yeo <dave.r.yeo@gmail.com> 
>>> wrote:
>>>
>>> <snip>
>>>> Probably missing a locale, os2cfg.h redefines toupper and tolower 
>>>> anyways so just wrap the #include <ctype.h> with #ifndef __EMX__ #endif.
>>> I ended up including ctype.h in the os2cfg.h file so that both Watcom 
>>> and GCC were happy.
>>>
>> So you got it building? If you set UNZIPOPT=-C, does it complain?
>>
> 
> Yes, but I think I already had it set... by clearing it then it no 
> longer complains.
> 
Carrying on two discussions about the same thing... ;-)

Indeed, if you don't then tell it to ignore case, it doesn't complain,
but I don't know whether it has any idea of case-sensitivity or not.

I'm going to ask Yuri for his diff. I feel like I'm asking for the
answers to last week's crossword, but this is driving me nuts!

-- 
Lewis
-------------------------------------------------------------
Lewis G Rosenthal, CNA, CLP, CLE, CWTS
Rosenthal & Rosenthal, LLC                www.2rosenthals.com
Need a managed Wi-Fi hotspot?                www.hautspot.com
visit my IT blog                www.2rosenthals.net/wordpress
-------------------------------------------------------------

--- news://freenews.netfront.net/ - complaints: news@netfront.net ---

[toc] | [prev] | [standalone]

Back to top | Article view | comp.os.os2.programmer.porting

csiph-web