🚀 go-pugleaf

In article <NICK.92Dec23151504@alphaville.osf.org> nick@osf.org (Nick Dokos) writes:
>I got the C version of dvitovdu from ymir.claremont.edu in the DRIVERS
>directory. It is tarred, compressed, btoa'ed and split in three
>pieces. I had to get btoa and atob (from ftp.uu.net comp.sources.misc,
>vol. 6 - is there a more recent version?)  and tried it. I do
>
>	cat the_pieces | atob > dvitovdu.tar.Z
>
>and try to uncompress. Uncompress complains about illegal input. I btoa
>the result and compare it with the original - they are different.
>
>So either my btoa/atob is bad or the files in ymir are bad. I'd
>appreciate hearing from anyone who has gotten past this?

I guess your atob is bad.  I used atob on a Sun ipc, but I did not note
the version.  The files in ymir are all right except for dummy sccs id
string declarations which should be in the form

   static char sccsid_modulename[]=".....";

rather than

   static char *sccsid_modulename[]=".....";

The following tiny ed script saved in a file called ed.scr

   s/\*sccs/sccs/g
   w
   q

and the C shell command

   foreach i (*.c)
    ed $i < ed.scr
   end

will solve the problem.

Another compiler error I noticed on a BSD system
is the declaration of strlen() in "def.h" as

    extern int strlen();

rather than

    extern size_t strlen();


>
>I also tried to look in the DRIVERS directory in ftp.shsu.edu. But ftp
>complains that the directory cannot be opened. I have no problems with
>other directories. George, can you check?
>
>Of course, dvitovdu is what I am really after. Archie revealed nothing.
>Can anybody point to an anonymous ftp site other than the above? Or is
>anyone able to provide me the source through e-mail?

I recommend that you try to use another atob implementation.
The compressed archive takes approximately 200 kilobytes.
I will have to make a kit if you want my modified copy.

If you are interested, please post on this newsgroup or
send me e-mail for other possible bugs
and a CGA Kermit VDU interface module implementation.

I would appreciate some assistance from interested readers before
entering a moral obligation towards them.  I feel this is an overwhelming
responsibility because

  (1) documenting takes effort and time,
  (2) my main occupation is electrical engineering,
      I am not supposed to be debugging and
      maintaining such software, which is only tangent
      to my occupation
  (3) I have a painfully slow modem connection and a low-resolution
      CGA monitor with limited capabilites, and
  (4) VDU interfaces for TEK4010-emulating terminals
      seemed not to work on the PC Kermit with a CGA monitor
      except the one I added, i. e. the CGA Kermit VDU interface,
      i. e. I could accomplish a very limited testing of dvi2vdu.
  
For the reasons listed above, I am practically unable to dedicate
the necessary time and effort single-handed to maintain a copy that
deserves to be placed on the public domain as yet.  At best, I might
contact the authors if I feel it is worth the noise I might generate.
On the other hand, responsible and reasonable requests are always
welcome.

BTW, my efforts about the Modula-2 version of dvitovdu could not
go beyond getting the following message under ULTRIX

ld:
Object file format error in: aed512vdu.o: relocation entry for non-relocatable symbol in .text section for entry 20

 , which I do not usually encounter with C compilers.


>
>Thanks,
>
>--
>Nick Dokos	nick@osf.org	
>OSF Systems Engineering
>
>Member of the League for Programming Freedom - write to lpf@uunet.uu.net

Best Regards,

Mustafa Kocaturk

-- 
Mustafa Kocaturk   mustafa@seas.smu.edu  EE Dept., Room 305A, Caruth Bldg.
Home: 214-706-5954  Office: 214-768-1475  SMU Box 753190, Dallas, TX 75275