AT&T vs. termcap

[Brandon Allbery]

Expires:

Quoted from <29000001 at ndm20.UUCP> ["Re: tgetstr on system V R2 (HELP!)"], by lvs at ndm20.UUCP...
+---------------
| 
| 
| >    I am trying to get a recently posted reminder program (rs)
| >to run on my machine (VAX 11/780 running Sys V R2).  The program
| >seems to run okay, except that all calls to tgetstr() do not
| >return the correct escape sequence (they return null).
| >
| >    The sequence being used is:
| >	tgetent(bp, terminal name)  /* works fine - return of 1 */
| >	tgetnum("co")		    /* works fine - returns 80 */
| >	tgetstr("so",&some_buffer) or
| >	tgetstr("bl",&some_buffer) or anything else
| >				    /* returns 0's in buffer */
| >
| >    Does anyone know why it doesn't?  Am I doing something wrong?
| >Please respond via email.
| >
| >Lee Vallone		AT&T Information Systems	Merlin
| >{... ihnp4, mtuxo}!mtsbb!lav
| 
| No, you are not doing anything wrong, AT&T is!  They  have broken the
| way  tgetent  and  tgetstr  work.   The only  help you  get from AT&T
| software support is "Sorry,  we don't  support termcap  anymore".  Oh
| well, so it goes.
| 
| The original definition for tgetstr was
| 
|    char *tgetstr( char *code, char **bufptr )
| 
| tgetstr would  get the  string you  desired (if  it can  find it) and
| place it in your buffer, supplied  by bufptr.   It  would then update
| bufptr  to point  just beyond  the string  just added  and return the
| original value  of the  buffer pointer  (unless the  code couldn't be
| found in which case it returns NULL).
| 
| Well, so much for the way it  SHOULD work.   The  version supplied by
| AT&T in SVR2 is just plain broken.  It  does nothing  with the buffer
| you supply  it, nor  does it  update the  pointer.   The only usefull
| information is the return value.  If it is not NULL it  will point to
| the string you requested.  Therefore, you must save  the return value
| from the call, that is you can't use bufptr because  it doesn't point
| to the string.  If you want the strings in a buffer you  will have to
| copy them into it yourself.
+---------------

I always declare the buffer as automatic within a function and copy the
result, to be sure.

Please note that the tget* functions are a compatibility aid only, and you
really should not use them under System V; they may go away.

+---------------
| Alas, it seems AT&T has broken termcap and instead of fixing it, they
| decided not to support it.
+---------------

Huh???  I hate to tell you, but what you're so blithely calling termcap is in
reality terminfo.  Read TERM(4) in the UNIX Programmer's Reference Guide.  The
current version of terminfo supports access of terminfo capabilities by the
old termcap names via compatibility calls to tget*; however, the two work in
totally different ways.  (Tgetstr() doesn't need a buffer; terminfo uses an
internal buffer instead.  Which is a step forward in my opinion.)

+---------------
| BTW, the machine I have checked this  on is  an AT&T  PC6300PLUS.  It
| exibited the same problems you described.  It  now works  by doing it
| the way I described above.
+---------------

Read TERM(4) and rewrite using terminfo.  You won't be sorry; terminfo is much
cleaner than termcap.

++Brandon
-- 
  ---------------- /--/	Brandon S. Allbery		UUCP:
 /              / /|\/	Tridelta Industries, Inc.       decvax!cwruecmp!ncoast!
----    -------- /-++	7350 Corporate Blvd.		    tdi2!brandon
   /   / /---,  /--/	Mentor, Ohio 44060		PHONE:  (home)
  /   / /    / /  /	     -- HOME --			+1 216 974 9210
 /   / /    / /  /	6615 Center St. Apt. A1-105	ARPA:  ncoast!allbery%
----  /----~ /--/	Mentor, Ohio 44060-4101		case.CSNET at csnet-relay
-------------------------------------------------------------------------------
			Space -- The Final Frontier