[comp.protocols.iso.dev-environ] ISODE Vol.5 doc error?

PWW@BNR.CA (Peter Whittaker, P.W.) (10/11/90)

My apologies if this has already been pointed out, but I've only
been reading this group for a few months....

According to the QUIPU manual (ISODE Vol. 5, Ver. 6.0) page 155,
dsap_init will ignore all arguments except a "-c dsa_name" pair.

This isn't strictly the case:  dasp_init will ignore arguments
that occur after "-c dsa_name" IF they are not "-flag" style arguments
(i.e. if they are "flag" style), and will reject (i.e. generate an error
message and exit your application) any "-flag" style arguments where
"-flag" is not one of the ones listed below.

In fact, dsap_init will ignore ANY argument that occurs after a
"flag" style argument, including the "-c dsa_name" argument.
This is due to a function (tai_args()) called by dsap_init;
To be safe and sure, dsap_init should only be passed an argv-style
argument of the form:

your_appl_name :-c dsa_name: :-t dsaptailor_style_tailor_file:
               :-T oidtable_format_file:

OR

two NULL pointers (in which case the only tailoring dsap_init
will do is to read your $HOME/.isode_tailor file, if it exists).

(as usual, the :...: indicate that these arguments are optional).

What will dsap_init do with your options?
-----------------------------------------

If you do not pass dsap_init(NULL, NULL), then it will:

1) look in your home directory for .your_application_name_tailor,
   an isodetailor-style file, which it will read for various
   tailoring options (log level settings, ETCDIR redefinitions).

   If the HOME environment variable is not set, it will try to find
   this file in the local directory.

   If it can't find it, it ignores it.

2) set dsa_address to the dsa_name followinng the '-c' flag;
   subsequent binds will connect to that dsa.

3) read the dsaptailor style file specified after the '-t' flag,
   which it will read for dsa addresses and names.

4) read the oidtable specified after '-T' (instead of ETCDIR/oidtable).

In case of inaccuracies or omissions in the above, please e-mail the
correct information to me, and I will repost this article.

Thanks,

Peter W.
pww@bnr.ca