-=> Maurice Kinal wrote to Tony Langdon <=-

 MK> -={ Thursday, 28 April 2016, 10:37:56.896639632 +1000 }=-

 MK> Hey Tony!

 TL> It takes me a little while to understand the nuances of scripts.

 MK> Sounds great.  I had to liberate the posted parts from a bash function
 MK> file that I source from other script(s) so I fully appreciate your
 MK> point.  ;-)

Fair enough. :)

 MK> Anyhow take your time and if you have any questions/suggestions I am
 MK> all ears.

 TL> but do we want to keep track of any individual components such as
 TL> headers, kludge lines, etc?

 MK> Yes we do.  That is the goal and other than possible splitting of the
 MK> kludges from the msg_body (fifth field) I believe the current format is
 MK> a very good foundation.  We may wish to translate the nulls to a more
 MK> suitable control code, such as the aptly named 'field seperator' (aka
 MK> 0x1c) but gawk doesn't seem to mind nulls if indeed we find gawk to be
 MK> a keeper in scripting.  bash string handling has improved over the
 MK> years and might be fun to replace sed and tr calls ... if indeed we
 MK> want to.

I use Bash subscring functions a lot, but still use sed a bit for those quick
and dirty edits.  I haven't had much to do with tr, actually.

 TL> you might need to modify the format a bit

 MK> No doubt.  Still I think it is a solid foundation to build from but I
 MK> will wait for your later assessment.

Any chance of getting my hands on scripts and sample data?  I'm limited by the
fact that I can't see this in action, so I can't check my assumptions (which if
I get wrong, the model in my head is totally bogus! :) ).


... Is truth not truth for all? Natira, stardate 5476.4.
--- MultiMail/Win32 v0.49
 * Origin: Freeway BBS - freeway.apana.org.au (3:633/410)