Hi! Richard,

Have you had any luck with any other utils, mate?

On 09 Oct 07 14:01, you wrote to me:

 RW> IF you've an idea how it could be done with something from the
 RW> collection of bfds utils indeed post it, I'd love to give it a try.

We've had a few dramas this past week and I'm short of time, all the time. 
So, I'll offer these comments on my contemplations on the matter thus far.

These are only 5 Fidonet zones effectively these days, numbered 1->5, so
the zone 6 ZC entry could be used as an 'end of data' marker, since there
are no nodes listed for the zone.  I've done some research with the current
nodelist (#278) and have found that there are 309 nets, numbered from
102->904.  However there are 6 duplicates +/- 1 (none within the same
zone, luckily).  Hence the use of a temp "zonefile" herewith,
thus ensuring no dupes would be encountered...

-----88-----
Pseudocode:
-----------

INPUT BATCH.FILENAME TARGET.ADDRESS NODELIST.FILE
BEGIN
  IF NODELIST.FILE DOESN'T EXIST
    WE HAVE AN ERROR CONDITION
  ELSE
    SEPARATE THE [ZONE]:[NET]/[NODE] ELEMENTS FROM THE TARGET.ADDRESS
    IF ([ZONE] LESS THAN 1 -OR- GREATER THAN 5)
    -OR- ([NET] LESS THAN 100 -OR- GREATER THAN 999)
      WE HAVE AN ERROR CONDITION
    ELSE
      CREATE "ZONEFILE" FROM NODELIST.FILE
      USE THE "ZONEFILE"
        IF TARGET [NET] DOESN'T EXIST
          WE HAVE AN ERROR CONDITION
        ELSE
          PROCESS EACH LINE IN THE "ZONEFILE"
            IF THIS IS THE TARGET [NODE]
              READ THE NODELIST ENTRY'S DETAILS
              SET A TARGET.ADDRESS FOUND FLAG TO TRUE
            ELSE
              BUMP A LINE COUNTER
            ENDIF
          LOOP UNTIL WE FIND THE NEXT NET -OR- TARGET.ADDRESS FOUND
          -OR- WE RUN OUT OF LINES
        ENDIF
      END (USE)
      DELETE THE "ZONEFILE"
    ENDIF
  ENDIF
  IF TARGET.ADDRESS FOUND
      OUTPUT THE NODE'S DETAILS
  ENDIF
  IF WE HAVE AN ERROR CONDITION
    OUTPUT ERROR MESSAGE
  ENDIF
END
-----88-----

(N.B I've chopped some of the P-code to get it to fit in the editor, so if
there's some inconsistancy then... it's may fault anyway. :)

Is there a faster way to achieve the goal?  Maybe.  But we're talking about
batch stuff so we're stuck with linear searches, etc.  I've laid out the
P-code to take advantage of DOS's 'find' command, twice.  Still, the
slowest processes will be the creation of a zonefile, and, then the final
search through it looking at net/node entries.  I still envisage a run-time
of maybe 30-40 seconds on my 500Mhz machine (watch for the run-time to
double and add a half again, in real life :).

If you have any thoughts on a final version, please let me know.  I'll get
started on the batch next week.

Cheers,
Paul.

... Move your vowels every day or you'll get consonated.
--- GoldED+/LNX 1.1.5-31012