I been asked to improve the sprint exchange parser to make it more
smart (along the lines of the SS exchange parser).
It is already about as smart as it can be given the syntax of data.
The sprint exchange doesn't lend itself to some of the tricks used
for the SS.
The current parser will take exchanges like this:
1 Bert OR
Bert 1 OR (the name Bert probably came from the TRMASTER database)
Bert OR 1 (Bert OR came from TRMASTER or previous QSO on another band)
xxx 1 Bert OR ( doesn't matter what xxx is)
xxx OR 1 Bert ( doesn't matter what xxx is)
1 Bert Or 2 ( uses 2 for the number )
xxx yyy 1 Bert OR (doesn't matter what xxx or yyy are }
The only change I can see to make is to allow more entries like xxx
and yyy if the whole exchange is repeated:
xxx yyy zzz zzz yyy 1 Bert OR (any number of xxx, yyy or zzz entries)
The problem with adding more tricks is demonstrated by considering
exchanges like the following:
I could add more tricks if you could guarantee me that someone named
Al won't get on from Pennsylvania, and nobody from Alabama will ever
use the name "Pa" or "Mo".
One other change would be to allow this:
Bert Or 1 2 3 ( uses the last number entered )
I have put the two changes into what will become 6.30. Sorry this
release is taking some time. So many people have asked for things
that I want to get as much of it into the release as I can.
Tree
PS: Anyone know what computer game I stole a magic word from in my
example?
--
FAQ on WWW: http://www.contesting.com/trlogfaq.html
Submissions: trlog@contesting.com
Administrative requests: trlog-REQUEST@contesting.com
Problems: owner-trlog@contesting.com
Feature Wishlist: http://web.jzap.com/n6tr/trwish.html
|