Automatic Generation of Soaring Forecasts

["John C. Peterson" <jaypee@netcom.com>]

   I wanted to start a separate thread that is somewhat related to
the previous discussion about collating weather data available from the
Internet for preflight planning.  I'm not sure how many glider pilots
there are on the list yet, but let me also throw out some thoughts
about another pet project I have been fiddling with lately, namely
constructing your own Soaring Forecasts from Internet data.

   Over the last year or two, I've noticed that the amount of quality
weather data available over the net (forecasts and observations) that
are suitable for preparing your own Soaring Forecasts has really grown.
Some of you may be familiar with Kevin Ford's Thermal Index Calculator,
and for some areas (like Southern California) the NWS now provides a
web based "ready to go" soaring forecast that is updated twice a day.
In our area, the state water district even has some real time stations
that can provide temperature, wind speed and direction.

   One of my pet gripes about web based information is that the process
of "surfing" for information is really pretty time consuming and in
general not very constructive. I'm much more interested in software
that "works for me" not the other way around.

   So here's my basic idea: write a script that you run from crontab
on a regular basis (say twice a day) that grabs the information and
formulates a Soaring forecast of it's own (say thermal, ridge and wave
forecasts). The script then compares the forecast conditions against some
user configurable threshold, and if things look like they are going to be
"booming" that day, it sends you an alert by e-mail: "Drop what you are
doing, and get your ass out to the gliderport!"

   It's actually not that hard to implement, in fact I've already cobbled
together a working version, though it's perhaps not polished enough for
general release.  There are several command line URL "grabbers" (like
curl, wget, or lynx with -dump) that can be used to fetch the pages. Now
comes the slightly messy part: parsing the information from the pages. My
current implementation uses tools like awk, sed, etc.  It works fine, but
I'm not real satisfied with this approach.

    This is an area where I would appreciate some of your thoughts.  I'm
well versed in all the string manipulation tools and languages; awk, sed,
tcl, perl, icon, etc. However, it would be nice if this problem could be
implemented in a manner that doesn't require the user to be an experienced
programmer to set it up. There's another issue as well.  You can bet that
the format of the pages will change every so often, so implementations
that are more complex will be a bigger pain in the rear to maintain.
Are there any efforts out there to create a user friendly (e.g. usable
by non tcl, perl hackers) software package for extracting information
from formatted documents????

Best Regards, John

-- 
 ___|___  | John C. Peterson, KD6EKQ | Installing Micro$oft Windows on a PC is
  -(*)-   | mailto:jaypee@netcom.com | like putting wagon wheels on a Porsche!
  o/ \o   | San Diego, CA   U.S.A    | See http://www.linux.org/ for info.
-
Archives of linux-aviation: http://mail.nl.linux.org/lists/linux-aviation/
To unsubscribe: send the command "unsubscribe linux-aviation" in the body
of a mail message to <Majordomo@mail.nl.linux.org>.