I take advantage of the power of the pocket queries at Geocaching.com, but there are a few minor annoyances that I find with them and try to cope with the best I can. The largest problem is that there are way more than 500 waypoints in my area. I can easily travel 20 miles, and there are 500 within 18 miles of my home. Also, I like to keep a little page with information about the cache on my laptop if possible, so I can view the cache info at the cache.

I also believe that having more of a good thing is also good. I download the caches from Navicache on a weekly basis and they are also added to my master .gpx file.

I solve these issues by making multiple pocket queries that are sent automatically to my email account. My Linux server manages the caches and merges the files together. Old pocket queries get removed automatically, so I can rename them and make special occasion queries without worry about filling up my hard drive. Individual HTML files are created for each cache and Microsoft Streets and Trips CSV files are created (as soon as I get DeLorme's software, M$'s garbage product will be removed from my computer). URLs in the CSV file are altered to point to my local hard drive.

I don't do any extra work besides downloading a .zip file whenever I like and the points are updated. The HTML files are stripped for size, so you can make a Plucker file or keep them on your laptop/handheld without much worry. This whole process is accomplished with a few tools. You will need procmail, shell access, PHP4 (as a CGI processor), gpsbabel, zip, and some other standard command-line tools.

procmailrc.txt

The procmail rules to process only incoming Pocket Queries. Rename to procmailrc and edit it (if needed) to match your directory structure.

geocache_shell.txt

The shell script that does the processing. Rename to geocache_shell. Change all references to ~fidian/public_html/geocaching to whatever directory you want the processing done in.

geocache_regen.txt

The regen script will regenerate all of the cache files. Useful if you do not want geocache_shell to regenerate the caches automatically and if you would rather use cron to do it every morning or something. Again, change directories as appropriate.

gpx2html.txt

This PHP script will read in a GPX file and write out an individual HTML file for each waypoint. Rename to gpx2html.php and put into the processing directory you set in geocache_shell.

navicache.txt

Rename this to navicache and edit it to suit your needs. The script retrieves caches from navicache and then re-runs geocache_regen. You'll need to change the paths in this file as well.

When this works, you will see new files in the processing directory. You should always see the gpx2html.php file. For each pocket query that was sent to your email account in the last 2 weeks, you will see the .zip file (from the pocket query), the .gpx file (the data of the .zip), and a .csv file (M$ S&T). There will be MANY directories that contain HTML files for each cache, which all get regenerated every time. Geocaching.gpx is created, containing all points from the other GPX files. Geocaching.zip has all of the HTML files and the generated CSV file.

Watch out! Known hazards:

• Don't include funky punctuation in your pocket query name. Only numbers, letters, and spaces are fine.
• Don't name any of your pocket queries Geocaching.
• If you include any caches from navicache, the resulting Geocaching.gpx file is not able to be read by Watcher, but the Geocaching.zip file will contain the individual .gpx files that you can load and use.