At Gallagher's urging I will offer the following suggestions:
On Nov 27, 2006, at 5:43 PM, Roy Mendelssohn wrote:
I apologize for the double post. I got my mail lists confused (old age is setting in), I originally posted to support-thredds but meant to post it here.
I am throwing out to the list a THREDDS related server problem that we are trying to figure out relatively clean ways to solve, in the hope that others may have some good ideas on how to accomplish this.
We are heavily invested in THREDDS (http://oceanwatch.pfeg.noaa.gov/ thredds/catalog.html) in particular the cataloging and the data aggregation. Most of the links on that page are not single files but many files (sometimes in the thousands) being aggregated through time. This is a very useful feature for the type of data we have.
Also, on an in-house server we have been testing Roberto De Almeida's pydap implementation of OPeNDAP (http://pydap.org/). This server has a lot of nice features that we are interested in, such as his wms service and kml service from an OPeNDAP URL type expression.
So what is the issue. Clearly, on web pages that are developed in house, where a user just clicks on a link, if we have both Pydap and THREDDS running on the same machine, we can change the links to point to the appropriate server. But for the more general public, or for people using the links in scripts etc, we would like to have just one basic link that depending on the file ending would go to the correct place.
So right now if I wanted the second time period, full extent of one of the files the URL would be:
If I wanted an ascii file it would be:
but what I would like to do is that if I give it:
and it would know to switch-over to the pydap server and produce the kml output (while still being able to view the file as a single file through the THREDDS aggregation).
Now as given, it is unlikely to work since THREDDS is in the URL. But suppose we had something like:
and want it so that in the ascii case it goes to THREDDS and in the kml case it goes to Pydap.
Ideas we had:
1. Monkey with the appropriate section in the THREDDS java to redirect when it recognizes the new service type, much as it does with WCS services. Downside is we would have to monkey with every revision and could mess up things in unexpected ways in the process.
2. At the apache or tomcat level, have the redirection made (is this possible?) based on file type.
3. Some other solution?
A question I pose to Roberto (as he is copied) but may be of general interest is if there is a way for Pydap to serve OPeNDAP served data that is at another URL (even if on the same machine), much as THREDDS can. It is this that would simplify what it would take for us to use the aggregation capabilities in THREDDS with Pydap in our own internal scripts as we can still views these as one file - or does anyone else have general suggestions as how we might combine the desirable features of each.
"The contents of this message do not reflect any position of the U.S. Government or NOAA."
Supervisory Operations Research Analyst
Environmental Research Division
Southwest Fisheries Science Center
1352 Lighthouse Avenue
Pacific Grove, CA 93950-2097
e-mail: address@hidden (Note new e-mail address) voice: (831)-648-9029 fax: (831)-648-8440 www: http://www.pfeg.noaa.gov/
"Old age and treachery will overcome youth and skill."
======================================================== Nathan Potter The OPeNDAP Project address@hidden http://opendap.org 541.752.1852 voice
=============================================================================== To unsubscribe thredds, visit: http://www.unidata.ucar.edu/mailing-list-delete-form.html ===============================================================================
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.