[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Dynamic generation of catalogs with catalogRef


Hi John,

I have found the above method to work for loading XML catalogs that are served from a remote location. This is excellent, because I can then use a script to generate those catalogs dynamically. I am planning to use this in order to generate catalogs on the fly from a database. It strikes me that this isn't a bad way of doing all kinds of things dynamically that people might want to hook in, as it's not coupled with the THREDDS code, and thus can be maintained independently.

The entry in the config file looks like

<catalogRef xlink:title="Dynamic Catalog" xlink:href="http://linux-dev.bom.gov.au:8020/thredds/catalog.html?catalog=http://foo.bom.gov.au:8040/dynamic.xml"; name ="dyn" />

I first tried to use
<catalogRef xlink:title="Dynamic Catalog" xlink:href="http://foo.bom.gov.au:8040/dynamic.xml"; name ="dyn" />

But this didn't work. (I think it was expecting that to be served by another DODS server). So, I tried to force it, and it worked.

This meets my requirements, but I thought I'd put it out for comment.


John Caron wrote:

Tennessee Leeuwenburg wrote:

Hi guys,

As some are aware, I have written an intermediate server for accessing files on an object-oriented database, with files in incompatible formats. For any given file, it works great.

I'd like to now go ahead and produce thredds catalogs from its indexes, but as new data is constantly being added, it would be nice if this was dynamic.

If I write a cgi script which would produce a thredds catalog, would the local server be able to properly interpret and use that? Or would it assume that it was a reference to another server's catalogs, rather than realising it is supposed to serve the data itself?


you can't dynamically modify the TDS catalogs. You can use the datasetScan feature to deal with dynamic files. However, that currently works only on directory scanners. We are working on a way to plug some code to find out what the datasets are. Seems like that should work. We can take the details offline.

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.