Hi Carl, Not sure I'm understanding fully what you are trying to do...You want to provide two access methods for each granule? One OPeNDAP, the other LAS?
I think John is right and you will need to regenerate (or hand edit) your catalog to add the LAS service/access information to each granule.
Is there an easy way to automatically get both the OPeNDAP url and the LAS id as in the url below ("var=6505")?
Hmm, strange, the URL below isn't working the way it did earlier. Ethan John Caron wrote:
Carl Wolfteich wrote:Hi John, Thanks for getting back to me. You asked:>How can the LAS access link be the same for each? How does LAS know which dataset you are referring to?I guess the LAS access link I had in mind was like this for NCEP Climatology:http://ferret.pmel.noaa.gov/NVODS/servlets/constrain?var=6505Such that I would have this general access link for the dataset at each granule level.If I understand what you mean, the LAS server would understand the entire dataset collection (named by, eg "NVODS") and allow the user to select granules within it?Im unclear how LAS and THREDDS will interact in this case. Who are you working with in LAS? we should probably bring them into the conversation.>What is the number of datasets and catalogs you are working with ?I have a main catalog that calls about 150 CatRefs, each corresponding to a particular dataset. The number of granules contained in each parent dataset can vary from just a handful (<10) to thousands.How do the granules differ within the collection dataset? Just by time, or ??Carl John Caron wrote:Hi Carl: Carl Wolfteich wrote:John,Your previous help has been very helpful as we continue to construct a thredds catalog for this ReASON project. I have a question though, concerning data access service elements. I have a main thredds catalog that calls a list of CatalogRef elements. Each CatalogRef element corresponds to a particular dataset *.xml file, which contains a primary dataset and nested datasets that correspond to the granule data, something like this:Example CatalogRef element: <catalog <service name="dods3" serviceType=Opendap ..... <service name="las1" serviceType=LAS ..... <dataset ID=... name=... <metadata inherited="true" <access serviceName="las1" urlPath=..... .... .... </metadata> <dataset ID="1980-3" name=... <access serviceName="dods3" urlPath=.... .... </dataset> <dataset ID="1980-4" name=... <access serviceName="dods3" urlPath=.... ..... </dataset> <dataset ID="1980-5" name=... <access serviceName="dods3" urlPath=.... .... </dataset> </dataset> </catalog>At the granule level (e.g. 1980-3, 1980-4, etc) there is an access link (dods3) to the Opendap(DOds) dataset form and this link appears in the html for each granule, given its unique urlPath to the data as shown above. What I would like to do is also add a link for the LAS access (las1) for the data. The LAS access link however will be the same for each granule.How can the LAS access link be the same for each? How does LAS know which dataset you are referring to?As you can see above I put this access element in the parent dataset with the rest of the metadata that is then inherited down to the child datasets (granules). When I do this however, no LAS link appears in the html. I would like this link in both the parent dataset as well as at the granule level. Most of these CatRefs have already been automatically generated and I do not want to have to go back and regenerate them to put the LAS access link in each child dataset. I was hoping I could just put it near the top in the inherited metadata so it just gets included (handed down), but it does not seem to work this way. Is there some way I could do this, include this LAS access link in the inherited metadata above? thanks a lot,access elements are not inherited, so you cant go that route. in fact, i dont think you can add an access elelent to a metadata element and still validate.Under certain conditions, you can use a compound service. however, these dont transfer across catalog refs, they must be self contained in the catalog.Im guessing you may have to regenerate, but first i need to understand how LAS is interacting with the dataset.What is the number of datasets and catalogs you are working with ? Im cc'ing ethan davis who may be able to help also.
-- Ethan R. Davis Telephone: (303) 497-8155 Software Engineer Fax: (303) 497-8690 UCAR Unidata Program Center E-mail: edavis@xxxxxxxx P.O. Box 3000 Boulder, CO 80307-3000 http://www.unidata.ucar.edu/ ---------------------------------------------------------------------------
netcdf-java archives: