Hi all, I submitted a second patch to the opendapps module that is intended to augment the functionality for parsing metadata from OpenDAP endpoints:
https://issues.apache.org/jira/browse/OODT-366 Here is the patch description: ================================================================================================ This is a second patch targeted at extending the opendapps functionality for use in the CMDS JPL project. This patch contains functional modifications to four files: o OpendapProfileHandler: when harvesting metadata from multiple THREDDS catalogs, avoid stopping the process if any of the catalogs results in an error, rather keep harvesting all the other datasets o DatasetExtractor: parse all datasets in a THREDDS catalog, even if they don't have an explicit direct <access> sub-element, but rather have a "urlPath" attribute o DataseCrawler: * only select for metadata extraction those THREDDS datasets that have an opendap access URL, disregard the others * when the THREDDS catalog has no "authority" attribute, the dataset.getUniqueID() returns the string "null". In this case, use the dataset.getID() method, which returns the dataset ID with no authority prepended. o ProfileUtils: create an OODT <profElement> for each opendap variable found in the DDS stream, wether it is explicitly configured or not, to minimize manual handling of the configuration files. If the variable is configured, use the configuration spec to possibly rename it, and cast it as RANGED_ELEMENT_TYPE or ENUM_ELEMENT_TYPE. If the variable is not configured, assume it is of RANGED_ELEMENT_TYPE. Finally, all classes have additional log statements for debugging purposes. ================================================================================================== More patches will follow... Please let me know of any problems, thanks, Luca
