Harvest geospatial and temporal information into top-level dataset

[LucaCinquini]

To enable geospatial and temporal queries for datasets, that information needs to be harvested into the top-level dataset metadata. Currently, the CMIP5 THREDDS catalogs produced by the publisher only store geo/temporal information at the aggregation level (where it is correctly harvested). That information needs to be passed up from the aggregations to the top-level dataset.

[LucaCinquini]

Implemented as part of the summary metadata that is collected while parsing the full dataset hierarchy within a THREDDS catalog. For example, the following information is written into the top-level dataset:

"datetime_start"=2002-09-16T00:00:00Z
"datetime_stop"=2011-05-16T12:00:00Z
"east_degrees"=359.5
"north_degrees"=89.5
"south_degrees"=89.5
"west_degrees"=0.5

[LucaCinquini]

Note: in order for this information to be made available at an Index Node, the following actions are necessary:

o Update the Index Node software to the latest version
o Re-crawl the full hierarchy of THREDDS catalogs for all participating Data Nodes)

No Solr schema change is necessary.