Results
Results are always returned in XML-file. Such file will contain URL’s from which the generated files can be downloaded.
Search
Search returns metadata for available FITS-images or FITS-cubes which satisfied the search query conditions. The XML-response file for search is build up as follows:
<result>
<description>
<input>
<msg>
<DatacubeCount>
<survey>
<datacube>
<datacube>
...
</survey>
<survey>
<datacube>
<datacube>
...
</survey>
...
</result>
Found datacubes are grouped by surveys. For convenience, input parameters are repeated and count of found datacubes is also given. Each survey also contains its descriptive metadata: its name, species, transition including rest-frequency and velocity unit.
The datacube node provides the following information:
Overlap: quality of overlap (full-, partial-, no-overlap etc.)
PublisherDID: unique identificator of the datacube (used in cutout)
Datacube-bounds: in galactic sky coordintes and velocity on the spectral axis
Datacube-vertices: in galactic sky coordintes and velocity on the spectral axis
Based on this data, client selects which datacubes are needed, and can perform a cut to retrieve only the required data.
The PublisherDID serves as datacube identificator. Clients should consider it opaque and not interpret it, only store and use where required (cutout, merge).
Cutout
The XML in response has the same preamble as search (nodes from begining until first <survey>). Then two nodes:
CUT : FITS-filename from which cut originates with pixels coordinates
URL : this url downloads the cutout file
nullValues: optional node, which shows percentage of undefined pixels in the cut region
Merge
The XML contains the cubes identified by pubdid and lying in the requested region and being part of the sub-survey. It shows the cuts used from all files (one CUT-node for each). Finally, the URL-node contains the merged file ready to be downloaded by the client (like in in cutout).