[DAS2] [Bug 1808] New: <NAMESPACE>, region ids and sequence ids

[bugzilla-daemon at portal.open-bio.org]

http://bugzilla.open-bio.org/show_bug.cgi?id=1808

           Summary: <NAMESPACE>, region ids and sequence ids
           Product: BioDAS
           Version: 2
          Platform: All
        OS/Version: All
            Status: NEW
          Severity: normal
          Priority: P2
         Component: Spec - GET Requests
        AssignedTo: das2 at portal.open-bio.org
        ReportedBy: dalke at dalkescientific.com


In bug 1807 I asked how region ids are related to sequence ids.  I think they are
supposed to be the same, so to get the sequence for a given region you append
the region id to the sequence namespace.

How does the client know which namespace to use?  I think it's supposed to
be hard-coded into the client that "/sequence/" contains the genomic DNA.

Otherwise there is insufficient information in the versioned source request to
figure out which URL to use.  Consider this example from the spec

      <NAMESPACE id="volvox/1/sequence" description="A sequence or subsequence">
         <FORMAT id="fasta"   type="text/x-fasta" />
      </NAMESPACE>
      <NAMESPACE id="volvox/1/protein" description="A protein sequence or subsequence">
         <FORMAT id="fasta"   type="text/x-fasta" />
      </NAMESPACE>

There's not enough information for the client to figure out that the "volvox/1/sequence"
URL contains genomic sequences while the "volvox/1/protein" contains protein sequences.

I would rather have a 'type' field

  <NAMESPACE type="sequence" id="volvox/1/protein" ...>

where the type data is from a well-defined vocabulary.



------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.