[DAS2] Re: Apollo and DAS/2 priorities

Allen Day allenday at ucla.edu
Sat Feb 4 10:43:10 UTC 2006


There is a database server down, which is why I haven't posted the new
code to /codesprint yet.  Hopefully it will be back online tomorrow.

However, on my dev box I was able to make the server code serve up almost
all of what is described in Andrew's new_spec.txt file.  The large
remaining problems are:

* Properties ( <PROP/> elements ).  I still don't fully understand how
these work, if the previous implementation continues to be valid, or if
the implementation has been invalidated by the new document.

* Alternate default Content-Type header for the same command, e.g.

  /sequence/.../segment       # Content-Type: application/x-das-blah+xml
  /sequence/.../segment/chrM  # Content-Type: text/x-fasta

This is an artifact of an earlier design decision assumed Content-Type had
a single default and would only be modified if a ?format= parameter was
passed.  This is difficult to fix properly, so right now the fasta is
served up under the XML Content-Type.

-Allen


On Wed, 1 Feb 2006, Allen Day wrote:

> Okay, I will tag the current server and leave it at:
> 
> http://das.biopackages.net/das
> 
> I saw in the most recent commits by Andrew that the root-level "/das" is
> no longer needed, so I propose putting an updated server at:
> 
> http://das.biopackages.net/codesprint
> 
> If we're going to keep the current server in a "maintained but deprecated"  
> mode like this, I'll be making changes to the "new" server before Friday.
> 
> When the new version of IGB comes out we can then upgrade the current
> server.
> 
> Sound good?
> 
> -Allen
> 
> 
> On Wed, 1 Feb 2006, Helt,Gregg wrote:
> 
> > Yes, what Ed said, that's what I meant.  Updated server, but at a
> > different address.  Otherwise the current release of IGB will break when
> > trying to use the biopackages server.
> > 
> > Once our IGB code has caught up to the updated server, we can roll out a
> > new release to point to the new server instead of the old one.  But not
> > yet.
> > 
> > 	Thanks,
> > 	Gregg
> > 
> > > -----Original Message-----
> > > From: das2-bounces at portal.open-bio.org
> > [mailto:das2-bounces at portal.open-
> > > bio.org] On Behalf Of Ed Erwin
> > > Sent: Wednesday, February 01, 2006 3:45 PM
> > > To: Allen Day
> > > Cc: DAS/2
> > > Subject: Re: [DAS2] Re: Apollo and DAS/2 priorities
> > > 
> > > 
> > > Gregg asked me to say "No".  Please do not break the current server
> > that
> > > IGB is using.
> > > 
> > > Please make your changes on a server at a different URL.
> > > 
> > > Thanks
> > > Ed
> > > 
> > > Allen Day wrote:
> > > > That's what I was thinking too, but I was worried about the existing
> > > > Genoviz clients "in the wild" having the server suddenly break.
> > > >
> > > > So you're saying it's okay with you if those clients have a service
> > > > interruption?
> > > >
> > > > -Allen
> > > >
> > > >
> > > > On Wed, 1 Feb 2006, Helt,Gregg wrote:
> > > >
> > > >
> > > >>That would be great if you could update the biopackages server
> > before
> > > >>the code sprint starts!  Then client implementers will have a server
> > to
> > > >>test with.
> > > >>
> > > >>	thanks,
> > > >>	gregg
> > > >>
> > > >>
> > > >>>-----Original Message-----
> > > >>>From: das2-bounces at portal.open-bio.org
> > > >>
> > > >>[mailto:das2-bounces at portal.open-
> > > >>
> > > >>>bio.org] On Behalf Of Allen Day
> > > >>>Sent: Wednesday, February 01, 2006 2:42 PM
> > > >>>To: Andrew Dalke
> > > >>>Cc: DAS/2
> > > >>>Subject: Re: [DAS2] Re: Apollo and DAS/2 priorities
> > > >>>
> > > >>>I just looked over your changes, and will begin making the changes
> > to
> > > >>
> > > >>the
> > > >>
> > > >>>server repository today.
> > > >>>
> > > >>>I'd like to update the server at das.biopackages.net with my
> > changes
> > > >>
> > > >>on
> > > >>
> > > >>>Friday, unless there are objections.
> > > >>>
> > > >>>I'll be taking notes along the way and will post to the list if
> > > >>
> > > >>anything
> > > >>
> > > >>>in your document is unclear to me.
> > > >>>
> > > >>>At first glance, I agree -- the changes are minor.
> > > >>>
> > > >>>-Allen
> > > >>>
> > > >>>
> > > >>>On Mon, 30 Jan 2006, Andrew Dalke wrote:
> > > >>>
> > > >>>
> > > >>>>Allen:
> > > >>>>
> > > >>>>>Is the spec going to be in a stable state for the code sprint?
> > > >>
> > > >>I'd
> > > >>
> > > >>>>>like
> > > >>>>>to use this time to sync the server implementation with a stable
> > > >>>>>version
> > > >>>>>of the spec.  It looks like there have been many substantial
> > > >>
> > > >>changes.
> > > >>
> > > >>>>I have just (within the last few minutes) completed the first
> > draft
> > > >>>>of the update of the spec.
> > > >>>>
> > > >>>>It's not in HTML - that calls for too much work for this stage.
> > > >>>>It's text, in CVS under das/das2/new_spec.txt
> > > >>>>
> > > >>>>There are many parts which need clarification.  These are marked
> > > >>>>with a "XXX" along with my comments.
> > > >>>>
> > > >>>>The RNC files are in
> > > >>>>
> > > >>>>   das/das2/scratch/*.rnc
> > > >>>>along with some test XML files.  These XML files are not meant
> > > >>>>to be realistic.  They are meant more to check edge cases.
> > > >>>>
> > > >>>>I do no think there are major changes to the spec.  Most of the
> > > >>>>changes have actually trimmed things down, like getting rid of
> > > >>>>the "properties" subtree and merging the different "sources"
> > > >>
> > > >>requests
> > > >>
> > > >>>>into a single document.
> > > >>>>
> > > >>>>
> > > >>>>Here are the major interfaces
> > > >>>>
> > > >>>>$PREFIX/sequence - a "sources" request
> > > >>>>   This is the top-level entry point to a DAS 2 server.  It
> > returns
> > > >>
> > > >>a
> > > >>
> > > >>>>   list of the available genomic sequence and their versions.
> > > >>>>   [sequence-namespace]
> > > >>>>
> > > >>>>$PREFIX/sequence/$SOURCE - a "source" request
> > > >>>>   Returns the available versions of the given genomic sequence.
> > > >>>>
> > > >>>>$PREFIX/sequence/$SOURCE/$VERSION - a "versioned source" request
> > > >>>>   Returns information about a given version of a genomic
> > sequence.
> > > >>>>   Clients may assume that the sequence and assembly are constant
> > > >>
> > > >>for a
> > > >>
> > > >>>>   given version of a source. Note that annotation data on a
> > server
> > > >>>>   with curational write-back support may change without changing
> > > >>
> > > >>the
> > > >>
> > > >>>>   version.
> > > >>>>
> > > >>>>
> > > >>>>For a given version here are the sub-parts.  Note that I've gone
> > > >>
> > > >>ahead
> > > >>
> > > >>>>and split the query urls (segment, features and types each have
> > > >>
> > > >>query
> > > >>
> > > >>>>interfaces) from the base directory used as containers for the
> > > >>
> > > >>segments,
> > > >>
> > > >>>>features and types.
> > > >>>>
> > > >>>>  $VERSION/segments - the segments query URL; summarizes the
> > > >>
> > > >>top-level
> > > >>
> > > >>>>     segments in the data source
> > > >>>>
> > > >>>>  $VERSION/segment/$SEGMENT_ID - a segment query; used to get
> > > >>
> > > >>detailed
> > > >>
> > > >>>>     information about the identified segment
> > > >>>>
> > > >>>>  $VERSION/features - the feature filter query URL.  Features are
> > > >>>>    locatable annotations or experimental results.  The feature
> > > >>
> > > >>filter
> > > >>
> > > >>>>    URL supports query parameters to select a subset of the
> > features
> > > >>>>    based on position, feature type and other properties.
> > > >>>>
> > > >>>>  $VERSION/feature/$FEATURE_ID - a feature query; used to get
> > > >>
> > > >>detailed
> > > >>
> > > >>>>     information about the identified feature
> > > >>>>
> > > >>>>  $VERSION/types - the types query URL which returns a list of all
> > > >>>>    feature types.  Feature types include ontology and depiction
> > > >>>>    details for all features of the given type.
> > > >>>>
> > > >>>>  $VERSION/type/$TYPE_ID - details about the specified feature
> > type
> > > >>>>
> > > >>>>Oh, and there are internal conflicts which will be straightened
> > > >>>>out in the next draft.  These shouldn't be big.
> > > >>>>
> > > >>>>					Andrew
> > > >>>>					dalke at dalkescientific.com
> > > >>>>
> > > >>>>_______________________________________________
> > > >>>>DAS2 mailing list
> > > >>>>DAS2 at portal.open-bio.org
> > > >>>>http://portal.open-bio.org/mailman/listinfo/das2
> > > >>>>
> > > >>>
> > > >>>_______________________________________________
> > > >>>DAS2 mailing list
> > > >>>DAS2 at portal.open-bio.org
> > > >>>http://portal.open-bio.org/mailman/listinfo/das2
> > > >>
> > > > _______________________________________________
> > > > DAS2 mailing list
> > > > DAS2 at portal.open-bio.org
> > > > http://portal.open-bio.org/mailman/listinfo/das2
> > > _______________________________________________
> > > DAS2 mailing list
> > > DAS2 at portal.open-bio.org
> > > http://portal.open-bio.org/mailman/listinfo/das2
> > 
> _______________________________________________
> DAS2 mailing list
> DAS2 at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/das2
> 



More information about the DAS2 mailing list