 |
 |
Joe, I'm not sure if this conforms to what you're talking about, but have you seen the Library of Congress' OAI-ORE implementation for Chronicling America? http://chroniclingamerica.loc.gov/ http://chroniclingamerica.loc.gov/lccn/sn83030214.rdf -Ross. On Wed, Mar 10, 2010 at 1:44 PM, Joe Hourcle <[log in to unmask]> wrote: > Most of the examples I've seen of OAI-ORE seem to assume that you're > ultimately interested in only one object within the resource map -- > effectively, it's content negotiation. > > Has anyone ever played with using ORE to point at an aggregation, with the > expectation that the user will be interested in all parts, and automatically > download them? > > ... > > Let me give a concrete example: > > A user searches for some data ... we find (x) number of records > that match their criteria, and they then weed the list down to 10 > files of interest. > > We then save this request as a Resource Map, as part of an OAIS > "order". I then want to be able to hand this off to a browser / > downloader / whatever to try to obtain the individual files. > > Currently, I have something that can take the request, and create a tarball > on the fly, but we have the unfortunate situation when some of the data is > near-line and/or has to be regenerated -- I'm trying to find a good way to > effectively fork the request into multiple smaller request, some of which I > can service now, and some for which I can return an HTTP 503 status (service > unavailable) w/ a retry-after header. > > ... > > Has anyone ever tried doing something like this? Should I even be looking > at ORE, or is there something that better fits with what I'm trying to do? > > Thanks for any advice / insight you can give > > -Joe > > ----- > Joe Hourcle > Programmer/Analyst > Solar Data Analysis Center >