It's certainly a similar problem. I think the disseminator route would work,
but it also seems like it is overkill to write a definition, deployment, and
some intermediate wrapper to essentially pass through query string parameters.
There may be some cases where a pass-through behavior isn't desirable, but (at
the risk of increasing complexity) maybe it would need something like the
Apache HTTPD mod_rewrite qsappend/QSA flag
<http://httpd.apache.org/docs/2.2/mod/mod_rewrite.html>.. Ah well.
Chris
On 1/13/10 5:38 PM, "Benjamin Armintor" <[email protected]> wrote:
>
>> Why not use RESTful urls, rather than query string parameters? Will the
>> h264.code-shop software handle that?
>
> The pseudostreaming takes a similar approach as the w3c media fragments spec
> -- http://www.w3.org/TR/2009/WD-media-frags-20091217/ -- which recommends
> either URI queries (which many media players implement) or URI fragments,
> both of which are probably appropriate (and better semantically?) in this
> application.
>
> Chris
Hi Chris.
I think this is similar to the way that people have run Fedora over,
for example, Djatoka: Mapping the service calls to OpenURL URIs, etc.
But there's also some limitations in what data are available for URL
substitution (and the larger limitation of http bindings and url
substitutions themselves). I can imagine that being able to
substitute some value from the DC stream, or the object of a triple in
RELS-EXT, might make things better suited to this kind of service
proxying. As it stands, if the service you want to call isn't able to
make sense of a PID or call back to Fedora for the data via a
datastream URI, you're often stuck implementing intermediary wrappers.
Are these the kind of difficulties keeping you from going the
disseminator route? I think there's some ongoing conversation in the
community about the way services work, and it's be great to get your
input.
To get back to your philosophical question: Ideally (if I may), this
is the kind of thing that should happen through defined services.
Datastreams are intended to be a specific addressable resource (so
that, for example, checksumming is supported and understood to be
relevant to the resolved bytes).
regards,
Ben
ps: Cool stuff over there at WGBH!
------------------------------------------------------------------------------
Throughout its 18-year history, RSA Conference consistently attracts the
world's best and brightest in the field, creating opportunities for Conference
attendees to learn about information security's most important issues through
interactions with peers, luminaries and emerging and established companies.
http://p.sf.net/sfu/rsaconf-dev2dev
_______________________________________________
Fedora-commons-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/fedora-commons-developers
