I'm not sure if you meant the sources schema exactly as it is or to
make the <SOURCE> element the root element. I would vote for the
former (i.e. a sources document containing one source). It makes
things much easier for the client and spec not to have a separate
schema.
Regarding the XSL stylesheets, I can see where you're coming from. I
have some prototype stylesheets which have an option to show the XML
(prettied up), which I think is the best compromise.
On 15 Sep 2009, at 17:56, Jonathan Warren wrote:
My vote at the moment would be to have the source element.
On a side note: My vote would also be to not have nice human
readable stylesheets at all ;) Though I accept I may be a lone voice
on this one. Mainly because I want DAS to be as powerful as possible
while being easy for new users to get into ( some people who have
provided das sources for years have not known they could just right
click and select "show source" to get the xml and see that their new
elements are shown in the xml but are not in the stylesheet!!). It
justs adds a layer of complexity and confusion for little gain in my
opinion. The alternative would be to have raw xml as default and
stylesheet as an option - presumably that could be done? Maybe I
missed the whole point about putting them in? Surely people use das
in a browser for discovery and testing not actually viewing data on
a day to day basis?
On 15 Sep 2009, at 17:28, Thomas Down wrote:
On Tue, Sep 15, 2009 at 4:35 PM, Jonathan Warren
<[email protected]> wrote:
I've always had an issue with the commands like this http://www.ebi.ac.uk/das-srv/genomicdas/das/eqtl_rat_cis_fat
not being a valid das command as it's the most natural request for
a person new to das to make. So giving it a specific purpose and
response is a good idea.
What would you put there? HTML "welcome" information (which seems
to be ProServer's current behaviour)? DAS XML? (in which case...
the relevant SOURCE element, or something else)? DAS XML with some
strong encouragement for servers to attach a stylesheet to turn it
into something readable if you hit it with a browser?
Currently, Dazzle just returns an error, because it interacts
rather nastily with the way Dazzle supports hierarchical datasource
namespaces. Having said that, the original use case for these
hasn't really materialized, so it wouldn't be a big deal to strip
this out.
Thomas.
Jonathan Warren
Senior Developer and DAS coordinator
[email protected]
Ext: 2314
Telephone: 01223 492314
-- The Wellcome Trust Sanger Institute is operated by Genome
Research Limited, a charity registered in England with number
1021457 and a company registered in England with number 2742969,
whose registered office is 215 Euston Road, London, NW1 2BE.
_______________________________________________
DAS mailing list
[email protected]
http://lists.open-bio.org/mailman/listinfo/das