On 03/24/15 17:13, Jeffrey Witt wrote:
Andy, thanks so much,
I went with option 2. It seems to work great.
Check out http://sparql.scta.info <http://sparql.scta.info/>
(If you have time maybe you could quickly browse the above link and see if
you see any security concerns. I want people to be able to query the
database, but just not add or delete anything. I tried to add a rdf/xml file
through the interface and — as desired — it gave me an error. So it seems
like it providing the desired read only access.
Not a security issues per se but editing the HTTP pages to remove the
validators and the update panel makes it clearer.
The only other follow up is that I would love to set my url:
sparql.scta.info directly to the query endpoint.
Right now the query endpoint is sparql.scta.info/ds/query, where as
sparql.scta.info takes the user to the fuseki interface.
Can you (or anyone else) think of a way to set sparql.scta.info/ds/query to
the web root, so that queries can be sent directly to
http://sparql.scta.info.
I would prefer to get rid of the Fueseki web interface altogether, but not
sure how to or if I can do that.
With just Fuseki, I can't think of a way.
But you could add a reverse proxy (i.e. run httpd, nginx etc) on
sparql.scta.info and have it send "/?query=" to "/ds/query?query="
Andy, feel free to pass on this question. You’ve already helped me a ton. Or
just tell me to post this as a new question to the list-serv.
Thanks.
jeff
On 3/24/15, 10:36 AM, "Andy Seaborne" <[email protected]> wrote:
Hi there,
Re: option 1
The TDB tools are command line scripts - no java programming involved.
# One time:
tdbloader --loc=DB ... your data ...
# Start server each time.
./fuseki-server --loc=DB /ds
Re: option 2
./fuseki-server --loc=DB --update /ds
then
# one time
./s-post
puts it in a directory called "DB".
There is a time when the server is accessible for update though.
Andy
On 24/03/15 12:40, Jeffrey Witt wrote:
Hi Andy,
Thanks for your response. This is probably a newbie follow up, butŠ
I¹m curious about option two. How can I build a persistent database with
--update without using TDB. Would I start Fuseki without the --mem option?
If so, where is the data getting stored if not with the TDB tools?
If I can somehow make a persistent data store while using only Fuseki, and
avoid using TBD for the time being, that would be the easiest option for
me.
Thanks for your help.
jw
