On 2018-12-11 11:29, Roel Janssen wrote:
> On 11-12-18 01:32, swedebu...@riseup.net wrote:
snip
>> I now implemented sparql queries as well. See
>> https://gitlab.com/swedebugia/guile-wikidata
>>
>
> "guile-wikidata" looks cool! And I'm glad to see you've added SPARQL
> support as well. It would be great to integrate it with
> "guile-sparql" if you find the time for it. The "guile-sparql" code
> is quite short, and hopefully readable enough.. :)
Do you mean integrate only the SPARQL-part or the whole? I agree to the
first statement :) but the other wikidata specifics are perhaps better
kept on its own.
Actually the endpoint is a Blazegraph-server and my get-query code does
not seem to work right (see the bug mentioned in the latest commit).
Does guile-sparql work with Blazegraph? If yes we don't need the (buggy)
sparql-get-query-code I wrote.
I could not get it to work and rolled my own as a teach-yourself-scheme
learning-by-doing pet project :)
I tried with the file attached but got this because the driver does not
support URIs but only host, port, type, token:
...
In web/client.scm:
461:0 2 (http-post "http://http://query.wikidata.org/sparql:80…"; …)
181:7 1 (open-socket-for-uri _)
In unknown file:
0 (getaddrinfo "http" "http" 0 #<undefined> #<undefined> #)
ERROR: In procedure getaddrinfo:
In procedure getaddrinfo: Name or service not known
>
> I've been working on a web interface to work with RDF/SPARQL (written
> mostly in Guile Scheme, which uses "guile-sparql" as well). You might
> find it interesting:
> https://github.com/UMCUGenetics/sparqling-genomics
>
> The Guile Scheme parts are in the "web" subfolder.
Thanks for the tip :)
Additionally I have a suggestion for guile-sparql. Could you generate a
pdf from the texi-manuel and include in the repo?
--
Cheers
Swedebugia(use-modules (sparql driver)
(sparql lang))
(sparql-query
"SELECT ?item
WHERE
{
?item wdt:P31 wd:Q146.
}
LIMIT 10
"
#:host "http://query.wikidata.org/sparql";
;; #:port 80
#:type "text/xml"
;; #:token "..."
)
