RDFConnection.fetch() can't Parse DBpedia's IRI #1623
-
|
Hi, I want to use When I parse RDFConnection conn = RDFConnectionRemote.service("http://dbpedia.org/resource/Resource_Description_Framework").build();
Model model = conn.fetch();
model.write(System.out);I got an error: Version: apache-jena-libs 4.6.1 But rdflib.js can parse it, RDF2h's demo Whether Did I do something wrong? Best regards |
Beta Was this translation helpful? Give feedback.
Replies: 4 comments 8 replies
-
|
You should be able to do what you want much more simply like so: Model model = ModelFactory.createDefaultModel();
model.read("http://dbpedia.org/resource/Resource_Description_Framework"); |
Beta Was this translation helpful? Give feedback.
-
|
Dear @rvesse , Thank you very much for your reply!
But I still got an error: Did I do something wrong? Best regards |
Beta Was this translation helpful? Give feedback.
-
|
Could you add an explicit This kinda looks like DBPedia doing something weird with HTTP content negotiation and/or redirection i.e. it's DBPedia's fault not yours! |
Beta Was this translation helpful? Give feedback.
-
|
Thanks. But I'm looking for a common way. I'll try to get the return data through an http request first and then put it into |
Beta Was this translation helpful? Give feedback.
-
|
I was getting a different error from the report. I can't reproduce the That's an HTTP environment issue. Zero bytes came back from DBPedia or an intermediary - not even the HTTP header. Using works from here - there are several redirection steps, upgrade to HTTP/2 and bouncing between http and https. Any of these may be unsupported by an intermediate. The ".ttl" needs to be the data resource: The simplest Jena code I can find is giving 772 triples. As to why JS works, it may be because it ends up at the HTML page (no content negoitation) and parses out the RDFa (that's just a guess). |
Beta Was this translation helpful? Give feedback.
-
|
Dear @afs , Thank you very much for your reply! I don't want to do anything specific with DBpedia IRI: Because the rules are different for every linked data site. I'm looking for a common way. As you said, it should be due to the redirection, because I can parse HttpClient hc = HttpClient.newBuilder().version(HttpClient.Version.HTTP_1_1).build();
Graph g = HttpRDF.httpGetGraph("https://dbpedia.org/resource/Resource_Description_Framework");
Model model = ModelFactory.createModelForGraph(g);
model.write(System.out);Does I'll try to get the return data through an http request first and then put it into |
Beta Was this translation helpful? Give feedback.
-
|
Hi @chenkun19960427 - changing the URL is a way to investigate the situation. What we know so far:
This isn't enough to say what is happening so the suggestion is to try something simpler with less steps to narrow down the root cause. The stacktrace appears to be saying is that the first step (an HTTP 1.1 request) is not even forming a usable TCP connection. https://www.geeksforgeeks.org/java-net-socketexception-in-java-with-examples/ -->
There is no indication where the reset comes from - it seems unlikely that DBPedia itself is doing this given it works elsewhere. This is no Jena involved (I'm using java17 - I tried Java11 as well) - it's It's hard to debug through the JDK so the equivalent from the command line is: No java, lots of detail.
It works here so yes, as in my previous reply, Jena does follow redirects. Jena sets the JDK networking to follow all redirects by default. HttpEnv line 71. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks. Your reply is very detailed. This should be a problem with my network. I also got an error using curl command: Jena does follow redirects, meets my needs. |
Beta Was this translation helpful? Give feedback.
-
|
Hi,
You can take a look at our Linked Data browser:
https://github.com/AtomGraph/Web-Client
It supports content negatiation.
Martynas
atomgraph.com
…On Sat, 19 Nov 2022 at 09.27, chenkun19960427 ***@***.***> wrote:
Dear @afs <https://github.com/afs> , Thank you very much for your reply!
I don't want to do anything specific with DBpedia IRI:
replace `resource` with `data`, and add `.ttl` to the end
Because the rules are different for every linked data site. I'm looking
for a common way.
As you said, it should be due to the redirection, because I can parse
https://dbpedia.org/resource/Resource_Description_Framework correctly:
HttpClient hc = HttpClient.newBuilder().version(HttpClient.Version.HTTP_1_1).build();Graph g = HttpRDF.httpGetGraph("https://dbpedia.org/resource/Resource_Description_Framework");Model model = ModelFactory.createModelForGraph(g);model.write(System.out);
Does Apache Jena have a direct way to support redirection?
I'll try to get the return data through an http request first and then put
it into Apache Jena for parsing. This seems to parse all IRI (but I
haven't done it yet). I will reply in this discussion when I have the
results.
—
Reply to this email directly, view it on GitHub
<#1623 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAGPM5TAIBTEU3VJCV6BL3LWJCFOFANCNFSM6AAAAAASEJ3QAQ>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
Dear @namedgraph ,
But I also want to show RDF as a graph, just to make it a little bit more intuitive. Probably I haven't learned much about By the way, does |
Beta Was this translation helpful? Give feedback.
-
|
Yes Web-Client includes a basic SVG graph layout.
There is no public instance at the moment.
…On Sat, Nov 19, 2022 at 10:19 AM chenkun19960427 ***@***.***> wrote:
Dear @namedgraph <https://github.com/namedgraph> ,
https://github.com/AtomGraph/Web-Client is great.
But I also want to show RDF as a graph, just to make it a little bit more
intuitive.
Probably I haven't learned much about
https://github.com/AtomGraph/Web-Client yet.
By the way, does https://github.com/AtomGraph/Web-Client have an online
access address? I haven't found it yet.
—
Reply to this email directly, view it on GitHub
<#1623 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAGPM5UKOKYTZ43S3SWYELLWJCLT3ANCNFSM6AAAAAASEJ3QAQ>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
|
Thanks. I'll try Web-Client out and learn more about it. |
Beta Was this translation helpful? Give feedback.
I was getting a different error from the report.
Are you behind local proxy or firewall?
I can't reproduce the
Caused by: java.io.IOException: HTTP/1.1 header parser received no byteserror.That's an HTTP environment issue. Zero bytes came back from DBPedia or an intermediary - not even the HTTP header.
Presumably this is the
Connection reset- the local connection has been unexpected terminated.Using
curl:works from here - there are several redirection steps, upgrade to HTTP/2 and bouncing between http and https. Any of these may be unsupported by an intermediate.
T…