Hello Rob, On Wed, Jan 29, 2014 at 09:12:04AM -0800, Rob Vesse wrote: > >It's a complete minimal example. I have an application that broke due to > >this > >one when I upgraded to 1.0.1. > > Really, that is actually a real application query.
No - a complete minimal example. This is one of the real queries: PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#> PREFIX dcterms: <http://purl.org/dc/terms/> PREFIX lio: <http://purl.org/net/lio#> select ?img (SAMPLE(?x) as ?xs) ?p (SAMPLE(?xtype) as ?xtypes) where { { graph ?s { { ?x ?p <http://dbpedia.org/resource/Graph_(mathematics)>.?img a lio:Image. } UNION { ?x1 a <http://dbpedia.org/resource/Graph_(mathematics)>. ?x ?p ?x1.?img a lio:Image. } OPTIONAL { ?x a ?xtype }}} UNION { ?d1 <http://dbpedia.org/ontology/isPartOf> <http://dbpedia.org/resource/Graph_(mathematics)> FILTER isIRI(?d1) FILTER regex(str(?d1),"http://dbpedia.org/resource/";) graph ?s { { ?x ?p ?d1 } UNION { ?x1 a ?d1. ?x ?p ?x1 } } graph ?s { ?img a lio:Image. } OPTIONAL { graph ?s { ?x a ?xtype } } } UNION { SERVICE <http://127.0.0.1:3030/datasets/query> { ?d1 rdfs:subClassOf+ <http://dbpedia.org/resource/Graph_(mathematics)>. } graph ?s { { ?x ?p ?d1 } UNION { ?x1 a ?d1. ?x ?p ?x1 } } graph ?s { ?img a lio:Image. } OPTIONAL { graph ?s { ?x a ?xtype } } } FILTER(?p != <http://www.w3.org/1999/02/22-rdf-syntax-ns#type>) FILTER(?p != <http://www.w3.org/2002/07/owl#sameAs>) FILTER(?p != lio:hasVisualPart ) } group by ?img ?p if(strstarts(str(?x),"http://www.imagesnippets.com/imgtag/images/";) && contains(str(?x),"#Region"),"region",?x) The first and the second or third outer UNION - together with the FILTERs at the bottom seem to cause the error. Can I skip the time consuming step of producing a minimal example next time? It took me maybe an hour. > If developers who rely heavily on Jena are not subscribed to that list and > do not take the time to test release candidates in advance and tell us if > they break things for them we don't find out about these things until > after the release is done. That's OK for me. > As I said in a previous email while we try to provide as many tests as we > can think of (Jena as a whole has ~31K at the moment) How about this one ? http://www.brunni.de/pdp1/ > queries/data that exposes corner cases we haven't covered. If you have a > set of mission critical queries then contributing those as test cases > would be most appreciated and reduces the risks that we unintentionally > introduce regressions in the future. Why not use the minimal example as new test case ? > >BTW: We are not able to implement a feature for the same application > >because > >of https://issues.apache.org/jira/browse/JENA-494 > > Contributions are always welcome, you are free to go look at the code and > provide a fix yourselves. Yes. I am free to learn SPARQL Algebra, the inner workings and optimizations of it in Jena, to get more familiar with Java and Java IDEs and then to fix the things you guys overlooked. I'd prefer not to and to stay in Jena quality management. However, I could 1) pay my employee more familiar with Java to look into this and submit a fix 2) do a donation 3) put a bounty on it What would you prefer ? Regards, Michael Brunnbauer -- ++ Michael Brunnbauer ++ netEstate GmbH ++ Geisenhausener Straße 11a ++ 81379 München ++ Tel +49 89 32 19 77 80 ++ Fax +49 89 32 19 77 89 ++ E-Mail bru...@netestate.de ++ http://www.netestate.de/ ++ ++ Sitz: München, HRB Nr.142452 (Handelsregister B München) ++ USt-IdNr. DE221033342 ++ Geschäftsführer: Michael Brunnbauer, Franz Brunnbauer ++ Prokurist: Dipl. Kfm. (Univ.) Markus Hendel
pgpgaeKCfFmjE.pgp
Description: PGP signature
