Re: querying TDB by program

[Andy Seaborne <an...@apache.org>]

On 01/01/13 16:20, Jean-Marc Vanel wrote:
> So, I created a new database with apache-jena 2.7.4 .
>
> 2012/12/31 Andy Seaborne <an...@apache.org>
>
>> On 31/12/12 12:01, Jean-Marc Vanel wrote:
>>
>>> And when I did that , I found that, apparently, TDB 0.9.4 inside EulerGUI
>>> is not able to read a directory made with latest TDB from SVN (the query
>>> result is empty).
>>>
>>
>> Do you have a test case for this?
>> (what's the query,
>
>
>   SELECT DISTINCT ?p WHERE {?s ?p ?o} LIMIT 200
>
> what's the data?)
>
>
> I loaded this from dbPedia 3.8 :
>   $HOME/src/apache-jena-2.7.4/bin/tdbloader2 --loc ../tdb74 labels_en.ttl

Not your fault but this is depressingly full of bad URIs.  I wish they 
would %-encode carefully to standardise the fixup of IRIs like the (much 
better) .nt file.

(Yes - I know many tools ignore the issues but if they input is suspect, 
how can we create reliable output on the web?)

>
> The code was :
>               Dataset dataset = TDBFactory.createDataset(service.substring(
> 5 ));
>              Model db = dataset.getDefaultModel();
>               queryExecution = QueryExecutionFactory.create(query, db );

Aside: This is not not the best way because it does not directly go the 
TDB store.

>
> also tried :
>               Dataset dataset = TDBFactory.createDataset(service.substring(
> 5 ));
>              queryExecution = QueryExecutionFactory.create(query, dataset );

Aside: This is better.  It goes straight to TDB.

>
> then in both cases :
> final ResultSet result = queryExecution.execSelect();
>
> but : result.getRowNumber() == 0 ! :( .

See javadoc:

/** Return the "row" number for the current iterator item */
     public int getRowNumber() ;

as you have not iterated over the results, you are at row 0.

getRowNumber is the *current* row, not the count.


Try:

int x = ResultSetFormatter.consume(result)

>
> However : db.size() == (long) 9442421
>
> And tdbquery gives :
>
> ------------------------------------------------
> | p                                            |
> ================================================
> | <http://www.w3.org/2000/01/rdf-schema#label> |
> ------------------------------------------------
>
> This was tested with EulerGUI code, whose Maven dependencies are :
>
> <dependency>
>    <groupId>org.apache.jena</groupId>
>    <artifactId>jena-arq</artifactId>
>    <version>2.9.4</version>
> </dependency>
>
> <dependency>
>    <groupId>org.apache.jena</groupId>
>    <artifactId>jena-tdb</artifactId>
>    <version>0.9.4</version>
> </dependency>
>
> I understand that I should replace these with a single dependency , but I
> don't think this is the cause of my problem :

It's not.

>
> <dependency>
>      <groupId>org.apache.jena</groupId>
>      <artifactId>apache-jena</artifactId>
>      <version>2.7.4</version>

        <type>pom</type>

> </dependency>

There is no jar there (currently) - it's a POM of the modules.

>
>
>>
>>          Andy
>>
>
>
>

Re: querying TDB by program

[Jean-Marc Vanel <je...@gmail.com>]

2013/1/1 Andy Seaborne <an...@apache.org>

>
>> but : result.getRowNumber() == 0 ! :( .
>>
>
> See javadoc:
>
> /** Return the "row" number for the current iterator item */
>     public int getRowNumber() ;
>
> as you have not iterated over the results, you are at row 0.
>
> getRowNumber is the *current* row, not the count.
>

OK. That was just an attempt to debug.
But after that I tried to make use of it, but it seems that for ARQ queries
on remove SPARQL, result.getRowNumber() starts at 2 .


> Try:
>
> int x = ResultSetFormatter.consume(**result)
>

This helped me to learn that ResultSet is iterable just once.
That was my problem, because I used it once to record the columns datatypes
on first row, and then for the data rows.

Thanks for for answers .

-- 
Jean-Marc Vanel
Déductions SARL - Consulting, services, training,
Rule-based programming, Semantic Web
http://deductions-software.com/
+33 (0)6 89 16 29 52
Twitter: @jmvanel ; chat: irc://irc.freenode.net#eulergui