Re: negative test case question

Yes, that clears up what we should do for the test cases.

Thanks

On Thu, Feb 9, 2012 at 10:28 PM, Souripriya Das
<SOURIPRIYA.DAS@oracle.com>wrote:

> If you are going to materialize the RDF data (for all the million rows):
> the error will show up (b/c every row gets mapped).
> If you are using virtual RDF, you'll not hit any error until and unless
> you happen to access the bad row and try to construct an IRI.
>
> Thanks,
> - Souri.
>
> ----- Original Message -----
> From: juanfederico@gmail.com
> To: SOURIPRIYA.DAS@oracle.com
> Cc: public-rdb2rdf-wg@w3.org
> Sent: Thursday, February 9, 2012 11:08:59 PM GMT -05:00 US/Canada Eastern
> Subject: Re: negative test case question
>
> But you said initially that the failure of the mapping would depend on the
> data. I understand that if in the data, everything is an IRI, then the
> mapping succeeded. If everything was not an IRI, then the mapping would
> fail because all the triples are "wrong". But if the data is mixed. Is the
> mapping right or wrong?
>
>
> On Thu, Feb 9, 2012 at 9:53 PM, Souripriya Das <SOURIPRIYA.DAS@oracle.com>wrote:
>
>> If the SPARQL query you are executing does not need to return or use the
>> constructed IRI from that one bad row, then the query should succeed.
>> Otherwise it should fail with an error such as "illegal IRI in data" or
>> something along that line.
>>
>> Thanks,
>> - Souri.
>>
>> ----- Original Message -----
>> From: juanfederico@gmail.com
>> To: SOURIPRIYA.DAS@oracle.com
>> Cc: public-rdb2rdf-wg@w3.org
>> Sent: Thursday, February 9, 2012 10:37:08 PM GMT -05:00 US/Canada Eastern
>> Subject: Re: negative test case question
>>
>> Ok. I see.
>>
>> If I have one million rows and they all have IRIs for the FirstName
>> column, except for one row... then what happens? Or is this just an
>> implementation issue.
>>
>>
>> Juan Sequeda
>> +1-575-SEQ-UEDA
>> www.juansequeda.com
>>
>>
>> On Thu, Feb 9, 2012 at 9:16 PM, Souripriya Das <SOURIPRIYA.DAS@oracle.com
>> > wrote:
>>
>>> Whether the following mapping will fail or not will depend upon the data:
>>>
>>> <TriplesMap1>
>>>     a rr:TriplesMap;
>>>     rr:logicalTable [ rr:tableName "Employees" ];
>>>  rr:subjectMap [ rr:column "FirstName" ];
>>>  rr:predicateObjectMap
>>>     [
>>>       rr:predicate foaf:name;
>>>       rr:objectMap    [ rr:column "FirstName" ]
>>>     ]
>>>
>>> For example, it will not fail for the following data (with uncommon
>>> first names) in the EMPLOYEES table:
>>>
>>> FirstName
>>> -----------
>>> http://example.com/ns#John
>>> http://example.com/ns#Mary
>>>
>>> Thanks,
>>> - Souri.
>>>
>>> ----- Original Message -----
>>> From: bvillazon@fi.upm.es
>>> To: juanfederico@gmail.com
>>> Cc: richard@cyganiak.de, public-rdb2rdf-wg@w3.org
>>> Sent: Thursday, February 9, 2012 9:17:46 PM GMT -05:00 US/Canada Eastern
>>> Subject: Re: negative test case question
>>>
>>> Hi Juan
>>>
>>> Thanks for this
>>>
>>> On Feb 10, 2012, at 1:56 AM, Juan Sequeda wrote:
>>>
>>> Let me clarify
>>>
>>> The following mapping is correct:
>>>
>>> <TriplesMap1>
>>>     a rr:TriplesMap;
>>>     rr:logicalTable [ rr:tableName "Employees" ];
>>>  rr:subjectMap [ rr:column "FirstName"; rr:termType rr:BlankNode ];
>>> rr:predicateObjectMap
>>>     [
>>>       rr:predicate foaf:name;
>>>       rr:objectMap    [ rr:column "FirstName" ]
>>>     ]
>>>     .
>>>
>>> The following mapping should fail:
>>>
>>> <TriplesMap1>
>>>     a rr:TriplesMap;
>>>     rr:logicalTable [ rr:tableName "Employees" ];
>>> rr:subjectMap [ rr:column "FirstName" ];
>>>  rr:predicateObjectMap
>>>     [
>>>       rr:predicate foaf:name;
>>>       rr:objectMap    [ rr:column "FirstName" ]
>>>     ]
>>>     .
>>>
>>> right?
>>>
>>>
>>> According to the spec
>>> <<
>>>
>>> If the term map does not have a rr:termType property, then its term type<http://www.w3.org/2001/sw/rdb2rdf/r2rml/#dfn-term-type>
>>>  is:
>>>
>>>    - rr:Literal, if it is an object map<http://www.w3.org/2001/sw/rdb2rdf/r2rml/#dfn-object-map> and
>>>    at least one of the following conditions is true:
>>>       - It is a column-based term map<http://www.w3.org/2001/sw/rdb2rdf/r2rml/#dfn-column-valued-term-map>
>>>       .
>>>       - It has a rr:language property (and thus a specified language tag<http://www.w3.org/2001/sw/rdb2rdf/r2rml/#dfn-specified-language-tag>
>>>       ).
>>>       - It has a rr:datatype property (and thus a specified datatype<http://www.w3.org/2001/sw/rdb2rdf/r2rml/#dfn-specified-datatype>
>>>       ).
>>>    - rr:IRI, otherwise.
>>>
>>> >>
>>> So, if the FirstName column has "normal" first names … I think yes, it
>>> would fail
>>>
>>> Richard?
>>>
>>> Boris
>>>
>>>
>>>
>>>
>>> Juan Sequeda
>>> +1-575-SEQ-UEDA
>>> www.juansequeda.com
>>>
>>>
>>> On Thu, Feb 9, 2012 at 6:33 PM, Richard Cyganiak <richard@cyganiak.de>wrote:
>>>
>>>> Juan,
>>>>
>>>> On 9 Feb 2012, at 23:51, Juan Sequeda wrote:
>>>> > I'm coming up with r2rml mappings that should fail.
>>>> >
>>>> > The following is correct:
>>>> >
>>>> > <TriplesMap1>
>>>> >     a rr:TriplesMap;
>>>> >     rr:logicalTable [ rr:tableName "Employees" ];
>>>> >       rr:subjectMap [ rr:column "FirstName"; rr:termType rr:BlankNode
>>>> ];
>>>> >       rr:predicateObjectMap
>>>> >     [
>>>> >       rr:predicate foaf:name;
>>>> >       rr:objectMap    [ rr:column "FirstName" ]
>>>> >     ]
>>>> >     .
>>>> >
>>>> >
>>>> > If I take the rr:TermType rr:BlankNode, that should fail, right?
>>>>
>>>> I'm not sure what you mean. Do you mean, if you add “rr:termType
>>>> rr:BlankNode” to the object map, then it should fail?
>>>>
>>>> Why do you think so?
>>>>
>>>> Richard
>>>
>>>
>>>
>>>
>>
>