I started looking at it this AM.  I think the proper solution is to accept
"*" as a var in the selecthandler.  I will look at it over the weekend and
will add documentation (javadoc at least)



On Fri, Sep 19, 2014 at 10:14 AM, Andy Seaborne <a...@apache.org> wrote:

> On 19/09/14 05:18, Bruno P. Kinoshita wrote:
>
>> Hello Claude,
>>
>> I didn't understand what QueryBuilder was supposed to do at first, or how
>> to use it. Luckily there are tests in the project, kudos for writing those,
>> very helpful. I liked the idea, and for users familiar with Java Jooq, PHP
>> and Ruby ActiveRecord that's definitely an intuitive API.
>>
>> In order to test it, I first created some dummy data.
>>
>> PREFIX foaf:  <http://xmlns.com/foaf/0.1/>
>> PREFIX :  <http://example.org/>
>>
>> INSERT DATA
>> {
>> :bruno foaf:name "Bruno" .
>> :jorge foaf:name "Jorge" .
>> :bruno :brotherOf :jorge
>> }
>>
>> Then I retrieved the data with a simple SELECT.
>>
>> PREFIX foaf:  <http://xmlns.com/foaf/0.1/>
>> PREFIX :  <http://example.org/>
>>
>> SELECT *
>> {
>> ?a ?b ?c
>> }
>>
>> ---------------------------------
>> | a      | b          | c       |
>> =================================
>> | :bruno | foaf:name  | "Bruno" |
>> | :bruno | :brotherOf | :jorge  |
>> | :jorge | foaf:name  | "Jorge" |
>> ---------------------------------
>>
>> I tried to recreate the same query with QueryBuilder, but alas it didn't
>> work. I have probably made something really stupid [1]. My output is always:
>>
>> PREFIX  :     <http://example.org/>
>> PREFIX  foaf: <http://xmlns.com/foaf/0.1/>
>>
>> WHERE
>>    { ?a  ?b  ?c . }
>>
>> I just wanted to validate that I could recreate the same query with
>> QueryBuilder. But I couldn't figure how that works. Looking at
>> WhereClauseTest, I thought that addVar("*") would result in "SELECT *". The
>> prefixes is not keeping the order it was added, thus that's probably not an
>> issue.
>>
>> The SelectBuilder throws ParseException, which extends Exception. I think
>> **if** the Query Builder is supposed to be used by programmers for writing
>> their Web applications or similar, then perhaps it could extend a
>> RuntimeException?
>>
>> Just my 0.002 cents
>>
>> Thanks
>> Bruno
>>
>> [1] https://gist.github.com/kinow/b50a5d3b875f2155b7bb
>>
>
> Ditto.
>
> (I have just pushed a fix to Wherehandler.testTriple.)
>
>         Andy
>
>
>


-- 
I like: Like Like - The likeliest place on the web
<http://like-like.xenei.com>
LinkedIn: http://www.linkedin.com/in/claudewarren

Reply via email to