zorba-coders team mailing list archive

[Matthias Brantner <matthias.brantner@xxxxxxxxxx>]

Review: Needs Information

> > I don't understand why. In my example it wouldn't. The count functions could
> > count
> > the number of empty objects or objects that contain only the STREET field.
> 
> I've looked into the test and here are the issues:
> 1) The fn:count() does not have the %explores-json annotation. I've confused
> it with the JSONiq array size() function.
> 
> 2) Currently the dataguide does not distinguish between any of the fn:
> functions. So for example if you modify the query:
> 
>     for $obj in parse-json(f:read-text(fn:resolve-uri("citylots-small.json")))
>     group by $s := $obj.STREET
>     return {
>       street : $s,
>       count : count($obj),
>       seq : subsequence($obj, 1, 100)
>     }
> 
> then if you prune the input to have only the .STREET field you'll get an
> incorrect result. So currently the argument to subsequence() is "explored" and
> so the dataguide is "*" in this case. The same happens with count().
> 
> 3) fn:count() is a special case as it only cares for the number of objects and
> not their contents. I think another annotation could be added, which would be
> the opposite of %explores-json and which would indicate that the function
> ignores the details of the object. Or I could just add some code in the
> dataguide collection to handle fn:count() separately.
The way I see this is not a problem with the fn functions. The problem is related to the fact how the objects are used. In this case, the objects are used to construct a new object. The dataguide needs to handle that. Similar to serialization, this means that the entire object is needed.

-- 
https://code.launchpad.net/~zorba-coders/zorba/use-dataguide/+merge/176385
Your team Zorba Coders is subscribed to branch lp:zorba.