Hi Craig, 

Out of top of my head is 3D bounding box (range sub-setting) and slicing 
queries.
I write you if I needed more cases. 

So, currently, there is no way to support a 3D bbox query or slice in neo4j 
spatial except 
directly checking the axes values, right?

Thanks,
Alireza

On Thursday, September 18, 2014 4:31:25 PM UTC+2, Craig Taverner wrote:
>
> Hi Alireza,
>
> A few years ago we started refactoring the RTree to support n-dimensional 
> objects, but this work was not completed. So currently only 2D is 
> supported. Can you tell me more about your use case? I would like to know 
> about cases for 3D so we can think again about future support.
>
> Regards, Craig
>
> On Thu, Sep 18, 2014 at 1:29 PM, Alireza Rezaei Mahdiraji <
> [email protected] <javascript:>> wrote:
>
>>
>> Hi Craig, 
>>
>> Are we limited to only 2D bbox for querying? In general, are any 3D 
>> queries are supported in Neo4j spatial?
>>
>> Best,
>> Alireza
>>
>> On Friday, September 12, 2014 9:58:51 AM UTC+2, Craig Taverner wrote:
>>>
>>> That's your call. The subject has changed a lot, so you could write a 
>>> new one, but if you want all this discussion together, keep this one.
>>>
>>> On Thu, Sep 11, 2014 at 4:01 PM, Alireza Rezaei M <[email protected]> 
>>> wrote:
>>>
>>>> Thanks a lot Craig, 
>>>> I will look at it and perhaps come back with some questions. In that 
>>>> case, should I write in this post
>>>> or create a new one?
>>>>
>>>> Alireza 
>>>>
>>>> On Thu, Sep 11, 2014 at 3:55 PM, Craig Taverner <[email protected]> 
>>>> wrote:
>>>>
>>>>> Yes, look at the ones included in the library. See one I made as a 
>>>>> sample years ago that can encode a LineString as a chain of connected 
>>>>> nodes: https://github.com/neo4j-contrib/spatial/blob/
>>>>> master/src/main/java/org/neo4j/gis/spatial/encoders/
>>>>> SimpleGraphEncoder.java
>>>>>
>>>>> You could copy this and extent it to handle Point, LineString and 
>>>>> Polygon. The OSMGeometryEncoder is one that can handle those types, but 
>>>>> it 
>>>>> uses a more complex graph model you perhaps don't want to be burdened by. 
>>>>> I 
>>>>> think the graph model I suggested in my earlier mail, and the sample code 
>>>>> linked to here, are more appropriate for you.
>>>>>
>>>>> On Thu, Sep 11, 2014 at 3:50 PM, Alireza Rezaei M <[email protected]> 
>>>>> wrote:
>>>>>
>>>>>> Any example on how to build such customized GeometryEncoder? 
>>>>>>
>>>>>> Alireza
>>>>>>
>>>>>> On Thu, Sep 11, 2014 at 3:07 PM, Craig Taverner <[email protected]> 
>>>>>> wrote:
>>>>>>
>>>>>>>
>>>>>>> I want to build a graph out of the result of this query, is there 
>>>>>>>> any such 
>>>>>>>> solution already in Neo4j, i.e., to recieve a query result as yet 
>>>>>>>> another graph not 
>>>>>>>> just a list of node? 
>>>>>>>>
>>>>>>>
>>>>>>> This sounds very domain specific. What kind of graph do you want? I 
>>>>>>> would assume you would need to build whatever graph you want yourself.
>>>>>>>
>>>>>>> However, if the graph you are talking about is the graph structure 
>>>>>>> of the geometries themselves, then I recommend my previous suggestion, 
>>>>>>> with 
>>>>>>> a custom GeometryEncoder that allows the LineString and Polygon 
>>>>>>> geometries 
>>>>>>> to already be graphs. In that case your data model for a three point 
>>>>>>> Polygon could be 
>>>>>>> (p)-[:points]->(a)-[:first]->(b)->[:next]->(c)-[:last]->(a), 
>>>>>>> where (a), (b) and (c) could all be Point Geometries in your index, and 
>>>>>>> (p) 
>>>>>>> would be the polygon containing those three points. The bbox and gtype 
>>>>>>> for 
>>>>>>> each point is stored on the a,b,c nodes, and the bbox and gtype of the 
>>>>>>> entire polygon is stored on the (p) node, and your GeometryEncoder 
>>>>>>> knows 
>>>>>>> how to convert from graphs to Geometry objects and back.
>>>>>>>
>>>>>>> This way you control entirely the data model, while still being able 
>>>>>>> to use the spatial library for geometry searches.
>>>>>>>
>>>>>>> -- 
>>>>>>> You received this message because you are subscribed to a topic in 
>>>>>>> the Google Groups "Neo4j" group.
>>>>>>> To unsubscribe from this topic, visit https://groups.google.com/d/
>>>>>>> topic/neo4j/svcOw_S0l1A/unsubscribe.
>>>>>>> To unsubscribe from this group and all its topics, send an email to 
>>>>>>> [email protected].
>>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> -- 
>>>>>> Best Regards
>>>>>>
>>>>>> Alireza Rezaei Mahdiraji
>>>>>>
>>>>>>  -- 
>>>>>> You received this message because you are subscribed to the Google 
>>>>>> Groups "Neo4j" group.
>>>>>> To unsubscribe from this group and stop receiving emails from it, 
>>>>>> send an email to [email protected].
>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>>
>>>>>
>>>>>  -- 
>>>>> You received this message because you are subscribed to a topic in the 
>>>>> Google Groups "Neo4j" group.
>>>>> To unsubscribe from this topic, visit https://groups.google.com/d/
>>>>> topic/neo4j/svcOw_S0l1A/unsubscribe.
>>>>> To unsubscribe from this group and all its topics, send an email to 
>>>>> [email protected].
>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>
>>>>
>>>>
>>>>
>>>> -- 
>>>> Best Regards
>>>>
>>>> Alireza Rezaei Mahdiraji
>>>>
>>>>  -- 
>>>> You received this message because you are subscribed to the Google 
>>>> Groups "Neo4j" group.
>>>> To unsubscribe from this group and stop receiving emails from it, send 
>>>> an email to [email protected].
>>>> For more options, visit https://groups.google.com/d/optout.
>>>>
>>>
>>>  -- 
>> You received this message because you are subscribed to the Google Groups 
>> "Neo4j" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to [email protected] <javascript:>.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Neo4j" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to