[postgis-devel] st_dwithin, st_within, st_dcompletlywithin

Wed Oct 8 10:47:47 PDT 2008

I'ts a little boy. His name is Birger, an old nordic name so my working name on the patch is Birgers Boost :-)

About the patch:
I can see more possible functions that might be interesting.

A function that gives the answer to for example:
"How far away from France can you get in Belgia"
To answer that we have to store compare the shortest distance to geometry2 from each vertex in geometry1 and finaly find wich vertex in geometry1 that has the highest "mindistance" to geometry2.
Hope the idee is possible to understand. Is it of any use?

The thing is that those distance functions all do the same iterations point for point, so it would be very nice to have the possibility to get more information out of doing it once. There is no problem in storing the information for all the functions st_distance, st:max_distance, st_shortestline, st_longestline and eventually this new function. The extra time for storing the information seems to be very small compared to the effort to do the iteration. The problem is just to get it out. I will try to start this discussion in a new post, he discussion we started earlier in springtime about calculating the same thing many times. 

/Nicklas

2009-07-07 Paragon Corporation  wrote:

>Nicklas,> >Congratulations on the new baby!  Is it a boy orgirl?> >I'm okay with you just adding this to the current ticket sinceI agree they are all related.  Probably best to wait for Paul, Kevin, orMark to voice any issues/complaints, but I can't think ofany.> >Thanks,>Regina
>>From:postgis-devel-bounces at postgis.refractions.net[mailto:postgis-devel-bounces at postgis.refractions.net] On Behalf Ofnicklas.aven at jordogskog.no
>Sent: Tuesday, July 07, 2009 8:16AM
>To: PostGIS Development Discussion
>Subject: Re:[postgis-devel] st_dwithin, st_within, st_dcompletlywithin
>
>>
 I think ST_DFullyWithin sounds great.
>Is it ok if I put this to inticket #137. In the code they are all close related, st_distance,st_max_distace, st_shortestline, st_longestline, st_dwithin,st_dfullywithin
>
>I have a patch for the trunk that seems to work, but Ihaven't tested more than randomly. I will try to build cunit-tests but I thinkit is difficult to get the make check thing work at all. In windows I get somproblems and in ubuntu I get permission-problems. I'm new to Linux.
>
>Ialso just have become a father :-) two weeks ago so my sparetime for learningnew things has drastically decreased. 
>
>/Nicklas
>
>
>2009-07-07Paragon Corporation wrote:
>
>Paul and Nicklas,
>>
>>I think wehad suggested calling it ST_DFullyWithin or maybe that wasmy
>>idea.
>>
>>http://trac.osgeo.org/postgis/ticket/20
>>
>>ThoughI guess we closed this out when we put the short circuit infor
>>ST_DWithin. We probably should resurrect or add it as a separateticket.
>>Nicklas since you are working on it, you want to add it as a tracitem.
>>
>>Thanks,
>>Regina
>>
>>
>>
>>-----Original Message-----
>>From:postgis-devel-bounces at postgis.refractions.net
>>[mailto:postgis-devel-bounces at postgis.refractions.net]On Behalf Of Paul
>>Ramsey
>>Sent: Monday, July 06, 2009 7:23PM
>>To: PostGIS Development Discussion
>>Subject: Re: [postgis-devel]st_dwithin, st_within, st_dcompletlywithin
>>
>>It gets even moreconfusing if you note that some members of the WFS working
>>group insistthat the semantics of the WithinDistance OGC filter op are the
>>semanticsof your proposed function, rather than the semantics ofdwithin.
>>
>>P
>>
>>On Mon, Jul 6, 2009 at 4:19 PM,wrote:
>>> Hallo
>>>
>>>This is a semantic question.
>>> I'm working with the new functionsst_shortestline etc.
>>> I found that we can get another new functionfor free when we have 
>>> implemented st_max_distance. By using themax-calculation instead of 
>>> min-calculation the same way asst_dwithin, we insead get 
>>> "st_dcompletlywithin" the functionreturns true if the maximum 
>>> distance between the geometries isshorter than the third input.
>>>
>>> My question is (beside thequestion if this is something useble), what 
>>> to call the function. Ithink it easily can be confusing. I mean, 
>>> st_within and st_dwithinis two quite similar names wich is ok because 
>>> they don't reallyrelate to eachother in what they do, but if we add a 
>>> function thatdo the same as st_dwithin but whith the difference that 
>>> thegeometries has to be completly inside the given distance of 
>>>eachother, maybe we risk making it confusing. I guess 
>>>st_dcompletlywithin is maybe the most describing, but verylong.
>>>
>>> Greetings
>>>Nicklas
>>>
>>>
>>>
>>>_______________________________________________
>>> postgis-develmailing list
>>> postgis-devel at postgis.refractions.net
>>>http://postgis.refractions.net/mailman/listinfo/postgis-devel
>>>
>>>
>>_______________________________________________
>>postgis-develmailinglist
>>postgis-devel at postgis.refractions.net
>>http://postgis.refractions.net/mailman/listinfo/postgis-devel
>>
>>
>>
>>_______________________________________________
>>postgis-develmailinglist
>>postgis-devel at postgis.refractions.net
>>http://postgis.refractions.net/mailman/listinfo/postgis-devel
>>
>>.AVEN at JORDOGSKOG.NO>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-devel/attachments/20081008/a459a814/attachment.html>