Re: [postgis-users] shp2pgsql segment fault (trunk -r 8924 ) without .shp file

2012-01-26 Thread Paul Ramsey
svn update, problem should be fixed now P. On Thu, Jan 26, 2012 at 12:52 PM, Min Wang wrote: > HI >   I  built the trunk version 8924  on 64 bit machine. > >   The shp2pgsql seems gave me segment fault when there is no .shp file. but > if pass -n, then it is ok. > >   Here is what I did: > >  gdb

Re: [postgis-users] How to match lines between two similar graphs?

2012-01-26 Thread Charles Galpin
Yes, my mistake :) On Jan 26, 2012, at 3:51 PM, Nicolas Ribot wrote: > As far as I understand, shouldn't the self join in the query be > "table1 t1, table1 t2", in your example ? ___ postgis-users mailing list postgis-users@postgis.refractions.net http

[postgis-users] shp2pgsql segment fault (trunk -r 8924 ) without .shp file

2012-01-26 Thread Min Wang
HI I built the trunk version 8924 on 64 bit machine. The shp2pgsql seems gave me segment fault when there is no .shp file. but if pass -n, then it is ok. Here is what I did: gdb shp2pgsql (gdb) set args s 4269 -g the_geom -W latin1 ./tl_2010_11001_featnames.dbf (gdb) run Sta

Re: [postgis-users] How to match lines between two similar graphs?

2012-01-26 Thread Nicolas Ribot
Hi Charles, Yes indeed, finding the right segment to preserve direction and driving characteristics is not easy at all. Fortunately in my case, only statistical data on the routes were important, so some direction errors were not that bad. As far as I understand, shouldn't the self join in the qu

[postgis-users] shp2pgsql-gui and option -S (Generate simple instead of MULTI geometries) / Ticket #660?

2012-01-26 Thread Stefan Keller
Hi Recent PostGIS versions come with a nifty plugin for pgAdmin3 called "PostGIS Shapefile and DBF Loader" (shp2pgsql-gui). I see Revision 7713 in my case. A common stumbling block of shp2pgsql is that Shapefile LineStrings are imported as MulitLineStrings (I'd wish the default would be the other

Re: [postgis-users] How to match lines between two similar graphs?

2012-01-26 Thread Charles Galpin
Stephen, do you mind sharing what the source of the traffic disruption feed is? Mainly curious if it's one of the ones we use (email direct if needed). I had tried to do something similar a while ago but honestly did not make good progress (this was before I found st_HausdorffDistance and need

Re: [postgis-users] How to match lines between two similar graphs?

2012-01-26 Thread Stephen Woodbridge
On 1/26/2012 12:05 PM, Nicolas Ribot wrote: On 26 January 2012 17:16, Stephen Woodbridge wrote: On 1/26/2012 10:47 AM, Nicolas Ribot wrote: Hi All, I have an interesting problem I am trying to solve and would love some feedback on how to best go about it. I have road data from two different

Re: [postgis-users] How to match lines between two similar graphs?

2012-01-26 Thread Nicolas Ribot
On 26 January 2012 17:16, Stephen Woodbridge wrote: > On 1/26/2012 10:47 AM, Nicolas Ribot wrote: >>> >>> Hi All, >>> >>> I have an interesting problem I am trying to solve and would love some >>> feedback on how to best go about it. >>> >>> I have road data from two different vendor data sets. Bu

Re: [postgis-users] Bug in org.postgis.LineString concat() from Java API 1.3.3

2012-01-26 Thread Paul Ramsey
Thanks, I agree w/ your analysis, patched. On Thu, Jan 26, 2012 at 2:20 AM, Innerebner Markus wrote: > Hi all, > > for me this looks like an error: > > postgis.jdbc.LineString method concat (line70) > > WRONG: > int count = points.length + points.length - (cutPoint ? 1 : 0); > > CORRECT: > int co

Re: [postgis-users] How to match lines between two similar graphs?

2012-01-26 Thread Stephen Woodbridge
On 1/26/2012 10:47 AM, Nicolas Ribot wrote: Hi All, I have an interesting problem I am trying to solve and would love some feedback on how to best go about it. I have road data from two different vendor data sets. But this could also be one road network and a GPS track, so I think this is a pre

Re: [postgis-users] How to match lines between two similar graphs?

2012-01-26 Thread Nicolas Ribot
> Hi All, > > I have an interesting problem I am trying to solve and would love some > feedback on how to best go about it. > > I have road data from two different vendor data sets. But this could also be > one road network and a GPS track, so I think this is a pretty common use > case. > > Assumpt

Re: [postgis-users] How to match lines between two similar graphs?

2012-01-26 Thread Nicolas Ribot
> Hi All, > > I have an interesting problem I am trying to solve and would love some > feedback on how to best go about it. > > I have road data from two different vendor data sets. But this could also be > one road network and a GPS track, so I think this is a pretty common use > case. > > Assumpt

Re: [postgis-users] How to match lines between two similar graphs?

2012-01-26 Thread Matej Mailing
Hi, I am having somehow similar issue: I have a (single) table with the lines that represent cables on the ground. Those cables sometimes cross, sometimes they go in the same direction in some distance between them, sometimes one on another and my goal is to get the trenches (that are wide for exa

Re: [postgis-users] migration/upgrade advice

2012-01-26 Thread Paragon Corporation
Oops. My bad. I had a typo on the page. I'll fix. That should have read January 7, 2012. If you are using that one, then you should be good to go since I haven't released a new version yet. Just use the 9.1 binaries on that page and restore your backup and you should be good to go.

[postgis-users] Bug in org.postgis.LineString concat() from Java API 1.3.3

2012-01-26 Thread Innerebner Markus
Hi all, for me this looks like an error: postgis.jdbc.LineString method concat (line70) WRONG: int count = points.length + points.length - (cutPoint ? 1 : 0); CORRECT: int count = points.length + opoints.length - (cutPoint ? 1 : 0); can you check this please? thanks Markus I am using the ver

Re: [postgis-users] Recursive intersect

2012-01-26 Thread Nicolas Ribot
>> >> From: Nicolas Ribot >> >> Hi >> >> WITH RECURSIVE needs an UNION ALL to link the "non recursive" term >> with the "recursive" one. >> > > Wow, that at least runs... though I think the recursion might be infinite. > Thanks for the tip Nicolas. > Hi, Concerning the infinite iteration, one tr