Re: [OSM-dev] multipolygons/areas - consistency vs scalability vs topology

2011-04-14 Thread Frederik Ramm
Hi, On 04/15/2011 04:16 AM, Ben Supnik wrote: - Is there any chance of having an area primitive so scalable that it scales all the way up to the major continents? Right now continents are sort of an exception - you need to go pull down the pre-processed coastline shape files because the data pro

[OSM-dev] multipolygons/areas - consistency vs scalability vs topology

2011-04-14 Thread Ben Supnik
Hi Y'all, Josef's comments on relations reminded me that I was going to rant^H^H^H^H opine on areas/multipolygons. In particular, this got my attention, from the comments on Josef's diary: "Basically multipolygons-in-relations are a horrible hack. Hopefully API 0.7 will move them into an ar

Re: [OSM-dev] A new "object" element?

2011-04-14 Thread Ben Supnik
Hi Josef, WRT both your comment about a real 'object' construct and a 'dependency' construct: If polygons/multipolygons/areas were implemented as their own 'first class' construct in the OSM data model (and not as a relation of ways), would this change your view on these things? IMHO (wort

Re: [OSM-dev] libosmpbf

2011-04-14 Thread Peter Körner
Am 14.04.2011 16:31, schrieb Jochen Topf: Osmium is a C++ library and has always been. It should be possible to include it in any language you want. It is a header only library (like a lot of the boost libraries), so there are only header files, no libosmium.a or libosmium.so. Thats done so you

Re: [OSM-dev] libosmpbf

2011-04-14 Thread Jochen Topf
On Thu, Apr 14, 2011 at 03:58:20PM +0200, Peter Körner wrote: > Am 14.04.2011 14:32, schrieb Jochen Topf: > >On Thu, Apr 14, 2011 at 12:21:58PM +0200, Peter Körner wrote: > >>As far as i understood there's still some logic needed to assemble > >>osm objects from the pbf messages. > > > >Yes. Thats

Re: [OSM-dev] osm2pgsql slow on update import

2011-04-14 Thread Markus Wagner
Seconded. I have my database on a RAID1 pair of 5.4K RPM drives (which gives me a speed up in reads and data redundancy, but seems about the same as a single disk in terms of writes) and am using PostgreSQL 9.0. Before the fastupdate patch, one minute of diffs took between 2 and 5 minutes to app

Re: [OSM-dev] libosmpbf

2011-04-14 Thread Peter Körner
Am 14.04.2011 14:32, schrieb Jochen Topf: On Thu, Apr 14, 2011 at 12:21:58PM +0200, Peter Körner wrote: As far as i understood there's still some logic needed to assemble osm objects from the pbf messages. Yes. Thats for instance provided by Osmium. Osmium builds on the PBF stuff from Scott.

Re: [OSM-dev] libosmpbf

2011-04-14 Thread Jochen Topf
On Thu, Apr 14, 2011 at 03:03:00PM +0200, Oliver Tonnhofer wrote: > On 14.04.2011, at 12:10, Jochen Topf wrote: > > To make the PBF stuff from Scott easier to use from C++ I have added a > > Makefile > > and Debian config in my fork at https://github.com/joto/OSM-binary . This > > gives > > you a

Re: [OSM-dev] libosmpbf

2011-04-14 Thread Oliver Tonnhofer
Hi Jochen, On 14.04.2011, at 12:10, Jochen Topf wrote: > To make the PBF stuff from Scott easier to use from C++ I have added a > Makefile > and Debian config in my fork at https://github.com/joto/OSM-binary . This > gives > you a /usr/lib/libosmpbf.a and the include files in /usr/include/osmpbf

Re: [OSM-dev] libosmpbf

2011-04-14 Thread Jochen Topf
On Thu, Apr 14, 2011 at 12:21:58PM +0200, Peter Körner wrote: > Am 14.04.2011 12:10, schrieb Jochen Topf: > >To make the PBF stuff from Scott easier to use from C++ I have added a > >Makefile > >and Debian config in my fork at https://github.com/joto/OSM-binary . This > >gives > >you a /usr/lib/l

Re: [OSM-dev] libosmpbf

2011-04-14 Thread Peter Körner
Am 14.04.2011 12:10, schrieb Jochen Topf: To make the PBF stuff from Scott easier to use from C++ I have added a Makefile and Debian config in my fork at https://github.com/joto/OSM-binary . This gives you a /usr/lib/libosmpbf.a and the include files in /usr/include/osmpbf. I never really used

[OSM-dev] libosmpbf

2011-04-14 Thread Jochen Topf
Hi! To make the PBF stuff from Scott easier to use from C++ I have added a Makefile and Debian config in my fork at https://github.com/joto/OSM-binary . This gives you a /usr/lib/libosmpbf.a and the include files in /usr/include/osmpbf. If you use Debian/Ubuntu you can have it all in a libosmpbf-d

Re: [OSM-dev] relations and "dependancies"

2011-04-14 Thread Frederik Ramm
Hi, On 04/14/2011 10:12 AM, the.promena...@gmail.com wrote: I made another suggestion here - http://www.openstreetmap.org/user/ThePromenader/diary/13574 - that'll be all for this week ; ). Thanks for any input. I think you're mixing up the data model, the editing, the rendering, and the geoco

[OSM-dev] relations and "dependancies"

2011-04-14 Thread the.promena...@gmail.com
Hello, I made another suggestion here - http://www.openstreetmap.org/user/ThePromenader/diary/13574 - that'll be all for this week ; ). Thanks for any input. Cheers, Josef. ___ dev mailing list dev@openstreetmap.org http://lists.openstreetmap.org/lis

Re: [OSM-dev] A new "object" element?

2011-04-14 Thread Frederik Ramm
Hi, On 04/14/2011 08:52 AM, Frederik Ramm wrote: find all nodes in that area, then find all objects using these nodes, then check if any of them has a pharmacy tag, or the other way round, first find all objects world-wide that have a pharmacy tag, then find s/pharmacy/traffic light/ Bye Fred