#mapnik log: Friday 05, March 2010

2010 | 03

previous | next
00:03:19 *** filbertkm (~chatzilla@95.sub-75-226-31.myvzw.com) has joined #mapnik
00:19:50 *** mperry_ (~mperry@c-67-164-175-65.hsd1.co.comcast.net) has joined #mapnik
00:19:51 *** mperry has quit (Read error: Connection reset by peer)
00:19:51 *** mperry_ is now known as mperry
00:28:14 *** tcarobruce (~tim@c-98-210-194-147.hsd1.ca.comcast.net) has joined #mapnik
00:29:36 *** StormTide has quit (Quit: Leaving)
00:57:11 *** filbertkm has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.8/20100214235838])
01:11:46 *** dkb has quit (Quit: Leaving.)
01:26:58 *** cgs_bob_ has quit (Ping timeout: 248 seconds)
02:05:47 *** cgs_bob_ (~bob@122.135-78-65.ftth.swbr.surewest.net) has joined #mapnik
02:07:56 *** dkb (~dkb@66-219-8-179.ip.gvtel.com) has joined #mapnik
02:11:03 *** dkb1 (~dkb@66-219-8-179.ip.gvtel.com) has joined #mapnik
02:13:12 *** tcarobruce has quit (Quit: tcarobruce)
02:14:09 *** dkb has quit (Ping timeout: 252 seconds)
02:14:10 *** dkb2 (~dkb@66-219-8-179.ip.gvtel.com) has joined #mapnik
02:16:16 *** dkb2 has quit (Client Quit)
02:17:06 *** dkb1 has quit (Ping timeout: 248 seconds)
02:23:21 *** filbertkm (~chatzilla@173-10-147-57-BusName-washingtonDC.hfc.comcastbusiness.net) has joined #mapnik
03:49:19 *** dkb (~dkb@66-219-8-179.ip.gvtel.com) has joined #mapnik
04:01:06 *** filbertkm has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.8/20100214235838])
04:05:08 *** springmeyer has quit (Ping timeout: 246 seconds)
04:39:12 *** mperry has quit (Quit: mperry)
04:42:31 *** cgs_bob (~bob@122.135-78-65.ftth.swbr.surewest.net) has joined #mapnik
04:43:04 *** cgs_bob_ has quit (Ping timeout: 265 seconds)
04:48:24 *** gavinf has quit (Ping timeout: 245 seconds)
04:58:32 *** filbertkm (~chatzilla@128.241.235.189) has joined #mapnik
05:13:04 *** dkb has quit (Remote host closed the connection)
05:56:27 *** HounD (~HounD@unics1.grfc.ru) has joined #mapnik
05:56:37 *** HounD has parted #mapnik (None)
06:17:52 *** springmeyer (~springmey@c-76-28-156-154.hsd1.wa.comcast.net) has joined #mapnik
06:45:53 *** gavinf (~gavinf@196.211.2.133) has joined #mapnik
07:42:57 <nikq> Mapnik Trac: Ticket #522 (overflow/image corruption in octree with png256 + large pixel output) created | http://trac.mapnik.org/ticket/522
07:43:38 <nikq> Mapnik Trac: Ticket #522 (overflow/image corruption in octree with png256 + large pixel output) updated | http://trac.mapnik.org/ticket/522#comment:1
08:23:28 *** harobed (~stephane@pda57-1-82-231-115-1.fbx.proxad.net) has joined #mapnik
08:45:13 *** kredik_ has quit (Remote host closed the connection)
09:09:36 *** filbertkm has quit (Quit: ChatZilla 0.9.86 [Firefox 3.5.8/20100214235838])
09:41:08 *** HounD (~HounD@unics1.grfc.ru) has joined #mapnik
11:14:34 *** gavinf has quit (Read error: Connection reset by peer)
12:37:25 *** cgs_bob_ (~bob@122.135-78-65.ftth.swbr.surewest.net) has joined #mapnik
12:38:28 *** cgs_bob has quit (Ping timeout: 256 seconds)
13:12:33 *** gavinf (~gavinf@196.211.2.133) has joined #mapnik
13:21:19 *** dkb (~dkb@66-219-8-179.ip.gvtel.com) has joined #mapnik
13:48:49 *** darth_bitterman has quit (Quit: Leaving)
14:14:36 *** dkb has quit (Quit: Leaving.)
14:21:30 *** mperry (~mperry@c-67-164-175-65.hsd1.co.comcast.net) has joined #mapnik
14:23:52 *** HounD has parted #mapnik (None)
15:02:22 *** mperry has quit (*.net *.split)
15:07:43 *** mperry (~mperry@c-67-164-175-65.hsd1.co.comcast.net) has joined #mapnik
15:21:06 *** luneff (~yury@93.178.86.17) has joined #mapnik
15:28:25 *** dkb (~dkb@66-219-8-179.ip.gvtel.com) has joined #mapnik
16:01:40 *** ajashton (~aj@c-98-218-226-144.hsd1.dc.comcast.net) has joined #mapnik
16:06:41 *** ajashton has quit (Remote host closed the connection)
16:07:04 *** ajashton (~aj@c-98-218-226-144.hsd1.dc.comcast.net) has joined #mapnik
16:54:39 *** luneff has quit (Ping timeout: 245 seconds)
16:58:11 *** luneff (~yury@233.37.117.87.donpac.ru) has joined #mapnik
17:11:48 *** bitterman (~yury@233.37.117.87.donpac.ru) has joined #mapnik
17:15:06 *** luneff has quit (Ping timeout: 258 seconds)
17:20:44 *** bitterman is now known as luneff
17:39:50 *** tcarobruce (~tcarobruc@adsl-75-10-247-30.dsl.pltn13.sbcglobal.net) has joined #mapnik
17:42:31 *** ovnicraft (~ovnicraft@190.154.247.83) has joined #mapnik
17:50:49 *** harobed has quit (Quit: Ex-Chat)
18:36:53 *** dkb has parted #mapnik (None)
19:00:56 *** shoe has quit (Read error: Connection reset by peer)
19:21:31 *** luneff has quit (Ping timeout: 268 seconds)
19:21:40 *** luneff (~yury@93.178.95.253) has joined #mapnik
19:25:01 <timlinux> hi springmeyer
19:40:28 <racicot> hey timlinux
19:40:35 * racicot notes he is not springmeyer
19:41:42 *** bitterman (~yury@93.178.95.253) has joined #mapnik
19:41:48 <timlinux> racicot: I see you are not springmeyer :-)
19:41:52 <timlinux> racicot: and Hi!
19:42:42 <racicot> love to see what you are doing at linfiniti ... inspiring work
19:43:49 <timlinux> racicot: it will be if we can produce some fossgis geeks
19:44:07 <timlinux> its going to be a lot of work since we are starting from first principles
19:44:54 <racicot> you are up to the task.  let us (the community) know how we can help.  I know many are interested in what you are doing and watching closely
19:45:03 <timlinux> but I have high hopes for them
19:45:36 <timlinux> well most useful help is to be patient when poorly formulated questions are asked on irc :-)
19:45:39 *** luneff has quit (Ping timeout: 240 seconds)
19:46:11 <timlinux> racicot: how mapnik guru are you?
19:46:15 * racicot has a high IRC tolerance and welcomes questions
19:46:46 <racicot> I am not... I just try to learn by watching Dane... but I am a novice and quite uneducated on Mapnik to be honest
19:47:02 <timlinux> I had quite a few offers from people to come and do mentoring
19:47:42 <timlinux> ok I will wait for the chief geek :-)
19:48:42 * springmeyer ha! summoned timlinux 
19:51:56 <timlinux> hey springmeyer
19:52:11 <timlinux> finished your phone call?
19:52:15 <timlinux> :-)
19:52:16 <springmeyer> heh :)
19:52:23 <springmeyer> ya, no kidding
19:52:27 <timlinux> (the one from yesterday I mean :-)
19:52:33 <springmeyer> i know, lol
19:52:51 <timlinux> springmeyer: can you bear a couple of dumb questions?
19:52:53 <springmeyer> crazyness, back just back from train station drop off
19:52:59 <springmeyer> timlinux: go for it
19:53:32 <timlinux> so I'm preparing to embark on my next step of mapnik nirvana: use it for publishing maps instead of just via quantumnik
19:54:18 <timlinux> I googled around and found an old perrygeo blog post from 2006 saying there was work in progress to support getFeatureInfo requests
19:54:34 <timlinux> is that now supported with current releases
19:55:07 <timlinux> I want to do a follow up article on publishing a map using mapnik
19:55:13 <springmeyer> "publish" meaning print maps or WMS server?
19:55:22 <timlinux> wms server sorry
19:55:26 <springmeyer> ah, okay
19:55:28 <timlinux> though I am interested in print too
19:56:02 <timlinux> to publish via wms does mapnik have a build in server ?
19:56:09 <springmeyer> yes, so WMS & Mapnik...
19:56:29 <timlinux> I remember our chat some months ago you mentioned the usual way its done is via tile cache
19:56:35 <timlinux> or an apache  mod
19:56:41 <springmeyer> there is a WMS server written in python
19:56:48 <springmeyer> which supports GetFeatureInfo
19:56:54 <timlinux> ok
19:57:03 <springmeyer> in fact I just hacked in a fix last week so that it would work with QGIS
19:57:21 <timlinux> with qgis as a wms client you mean?
19:57:31 <springmeyer> because QGIS does not send i=foo&j=foo but rather X=foo&Y=foo
19:57:32 <springmeyer> yes
19:57:58 <timlinux> is the python wms server part & parcel of the python-mapnik package from apt?
19:58:30 <springmeyer> it has been part of Mapnik (as a python submodule), but the near-to-happen plan is to move it out of the Mapnik SVN
19:58:30 <timlinux> I also wanted to ask is there any migration path from a mapserver map file to a mapnik xml file?
19:58:46 <springmeyer> timlinux: nothing formal, but I could share scripts, yes
19:59:01 <springmeyer> er, wait, nm
19:59:01 *** filbertkm (~chatzilla@178.sub-75-197-133.myvzw.com) has joined #mapnik
19:59:04 * springmeyer blinks
19:59:12 <timlinux> :-)
19:59:15 <springmeyer> I only have scripts to do SLD -> mapnik
19:59:30 <timlinux> oh thats cool too
19:59:39 <timlinux> we still need to get qgis to produce sld
19:59:41 <springmeyer> not mapfile, maybe worth while writing such a thing when Mapserver finishes XML mapfile support
19:59:49 <springmeyer> timlinux: yes :)
20:00:06 <timlinux> and to read sld for that matter
20:00:16 <springmeyer> though its tricky of course because SLD is just the "styles" and not a full blown map definition
20:00:37 <timlinux> going back to publising for a minute
20:00:49 <springmeyer> anyway, big picture, good time to ask about WMS server because in last month it has just gotten attention
20:01:04 <springmeyer> but previously it has been fairly uncommonly used and unmaintained
20:01:09 <timlinux> what is your preferred way of publishing ? via tilecache? via that apache mod? via that python wms server you mentioned?
20:01:44 <springmeyer> mod_tile + OpenLayer XYZ layer
20:02:15 <springmeyer> but that of course is quite limiting for the broader reasons you'd want WMS
20:03:12 <springmeyer> so I do use the python WMS server
20:04:06 <timlinux> so I was thinking to use a tilecache (either mod_tile or tilecache) and then use mapserver to handle getfeature info requests for the same layers
20:05:00 <springmeyer> seems that if you were using MapServer for GetFeature I would just use Mapserver to render too
20:05:38 <springmeyer> honestly my most frequent combo for small projects in the past has been
20:06:00 <timlinux> yeah in the project we are planning to implement both mapnik and mapserver so the client can pick between em
20:06:01 <springmeyer> mapnik (rendering) + GeoDjango (for anything relating to querying info)
20:06:13 <springmeyer> timlinux: ah okay
20:06:19 <timlinux> that was another option  I had in mind
20:06:55 <springmeyer> well then likely best (if you want to compare the two - without cross switches to much)...
20:07:01 <timlinux> since I could return richer data iwth django that I can using wms and getfeature info
20:07:12 *** ovnicraft has quit (Ping timeout: 252 seconds)
20:07:40 <springmeyer> yes, I find getfeatureinfo only usefull when its automatically there, but ultimately replace with something else
20:08:41 <springmeyer> anyway, timlinux - main limitation of Mapnik's WMS server (called the 'ogcserver')
20:08:46 <timlinux> so you would use openlayers xyz + apache mod_tile + ol + django?
20:08:56 <springmeyer> is that it's setup...
20:09:06 <springmeyer> yes that would be a bomber setup for many purposes
20:09:19 <timlinux> and for print production?
20:09:26 <springmeyer> limitations are setup is not terribly easy and lacking GetLegendGraphic
20:10:09 <timlinux> ok but getlegendgraphic one could hand do for static maps
20:10:22 <springmeyer> yes
20:10:30 <springmeyer> timlinux: one more thing before chatting print...
20:10:36 <timlinux> sure
20:10:57 <springmeyer> this is public but yet to be advertised new home for the ogcserver:
20:10:59 <springmeyer> http://bitbucket.org/springmeyer/ogcserver
20:11:24 <springmeyer> mostly "yet to be advertised" because the location could change
20:12:06 <timlinux> is that teh python wms server you are hiving off from mapnik?
20:12:12 <springmeyer> yes
20:12:21 <timlinux> ok
20:12:41 <springmeyer> that is this: http://trac.mapnik.org/browser/trunk/bindings/python/mapnik/ogcserver
20:13:18 <springmeyer> but with lots of changes to actually work (mostly GetCaps and GetFeatureInfo) with clients: QGIS, ArcMap 9.2/9.3, uDig
20:13:53 <springmeyer> out of the box the ogcserver has always worked fine with Openlayers (WMS 1.1.1)
20:14:40 <springmeyer> notice the django mention. potentially GeoDjango could be primary interface in the future
20:15:19 <springmeyer> so you if you set up a django/geodjango app, and if mapnik is installed, then you've got WMS/TMS/ etc/etc for free
20:15:31 <timlinux> nice
20:15:36 * springmeyer done dumping brain :)
20:16:02 <timlinux> springmeyer: you can implement it as a django app you mean?
20:16:16 <timlinux> where getFeatureInfo etc are just django views?
20:17:43 <springmeyer> well, yes you could do anything. likely you'd re-write the ogcserver code that does the getFeatureInfo handling as a pretty django view, but still dispatch to Mapnik
20:18:05 <springmeyer> so your django app would be OGC compliant
20:19:03 <springmeyer> but I'd likely end up using Geodjango, so that I can spit back fancy json objects (that go much beyond what is in a GetFeatureInfo response)
20:19:08 * timlinux has been pondering this concept but kinda waiting on the sidelines for someone with enough geek cred to tamake something the rest of us can enjoy :-P
20:19:38 <timlinux> Yes I mentally insert Geo in the front whenever I say 'Django' :-)
20:20:03 <springmeyer> :)
20:20:13 *** Ldp__ (~thid@osm.xs4all.nl) has joined #mapnik
20:21:24 <timlinux> springmeyer: ok but since I am a small steps kinda bloke, I think I will try to tackle mapnik + mod_tile + ol xyz to start
20:21:35 <timlinux> or tilecache
20:21:54 <springmeyer> :) sounds good
20:22:30 <springmeyer> feel free to ask questions here. mod_tile can take some elbow grease to set up but it is a brilliant server
20:23:04 <timlinux> thanks
20:23:31 <timlinux> I usually dont mind mucking about with apache too much
20:24:08 * timlinux goes to scrounge on your mapnik page for pertinnet info
20:24:13 <timlinux> and pertinent even
20:26:02 <springmeyer> mishok13: around? is it still the case that a style created in the cloudmade style editor cannot be exported as Mapnik XML?
20:33:24 *** ajashton has quit (Ping timeout: 245 seconds)
20:33:44 *** ajashton (~aj@c-98-218-226-144.hsd1.dc.comcast.net) has joined #mapnik
20:35:41 *** ajashton has quit (Remote host closed the connection)
21:14:12 <nikq> Mapnik Trac: Ticket #522 (overflow/image corruption in octree with png256 + large pixel output) updated | http://trac.mapnik.org/ticket/522#comment:2
21:38:06 <Ldp__> springmeyer: mapnik does do reprojection of shapefiles, right?
21:38:26 <springmeyer> Ldp__: yes
21:39:11 <Ldp__> I've got a report with the 900913 coastline shapefiles, a 4326 osm2pgsql db, and a epsg:28992 <Map> srs dropping the coastlines
21:39:30 <Ldp__> (dutch) http://lists.openstreetmap.org/pipermail/talk-nl/2010-March/010756.html
21:41:27 <springmeyer> hmm, okay likely transforming 900913 into 28992 is failing
21:41:42 <Ldp__> 28992 is double stereographic
21:41:56 <Ldp__> but 4326 -> 28992 seems to work (the road network etc is there)
21:45:12 <springmeyer> just tried here:
21:45:20 <springmeyer> nik2img.py osm.xml -b 2.307 50.134 8.752 54.087 t.png -s EPSG:28992
21:45:25 <springmeyer> gives fully blank map
21:45:32 <springmeyer> (using 900913 osm db)
21:46:09 <springmeyer> I think that the processed_p will need to be clipped to be able to work in 28992
21:46:34 * springmeyer goes to use ogr2ogr to re-reprojected processed_p to 28992, expects to see failures....
21:56:34 *** ovnicraft (~ovnicraft@190.154.247.83) has joined #mapnik
22:08:02 *** filbertkm has quit (Ping timeout: 246 seconds)
22:09:45 <springmeyer> yep, heaps of errors emitted from ogr2ogr trying to transform processes_p from epsg 900913 to 28992
22:10:05 <springmeyer> so I'd guess that this is #308
22:10:05 <nikq> Ticket #308: Doesn't display feature if another (possibly invalid) feature is present, http://trac.mapnik.org/ticket/308
22:10:22 <springmeyer> which is a feature or a bug depending on your perspective
22:10:58 <Ldp__> ack
22:13:24 <springmeyer> in short proj4 bails if you can't transform coordinates
22:13:38 <springmeyer> and mapnik skips features after that fail
22:13:59 <springmeyer> processes_p is not really in good enough shape to re-project
22:14:20 <springmeyer> Ldp__: here's what I would recommend:
22:14:49 <springmeyer> write talk-nl fellow to try installing GDAL/OGR 1.7.1
22:15:01 <springmeyer> then re-project shapefile ahead of time
22:15:05 <Ldp__> looking at his resume, he'll definately have those already :)
22:15:25 <springmeyer> and change the srs in osm.xml and try rendering that way
22:15:34 <springmeyer> likely should be done anyway for all shapefiles for speed
22:15:39 <Ldp__> yup, ok
22:15:41 <springmeyer> ogr2ogr -t_srs EPSG:28992 -clipdst 12628.0541, 308179.0423, 283594.4779, 611063.1429 p_clip_amersfoort.shp processed_p.shp
22:16:08 <springmeyer> I got that bounding box from here:
22:16:09 <springmeyer> http://spatialreference.org/ref/epsg/28992/
22:16:14 <springmeyer> it may not be correct
22:16:21 <springmeyer> I would recommend getting bounds from
22:16:54 <springmeyer> Select ST_Extent(ST_Transform(way,28992)) from planet_osm_line;
22:18:14 <Ldp__> the issue here is that our planet-benelux covers more area than is normally guaranteed by 28992, but he'll figure that out
22:27:13 <Ldp__> done
22:30:08 <springmeyer> ah, okay, thx
22:32:44 *** filbertkm (~chatzilla@220.sub-75-196-1.myvzw.com) has joined #mapnik
23:48:18 *** springmeyer has quit (Read error: Connection reset by peer)
23:48:35 *** springmeyer (~springmey@c-76-28-156-154.hsd1.wa.comcast.net) has joined #mapnik
23:51:40 *** filbertkm has quit (Remote host closed the connection)