00:40:33 *** drewby (n=dfaubion@128-8-138-156.umd.edu) has joined #mapnik
00:40:39 * drewby is excited.
00:40:44 * drewby is compiling.
01:02:12 *** springmeyer has quit ()
01:08:52 <drewby> alright
01:09:00 <drewby> I got the sharp line exploding problem fixed
01:09:11 <drewby> but there's still a lot of issues
01:09:16 <drewby> with the continuity of the lines
01:18:01 <nikq> Mapnik Trac: sharp_spike_fix.patch attached to Ticket #180 | http://trac.mapnik.org/attachment/ticket/180/sharp_spike_fix.patch
01:29:32 *** drewby has quit ("Leaving.")
01:41:14 *** DCREdD (n=blah@cpc2-colc4-0-0-cust771.colc.cable.ntl.com) has joined #mapnik
01:46:32 *** cgs_bob has quit (Read error: 110 (Connection timed out))
01:52:47 *** tcb has quit ()
02:02:18 *** DCREdD has quit (Read error: 110 (Connection timed out))
02:19:07 *** cgs_bob (n=bob@122.135-78-65.ftth.swbr.surewest.net) has joined #mapnik
02:53:07 *** springmeyer (n=springme@207.170.245.34) has joined #mapnik
03:10:37 *** springmeyer has quit (Read error: 110 (Connection timed out))
03:12:51 *** jctull (n=jctull@ppp-71-142-138-235.dsl.renocs.pacbell.net) has joined #mapnik
05:10:16 *** StormTide (n=Kevin@S010600131042729b.du.shawcable.net) has joined #mapnik
05:10:57 <StormTide> im getting a segfault trying quauntumnik, any usual suspects?
05:25:16 *** cgs_bob has quit (Read error: 104 (Connection reset by peer))
05:25:26 *** cgs_bob (n=bob@122.135-78-65.ftth.swbr.surewest.net) has joined #mapnik
06:06:50 *** HounD (n=HounD@unics1.grfc.ru) has joined #mapnik
06:13:50 *** jctull has quit ()
06:28:58 *** cgs_bob_ (n=bob@122.135-78-65.ftth.swbr.surewest.net) has joined #mapnik
06:29:08 *** cgs_bob has quit (Read error: 104 (Connection reset by peer))
06:29:21 *** cgs_bob_ is now known as cgs_bob
06:40:28 *** DCREdD (n=blah@cpc2-colc4-0-0-cust771.colc.cable.ntl.com) has joined #mapnik
06:54:36 *** racicot has quit (Read error: 104 (Connection reset by peer))
06:55:29 *** racicot (n=chatzill@dsl-209-166-85-189.whidbey.net) has joined #mapnik
08:15:42 *** DCREdD has quit (Read error: 110 (Connection timed out))
08:15:45 *** DCREdD (n=blah@cpc2-colc4-0-0-cust771.colc.cable.ntl.com) has joined #mapnik
08:59:29 *** mperry_ (n=mperry@c-67-164-175-65.hsd1.co.comcast.net) has joined #mapnik
08:59:29 *** mperry has quit (Read error: 104 (Connection reset by peer))
08:59:34 *** mperry_ is now known as mperry
10:14:07 *** mperry_ (n=mperry@c-67-164-175-65.hsd1.co.comcast.net) has joined #mapnik
10:14:07 *** mperry has quit (Read error: 104 (Connection reset by peer))
10:14:12 *** mperry_ is now known as mperry
10:47:24 *** mperry_ (n=mperry@c-67-164-175-65.hsd1.co.comcast.net) has joined #mapnik
10:47:24 *** mperry has quit (Read error: 104 (Connection reset by peer))
10:47:29 *** mperry_ is now known as mperry
11:00:36 *** D3f0 (n=D3f0@www.transpa-sa.com.ar) has joined #mapnik
11:08:14 *** Ldp__ (n=thid@osm.xs4all.nl) has joined #mapnik
11:08:14 *** Ldp__ has quit (Remote closed the connection)
11:15:44 *** mperry_ (n=mperry@c-67-164-175-65.hsd1.co.comcast.net) has joined #mapnik
11:15:44 *** mperry has quit (Read error: 104 (Connection reset by peer))
11:15:49 *** mperry_ is now known as mperry
11:20:15 *** Ldp__ (n=thid@osm.xs4all.nl) has joined #mapnik
11:40:57 *** Phurl is now known as PhurlIpv4
12:25:41 *** Genscher (n=Richard@dslb-092-075-121-151.pools.arcor-ip.net) has joined #mapnik
12:45:04 <nikq> Mapnik Trac: Ticket #503 (Add tolerance to the features_at_point() query in the PostGIS plugin) created | http://trac.mapnik.org/ticket/503
12:47:13 *** DCREdD has quit (Read error: 110 (Connection timed out))
13:09:38 *** DCREdD (n=blah@cpc2-colc4-0-0-cust771.colc.cable.ntl.com) has joined #mapnik
13:37:54 *** DCREdD has quit (Read error: 110 (Connection timed out))
14:15:47 *** mperry_ (n=mperry@c-67-164-175-65.hsd1.co.comcast.net) has joined #mapnik
14:15:47 *** mperry has quit (Read error: 104 (Connection reset by peer))
14:15:52 *** mperry_ is now known as mperry
14:48:07 *** HounD1 (n=HounD@unics1.grfc.ru) has joined #mapnik
15:03:36 *** HounD has quit (Read error: 113 (No route to host))
15:04:13 *** HounD1 has parted #mapnik ()
15:39:51 *** springmeyer (n=springme@c-76-28-156-154.hsd1.wa.comcast.net) has joined #mapnik
15:43:48 <springmeyer> StormTide: segfault?
15:47:09 *** drewby (n=dfaubion@128-8-138-156.umd.edu) has joined #mapnik
16:06:02 *** jctull (n=jctull@ppp-71-142-138-235.dsl.renocs.pacbell.net) has joined #mapnik
16:29:53 *** mak (n=mak@vobster.nepharia.org) has joined #mapnik
16:40:23 *** jctull has quit ()
17:00:40 *** D3f0 has quit ("Saliendo")
17:25:03 *** jctull (n=jctull@75.0.7.127) has joined #mapnik
17:31:42 *** cgs_bob has quit (Read error: 110 (Connection timed out))
17:48:33 *** cgs_bob (n=bob@194.sub-75-210-13.myvzw.com) has joined #mapnik
18:04:28 <StormTide> springmeyer, turns out mapnik was just the trigger for a prob in gdal
18:05:01 <StormTide> Program received signal SIGSEGV, Segmentation fault.
18:05:01 <StormTide> 0xb63a38c0 in S57Reader::ApplyObjectClassAttributes(DDFRecord*, OGRFeature*) ()
18:05:01 <StormTide> from /usr/lib/libgdal1.6.0.so.1
18:05:04 <StormTide> was the fault
18:08:44 *** tcb (n=tcarobru@adsl-75-10-247-30.dsl.pltn13.sbcglobal.net) has joined #mapnik
18:10:02 <springmeyer> ah, crazy
18:10:11 <springmeyer> so did you resolve it StormTide ?
18:10:16 <StormTide> not really
18:10:20 <springmeyer> ah
18:10:20 <StormTide> it occurs on mac too tho ;)
18:10:27 <StormTide> the workaround was to save to shp
18:10:33 <StormTide> then use the shp with mapnik
18:10:36 <springmeyer> how did you install mapnik/gdal/qgis ?
18:10:37 <StormTide> and everything was happy
18:10:50 <StormTide> on ubuntu via apt ... on the mac via the dmg's
18:10:59 <springmeyer> I've never seen this issue before
18:11:09 <springmeyer> can you post the mac error?
18:11:19 <StormTide> you can reproduce it... lemme point you at the right s57 file
18:11:19 <StormTide> sec
18:11:23 * springmeyer is on a call but will take a look in a bit
18:11:28 <springmeyer> k
18:11:43 <StormTide> <StormTyde> Program received signal EXC_BAD_ACCESS, Could not access memory.
18:11:43 <StormTide> <StormTyde> Reason: KERN_PROTECTION_FAILURE at address: 0x0000024f
18:11:43 <StormTide> <StormTyde> 0x0073cdc9 in S57Reader::ApplyObjectClassAttributes ()
18:11:50 <StormTide> was the mac issue
18:13:07 <StormTide> http://www.charts.noaa.gov/ENCs/Agreement.shtml?US5WA43M <--- thats the enc file i was using
18:13:54 <StormTide> basically, loaded qgis, loaded the depare layer, and went to quantumnik->create mapnik canvas, and boom, segfault
18:14:17 <springmeyer> ah, interesting
18:15:06 <springmeyer> what is likely happening is a conflict between gdal opening the file twice, once through qgis's gdal usage, once through mapnik's gdal usage...
18:15:26 <StormTide> maybe you can help me with another issue though. i'm trying to take this s57 depare layer... and convert it to tiles in the OSM tile structure...
18:15:34 <StormTide> ive been told mapnik is the solution for this
18:15:37 <StormTide> but im rather lost
18:15:42 <dodobas> yello
18:15:50 <springmeyer> sure, in a bit
18:16:01 <springmeyer> if dodobas does not beat me too it :)
18:17:00 <dodobas> err, what seems to be the problem
18:17:33 <StormTide> lack of my knowing much about gis i think ;)
18:17:53 <dodobas> StormTide: ahh, fill me in...
18:18:11 <dodobas> ill try to help
18:18:25 <StormTide> ok... so i'm a php dev... not a gis guy... i'm trying to create an iphone app using a library called route-me...
18:18:33 <StormTide> it can take tiles in the OSM tile format
18:18:52 <StormTide> im then trying to take these noaa s57 files, and get them converted to the osm tile format
18:18:59 <StormTide> which is where im told mapnik can help
18:19:11 <StormTide> but i know like sooo little about gis and mapnik ;)
18:19:56 <dodobas> ok, what format are this noaa s57 files in?
18:20:07 <StormTide> by first converting to a shp file, i can avoid the s57 core dump (noted above)... but i still cant seem to find the right commands to make the tiles
18:20:13 <StormTide> the format is s57?
18:20:23 <StormTide> http://www.charts.noaa.gov/ENCs/Agreement.shtml?US5WA43M is the file
18:20:36 <StormTide> (at the bottom)
18:21:59 <dodobas> hmm...
18:22:46 <StormTide> http://www.mailinglistarchive.com/html/route-me-map@googlegroups.com/2009-07/msg00031.html <-- explains kinda how i get the OSM tiles into route-me and the zoom/col/row.png format thats expected
18:22:53 <dodobas> im checking it out...give me a minute
18:26:44 <dodobas> StormTide: i've tried converting s57 file to SHP file format...but...
18:26:54 <dodobas> ERROR 6: Can't create fields of type StringList on shapefile layers.
18:26:54 <dodobas> ERROR 6: Can't create fields of type IntegerList on shapefile layers.
18:27:07 <drewby> I got the tiger data for north carolina integrated.
18:27:13 <drewby> oops wrong window
18:27:34 <StormTide> hrm strange... i just went into qgis and right clicked the depare layer, and made it a shp file in the esri format
18:27:45 <dodobas> so...its' possble that's the reason why ogr driver in mapnik does not work
18:28:31 <StormTide> maybe, but i can generate the mapnik preview on the shp file
18:28:44 <StormTide> its how i go from that 'preview' to generating the tile structure that im confused about
18:29:27 <dodobas> aha...
18:29:40 <dodobas> you are using quantumnik?
18:29:57 <StormTide> yes
18:31:12 <dodobas> to generate tiles should use a script
18:31:40 <StormTide> thats where i get confused
18:32:16 <StormTide> how do i go from what i see in 'mapnik rendered map' to the zoom/col/row.png format
18:32:28 <StormTide> er /row/col.png
18:32:37 <dodobas> first of all, to generate a map mapnik needs a XML map file
18:33:37 <StormTide> http://www.phpriot.com/2635 <-- likE THAT?
18:34:14 <dodobas> there is also a script called 'generate_tiles.py'
18:34:27 <dodobas> in osm mapnik distribution
18:34:29 <dodobas> http://wiki.openstreetmap.org/index.php/Mapnik#Rendering_tiles
18:34:59 <dodobas> yes like that
18:36:28 <dodobas> there are also other solutions for tile generation, osm 'mod_tile' or 'tilecache', probably others
18:37:06 <StormTide> ok so i grabbed the svn, now have generate_tiles.py....
18:37:27 <StormTide> do you offhand know how to pass it my xml file?
18:37:43 <StormTide> it _seems_ to want a postgis server?
18:38:26 <dodobas> check the bottom of the script
18:38:59 <dodobas> it reads it from MAPNIK_MAP_FILE env variable
18:39:52 <dodobas> there are also examples on how to start rendering a specific area
18:39:57 *** tcb has quit ()
18:40:02 <springmeyer> StormTide: perhaps useful: http://linfiniti.com/2009/08/harnassing-mapnik-for-high-quality-map-rendering/
18:40:53 <dodobas> yeah tilelite, forgot about that one :)
18:41:38 <StormTide> tilelite eh
18:41:41 <StormTide> interesting
18:43:37 <StormTide> haha, ok run _that_ with nice
18:45:40 *** tcb (n=tcarobru@adsl-75-10-247-30.dsl.pltn13.sbcglobal.net) has joined #mapnik
18:46:01 <springmeyer> ah, if you are using map2sqlite then use generate_tiles.py...
18:46:27 <springmeyer> sorry to confuse
18:47:14 <StormTide> no worries
18:47:22 <StormTide> this will be good for testing anyway
18:47:31 <springmeyer> yep, that is the point
18:47:53 <springmeyer> to test before sending a tile generation job that takes days...
18:48:11 <StormTide> haha, i was wondering when this was gonna finish ;)
18:48:20 <StormTide> is the bbox a lat/long pair?
18:48:41 <springmeyer> two lon/lat pairs
18:48:45 <dodobas> springmeyer: (OFFTOPIC) i did manage to solve that django issues (ldap + db), i ended up craating a copy of ldap in 'normal' db, and then i use pre_save,post_save and post_delete to keep ldap in sync, only trouble is that if a record is changed in ldap (by some other method) i have to manually sync LDAP to db again
18:48:51 <springmeyer> yes, lon/lat not lat/lon
18:49:13 <springmeyer> dodobas: good stuff
18:49:44 <dodobas> but as only a can change records directly in ldap...its a minor issue :D
18:49:53 <dodobas> s/a/I
18:50:40 *** alarmschaben (n=armin@alarmschaben-1-pt.tunnel.tserv5.lon1.ipv6.he.net) has joined #mapnik
18:51:40 <StormTide> is there any easy way in qgis to see the latlng coords of the current viewport
18:51:57 <dodobas> not that i know of :)
18:52:13 <alarmschaben> Hi! Could someone give me a quick pointer on how I can center and zoom my map when I just have a point and a zoom factor?
18:52:24 <springmeyer> Plugins > Qnik > View Live XML (look at bottom)
18:53:22 <StormTide> aha
18:53:24 <StormTide> -123.349671578 48.4577389545 -123.289028422 48.4914010455
18:53:25 <StormTide> ;)
18:53:46 <springmeyer> shhh secret
18:54:14 <StormTide> heh
18:57:26 <StormTide> hazaa, i have tiles
18:57:32 <StormTide> thanks guys!
18:57:54 <StormTide> will have to play with getting it into a sane format, but i think im on my way.
18:59:06 *** drewby has quit ("Leaving.")
19:01:26 <springmeyer> StormTide: good good
19:09:54 <StormTide> springmeyer, would be nice if i didnt have to go to shp as a intermediate file tho
19:10:42 <springmeyer> i agree
19:27:36 *** alarmschaben has parted #mapnik ()
19:39:20 <StormTide> any idea with mapnik on how to render out data from the map.. like say depth values?
19:40:06 <springmeyer> you mean as labels? or other shapes?
19:40:11 <StormTide> as labels
19:40:43 <springmeyer> yes, Mapnik supports labels
19:42:26 <springmeyer> if you are still using QGIS just right click on the layer in the legend, click properies, then click the labels tab
19:42:31 <springmeyer> and style it up
19:42:44 <springmeyer> which Quantumnik will translate into a 'TextSymbolizer'
19:43:01 *** RichardsDesk_ (n=Richard@dslb-092-075-125-152.pools.arcor-ip.net) has joined #mapnik
19:43:15 <StormTide> gotcha
19:43:40 <springmeyer> for all the gory complexity:
19:43:46 <springmeyer> `TextSymbolizer
19:43:46 <nikq> http://trac.mapnik.org/wiki/TextSymbolizer
19:44:28 <StormTide> is there a trick to getting mapnik to render a graduated symbology... mode quantities with say 4 classes... it seems to work selectively compared to what i see in qgis
19:44:53 <springmeyer> selectively?
19:44:59 <StormTide> lemme take a screenshot
19:44:59 <StormTide> sec
19:45:08 <springmeyer> if the mapnik render does not look correct then its a bug in Qnik
19:47:21 <StormTide> http://www.stormtide.ca/qgis.png vs http://www.stormtide.ca/mapnik.png
19:48:46 <springmeyer> StormTide: yep, bug. can you send me the XML that is produced from the Plugins > Qnik > View live xml
19:48:47 <springmeyer> ?
19:48:49 <springmeyer> er...
19:48:54 <springmeyer> http://dpaste.com
19:48:57 <StormTide> yep sec
19:49:10 <StormTide> http://www.stormtide.ca/layerproperties.png <-- props
19:49:16 <StormTide> sec on the xml
19:49:55 <StormTide> http://www.phpriot.com/2636
19:53:02 <StormTide> maybe im doing this really wrong, but the qgis view is basically what im going for
19:54:45 <springmeyer> what was the command you used to convert the ENC to the shapefile?
19:54:54 <springmeyer> ogr2ogr ? or using QGIS?
19:55:13 <StormTide> i loaded the enc in qgis, loading only the depare layer, then from the left i right clicked and went export as shpfile
19:55:23 <StormTide> er save as
19:55:26 <springmeyer> k
19:55:28 <StormTide> then i loaded a new project, and added the shp
19:55:44 <StormTide> added the symbology
19:55:54 <StormTide> and the render doesnt look quite right
20:00:28 <StormTide> am i doing something nonstandard? ... like would you have used a different process to get from s57 to mapnik? maybe postgis first instead of shp?
20:00:57 *** Genscher has quit (Read error: 110 (Connection timed out))
20:05:22 <springmeyer> the shape export out of qgis looks like it is failing to bring through attributes for me ....
20:05:42 <StormTide> hrm, they're there for me... i think
20:06:32 <StormTide> www.stormtide.ca/identify.png
20:06:54 <StormTide> its the drval1 and drval2 that im after
20:10:02 <StormTide> if i up the classification to like 15 types
20:10:07 <StormTide> it seems to render out sorta right
20:10:15 <StormTide> but theres a lot of duplicated filters
20:10:39 <StormTide> interestingly, theres no color agreement tho when i do that
20:20:25 <StormTide> springmeyer, also, it seems that quantumnik is emitting the xml in a weird order... trying to draw the text first, and then the shapes. which, if i understand how mapnik draws, is probably accounting for why i cant see the text
20:20:47 <Ldp__> Painter's Algorithm
20:21:16 <StormTide> does the order of the <style> tags matter or just the <StyleName> within the <Layer>
20:21:19 <springmeyer> StormTide: nope, labels are drawn second, in order they appear on the layer
20:21:23 <Ldp__> the latter
20:21:28 <StormTide> nm then
20:21:29 <StormTide> hrm
20:21:36 <Ldp__> the ordering of the <Layer> and then the <StyleName> therein
20:21:47 <springmeyer> StormTide: can you post your shapefile?
20:21:54 <StormTide> sure
20:21:55 <StormTide> sec
20:27:43 <StormTide> http://www.stormtide.ca/43shpdepare.shp
20:27:52 <StormTide> er
20:27:53 <StormTide> sec
20:28:39 <StormTide> there
20:29:22 *** drewby (n=dfaubion@128-8-138-156.umd.edu) has joined #mapnik
20:29:25 <drewby> StormTide: what's at those coordinates?
20:29:36 <drewby> from an hour and a half ago
20:29:50 <StormTide> cadborough bay victoria bc
20:30:24 <StormTide> (Royal victoria yacht club ;])
20:31:09 <StormTide> its just a testing area for me really... once i sort out this mapnik stuff i'll apply the file to all the noaa maps
20:32:01 <drewby> You're lucky.  I'm on roads
20:33:12 <StormTide> heh
20:33:29 <StormTide> then you've got it easy, no wind or current to worry about ;)
20:33:34 <StormTide> *hides*
20:33:59 <drewby> do you have to render wind and current?
20:34:18 <StormTide> render not so much, but part of this app is calculating tacking time to destination
20:34:22 <StormTide> which does require those calcs
20:35:23 <drewby> we have a million road-rendering issues, but the only water-relating rendering issue was when our populated areas layer extended beyond the bounds of the ocean. yeah I suppose not so much math for our display data, but plenty of math trying to generate it ><
20:36:29 <StormTide> ;)
20:36:57 <StormTide> how does mapnik decide where to place labels?
20:37:18 <StormTide> like theres hundreds of data points on this layer... but very few print
20:37:34 <StormTide> and, interestingly, totally different ones than show in qgis
20:37:58 <drewby> not a question for me 0.0 I have no idea
20:48:10 <springmeyer> StormTide: okay, I see problem
20:48:17 <StormTide> im listenin
20:49:01 <springmeyer> looks like a simple regression in Qnik. Not sure when it happened but graduated classifcation is currently broken
20:49:26 <springmeyer> and Qnik maps it to unique values symbolization
20:49:30 *** mak has parted #mapnik ("bails")
20:49:38 <springmeyer> which is why only a few layers match up when rendered with Mapnik
20:50:04 <springmeyer> easy fix which I am about to make
20:50:06 <springmeyer> so, would be ideal if you could pull the latest sources for Qnik
20:50:20 <StormTide> if you can walk me through that... i'd be happy to
20:50:26 <springmeyer> will require removing the plugin code installed via the plugin manager and placing it yourself
20:50:32 <springmeyer> sure, not too bad
20:51:06 <StormTide> i can do it on either ubuntu or mac, whatever's easier for you
20:51:10 <springmeyer> see 'Installing Quantumnik' here: http://bitbucket.org/springmeyer/quantumnik/
20:51:31 <springmeyer> it has instructions for 'manual' right there, let me know if you have any questions
20:51:32 <StormTide> aha, its your page i followed to get this all setup ;)
20:51:39 <StormTide> sure sec
20:52:07 <drewby> springmeyer: I admire how responsive you are to questions/comments on the project.
20:53:50 <StormTide> should i rm -rf the old quantumnik installed by the fetch mechanism first?
20:55:39 <springmeyer> yep
20:55:47 <springmeyer> or "unintall' from qgis plugin interface
20:56:34 <StormTide> k i uninstalled, copied it over
20:56:37 <StormTide> try to do it
20:56:38 * springmeyer tips hat to drewby 
20:56:41 <StormTide> and i get a python error
20:57:04 <springmeyer> okay, I'm not surprised, there is a lot of new stuff in the latest code
20:57:16 <springmeyer> paste the error?
20:57:20 <StormTide> http://www.phpriot.com/2637
20:58:09 <springmeyer> StormTide: looks like you cloned right before I committed fix
20:58:15 <StormTide> haha
20:58:21 <StormTide> k lemme try again
20:58:23 <springmeyer> $ hg pull -u #inside that dir
20:58:53 <StormTide> that didnt work
20:58:53 <StormTide> sec
20:58:58 <StormTide> i think its cuz i copied it
20:59:39 <springmeyer> you need to install mercurial and 'clone' the repository
20:59:56 <StormTide> hazaa
21:00:01 <StormTide> nono i did that... ;)
21:00:02 <springmeyer> easy_install mercurial
21:00:10 <StormTide> was just the way i cloned then copied
21:00:14 <StormTide> anyway working now
21:00:16 <springmeyer> ah, I see
21:00:18 <springmeyer> k
21:00:49 <StormTide> its looking  a lot better, maybe you can help me understand a few things tho.
21:00:59 <StormTide> the lines appear a lot heavier when rendered vs displayed
21:01:09 <StormTide> the text depth labels are in different places and for different values
21:01:45 * springmeyer glad the problem is fixed ;)
21:01:56 <StormTide> ;)
21:02:14 <springmeyer> sure, there are going to be subtle differences in how QGIS and Mapnik display/render the data
21:02:26 <springmeyer> they are completely different rendering engines
21:03:03 <springmeyer> All Quantumnik does is give you a "head start" toward the map (style+layer) definitions you need to render data with Mapnik
21:03:17 <StormTide> so the next step is to tune the xml then i assume
21:03:28 <springmeyer> yes
21:03:29 <StormTide> do you use that tilelite or whatever when doing that/
21:03:40 <StormTide> or is there a way to do that within qgis
21:03:49 <StormTide> it doesnt appear i can edit the live xml
21:03:50 <springmeyer> Qnik does not (yet) expose any real ability to tune
21:03:59 <springmeyer> nope
21:04:23 <springmeyer> That is a feature I would be happy to add but its non-trival is a waiting on funding
21:05:23 <springmeyer> but what you can do
21:05:34 <springmeyer> is open the XML in a text editor
21:05:48 <springmeyer> and then load it back into QGIS
21:06:03 <springmeyer> Hitting ctrl-r in QGIS will reload it
21:06:15 <StormTide> aha
21:06:16 <StormTide> ok
21:06:18 <springmeyer> so you can go back and forth between editing in a text editor and refreshing live
21:06:19 <StormTide> that i can handle
21:06:27 <StormTide> and i can keep my changes in svn which'll be a bonus ;)
21:06:35 <springmeyer> yep
21:07:03 <StormTide> this xml isn't all that scary to me, coming from a web background... so i guess i'll just start playing with it from the DTD
21:07:27 <StormTide> i dont really understand how the text symbolizers decide where to pull text from tho for example
21:08:00 <springmeyer> you don't understand where they pull text from or the logic behind where the text is placed?
21:08:17 <StormTide> well, theres thousands of data points, but it shows like 5
21:08:28 <StormTide> why not all of em, or ...
21:08:49 <springmeyer> allow_overlap=True
21:11:56 <drewby> springmeyer: have you though about multi-threading the compilation process?
21:11:59 <drewby> thought*
21:12:14 <StormTide> http://www.phpriot.com/2638
21:12:19 <springmeyer> drewby: JOBS=N
21:12:21 <StormTide> another python bug btw
21:12:28 <drewby> springmeyer: hah, ty
21:12:29 <StormTide> not exactly sure how i caused that tho
21:12:35 <StormTide> was reloading the map
21:12:50 <springmeyer> drewby: also, FAST=True
21:13:09 <springmeyer> StormTide: thats not good ;)
21:14:19 <StormTide> seems to occur on pan after a refresh
21:14:20 <drewby> oh yeah, I like to see both those cores maxed, this is much better :-)
21:14:31 <springmeyer> drewby: rock on
21:14:47 <springmeyer> StormTide: there is known buggyness in Qnik between these two modes....
21:14:56 <springmeyer> one being the split/tab view
21:15:11 <springmeyer> and the other being the load xml from external mapnik mapfile
21:15:29 <springmeyer> I likely need to split the functionality into two different plugins
21:15:51 <springmeyer> for now just don't hit the blue globe to create the tabs when you are loading external XML
21:15:56 <springmeyer> make sense?
21:16:52 <StormTide> i think so
21:16:53 <StormTide> lemme try
21:17:38 <StormTide> ok ya, if i dont load my shpfile/tab at the same time
21:17:41 <StormTide> i dont get the error
21:17:52 <StormTide> also, the control-r shortcut doesnt work for me
21:18:14 <StormTide> (but the menu command does)
21:19:34 <springmeyer> bugger
21:23:54 <StormTide> cool, i think i get the hang of this line/styling thing
21:24:06 <StormTide> now just to figure out how to render the textual data different ;)
21:29:18 <springmeyer> allow_overlap=True
21:29:31 <StormTide> tried that, doesnt seem to show any more points
21:30:02 <springmeyer> I thought you said text?
21:30:10 <StormTide> points of text
21:30:19 <StormTide> lemme take another screenshot
21:31:13 <StormTide> www.stormtide.ca/text.png
21:31:26 <StormTide> see the 9.1 5.4 etc
21:31:32 <StormTide> there should be hundreds of those
21:31:43 <StormTide> least if it were doing what i wanted it to
21:32:04 <StormTide> not sure what mechanism controls when those values get written and when they dont
21:32:55 <StormTide> http://www.accessnoaa.noaa.gov/images/flapic4.jpg <-- would be an example showing depth numbers
21:33:35 <springmeyer> that depth information is being labeled from a different dataset I bet
21:33:51 <springmeyer> not a polygon dataset of contours but a point dataset layer
21:34:08 <springmeyer> it looks to me like you are labeling text using the polygon layer...
21:34:28 <springmeyer> which will only label the centroid of the features geometry
21:34:34 <StormTide> ohhh
21:34:45 <StormTide> that makes a lot more sense
21:35:07 <StormTide> so now i guess the obv question is how do i tell mapnik to use two shpfiles haha
21:35:19 <StormTide> (sorry im being a bit of a pest)
21:37:44 <springmeyer> load up your data in Qnik in a separate session and take a look at the XML it creates if you've added two layers to QGIS
22:11:55 <StormTide> springmeyer, can i expect qgis to hang if i add a ton of info from a bigish shapefile?
22:12:22 <StormTide> when i try to render these depths...
22:12:35 <StormTide> it just maxxes out the processor and stays that way
22:12:46 <springmeyer> no
22:12:53 <springmeyer> thats bad
22:13:17 <springmeyer> labeling can take a long time in QGIS but it should not hang for more than 3-4 seconds
22:13:34 <springmeyer> StormTide: did you convert to shapefile?
22:13:38 <StormTide> yes
22:13:45 <StormTide> www.stormtide.ca/test3.shp
22:13:57 <springmeyer> is it handing when switching to Qnik or just in normal QGIS mode?
22:14:10 <StormTide> http://www.perrygeo.net/wordpress/?p=1 <-- i followed those instructions until i got to a shapefile, then loaded it in qgis
22:14:16 <StormTide> when switching to qnik is when it hangs
22:14:19 <StormTide> qgis is fine
22:14:40 <StormTide> happens with or without it created in a tab
22:14:47 <springmeyer> ah okay
22:14:58 <springmeyer> yes, there is a reason that would hang mapnik
22:15:28 <springmeyer> not necessarily the problem here but I've noticed that QGIS 1.4 can create corrupted shapefiles
22:16:31 <springmeyer> but, in this case are you saying you followed those exact instructions on perrys' blog?
22:16:45 <springmeyer> e.g that you used ogr2ogr to create the shapefile... instead of QGIS?
22:17:32 <StormTide> yes
22:17:52 <springmeyer> then post the shapefile as a zip of each part.
22:18:49 <StormTide> as a zip?
22:19:00 <StormTide> as in the compressed archive?
22:19:07 <springmeyer> yep
22:19:24 <springmeyer> you can't just post the 'shp', need all of the files...
22:19:32 <StormTide> oh ok
22:19:35 <StormTide> sec
22:19:40 <StormTide> can you handle a tar.gz?
22:20:46 <StormTide> www.stormtide.ca/test3.tar.gz
22:27:29 *** drewby has quit ("Leaving.")
22:32:53 <StormTide> www.stormtide.ca/showingdepths.png works and is responsive (no slow map redraw)... until i go to mapnik
22:35:03 <StormTide> http://www.phpriot.com/2639 <-- it wants to generate that xml file
22:36:46 <springmeyer> StormTide: $ hg pull -u
22:37:45 <springmeyer> StormTide: btw, are you running latest Mapnik release (0.7.0), or no?
22:38:27 <StormTide> ii  python-mapnik                         0.6.1-0ubuntu4                             C++/Python toolkit for developing GIS applic
22:39:00 <StormTide> pulled the update, trying to l oad that xml file still kills er
22:39:01 <springmeyer> okay
22:39:12 <springmeyer> what!
22:39:41 <StormTide> i pull'd, opened a new qgis, and tried to load that xml file i just pasted
22:39:44 <StormTide> still hangs
22:40:06 <springmeyer> okay, try to get the XML out again for me?
22:40:15 <StormTide> http://www.phpriot.com/2639
22:40:28 <StormTide> or do you mean from the new ver?>
22:40:30 <StormTide> can do that
22:40:30 <StormTide> sec
22:41:02 <springmeyer> does not look like the update to qnik sources worked...
22:41:46 <StormTide> http://www.phpriot.com/2641
22:41:49 <StormTide> is from the new
22:42:07 <springmeyer> that looks better/right
22:42:12 <springmeyer> still hangs though?
22:42:14 <StormTide> lemme try loading that one
22:42:14 <StormTide> sec
22:42:43 <StormTide> no hang that time
22:42:47 <StormTide> but no text either
22:42:56 <StormTide> test3's labels dont render
22:43:14 <springmeyer> lovely
22:45:22 <nikq> Mapnik Trac: Ticket #504 (Support for Pointzm missing in shapefile driver) created | http://trac.mapnik.org/ticket/504
22:45:53 <nikq> Mapnik Trac: Ticket #504 (Support for Pointzm missing in shapefile driver) updated | http://trac.mapnik.org/ticket/504#comment:1
22:46:23 <nikq> Mapnik Trac: test3.tar.gz attached to Ticket #504 | http://trac.mapnik.org/attachment/ticket/504/test3.tar.gz
22:54:16 <springmeyer> StormTide: you sure?
22:54:22 <springmeyer> do just a few show up if you zoom in?
22:54:24 <StormTide> that they dont render
22:54:26 <StormTide> ?
22:54:27 <StormTide> se
22:54:27 <StormTide> c
22:55:09 <StormTide> hrm
22:55:16 <StormTide> at the zoom level i had it, none show
22:55:18 <StormTide> if i zoom out
22:55:22 <StormTide> i can faintly see a few
22:55:28 <StormTide> certainly not the thousands of em tho
22:55:42 <springmeyer> but if you zoom in you see more and more right?
22:55:43 <StormTide> the higher the zoom, theres none
22:55:46 <StormTide> no
22:55:47 <StormTide> i see less
22:55:54 <springmeyer> odd
22:56:15 <StormTide> i have to have nearly the whole map in view to see any
22:57:06 <springmeyer> lets make sure we mean the same thing by zoom in and out...
22:57:54 <StormTide> zoom in == see less of the map in more detail
22:58:11 <StormTide> (objects appear larger)
22:58:41 <springmeyer> right, same here
22:58:44 <StormTide> <!-- nik2img.py mapnik.xml out.png -d 1418 324 -e -123.320510313 48.4445812531 -123.252829801 48.4599718781 -->
22:58:44 <StormTide>  for example i see no markers
22:58:47 <springmeyer> so, two things...
22:58:53 <springmeyer> if you zoom out a LOT
22:58:55 <StormTide> <!-- nik2img.py mapnik.xml out.png -d 1418 324 -e -123.557392107 48.3907140656 -123.015948007 48.5138390656 -->
22:58:59 <StormTide> is where i start to see some
22:59:06 <springmeyer> then label locations will "collide" more
22:59:09 <springmeyer> which drops labels
22:59:23 <springmeyer> if you zoom in they will not collide, and with more frequency will show up
22:59:26 <springmeyer> but likely...
22:59:30 <springmeyer> what you are describing
22:59:33 *** klabatter (i=4e361ab0@gateway/web/freenode/x-vunppguxjejzzbcr) has joined #mapnik
22:59:33 <StormTide> ok, well its doing the opposite of that
22:59:35 <springmeyer> is that you are zooming so close
22:59:52 <Ldp__> test that theory with allow_overlap="true" ?
22:59:53 <StormTide> i know where the labels should be based on the qgis output
23:00:01 <StormTide> and they're not there
23:00:04 <springmeyer> that you are in view of less points
23:00:06 <springmeyer> the "data" behind the labels
23:00:11 <springmeyer> screenshot?
23:00:17 <StormTide> sure sec
23:02:08 <StormTide> www.stormtide.ca/nolabelsqnik.png www.stormtide.ca/labelsqgis.png
23:03:22 <StormTide> www.stormtide.ca/somelabelsqnic.png
23:03:26 <klabatter> wow @StormTide, that looks gread
23:03:32 <springmeyer> hmm, paste xml for that one?
23:04:15 <StormTide> http://www.phpriot.com/2642
23:04:28 <StormTide> klabatter, thank springmeyer he's been helping me all day haha
23:04:40 <StormTide> *loves irc*
23:04:55 <klabatter> the mapnik-channel is gread :-)
23:05:28 <klabatter> great
23:05:41 <StormTide> ive been helping to run a similar #php on efnet channel for about a decade now ;)
23:08:22 <StormTide> springmeyer, the test3 properties in qgis are symbology set to 0 size...labels on depth with a font size of 6
23:10:48 <springmeyer> StormTide: baffled, I see labels just fine
23:10:54 <springmeyer> although they sure are tiny :)
23:11:11 <springmeyer> StormTide: I know of no bug in Mapnik 0.6.1 that would cause this
23:11:16 *** D3f0 (n=D3f0@190.177.81.33) has joined #mapnik
23:11:26 <klabatter> just a little question: How can i tell mod_tiles to render new tiles after importing changesets into postgres-database?
23:11:28 <springmeyer> but I am running 0.7.0 and I would recommend you upgrade on ubuntu
23:11:55 <StormTide> springmeyer, hrm
23:12:53 <Ldp__> klabatter: you want to rerender all changed tiles ?
23:12:57 <StormTide> springmeyer, are there newer mapnik packages that i could try?
23:13:16 <klabatter> yes, if possible...
23:13:33 *** myselfhimself (n=chatzill@85-171-136-33.rev.numericable.fr) has joined #mapnik
23:13:34 <springmeyer> `UbuntuInstallation
23:13:34 <nikq> http://trac.mapnik.org/wiki/UbuntuInstallation
23:13:38 <springmeyer> ^^
23:13:39 <Ldp__> klabatter: try out render_expired in the mod_tile repo. I've been trying it tonight, for the first time, but it seems to work fine.
23:13:41 <myselfhimself> hi
23:14:21 <StormTide> ah crap, do i have to setup 8.4 postgis?
23:14:40 <Ldp__> StormTide: no
23:15:14 <StormTide> should i remove the mapnik package from ubuntu before installing from src or will they co-exist
23:15:30 <Ldp__> remove all traces
23:15:52 *** jctull has quit ()
23:15:54 <StormTide> apt-get remove libmapnik0.6 mapnik-utils python-mapnik good enough?
23:16:31 <klabatter> is it in the renderd.conf :Ldp__?
23:16:47 <Ldp__> StormTide: http://pastebin.com/d2264c2ec these are what Scons install, so check to see if they are really gone after the apt-get remove
23:17:06 <springmeyer> StormTide: yep
23:17:15 <Ldp__> /usr or /usr/local or wherever Ubuntu put them
23:17:41 <myselfhimself> StormTide: what's your ubuntu version ?
23:17:46 <Ldp__> klabatter: http://trac.openstreetmap.org/browser/applications/utils/mod_tile/render_expired.c
23:17:47 <StormTide> karmic fully updated
23:19:00 <Ldp__> klabatter: in short: use osm2pgsql with '-e 15 -o expired.tiles.file' and then pipe that file into render_expired with suitable options for your layers
23:19:26 <Ldp__> klabatter: repeat for every renderd.conf layer you want to rerender
23:19:31 <myselfhimself> StormTide: run locate libicu | grep ".so" ; what versions do you see ?
23:19:35 <StormTide> *compiling in one core*
23:19:46 <Ldp__> klabatter: repeat piping that file into render_expired, not run osm2pgsql, to be completely clear on that
23:19:55 <klabatter> Ldp__: thanks i will try it tomorrow. It's too late for today ;-)
23:20:14 <Ldp__> klabatter: assuming you use the regular metatile scheme, of course
23:20:27 <myselfhimself> StormTide: have you checked your libboost version fits the requirements before compiling mapnik ?
23:20:31 <StormTide> myselfhimself, http://www.phpriot.com/2644
23:20:47 <StormTide> myselfhimself, i didnt, but the configure worked... its compiling now
23:21:24 <myselfhimself> StormTide: ok
23:21:39 <myselfhimself> StormTide: I think that you'll be ok
23:21:49 <klabatter> is there a way to re-render all tiles without deleting them?
23:22:53 <myselfhimself> StormTide: I had problems on Ubuntu 8.04 because libicu was too old in the ubuntu repos, and libboost too and libboost depended on libicu, so I had to remove packages and compile both libicu then libboost from recent svn/tarballs before compiling a correct mapnik version
23:22:55 <StormTide> wonders if python has a -j 4 equiv to use all cores to compile heh
23:23:07 <myselfhimself> klabatter: what's your setup ? do you use mod_tile / renderd ?
23:23:19 <klabatter> yes, mod_tile / renderd
23:23:33 <myselfhimself> klabatter: then just touch /your/tiles/folder
23:23:37 <Ldp__> klabatter: touch the planet_update file
23:23:40 <Ldp__> not the folder
23:23:44 <myselfhimself> yes sorry
23:23:51 <myselfhimself> the default name is planet-import-complete
23:24:04 <Ldp__> then it will rerender on demand, not prerender them all
23:24:06 <myselfhimself> in my case that was touch /var/www/tiles/planet-import-complete
23:24:47 <springmeyer> StormTide: send the object JOBS=4
23:24:49 <Ldp__> it's in render_config.h: #define PLANET_TIMESTAMP "/planet-import-complete" (relative to your tile dir)
23:25:07 <klabatter> hm, there is no planet-import-complete in our tiles-dir
23:25:15 <StormTide> ok compiled
23:25:19 <StormTide> import mapnik works
23:25:25 <StormTide> any way to check the version now that its not packages?
23:26:02 <klabatter> should i create it?
23:26:26 <StormTide> ok, so, with new mapnik i see a few more labels
23:26:28 <Ldp__> klabatter: yes, and any time you update the timestamp, it will rerender any metatile that's older
23:26:31 <StormTide> still not as many as i should
23:26:57 <StormTide> and the font doesnt get bigger as i zoom in which is annoying heh
23:27:25 <myselfhimself> Ldp__: this timestamp thing is cool, but it doesn't work per-tileset ; that's a pity
23:27:36 <Ldp__> it's global
23:27:36 <klabatter> Ldp__ yea! it's working!
23:28:01 <Ldp__> myselfhimself: use render_expired if you want to rerender/delete per tileset
23:28:06 <StormTide> springmeyer, if i turn on the markers so they're lil dots
23:28:11 <StormTide> i see all the right dots
23:28:14 <StormTide> but some dont have text
23:28:16 <myselfhimself> Ldp__: aw.. I didn't know this one
23:28:18 <StormTide> whats that mean?
23:28:47 <Ldp__> myselfhimself: there's one drawback: if you delete the metatiles, mod_tile can't serve those old ones when the load is too high to render fresh ones
23:28:48 <springmeyer> try removing the <PointSymbolizer> altogether from the XML
23:31:32 <StormTide> springmeyer, leaving the style or nuke the whole style block an all references to that
23:31:41 <StormTide> eg only leave test3_layers?
23:31:43 <StormTide> er
23:31:46 <StormTide> test3_labels
23:31:51 <springmeyer> sure, no matter
23:32:30 <StormTide> hazaa
23:32:33 <StormTide> that did the trick
23:33:08 <springmeyer> okay, so that was happening because the points overlapped the text then
23:33:17 <klabatter> ok, enough for today. thank you for the fantastic help!
23:33:40 <StormTide> springmeyer, hrm but the points layer comes first in the file...
23:34:04 <springmeyer> StormTide: just pushed a small change to Qnik so that if you set point outlines and fill to None, they will not render
23:34:14 <StormTide> *tries update
23:35:15 <StormTide>  springmeyer seems to be working about the none thing
23:35:29 <springmeyer> huh?
23:36:09 <StormTide> the update works
23:36:17 <springmeyer> ah, good
23:36:28 *** klabatter has quit ("Page closed")
23:36:38 <StormTide> springmeyer, interesting,... in qgis... the icons appear to overlay the text... but the labels are set to 'over'
23:36:45 <StormTide> (if i turn them on that is)
23:37:38 <springmeyer> the "Placement" parameter is not honored by Qnik
23:37:57 <springmeyer> it could be of course but I've never added support for it
23:38:26 <springmeyer> that would be a good one to log here: http://bitbucket.org/springmeyer/quantumnik/issues/
23:38:41 <StormTide> ok, whats the xml verbiage to overlay labels (eg so they dont hide on collision)
23:38:57 <springmeyer> allow_overlap = True
23:39:12 <springmeyer> clear_label_cache=True (on the Layer)
23:39:13 <StormTide> http://bitbucket.org/springmeyer/quantumnik/issue/5/ensure-labels-are-placed-last-while-avoiding-unnecessary-layer
23:39:19 <StormTide> looks like its already logged
23:39:27 <springmeyer> nope, separate issue
23:40:06 <springmeyer> er, well kinda :)
23:40:18 <springmeyer> "over" means to me paint them last
23:40:31 <springmeyer> but collision avoidance is a separate issue
23:40:46 <springmeyer> as is where the default placement around the point is set to...
23:41:23 <StormTide> k kew
23:42:08 <StormTide> i'm going to go peruse the docs for the text options, see if i cant figure out how to get the font to scale with the map properly
23:43:57 <StormTide> thanks for all your help with these lil issues
23:44:20 <StormTide> pretty sure i can force mapnik to render what i need to outta this ;)
23:44:31 <springmeyer> ya, so thats MinMaxScaleDenominators
23:46:56 *** jctull (n=jctull@ppp-71-142-138-235.dsl.renocs.pacbell.net) has joined #mapnik
23:47:22 <StormTide> hrm
23:48:46 <StormTide> is there like any kind of 'normal' template for this kinda stuff
23:49:17 <StormTide> like, essentially, i'm going to be gettin output to raster, OSM style tiles... so will have certain defined zoom levels etc
23:49:36 <StormTide> someone's got to have already figured out what size text corresponds to what scale etc
23:50:06 <springmeyer> http://svn.openstreetmap.org/applications/rendering/mapnik/inc/entities.xml.inc
23:50:09 <StormTide> like in html/css we've got h1 h2 h3... tags which come configured in some stock sort of way in a browser
23:50:45 <StormTide> aha now we're talkin
23:50:51 <Ldp__> you're welcome :)
23:51:25 <StormTide> suppose i better check the osm license before i just yank this...
23:51:41 <Ldp__> I wrote it, you can use it
23:51:59 <Ldp__> the OSM license pertains to the data in the OSM database itself
23:53:08 <StormTide> k kew
23:54:36 <StormTide> Ldp__, is this run through some sort of preprocessor?
23:54:41 <StormTide> to resolve the entity names?
23:56:01 *** RichardsDesk_ has quit (Client Quit)
23:56:46 <springmeyer> StormTide: Mapnik understands the entities through its use of libxml2 to parse the XML in general
23:57:09 <StormTide> ah
23:57:45 *** dodobas has quit (calvino.freenode.net irc.freenode.net)
23:57:49 *** dodobas (n=dodobas@open.geof.hr) has joined #mapnik
23:57:54 <springmeyer> see the osm.xml for how those are included
23:57:57 <springmeyer> also:
23:58:05 <springmeyer> `ManagingLargeXmlFiles
23:58:05 <nikq> http://trac.mapnik.org/wiki/ManagingLargeXmlFiles
23:58:41 <StormTide> cool
23:59:10 <StormTide> prob too much to hope for but know of anyone who's developed a marine style sheet?