00:01:50 <springmeyer> wonxly: we should find a way to check for it within the SConstruct...
00:05:16 * springmeyer is not positive that is the problem however...
00:05:24 * springmeyer still on the phone..
00:08:32 <wonxly> ok, i appreciate the help though.  i am rebuilding boost.  i think it was looking for icu in the wrong place ;)
00:11:06 *** aub has quit ()
00:11:28 <wonxly> hmm no it looks like it was right before.  and it was building boost with icu:
00:11:29 <wonxly> Note: Building Boost.Regex with Unicode/ICU support enabled
00:11:31 <wonxly>     Using ICU in  /usr/include
00:54:52 <springmeyer> wonxly: still around?
00:55:45 <wonxly> yeah i'm here ;)  i downloaded icu 3.8 and am compiling that. is it possible that icu that ships with solaris 10 is not compatible with mapnik?
00:55:54 <springmeyer> yes
00:56:01 <springmeyer> you need icu && icudata
00:56:25 <wonxly> ok, thanks for your help springmeyer ;)
00:56:32 <springmeyer> for example 'libicucore' is shipped on mac osx but does not work
00:56:51 * springmeyer has never used solaris but its been tricky to build in the past for others
00:57:01 <springmeyer> parts because few have done it :)
00:57:41 <springmeyer> wonxly: are you building from mapnik trunk?
00:57:49 <wonxly> yeah i have definitely had to jump through a few hoops.  but it will be worth it once everything is working ;)
00:57:59 <wonxly> yes
00:58:09 <springmeyer> okay
00:58:14 <springmeyer> one note about scons...
00:58:17 <wonxly> i think pulled things down from svn yesterday
00:58:38 <springmeyer> if you cancel out of the build with cntrl-c (or whatever halts a terminal process)...
00:58:49 <springmeyer> that can screw up the scons test targets
00:59:17 <wonxly> yeah i saw something about that somewhere...it recommended blowing away that temp directory?
00:59:22 <wonxly> i tried that too ;)
00:59:27 <springmeyer> ya, okay...
00:59:28 <springmeyer> shoot
00:59:30 <springmeyer> :)
00:59:50 <wonxly> lol, def a good idea though ;)
00:59:51 <springmeyer> can you paste the config.log error?
01:00:00 <wonxly> yeah sure...
01:00:06 <springmeyer> http://dpaste.com
01:02:00 <wonxly> g++ -o .sconf_temp/conftest_17 .sconf_temp/conftest_17.o -Lsrc -L/opt/csw/lib -L/usr/lib -L/usr/local/lib -L/usr/sfw/lib -lfreetype -lxml2 -lz -lpthread -lm -lsocket -lnsl -lm -lltdl -lpng -ltiff -lz -ljpeg -lproj -licuuc -licudata -lboost_system-mt -lboost_filesystem-mt -lboost_regex-mt
01:02:01 <wonxly> Undefined                       first referenced
01:02:03 <wonxly>  symbol                             in file
01:02:04 <wonxly> icu::Collator::createInstance(icu::Locale const&, UErrorCode&)/usr/lib/libboost_regex-mt.so
01:02:06 <wonxly> icu::Locale::Locale(icu::Locale const&)/usr/lib/libboost_regex-mt.so
01:02:07 <wonxly> icu::Locale::Locale()               /usr/lib/libboost_regex-mt.so
01:02:09 <wonxly> icu::Locale::~Locale()              /usr/lib/libboost_regex-mt.so
01:02:10 <wonxly> ld: fatal: Symbol referencing errors. No output written to .sconf_temp/conftest_17
01:02:12 <wonxly> collect2: ld returned 1 exit status
01:02:13 <wonxly> scons: Configure: no
01:02:48 <wonxly> sorry, you wanted me to point you to that dpaste link didn't you?
01:02:56 <wonxly> http://dpaste.com/5608/
01:03:05 <springmeyer> ya, no worries tho
01:04:38 <springmeyer> wonxly: and thats on the boost_regex test?
01:04:50 <wonxly> yes
01:05:16 <springmeyer> so the ICU tests the precede it succeed? ie....checking for icu..... yes
01:05:27 <springmeyer> the/that precede it
01:05:46 <wonxly> yeah let me paste the whole test in there so you can see all of it and i'll send you the dpaste link
01:06:43 <springmeyer> okay. that is an odd error :)
01:07:09 <wonxly> http://dpaste.com/5615/
01:08:14 <springmeyer> what boost version? and how did you install?
01:08:51 <wonxly> boost 1.37...and i had to compile it and install it
01:09:56 <wonxly> i'm trying to compile mapnik now with the icu 3.8 ....
01:11:20 <wonxly> same problem :(
01:11:53 <springmeyer> I wonder if we have the library name wrong
01:13:11 <springmeyer> assuming icu 3.8 went in /usr/lib/ (for example) then what does $ ls -l /usr/lib/libicu* give?
01:14:36 <wonxly> http://dpaste.com/5620/
01:16:42 <springmeyer> hmm, okay thx, looks good
01:16:49 <springmeyer> and so what does this give?
01:17:16 <springmeyer> $ ldd /usr/lib/libboost_regex*
01:18:42 <wonxly> http://dpaste.com/5624/
01:18:57 <wonxly> it looks like you found it ;)  looks like i need to recompile boost now that i have the new icu
01:23:48 <springmeyer> perhaps
01:24:20 <springmeyer> although I wonder if this is a side-effect of the  -pthread fix
01:24:51 <wonxly> it could be...
01:24:58 * springmeyer notes 'libpthread.so.1 =>       /lib/libpthread.so.1'
01:25:52 <springmeyer> but we're not putting /lib  the -Lpaths...
01:33:01 <springmeyer> wonxly: try sticking in the SConstruct: env.Append(LIBS='-licui18n') before the boost check
01:33:14 <wonxly> ok
01:33:26 *** springmeyer has parted #mapnik ()
01:33:41 *** springmeyer (n=dane@c-24-19-50-92.hsd1.wa.comcast.net) has joined #mapnik
01:33:45 <springmeyer> https://svn.boost.org/trac/boost/ticket/1083
01:34:02 <springmeyer> kinda of an unclear ticket but worth a shot
01:43:08 <wonxly> ok after adding the env.Append(LIBS = '-licui18n') i still get the same error
01:46:05 <springmeyer> wonxly: are you setting ICU_LIBS and ICU_INCLUDES ?
01:48:12 <wonxly> no, i will try that now.  should it be ICU_INCLUDES=/urs/include and ICU_LIBS=/usr/lib ?
01:48:25 <springmeyer> ya
01:48:38 <springmeyer> but I just realized that they are not respected
01:48:56 <springmeyer> committing the fix in one moment...
01:50:20 <nikq> Mapnik Trac: Changeset [992]: scons: add the paths specified for ICU_LIBS/INCLUDES | http://trac.mapnik.org/changeset/992
01:50:36 <springmeyer> that won't help you, but good thing to finally notice
01:50:57 <springmeyer> (won't help you because those paths are likely already being added)
01:51:39 <wonxly> ok gotcha...i am still rebuilding boost as we speak...i am crossing my fingers that it will magically help? lol
01:52:05 <springmeyer> okay :)
02:13:15 <wonxly> its taking forever to build but i did find one of the newly built .so files and ran ldd on it and it is using the later icu now at least ;)
02:19:01 *** aub (n=aubrey@216.156.96.6) has joined #mapnik
03:00:48 *** rcoup has quit ()
03:46:53 <wonxly> springmeyer, any chance you are still around?
03:48:55 * springmeyer saunters back in
03:49:39 <wonxly> after installing icu 3.8.1 and recompiling boost i got past my last problem...but now i get a different error...check this out...
03:50:28 <wonxly> http://dpaste.com/5708/
03:50:58 <springmeyer> awesome
03:51:13 <springmeyer> that should be an easy one :)
03:51:37 <springmeyer> can you re-run...
03:51:49 <springmeyer> $ python scons/scons.py configure
03:52:07 <wonxly> yeah sure, any options or just configure?
03:52:19 <springmeyer> it will inherit past options
03:52:23 <springmeyer> so, nope
03:53:54 <springmeyer> and then in another console open a python interpreter...
03:54:08 <w0lfie_> hm what was i going to ask you
03:54:32 <w0lfie_> oh well :)
03:54:38 <springmeyer> in the python interpreter type...
03:54:39 <wonxly> ok got a python interpreter opened up and finished executing the configure
03:54:41 <springmeyer> >>> from distutils.sysconfig import get_python_version; print get_python_version()
03:54:46 <springmeyer> let me know what that gives
03:55:18 <springmeyer> I'm guessing you're running python2.4?
03:55:29 *** rcoup (n=rcoup@ip-118-90-67-87.xdsl.xnet.co.nz) has joined #mapnik
03:57:14 <w0lfie_> oh right
03:57:25 <wonxly> i think i must be putting that in wrong it says ImportEror: No module named distutils.sysconfig
03:57:41 <wonxly> but from the command line doing python -V i get Python 2.3.3
03:57:43 <springmeyer> nope, thats possible
03:57:50 <springmeyer> shoot, okay
03:57:53 <w0lfie_> mod_tile! thats what i was going to ask about
03:57:59 <springmeyer> mod_tile!
03:58:04 <w0lfie_> yes!
03:58:14 <w0lfie_> hurrah!
03:58:23 <springmeyer> so wonxly: can you do..
03:58:30 <w0lfie_> i can't think
03:58:32 <springmeyer> import distutils #?
03:58:51 <w0lfie_> i was trying to figure out what mod_tile does actually
03:58:59 <w0lfie_> i mean its obviously an apache module that.. tiles (bear with me)
03:59:18 <w0lfie_> but how does one utilize it, access the tiles directly? or via WMS still, etc?
03:59:18 <wonxly> ImportError: No module named distutils
03:59:31 * springmeyer did not realize that (recently) adding the usage of distutils would break python 2.3 compatibility
03:59:38 * springmeyer is to blame here
03:59:52 * springmeyer goes to look for when distutils was added...
04:00:05 <w0lfie_> basically i'm trying to figure out whether i should use tilecache or mod_tile..
04:01:02 <springmeyer> w0lfie_: ya so TileCache rocks, and so does Mod_tile
04:01:11 <wonxly> lol thanks for your help...i have been banging my head on the desk a few days over this ;)
04:01:17 <springmeyer> but they are very different :)
04:01:32 <w0lfie_> okay so to use TileCache i have to implement the extensions to my WMS client
04:01:43 * migurski de-lurks for a minute
04:01:45 <w0lfie_> and mod_tile.. is basically giving me just tiles on a server, but it intelligently manages them?
04:01:51 <w0lfie_> is that a fair summary
04:02:16 <springmeyer> migurski is my new unicorn ;)
04:02:29 <migurski> =)
04:02:35 * springmeyer fails at finding when distutils came around...
04:02:48 <migurski> I'm sitting here with my colleague Shawn planning our Monday E-Tech tutorial on making maps
04:02:59 <springmeyer> awesome
04:03:06 <migurski> we just banged our heads against mapnik SRS stuff for a while, I'll need to pick your brain about it
04:03:17 <springmeyer> oh, you bet
04:03:32 <migurski> esp how nik2img deals with it
04:03:46 <w0lfie_> you're trying to figure out when distutils was released?
04:04:00 <springmeyer> w0lfie_: yep, you know?
04:04:16 <w0lfie_> well depends which version, 0.9 came out in 2000
04:04:17 <springmeyer> weird: this indicates 2.2.x: http://www.python.org/doc/2.2.3/lib/module-distutils.html
04:04:26 <w0lfie_> which is um.. python 2.0 something
04:04:27 <migurski> anyway, I don't want to get fully into it right now because you guys are talking about other stuff, but I'm totally flummoxed by the multiple levels of (re)projection
04:04:42 <springmeyer> okay, then wonxly: perhaps you've got a custom python install on this solaris thingy
04:04:54 <w0lfie_> i don't know if that helps! but basically it's been around Forever (tm)
04:05:02 <springmeyer> k
04:05:22 <springmeyer> so wonxly: to sum up them, I need to add a check/catch for distutils if the user does not have it...
04:05:35 <w0lfie_> hm, yes after review they were a standard part of python in 2.0
04:05:36 <springmeyer> but for now wonxly: perhaps you can install setuptools and get on your way?
04:05:57 <w0lfie_> they existed earlier than that but were not part of python per se
04:06:03 <springmeyer> w0lfie_: I'd start with TileCache since it is very easy to set up...
04:06:09 *** migurski has quit ()
04:06:12 <springmeyer> migurski: at hand
04:06:15 <springmeyer> ack!
04:06:24 <w0lfie_> hm, i'm more concerned with client side
04:06:37 <w0lfie_> right now my WMS implementation is a subclass of a dynamic map class
04:06:40 <springmeyer> flummoxed? oh no.
04:06:43 <w0lfie_> i dunno how tiled wms will fit into that..
04:06:45 <wonxly> ok cool, thanks for your help springmeyer...i will try that
04:09:02 <springmeyer> wonxly: or revert svn to before r866
04:09:03 <nikq> http://trac.mapnik.org/changeset/866, at , by artem: + scons_python_usability_improvements2 (springmeyer)
04:09:44 <wonxly> ok i will give that a go first...it might be faster than trying to install setuptools on this solaris box ;)
04:10:04 <springmeyer> well setuptools should be easy
04:10:41 <springmeyer> but it will upgrade distutils so r866 is a backup if that does not go well :)
04:10:42 <nikq> http://trac.mapnik.org/changeset/866, at , by artem: + scons_python_usability_improvements2 (springmeyer)
04:10:57 <springmeyer> .g setuptools
04:10:58 <nikq> springmeyer: http://peak.telecommunity.com/DevCenter/setuptools
04:11:43 * springmeyer is facinated to know if mapnik will work with python 2.x :)
04:12:18 <wonxly> yeah i was just going to avoid setuptools because it said "setuptools is a collection of enhancements to the Python distutils (for Python 2.3.5 and up on most platforms; 64-bit platforms require a minimum of Python 2.4) since the Solaris 10 install had python 2.3.3
04:12:42 <springmeyer> ah, yup
04:12:57 <springmeyer> and also cuz you really want to test mapnik against 2.3.3 :)
04:13:20 <wonxly> yeah you are on to me ;)
04:14:23 <springmeyer> w0lfie_: ya, you rolled a custom js implementation of WMS into esri fu huh?
04:14:40 <springmeyer> the cool thing about TileCache is the flexibility
04:15:14 <springmeyer> you can set it up as a caching interface between a client and a WMS server
04:16:12 <springmeyer> or you use tilecache in a strict tiling mode where it expects requests not by bounding box (aka WMS) but by tile uri, ie /0/3/5/6.png
04:25:56 <wonxly> ok with r866, after adding env.Append(CXXFLAGS = '-pthreads') into SConstruct again for gcc i still get a similar error:
04:25:57 <nikq> http://trac.mapnik.org/changeset/866, at , by artem: + scons_python_usability_improvements2 (springmeyer)
04:26:47 <springmeyer> ack. a distutils error?
04:27:03 <springmeyer> make sure to svn up -r <866
04:27:05 <wonxly> http://dpaste.com/5731/
04:27:21 <wonxly> oh i need to be less than 866? lol
04:27:33 <wonxly> my bad
04:27:47 <springmeyer> ya, sorry
04:27:55 * springmeyer could have been more clear :)
04:28:07 <wonxly> no you def said before 866...lol
04:36:11 <springmeyer> wonxly: you are building the g++ not the sun compiler, right?
04:36:20 <wonxly> right
04:36:34 <wonxly> ok i def got farther but i still had errors...want to see them?
04:36:43 <springmeyer> yep
04:37:07 <wonxly> is boost python required?
04:37:31 <springmeyer> yes, for the python bindings
04:37:54 <wonxly> oh ok...that didn't build correctly with boost so i don't have that...i got a bunch of errors related to that ithink
04:38:21 <springmeyer> r923 added checking for boost python :)
04:38:22 <nikq> http://trac.mapnik.org/changeset/923, at , by dane: scons: make sure to check for boost_python (closes #229), improve checking of boost version (closes #236), and fetch ABI version from mapnik/version (closes #72), as well as adding PathAccept? for all
04:39:12 <springmeyer> wonxly: you can do $ python scons/scons.py BINDINGS=None
04:39:24 <wonxly> ok
04:39:30 <springmeyer> to build mapnik c++ core while you wait for boost to recompile with python
04:41:26 <wonxly> did the bindings option get added after r865? http://dpaste.com/5740/
04:41:27 <nikq> http://trac.mapnik.org/changeset/865, at , by dane: + apply occi-input-plugin-2.patch (kunitoki) (#212)
04:42:15 <springmeyer> ah, no sorry..
04:42:21 <springmeyer> BINDINGS=''
04:42:24 <springmeyer> try that?
04:42:30 <wonxly> oh ok gotcha...
04:42:57 * springmeyer is about to nuke that ListVariable option that scons has, so troublesome..
04:48:16 <w0lfie_> springmeyer, flex, not JS; I'm starting to wonder if I should use OpenLayers for this app but I'm not experienced with JS really
04:50:02 <springmeyer> well, that flex is slick of course
04:50:04 <w0lfie_> either that or use modest maps
04:50:05 <springmeyer> but...
04:50:51 <springmeyer> OpenLayers is top notch, the lead developer also wrote TileCache, and they are both open source so your non-profit won't have trouble finding future devs
04:51:22 <w0lfie_> yeah i've no doubt openlayers is excellent
04:51:34 <springmeyer> ya, I'm starting to learn modestmaps
04:51:48 <w0lfie_> but i'm not experienced nor generally a fan of JS
04:52:11 <springmeyer> ah, okay.
04:52:12 <w0lfie_> of course i don't want to penalize them for that
04:52:19 <w0lfie_> but there is still that ;)
04:52:27 <w0lfie_> quite possibly i will take a serious look at modest maps
04:52:53 <w0lfie_> and end up going that way
04:52:53 <springmeyer> ya, I'm not experiences nor generally a fan of flex :)
04:53:18 <springmeyer> and can't type correctly, ever
04:53:20 <w0lfie_> yeah i think you can do a great job with either
04:53:27 <w0lfie_> i'm not one of those language/implementation nazis
04:53:51 <w0lfie_> except when it comes to php *cough*
04:55:16 <springmeyer> wonxly: how goes it?
04:55:38 <springmeyer> I've got a commit ready here for handling lack of distutils...
04:55:40 <wonxly> it is compiling quite nicely actually :)  still going, a little slow because i'm running it in a VM
04:55:51 <springmeyer> get cha back on svn head :)
04:56:04 <wonxly> awesome
04:56:05 <springmeyer> VM? like virtual box?
04:56:14 <wonxly> yeah i'm running it on virtual box actually
04:56:48 <springmeyer> so, why solaris then instead of debian or something?
04:56:50 <wonxly> i just did a Solaris 10 install in there and am trying to get mapnik up and running
04:57:05 * springmeyer is clueless about solaris so I'm truly curious
04:57:43 <wonxly> because my plans are to install it on production machines at work...they all run solaris...
04:57:48 <springmeyer> debian packages make mapnik literally a 20 minute auto-install
04:57:52 <springmeyer> ah, great
04:58:03 <wonxly> ohh i got an error :(
04:58:17 <springmeyer> and secretly you wanted to help the mapni project be able to say we support solaris :)
04:58:28 <wonxly> lol exactly
04:58:54 <wonxly> yeah i built everything the other day on my linux box and it was very smoothe
04:59:54 <springmeyer> :) ah good
05:00:40 <wonxly> i think mapnik is awesome...so if i can help out even better
05:00:43 <wonxly> http://dpaste.com/5750/
05:00:54 * springmeyer has had virtualbox going with fedora lenny and intrepid on mac osx for the last week while he has tweaked the build scripts
05:01:11 <springmeyer> so they work perfectly on linux and blow up on solaris :)
05:01:19 <wonxly> lol
05:02:03 <wonxly> looks like it doesn't like some linker options
05:02:10 <springmeyer> okay, that should also be a fairly easy one
05:02:26 <w0lfie_> i have mapnik running in a xen
05:02:33 <springmeyer> jump into src/SConscript and start pulling the plugs on those linux flags!
05:02:46 <w0lfie_> still thinking how we can use it at work
05:03:13 <w0lfie_> may end up using it to fake esri tile services
05:03:23 <w0lfie_> since all you need to do is create the geojson file and have the correct tiling scheme
05:03:48 <springmeyer> ..hmm I'm wrong: http://trac.mapnik.org/browser/trunk/src/SConscript#L38
05:03:58 <springmeyer> there is already solaris checking in that file it seems
05:04:24 <w0lfie_> are you running solaris or opensolaris
05:04:29 <wonxly> solaris
05:04:38 <springmeyer> err I guess that's just sun CC checking, but still seems like the flags are not being added there
05:04:50 <springmeyer> likely back in the main SConstruct in the root
05:05:34 <wonxly> is there any easy way to see my env['PLATFORM'] and env['CXX']??
05:05:45 <springmeyer> ya.
05:06:02 <wonxly> sorry not very familiar with python ;)
05:06:03 <springmeyer> just stick this somewhere in the SConstruct....
05:06:12 <springmeyer> import pdb;pdb.set_trace()
05:06:34 <springmeyer> will pause the build script and you can get variables out by typing them in the python interpreter
05:07:14 <wonxly> cool...now i just type something to print them?
05:07:32 <springmeyer> ya, just type
05:07:33 <springmeyer> env['PLATFORM']
05:09:12 <wonxly> hmm i guess i need to put that after the env stuff is setup? lol i get NameError: name 'env' is not definied
05:09:26 <springmeyer> ya, :)
05:11:25 <nikq> Mapnik Trac: Changeset [993]: scons: add fallback for contructing python install environment if  ... | http://trac.mapnik.org/changeset/993
05:12:07 <w0lfie_> what would really be ideal is a WMS tiling solution that didn't require client-side changes
05:14:01 <springmeyer> w0lfie_: explain?
05:14:56 <springmeyer> wonxly: odd thing about this error http://dpaste.com/5750/ is that the 'fatal' flags are not actually in that g++ line
05:15:03 <springmeyer> or at least I don't see 'em
05:15:24 <w0lfie_> because 'regular' WMS is one protocol that pretty much all our applications can use
05:15:28 <springmeyer> -dn ?
05:15:36 <wonxly> you're right..maybe it was farther up?
05:15:39 <w0lfie_> i guess we could basically just pre-render tiles then serve them out via WMS
05:15:45 <w0lfie_> hm
05:15:49 <springmeyer> I guess... maybe
05:16:46 <springmeyer> w0lfie_: that is what WMS-C is all about... right?
05:16:54 <wonxly> wow i'm struggling where to put this import statement...i put it right before #### Builds #### and i still get name 'env' is not defined
05:16:57 <w0lfie_> yeah but that requires the client to implement something
05:17:49 <springmeyer> WMC-C? no, just required a WMS request to TileCache which handles dynamically generated the images if it is not cached, or fetching it from the cache if it is already renderd
05:17:56 <springmeyer> required/requires
05:18:02 <w0lfie_> oh
05:18:09 <w0lfie_> i thought you had to implement extensions
05:18:29 <w0lfie_> on the client side
05:18:51 <springmeyer> w0lfie_: well in OpenLayers you don't :)
05:18:56 <w0lfie_> heh
05:19:11 <springmeyer> because its easy to contraint the client to a set of standard scales, etc...
05:19:18 <w0lfie_> well if only had web clients to worry about then that'd be fine
05:19:24 <w0lfie_> but we have a lot of things ingesting data than browsers
05:19:42 <dukeku_> general gis question: what would a plain "geographic projection" be? just lat/lon?
05:19:54 <springmeyer> dukeku_: ya thats the jargon
05:19:58 <dukeku_> sweets
05:20:05 <dukeku_> messing with srtm
05:20:05 <springmeyer> skittles
05:20:11 <springmeyer> you roc
05:20:58 <dukeku_> what's your favorite projection, springmeyer
05:21:30 <w0lfie_> re-reading the WMS-C info.. it looks like it's just a way to serve tiles via WMS
05:21:37 <springmeyer> dukeku_: nudge artem to release his bezier curves if you are doing countours
05:21:39 <w0lfie_> and you need clientside changes
05:21:50 <dukeku_> springmeyer: i absolutely love bezier curves
05:21:52 <w0lfie_> maybe i'm not reading this correctly
05:21:55 <dukeku_> math behind them is totally a+ awesome
05:21:55 <springmeyer> .g mapnik-devel curves
05:21:56 <nikq> springmeyer: http://lists.berlios.de/pipermail/mapnik-devel/2008-January/000367.html
05:22:04 <dukeku_> but i'm going for raster images ;)
05:22:07 <springmeyer> agg does 'em
05:22:10 <springmeyer> oh
05:22:17 <dukeku_> well, i want them for contours as well
05:22:24 <springmeyer> yeh!
05:22:25 <dukeku_> but atm i need a background for basic maps
05:22:41 <dukeku_> i figure elevation is a good start
05:22:42 <springmeyer> dukeku_: you see that patch from martîn?
05:22:56 <dukeku_> on mapnik-devel?
05:23:03 <dukeku_> i think i did
05:23:03 <springmeyer> adds crazy shading symbolizers to rasters
05:23:08 <dukeku_> iirc that's why i'm doing this
05:23:10 <dukeku_> yeah that sounds right :P
05:23:18 <springmeyer> yep
05:23:19 <dukeku_> something sparked me off
05:23:24 <dukeku_> i was just going to use hillshade
05:23:51 <dukeku_> .g mapnik-devel shade
05:23:52 <nikq> dukeku_: http://osdir.com/ml/gis.openstreetmap.devel/2007-01/msg00234.html
05:24:24 <dukeku_> springmeyer: is there any way to fill in gaps in curves?
05:24:29 <dukeku_> i'll grab an example...
05:25:23 <springmeyer> huh? :)
05:25:46 <dukeku_> hold on :)
05:25:56 <dukeku_> i'm sure it's a data issue but i'm holding out hope that there's a way around it
05:26:30 <dukeku_> http://oool.org/breaks.png
05:26:38 <springmeyer> wonxly: svn up to head then put it at the bottom of the configure step
05:26:41 <dukeku_> rendered w/cairo, zoomed in for "maximum effect"
05:26:55 <springmeyer> or anywhere under #416
05:26:56 <nikq> Ticket #416: no such ticket. (list index out of range)
05:26:56 <springmeyer> http://trac.mapnik.org/browser/trunk/SConstruct#L416
05:26:58 <wonxly> ok
05:27:25 <springmeyer> wonxly: also dpaste your pthread so I can add it and lessen the burden on your copy and paste fu :)
05:28:25 <springmeyer> dukeku_: ya so that is a polygon right?
05:28:48 <dukeku_> yeah
05:28:53 <dukeku_> er, no
05:28:55 <dukeku_> it's a line
05:29:01 <dukeku_> that i've made the width gigantic on
05:29:13 <springmeyer> ah, facinating
05:29:17 <springmeyer> so two lines
05:29:21 <dukeku_> yeah
05:29:27 <springmeyer> one massive and yellow/orange
05:29:35 <dukeku_> at that particular point multiple lines are being drawn, but the huge width merges them all ogether
05:29:39 <dukeku_> together
05:29:40 <springmeyer> the other really massive and black?
05:29:46 <dukeku_> yes, slightly wider
05:30:11 <springmeyer> huh, I don't think that mapnik lines should be 'cracking' like that
05:30:33 <springmeyer> seems really odd
05:30:44 <dukeku_> they're roads from the 2008 tiger data
05:30:51 <springmeyer> dukeku_: care to indulge just how big those stroke-widths are? :)
05:31:21 <dukeku_> well, they are being drawn by LineSymbolizer so i'd assume they're lines :)
05:31:58 <dukeku_> springmeyer: http://dpaste.com/5763/
05:32:23 <dukeku_> the image i linked was zoomed in about 300% with acrobat
05:32:59 <dukeku_> but it appears if i don't use cairo and render to png as well
05:34:10 <dukeku_> http://oool.org/bigline.png cropped from the png
05:34:11 <springmeyer> k
05:35:01 <springmeyer> so it would be good to know for sure your geometry type
05:35:08 <springmeyer> do something like...
05:35:09 <w0lfie_> but yeah springmeyer.. heh it's funny. We're actually using ESRI to save money over some other solutions
05:35:12 <w0lfie_> as crazy as that sounds ;)
05:35:47 <springmeyer> $ psql -U mapnik gis
05:35:57 <dukeku_> springmeyer: i got it, hold on :)
05:36:27 <dukeku_> gis=> select distinct ST_GeometryType(the_geom) from tl_oregon_multnomah_edges;
05:36:27 <dukeku_>   st_geometrytype
05:36:27 <dukeku_> --------------------
05:36:27 <dukeku_>  ST_MultiLineString
05:36:27 <dukeku_> (1 row)
05:36:29 <springmeyer> # select GeometryType(the_geom) from tl_oregon_multnomah_edges limit 1;
05:36:45 <springmeyer> okay
05:36:52 <dukeku_> i'm getting on top of my postgis usage
05:37:09 <springmeyer> so they are lines, cool
05:37:13 <wonxly> not sure what I am doing wrong, on line 417 i added import pdb: pdb.set_trace() right after env is set, but when i get the prompt and type env['PLATFORM'] it tells me name 'env' is not defined
05:37:31 <springmeyer> try setting the stroke-caps
05:38:07 <springmeyer> and stroke-linejoin...
05:38:07 <springmeyer> http://trac.mapnik.org/wiki/LineSymbolizer
05:39:11 <springmeyer> import pdb;pdb.set_trace()
05:39:18 <dukeku_> springmeyer: suggested values?
05:39:21 <springmeyer> not ; not :
05:39:26 <springmeyer> er
05:39:30 <springmeyer> note ; not :
05:39:37 <dukeku_> trying round for both
05:39:42 <springmeyer> cool
05:40:12 <springmeyer> wonxly: I'm thinking that the script is failing on a syntax error there perhaps?
05:40:31 <wonxly> yeah sorry, that was a typo here, i have semicolon...it is correctly giving me the (Pdb) prompt
05:40:43 <springmeyer> really? huh
05:40:54 <wonxly> but at the prompt i type env['PLATFORM'] and it doesn't like it...i don't have to use some print command?
05:41:45 <springmeyer> oh, I bet this might be some problem with an older versions of python
05:42:00 <wonxly> http://dpaste.com/5769/
05:42:14 <springmeyer> wonxly: perhaps just stick print statements into SConstruct then...
05:42:17 * springmeyer looks
05:43:51 <dukeku_> springmeyer: still there
05:44:13 <dukeku_> tried with linejoin as round, miter and linecap as butt, round
05:44:21 <springmeyer> k
05:44:44 <springmeyer> dukeku: can you file a ticket?
05:44:49 <dukeku_> sure
05:44:57 <dukeku_> let me try once more with square
05:45:00 <springmeyer> it may be unavoidable but I'm sure others will run into it
05:45:35 <dukeku_> setting it to butt *seems* to make a minor difference but they're still visible
05:45:48 <springmeyer> dukeku_ then after that try something like...
05:46:09 <dukeku_> nope, it's just me
05:46:13 <dukeku_> did a side-by-side comparison
05:46:13 <wonxly> ok the print statements in the script work:  env['PLATFORM' is sunos   and env['CXX'] is g++
05:46:33 <wonxly> does the case matter?  i think that if was lookingn for a capital S
05:46:38 <springmeyer> <Parameter name="table">(select ST_Buffer(the_geom,N) from tl_oregon_multnomah_edges) as newtable</Parameter>
05:47:16 <wonxly> SunOS and looking for cc instead of g++
05:47:26 <springmeyer> wonxly: I'm not following...
05:47:34 <springmeyer> oh, not sure..
05:48:01 <wonxly> i think maybe the SConstruct is not looking for Solaris correctly
05:48:12 <springmeyer> likely
05:48:18 <springmeyer> okay, ya, the case matters
05:48:44 <dukeku_> springmeyer: should i be using some number for N?
05:48:56 <springmeyer> yes, some integer
05:49:03 <springmeyer> mess around with it
05:49:16 <springmeyer> .g Postgis SVN documentation ST_Simplify
05:49:17 <nikq> springmeyer: http://postgis.refractions.net/documentation/manual-svn/ST_Simplify.html
05:49:45 <springmeyer> wonxly: in a python interpreter what do you get with...
05:49:50 <springmeyer> import platform
05:49:58 <springmeyer> print platform.uname()
05:50:46 <wonxly> http://dpaste.com/5773/
05:50:56 <springmeyer> .g Postgis SVN documentation ST_Buffer
05:50:57 <nikq> springmeyer: http://postgis.refractions.net/documentation/manual-svn/ST_Buffer.html
05:50:58 <wonxly> interesting..there it says SunOS
05:51:00 <springmeyer> ... I meant
05:51:28 <dukeku_> springmeyer: playing around with the value isn't doing anything besides making a lot of streets dissapear
05:51:56 <springmeyer> oh, sorry dukeku_: ya buffer will turn your lines into polygons :)
05:52:04 <dukeku_> oh :)
05:52:08 <springmeyer> so you'll have to modify your xml
05:52:24 <dukeku_> so that's a workaround?
05:52:30 <springmeyer> he he
05:52:37 <dukeku_> i'll still file that bug ;)
05:52:55 <springmeyer> yep
05:53:24 <springmeyer> .g Postgis SVN documentation ST_LineMerge
05:53:25 <nikq> springmeyer: http://postgis.refractions.net/documentation/manual-svn/ST_LineMerge.html
05:53:41 <springmeyer> that one might help two and will keep 'em as lines
05:54:28 <dukeku_> springmeyer: mapnik version information would be where?
05:54:47 <springmeyer> nowhere
05:54:51 <dukeku_> darn
05:54:55 <springmeyer> in trunk we've added
05:54:59 <springmeyer> >>> import mapnik
05:55:07 <springmeyer> >>> mapnik.mapnik_version()
05:55:10 <dukeku_> ah
05:55:13 <dukeku_> i should update :)
05:55:18 <dukeku_> i'm on 1/06
05:55:22 <dukeku_> rev781
05:55:23 <springmeyer> but if you knew that you would know the version :)
05:55:30 <springmeyer> ah ah
05:55:38 <dukeku_> doing an svn up
05:55:43 <dukeku_> wow
05:55:45 <dukeku_> 200 revisions
05:55:46 <springmeyer> watch out dude
05:55:58 * springmeyer has been mucking about
05:56:06 <dukeku_> i feel like i'm in the past
05:56:13 <dukeku_> springmeyer: i have faith in trunk
05:56:19 <dukeku_> at least faith that it'll be working eventually :P
05:56:22 <springmeyer> nah you aren't running solaris
05:56:26 <dukeku_> hehe
05:56:27 <dukeku_> sadly no
05:56:28 * springmeyer ducks
05:56:34 <dukeku_> sun never sent me my free solaris dvd
05:56:39 <dukeku_> jerks, not giving me something i never even paid for
05:57:13 <springmeyer> dukeku_: trunks builds scripts behave differently as they now have a 'configure' stage
05:57:20 <springmeyer> .g UsingScons
05:57:20 <nikq> springmeyer: http://www.scons.org/refer.php
05:57:24 <springmeyer> ack
05:57:31 <springmeyer> .g Mapnik wiki UsingScons
05:57:31 <nikq> springmeyer: http://trac.mapnik.org/browser/trunk/INSTALL
05:57:44 <dukeku_> i really should be studying
05:57:45 <dukeku_> oh well
05:57:47 <dukeku_> maps are way more fun
05:57:57 <springmeyer> trac.mapnik.org/wiki/UsingScons
06:01:20 <nikq> Mapnik Trac: Ticket #254 (Large-width lines have "breaks") created | http://trac.mapnik.org/ticket/254
06:01:51 <nikq> Mapnik Trac: breaks.png attached to Ticket #254 | http://trac.mapnik.org/attachment/ticket/254/breaks.png
06:02:11 <nikq> Mapnik Trac: bigline.png attached to Ticket #254 | http://trac.mapnik.org/attachment/ticket/254/bigline.png
06:15:35 <wonxly> i'm confused... lol
06:15:37 <wonxly> http://dpaste.com/5786/
06:16:02 <wonxly> still complaining about these options but i don't see them getting set?
06:18:31 <springmeyer> ugh.
06:18:49 <springmeyer> wonxly: so src/SConscript also have logic in it
06:19:06 <springmeyer> that checks for both SunOs and the Sun CC
06:19:12 <wonxly> yeah i don't think my changing from SunOS to sunos worked either
06:19:12 <springmeyer> vs just g++
06:19:27 <springmeyer> so you'll need to hack at that too
06:19:53 <springmeyer> right, it shouldn't be needed since you are building with g++ you would think
06:20:55 * springmeyer looks and realized that the one time that mapnik was built on solaris...
06:20:56 <springmeyer> http://trac.mapnik.org/ticket/63
06:21:17 <springmeyer> bu jbronn it looks like he did it with the Sun compiler based on that ticket comment
06:22:04 <springmeyer> so wonxly: I think its safe to assume that nobody has yet upgraded the build to work with Solaris + g++
06:25:11 <wonxly> ok, i am going to play around with it some more and see if i can get it working ;)  if i can get the C++ core to compile is that all i need to get it working with mod_tile?
06:25:37 <springmeyer> yes.
06:26:10 <springmeyer> mod_tile has a cool new python rendering daemon but besides that it should work well without python
06:26:48 <springmeyer> wonxly: please post to the list too, likely more compiler gurus there
06:28:45 <wonxly> ok, thanks for your help.  i am hoping to get mod_tile working with apache and then use openlayers ;)
06:30:31 <springmeyer> cool
06:31:36 <nikq> Mapnik Trac: Changeset [994]: scons: fix typo | http://trac.mapnik.org/changeset/994
06:39:42 <dukeku_> what's the default source srs for gdal?
06:40:57 <springmeyer> WGS84 or '+proj=latlong +datum=WGS84'
06:41:15 <springmeyer> thats the default for all layers and the map, no different for gdal
06:41:26 <springmeyer> or are you talking about gdal independent of mapnik?
06:42:04 *** wonxly has quit (Remote closed the connection)
06:42:23 <dukeku_> independent of mapnik :)
06:42:46 <dukeku_> i'm guessing it'd be latlong/WSG84
06:43:01 <springmeyer> then there is no such thing as a default I don't think
06:43:32 <dukeku_> so would a "geographic projection" be just latlong?
06:43:41 <springmeyer> yes
06:43:45 <dukeku_> gotcha
06:46:50 <dukeku_> i'm really starting to like postgis
06:48:22 <CIA-6> mapnik-utils: migurski * r585 /trunk/serverside/cascadenik/ (cascadenik/compile.py test.py tests/test.py): Changed test_combinations() into a recursive function so that the old 15-test limit and excessive memory use could go away
06:56:22 *** migurski (n=migurski@dsl081-049-227.sfo1.dsl.speakeasy.net) has joined #mapnik
07:04:22 *** migurski_ (n=migurski@dsl081-049-227.sfo1.dsl.speakeasy.net) has joined #mapnik
07:11:12 <CIA-6> mapnik-utils: dane.springmeyer * r586 /trunk/nik2img/ (CHANGELOG.txt nik2img.py): Begin the upgrade to mapnik trunk adding support for the new ProjTransform class exposed in python
07:11:17 <springmeyer> migurski: $ nik2img.py -m map.mml -o image.png -p epsg:900913 -e lon,lat,lon,lat
07:11:35 <springmeyer> that would force the map to be in mercator
07:12:02 <springmeyer> and the -e flag expects a bbox group in long/lat
07:12:43 <springmeyer> perhaps you expected lat/long ?
07:12:49 <springmeyer> migurski_
07:12:57 <migurski_> oh hey
07:13:16 <migurski_> no, we definitely did long/lat
07:13:21 <migurski_> and tried the -p thing
07:13:26 <springmeyer> okay
07:13:30 <migurski_> but we were also dealing with extents simultaneously
07:13:32 <springmeyer> the -p thang :)
07:13:38 <migurski_> hehehe
07:13:44 <migurski_> I think there were like five moving pieces all at once
07:14:08 * springmeyer is excited to switch to optparse and drop all the single char flags
07:14:14 <migurski_> mmmm YES
07:14:33 * springmeyer started with those since nik2img was intially meant to be a clone of shp2img
07:14:49 <springmeyer> until I actually started finding it somewhat useful
07:14:54 <migurski_> btw not sure if this interests you but I finally got around to dealing with that crude hack limitation on test counts in cascadenik: http://code.google.com/p/mapnik-utils/source/detail?r=585
07:15:11 <springmeyer> are you calling me crude and old?
07:15:24 <springmeyer> 'old 15-test limit'...
07:16:52 <springmeyer> seriously though, I feel proud to have been hanging with you for a while now that we can refer to the 'past' :)
07:16:53 <migurski_> well the conditions that made it necessary =)
07:16:59 <migurski_> haha
07:17:00 <migurski_> yeah
07:17:00 <springmeyer> he he
07:17:04 <migurski_> "back in the summertime"
07:17:11 <springmeyer> oh oh
07:17:25 * migurski_ cues up some fresh prince
07:17:32 <springmeyer> lol
07:17:53 <springmeyer> anyway, that's excellent :)
07:18:10 <dukeku_> yeah i got on sneaks, but i need a new pair
07:18:38 <migurski_> hehe
07:18:52 <migurski_> so tomorrow is BAUGOS mtg #1 btw
07:18:52 <nikq> Ticket #1: no such ticket. (list index out of range)
07:19:10 <springmeyer> BAUGOS - nice!
07:20:16 <migurski_> yeah jlivni's cuing it up
07:20:30 <springmeyer> love it
07:20:46 <migurski_> I'm trying to figure out how much of this loony printed-OSM experiment I've got on the front burner I can finish by then
07:21:43 <springmeyer> by BAUGOS? ah tons of time :)
07:22:11 <springmeyer> printed-OSM - I should fly down for this one
07:22:57 *** migurski has quit (Read error: 110 (Connection timed out))
07:23:41 <migurski_> it'd be great to have you down here
07:23:49 <migurski_> do you ever visit?
07:24:11 <springmeyer> last time was WhereCamp, and that'll perhaps be the next
07:25:36 <springmeyer> migurski_: not sure about your etech goals but i've got another mapnik-junk type tool thats targeted at easy rendering
07:25:46 <migurski_> oh?
07:26:16 <springmeyer> not that it is ready for anything yet but its with new users in mind who want to run with osm data
07:27:22 <springmeyer> I'm thinking that combined with a chunk of osm in a rtree/sqlite db and TileLite it'll be freeking easy to get a server running
07:27:23 <migurski_> what will it do?
07:27:23 <springmeyer> http://bitbucket.org/springmeyer/tilelite/src/tip/README.txt
07:28:30 <springmeyer> essentially a bridge to something more fancy like TileCache or your 'contraptions' <-- your term :)
07:28:39 <migurski_> =)
07:28:51 <springmeyer> I'd like to think of it as a mapnik gateway drug
07:28:53 <migurski_> hm nice
07:29:08 <migurski_> I was just messing with tilecache today to see how easy it is to drop in a new module
07:29:14 <migurski_> this seems related
07:29:20 <springmeyer> yes yes
07:29:38 <migurski_> tilecache suffers from some of the same goonie frameworkitis as openlayers, but it also gets the job done =)
07:29:46 <springmeyer> my long term thinking is that once I can add full pickle/deepcopy support to mapnik python objects...
07:30:03 <springmeyer> I'll swing back around and know exactly how to add to TileCache
07:30:20 <dodobas> good night springmeyer  :D
07:30:41 <springmeyer> good morning old friend :p
07:30:46 <dodobas> i have 0(zero) exams left
07:30:53 <springmeyer> and with that, I should really go :)
07:31:01 <springmeyer> yahoo! congrats dodobas
07:31:03 <dodobas> only my thesis
07:31:26 <dodobas> Specification and implementation of spatial networks
07:31:27 <springmeyer> migurski_: pickling is actually coming along since we last talked
07:31:32 <springmeyer> cool
07:31:38 <migurski_> I've noticed a number of commit messages
07:31:51 <migurski_> it's exciting, I'd love to see it Just Work™
07:32:25 <springmeyer> I got a 'boost' of inspiration when I realized that boost::python pickling handles both true pickling and also copy/deepcopy support
07:32:45 <migurski_> still frightened of boost
07:33:03 <springmeyer> :)
07:33:42 <springmeyer> hey, we should chat more on srs's sometime soon as well
07:33:50 * springmeyer has gotta roll
07:34:03 <migurski_> okay
07:34:05 <migurski_> have a good night
07:34:09 <migurski_> thanks again for your help
07:34:14 <springmeyer> likewise!
07:34:30 <migurski_> and yeah you should def come to where camp
07:34:54 <springmeyer> looking forward
07:35:09 <springmeyer> seeya
07:35:23 <migurski_> bye!
07:41:23 *** xcacou (n=aga@AToulouse-157-1-70-225.w81-50.abo.wanadoo.fr) has joined #mapnik
08:30:47 *** aba (i=sam@libya.turmzimmer.net) has joined #mapnik
09:10:46 *** migurski_ has quit ()
09:24:41 <dukeku_> osm takes *forever* to put into postgis
10:23:30 *** rcoup has quit ()
11:27:30 *** adakkak_ (n=adakkak@gres110.lis.illinois.edu) has joined #mapnik
12:24:01 <crschmidt> only about 12 hours?
12:24:23 <crschmidt> if it's slower than that, you need better hardware :)
13:05:43 *** kunitoki (n=kraken@host59-41-dynamic.21-79-r.retail.telecomitalia.it) has joined #mapnik
13:44:24 *** sanjiv (n=chatzill@59.180.154.226) has joined #mapnik
13:48:43 *** sanjiv has quit (Read error: 104 (Connection reset by peer))
13:50:38 *** _closure_ is now known as _closure
14:09:03 *** sanjiv (n=chatzill@59.180.154.226) has joined #mapnik
14:48:06 *** D3f0 (n=defo@190.176.247.242) has joined #mapnik
15:36:57 *** xcacou has quit (Remote closed the connection)
15:52:19 *** sanjiv has quit (Remote closed the connection)
16:24:51 <springmeyer> dukeku_: why not just import a small chunk by downloading say oregon from cloudmade?
16:36:04 <aub> noob question... i'm trying to use nik2img to make images centered around a lat/lng point. if i use a wgs84 projection and --zoomto=0,50 it works fine. if i switch to the google projection with the same arguments i get an empty map and lots of nan in the output. any ideas?
16:44:33 <springmeyer> aub: you found a bug :)
16:45:05 <aub> springmeyer: cool! actually it's a mercator vs. latlong issue, i think
16:45:19 <springmeyer> aub: workaround is to specify a zoom level as well
16:45:24 <springmeyer> something like...
16:46:01 <springmeyer> nik2img.py -m yourmap.xml -o yourimage.png -p epsg:900913 --zoomto 0,50,10
16:48:10 <aub> ok, that does seem to improve things. i had tried that, but used level 1, which fails in the same way
16:48:30 <springmeyer> yup, i noticed that too
16:50:40 <aub> springmeyer: much improved.
16:51:14 <springmeyer> aub, you running mapnik trunk?
16:51:43 <aub> springmeyer: as of a couple weeks ago, yes. happy to build latest if that helps
16:53:20 <springmeyer> okay. well just curious since nik2img trunk is only compatible with mapnik trunk at this point
16:54:03 <aub> springmeyer: i'll pull the latest now and check it out
16:54:09 <crschmidt> it's possible that at the whole world level, your projection infor is wrapping or something
16:54:20 <crschmidt> I don't know if that's what '0' means
16:54:28 <crschmidt> or anything about nik2img
16:58:07 <springmeyer> ya, whats happening is a bbox expantion in nik2img that pushes past 90, which won't work with mercator
16:58:23 <springmeyer> just faulty code :)
16:59:25 <springmeyer> then when Mapnik attempts to reproject an Envelope() with any values greater than 180 or 90 it gives back 'inf' as an answer
17:00:07 *** migurski (n=migurski@dsl081-049-227.sfo1.dsl.speakeasy.net) has joined #mapnik
17:10:49 *** migurski has quit ()
17:23:20 *** migurski (n=migurski@dsl081-049-227.sfo1.dsl.speakeasy.net) has joined #mapnik
17:37:46 *** migurski has quit ()
18:14:38 *** migurski (n=migurski@h-68-165-1-62.snvacaid.covad.net) has joined #mapnik
18:41:26 *** rcoup (n=rcoup@ip-118-90-67-87.xdsl.xnet.co.nz) has joined #mapnik
20:05:55 *** D3f0 has quit (Read error: 104 (Connection reset by peer))
20:51:40 *** audifahrer_ (n=andreas@p57AF6D53.dip.t-dialin.net) has joined #mapnik
20:51:43 <audifahrer_> Hello
20:54:43 <audifahrer_> I like to develop a mapnik plugin that reads data from a socket. What is "allowed" in a plugin. I think in respect of platform compatibility. Should I use boost functions to access a socket? Not sure if there's something like that. If not I#ve to work with some #ifdef WIN32 stuff.
20:54:54 *** adakkak_ has quit ("Ex-Chat")
21:00:14 <dukeku_> springmeyer: i'm only importing a box around north america
21:30:34 <CIA-6> mapnik-utils: dane.springmeyer * r587 /trunk/nik2img/ (CHANGELOG.txt nik2img.py): quick fix to max_resolution and zoom levels code
21:36:40 <CIA-6> mapnik-utils: dane.springmeyer * r588 /trunk/nik2img/tests/using_nik2img.txt: Fix state level zoom
21:39:41 <springmeyer> hey dukeku_: okay
21:40:01 <springmeyer> aub: so I just had a moment to try to fix the zoom levels in nik2img
21:40:20 <aub> springmeyer: great, thanks
21:40:23 <springmeyer> I'm not sure who wrote that previous code but I do know they were smoking something :)
21:40:42 <springmeyer> or at least it seems like it
21:40:47 <aub> nice.
21:41:22 <springmeyer> aub: so if you just specify a lon,lat it will now default to zoom level 0
21:41:42 <springmeyer> which should be roughtly equivalent to a google maps zoom level 0 for the world extent
21:42:10 <springmeyer> if you specify a map output size of 256 pixels
21:42:30 <aub> aok, cool. man, i thought i must have been doing something real stupid there :)
21:42:48 <springmeyer> ie with -s 256,256 then the resulting tile should more closely match a given zoom level for a gmaps tile
21:42:55 *** audifahrer_ has quit ("Verlassend")
21:43:17 <springmeyer> naw, I was doing something stupid :)
21:43:36 <aub> well, thanks so much for the help
21:43:43 <springmeyer> and it could likely be improved on, and I've barely tested, and...
21:43:45 <springmeyer> :)
21:43:46 <springmeyer> np
21:50:40 *** rcoup has quit ()
21:54:58 <dukeku_> springmeyer: it's just on a slow old box :)
22:23:54 <dukeku_> does anyone know of an easy way to fill srtm voids?
22:27:28 <nikq> Mapnik Trac: Changeset [995]: + accept empty bboxes for point geom | http://trac.mapnik.org/changeset/995
22:30:40 *** wonxly (n=wonxly@49.165.189.72.cfl.res.rr.com) has joined #mapnik
22:33:18 *** rcoup (n=rcoup@ip-118-90-58-136.xdsl.xnet.co.nz) has joined #mapnik
22:40:25 *** adakkak has quit (Remote closed the connection)
22:53:56 *** wonxly has quit (Remote closed the connection)
23:05:25 *** weizhuo (n=chatzill@nat/yahoo/x-0fd8704318f6cdd1) has joined #mapnik
23:46:25 *** aub has quit ()