#GRASS IRC Log - 2008-05-21

For logs after Feb 3, 2007, all times are GMT-8. Prior logs are GMT-9.
Back to Logs
00:35:55 CIA-10: neteler * r31458 /grass/trunk/ ( 4 files in 3 dirs ): sync'ed with devel_grass6
00:37:00 sigq: grassfeed: Changeset [31458]: sync'ed with devel_grass6 <https://trac.osgeo.org/grass/changeset/31458>
01:56:33 Jazon: hi everyone :- )
01:56:46 : <_wolf_> i
01:56:48 : <_wolf_> hi
01:57:54 Jazon: I have a vector layer consisting of thousands of polygons that defines a catchment area. I want to divide this area evenly by, say, 100, and create a new vector layer of 100 polygons each with a centroid. how can i approach this?
01:58:03 Jazon: ( I also have this layer as a raster if that helps )
01:58:21 Jazon: _wolf_: by the way, that sshfs works like a dream.... :- )
01:58:35 : <_wolf_> glad to hear it : )
01:59:30 : <_wolf_> as for your problem... what do you mean divide evenly?
01:59:44 Jazon: essentially, I have a cropland catchment area defined, and I want to break it up into individual plots of a given area, so I can calculate likely walking routes to and from the plots ( or fields in other words )
02:00:10 : <_wolf_> Jazon: have you thought of working with rasters?
02:00:16 : <_wolf_> instead of vector space
02:00:23 Jazon: yup, I certainly can. I have same data in both formats
02:00:43 Jazon: in fact, it started as a rasster
02:00:58 Jazon: so if i dont need vectors even better
02:01:01 Jazon: saves me a step
02:01:22 : <_wolf_> so is this raster map some kind of a landuse map?
02:01:44 Jazon: yes
02:01:47 Jazon: binary
02:02:43 : <_wolf_> and you want to calculate walking routs around the watery areas?
02:03:03 Jazon: no
02:03:13 ct529: is grass grid ready?
02:03:20 Jazon: I want to see how people are getting to their plots of land
02:03:33 : <_wolf_> ct529: grid is GRASSs strongest point
02:03:49 : <_wolf_> ah, ok, you have a dem also?
02:03:51 ct529: _wolf_: hi wolf : )
02:03:52 Jazon: yes
02:04:07 Jazon: i use r.walk for cost surface
02:04:12 Jazon: r.drain for the walking path
02:04:23 ct529: _wolf_: are all the library thread safe?
02:04:25 : <_wolf_> ah, now I see
02:04:27 Jazon: but I want to do it for each individual plot of land
02:04:29 ct529: s/library/libraries
02:04:33 Jazon: ;- )
02:04:35 : <_wolf_> you want the starting points generated automatically
02:04:40 Jazon: yes
02:04:48 : <_wolf_> hmm
02:05:05 : <_wolf_> ct529: hi. no, sorry GRASS is not thread safe
02:05:22 : <_wolf_> at all
02:05:29 ct529: _wolf_: so, how do you manage running on the grid?
02:05:50 : <_wolf_> ct529: sorry I then misunderstood what you meant by grid. please explain
02:05:50 Jazon: _wolf_: there a couple of modules that are optimized for threading. but i forget waht ones they were
02:06:14 Jazon: ( i think )
02:06:24 : <_wolf_> Jazon: but still in general not thread safe ( at least that seems to be what Glynn preaches )
02:06:31 Jazon: very very true
02:06:41 ct529: _wolf_: when you have a computer grid, like 100 processors with their memory and so on
02:07:10 ct529: _wolf_: and yu run an application over all of them at the same time ( sorry for the ridiculous explanation, but let's make it short )
02:07:16 : <_wolf_> ct529: ah you mean concurrent calculation
02:07:24 : <_wolf_> sorry, no support really
02:07:33 ct529: _wolf_: what grid did you mean?
02:07:34 : <_wolf_> unless you can split your maps
02:07:38 Jazon: _wolf_: i suppose i could just generate the correct number of points and randomly place them within the confines the catchment area
02:08:26 : <_wolf_> Jazon: I think that is maybe the easiest
02:09:14 : <_wolf_> Jazon: if you care to program, you could add support for giving a raster map of starting areas to r.drain
02:09:43 Jazon: that is what I had in mind actually...
02:09:48 ct529: _wolf_: when you said that grid is grass strngest point, what did you mean?
02:09:59 Jazon: grid == raster ct529
02:10:29 Jazon: ( at least i think thats what he meant )
02:10:34 : <_wolf_> ct529: what Jazon said
02:10:37 Jazon: heh
02:11:59 : <_wolf_> Jazon: I'll be happy to test your area mod to r.drain : )
02:25:30 ct529: thanks!
02:25:45 ct529: is there an OS gis that is grid ready?
02:32:03 : * _wolf_ only uses GRASS
02:36:34 ct529: the problem is when you use a shading algorithm
02:36:49 ct529: the new one does take ages if you have a large surface
02:37:10 ct529: or is there a way that I am not aware of?
02:37:21 : <_wolf_> what resolution do you use?
02:40:26 ct529: 25
02:40:32 ct529: sorry, 25m
02:41:13 : <_wolf_> and a huge area? Usually I have seen problems when people by mistake run with a resolution 1m
02:41:53 ct529: _wolf_: yes fairly huge .... GB
02:42:10 : <_wolf_> I suppose you could split the area into n tiles, and concurrently run r.hillshade on the fragments and then merge them together
02:42:42 ct529: _wolf_: that is manual
02:42:50 : <_wolf_> you can script it ; )
02:43:03 : <_wolf_> write once execute a thousand times
02:43:51 : <_wolf_> but you are right it is a workaround or a hack
02:44:42 ct529: _wolf_: strange we have not thought of making grass distributed computing ready from scratch ....
02:45:21 : <_wolf_> In those days distributed computing was not really common yet
02:45:27 : <_wolf_> no fast network etc
02:45:35 ct529: _wolf_: you are right
02:46:30 : <_wolf_> another problem is the fact that GRASS raster maps are decompressed on the fly which prevents random access
02:49:35 ct529: _wolf_: have you ever reverse engineered a .dll?
02:53:01 : <_wolf_> ct529: no, fortunately ; )
02:56:33 ct529: _wolf_: :(
04:41:57 CIA-10: neteler * r31459 /grass/trunk/imagery/i.class/main.c: fix interactive usage problem ( thanks to Glynn )
04:41:58 sigq: grassfeed: Changeset [31459]: fix interactive usage problem ( thanks to Glynn ) <https://trac.osgeo.org/grass/changeset/31459>
04:59:25 CIA-10: epatton * r31460 /grass/trunk/raster/r.in.mat/description.html: Update link to OSGeo bugtracker in manpage
05:00:42 CIA-10: epatton * r31461 /grass/branches/develbranch_6/raster/r.in.mat/description.html: Update link to OSGeo bugtracker in manpage; merge from trunk, r31460.
05:00:43 sigq: grassfeed: Changeset [31461]: Update link to OSGeo bugtracker in manpage; merge from trunk, r31460. <https://trac.osgeo.org/grass/changeset/31461> || Changeset [31460]: Update link to OSGeo bugtracker in manpage <https://trac.osgeo.org/grass/changeset/31460>
05:14:14 CIA-10: neteler * r31462 /grass/branches/develbranch_6/ ( include/gis.h lib/gis/parser.c ): added G_OPT_I_SUBGROUP ( merge from trunk )
05:16:09 CIA-10: neteler * r31463 /grass/trunk/imagery/i.class/main.c: message cosmetics for interactive mode
05:16:10 sigq: grassfeed: Changeset [31463]: message cosmetics for interactive mode <https://trac.osgeo.org/grass/changeset/31463> || Changeset [31462]: added G_OPT_I_SUBGROUP ( merge from trunk ) <https://trac.osgeo.org/grass/changeset/31462>
05:16:31 CIA-10: neteler * r31464 /grass/branches/develbranch_6/imagery/i.class/main.c: message cosmetics for interactive mode ( merge from trunk )
05:17:40 CIA-10: neteler * r31465 /grass/branches/releasebranch_6_3/imagery/i.class/main.c: fix interactive usage problem ( thanks to Glynn ); message cosmetics for interactive mode ( merge from trunk )
05:40:56 CIA-10: neteler * r31466 /grass/trunk/ ( include/gis.h lib/gis/parser.c ): added G_OPT_I_SUBGROUP
05:42:19 sigq: grassfeed: Changeset [31466]: added G_OPT_I_SUBGROUP <https://trac.osgeo.org/grass/changeset/31466>
10:35:12 CIA-10: epatton * r31467 /grass/trunk/raster/r.in.poly/description.html: HTML tag cleaning, cosmetics
10:35:53 sigq: grassfeed: Changeset [31467]: HTML tag cleaning, cosmetics <https://trac.osgeo.org/grass/changeset/31467>
10:36:34 CIA-10: epatton * r31468 /grass/branches/develbranch_6/raster/r.in.poly/description.html: HTML tag cleaning, cosmetics; ( merge from trunk: r31467 )
10:37:30 sigq: grassfeed: Changeset [31468]: HTML tag cleaning, cosmetics; ( merge from trunk: r31467 ) <https://trac.osgeo.org/grass/changeset/31468>
13:16:35 CIA-10: neteler * r31469 /grass-web/trunk/screenshots/vector_61.inc: URL fix
13:18:15 sigq: grassfeed: Changeset [31469]: URL fix <https://trac.osgeo.org/grass/changeset/31469>
13:26:06 doktoreas: anyone have built grass from svn with wxpython interface?
13:26:16 doktoreas: I have got this error: Details: 'NoneType' object has no attribute 'OpenMap' ( /usr/local/lib/libgdi.so: undefined symbol: Py_InitModule4 )
13:26:27 doktoreas: when selecting the vector to edit
13:27:38 : <_wolf_> sorry not in a while. trunk or 6.4?
13:30:23 doktoreas: trunk
13:30:49 doktoreas: but it happen also with 6.3.0 package in Ubuntu
13:31:32 doktoreas: http://www.nabble.com/wxgui-digitizer-td17115219.html#a17115219
13:31:45 doktoreas: seem this
13:32:30 : <_wolf_> ah yes, you need to do some symlinking to get it working
13:32:44 : <_wolf_> at least that is how it was back whenever
13:33:00 doktoreas: the symlink is up, as posted on the README
13:33:25 doktoreas: lrwxrwxrwx 1 root root 64 2008-05-20 18:01 libgdi.so -> /usr/lib/python2.4/site-packages/wx-2.8-gtk2-unicode/wx/_gdi_.so
13:35:22 doktoreas: _wolf_: solved
13:35:28 doktoreas: it linked to a wrong version
13:35:37 : <_wolf_> doktoreas: yay : )
13:35:38 doktoreas: : ) now can go to sleep
13:35:40 doktoreas: bye
13:35:45 : <_wolf_> bye
13:35:54 : * _wolf_ also goes to sleep
16:17:44 WRichard: hi everyone
23:22:59 CIA-10: glynn * r31470 /grass/trunk/raster/r.null/mask.c: Allow setnull=nan
23:23:01 sigq: grassfeed: Changeset [31470]: Allow setnull=nan <https://trac.osgeo.org/grass/changeset/31470>
23:46:20 CIA-10: glynn * r31471 /grass/trunk/general/g.gisenv/main.c: Remove bogus quotes from example usage of set= option
23:46:22 sigq: grassfeed: Changeset [31471]: Remove bogus quotes from example usage of set= option <https://trac.osgeo.org/grass/changeset/31471>
23:48:08 CIA-10: glynn * r31472 /grass/branches/develbranch_6/general/g.gisenv/main.c: Remove bogus quotes from example usage of set= option ( merge r31471 from trunk )