Hey,
I'm not an expert,
I would say the first thing would be to create a table with this query,
then try to load the result into qgis.
Another option is to encapsulate your group by into a sub querry, QGis
might be confused by the number of rows.
Try :
SELECT *
FROM (
SELECT row_number() over()
table as plylines, but many (not all) of
them present segments that goes to 0,0 x,y..
Look at this
https://dl.dropboxusercontent.com/u/15885512/strangequeryresult1.jpg
and this
https://dl.dropboxusercontent.com/u/15885512/strangequeryresult1.jpg
2014-03-28 10:21 GMT+01:00 Rémi
SELECT postgis_full_version()
error : no postgis
Cheers,
Rémi-C
2014-03-28 12:14 GMT+01:00 Morten Sickel mor...@sickel.net:
spatial_ref_sys should have been created and populated when postgis was
installed. Double check that that has been done correctly.
Morten
3DCityModeler skrev:
You don't need to do that,
just
'CREATE EXTENSION postgis;'
assuming you have a decently recent postgres and postgis
Cheers,
Rémi-C
2014-03-28 12:26 GMT+01:00 Robert Burgholzer rburg...@vt.edu:
Loojs like postGIS did not get installed. Two tables are created and
populated as part of the
promising :
5 seconds to load everything, build topology, compute face, output faces,
impressive!
I shall have CGAL benchmark soon.
Cheers,
Rémi-C
2014-03-28 16:04 GMT+01:00 Sandro Santilli s...@keybit.net:
On Tue, Mar 25, 2014 at 03:13:13PM +0100, Rémi Cura wrote:
Back after some other
Ok,
thanks for the tip !
An idea about setting multiple band at once?
Cheers,
Rémi-C
2014-03-24 20:05 GMT+01:00 Bborie Park dustym...@gmail.com:
setvalues with a standard array is faster than an array of geomvals.
-bborie
On Mon, Mar 24, 2014 at 11:59 AM, Rémi Cura remi.c...@gmail.com
information from geos (all methods are private).
Cheers,
Rémi-C
2014-03-21 10:40 GMT+01:00 Sandro Santilli s...@keybit.net:
On Thu, Mar 20, 2014 at 06:46:51PM +0100, Rémi Cura wrote:
Yep,
We tried a modified version of AddNode without check (because I already
delete duplicates nodes before calling
Hey,
somebody knows a way to set multiple bands at once?
It is about rasterizing points (into raster).
Currently it is extremely slow (50*50 pixels, 1 band, 300ms using
SetValues, *10 = 3 sec / tile).
Somebody knows if setvalues with straight array is faster than using
array[geom, value]?
, the other 70 sec being for line insertion.
I don't understand your last question, do you mean just fill the table
manually without computing adjacencies?
Cheers,
Rémi-C
2014-03-19 19:08 GMT+01:00 Sandro Santilli s...@keybit.net:
On Wed, Mar 19, 2014 at 06:43:21PM +0100, Rémi Cura wrote
It is a little bit hard to understand what you mean..
As a summary :
_you have lot's of rasters.
_Each one as a position (and a date associated?).
You need to get a raster based on its position (where st_intersects(..))),
but it is not sufficient and you would like too also filter by date.
You
Hey,
We tried several way to load an already topologically correct shapefile
into postgis topology .
10k lines loaded
_Using totopogeom : 750 sec
_Using addlinestring : 246 sec
_Manually using addnode and st_addedgemodface : 86 sec.
Still very slow :-/
Considering manually filling table
Hey,
you can do it with pointcloud (https://github.com/pramsey/pointcloud)
Currently the database I use with 600 millions points is doing fine.
In your case the problem is not : i want to render 500 millions points, but
more : i want to render the adapted number of points given a scale (the
Again:
of course the postgres postgresql.conf file is correctly configured (shared
memory, working memory, etc :
http://www.postgresql.org/docs/9.3/static/runtime-config-resource.html)
It is very well explained in the doc linked...
Cheers,
Rémi-C
2014-03-18 14:28 GMT+01:00 Humberto Cereser
If you _know_ that your data is topological ready (no intersection).
You can fill the topology table way faster using set query (rather than
the actual line by line algorithm).
the first step would be to compute the pair of all connected segments, then
extract from it node, node-segment
Hey,
I am faced with a problem I can't solve :
I have a raster obtained by using st_setvalues and several points.
The raster has 2 bands, half of the pixel being set.
Using the st_pixelaspolygons and the ST_SummaryStats I am sure that some of
the pixels of the raster are set.
Now when I use
1.10.1...
http://trac.osgeo.org/gdal/wiki/frmts_wtkraster.html
On Fri, Mar 7, 2014 at 7:42 AM, Rémi Cura remi.c...@gmail.com wrote:
Hey,
I am faced with a problem I can't solve :
I have a raster obtained by using st_setvalues and several points.
The raster has 2 bands, half of the pixel
http://www.postgresql.org/docs/9.3/static/plpgsql.html
I'm not sure you need plpgsql for this, looks like you could do in sql only
(maybe even no need of CASE WHEN)
Cheers,
Rémi-C
2014-03-05 17:38 GMT+01:00 mane moshref many_ya...@yahoo.com:
Hi all,
I am quite new to pl/pgsql and completely
Hey,
you could use postgres windows function and group by on parcel_id,
coverage_id.
If using Union, overlapping polygons while be merged, and non overlapping
will only be grouped into multi or geom collection.
You might struggle to keep the id.
Cheers,
Rémi-C
2014-03-03 22:52 GMT+01:00 mane
If you now you want to get only polygon, better use
ST_CollectionExtract(...,3),
instead of explicitly getting the type, cheking ...
Cheers,
Rémi-C
2014-02-27 22:50 GMT+01:00 Hugues François hugues.franc...@irstea.fr:
Hello,
Maybe I don't understand your problem but in my mind you can
Maybe you can try to separate raster2pgsql and psql .
First use raster2pgsqland output it to a file on you filesystem.
Then use psql
to execute this file.
You could see how much time each step takes, and how much space the sql
temp file is.
Maybe you could use the -Y flag to make it faster?
not know 'lag'. After play
with that function I have success:
http://pastebin.com/eC4PGq3Q
I have no sure if can appear another combination/sequence of order draw
that can cause a false reverse but, for now, I have no problems...
Em 25-02-2014 14:46, Rémi Cura escreveu:
OK,
first you may
Hey,
I don't understand your problem.
If you have multilinestring, you need to break it to linestring with an id
per multilinestring.
If you have linestring, you just need to keep an id for each line and an id
per point (given in path).
Then you have several option to generate a ordered set of
...
Em 25-02-2014 11:11, Rémi Cura escreveu:
You can use
http://postgis.refractions.net/docs/ST_Reverse.html
Or simply ORDER BY DESC for your path
Cheers,
Rémi-C
2014-02-25 11:49 GMT+01:00 Pedro Costa pedrocostaa...@sapo.pt:
Hi Rémi,
Thanks for your awnser.
My problem
(r1.geom),1) THEN 1
ELSE 0
END AS reverse,
r1.seq AS r1,
r2.seq AS r2,
r3.seq AS r3
FROM path r1, path2 r2, path2 r3
WHERE r1.seq = r2.seq + 1 AND r2.seq = r3.seq + 1 AND r1.seq 0 AND
r2.seq 0
Em 25-02-2014 13:50, Rémi Cura escreveu:
I still don't understand what
Hey,
could you please explain what you mean by Congrats on the availability of
postgis in de postgresql apt reporsitory?
It would be very great,
but I can't find postgis package in
http://apt.postgresql.org/pub/repos/apt/dists/
Thanks,
Rémi-C
2014-02-20 16:45 GMT+01:00 Willy-Bas Loos
So cool ^^
Going to simplify a lot install and upgrades !
Cheers,
Rémi-C
2014-02-20 17:47 GMT+01:00 Willy-Bas Loos willy...@gmail.com:
On Thu, Feb 20, 2014 at 4:45 PM, Willy-Bas Loos willy...@gmail.comwrote:
Since the package names are equal to those in debian and ubuntu
Hey,
short suggestion,
you may want to try to not use a case for perf reasons, in perticular if
it is callen often
You could do something equivalent like this :
floor(rast/61440).
If rast is an int you may even skip the floor (don't have postgres here to
test it).
Cheers,
Rémi-C
2014-02-13
/uploads/75091
Do you have an idea to solve this new problem?
Cheers,
Richard L
Le 10/02/2014 13:56, Rémi Cura a écrit :
simplify
simplify preserve topology
polygonise then filter on area,
opening operator (negativ buffer of X then positiv buffer of X)
...
Cheers,
Rémi-C
2014-02
simplify
simplify preserve topology
polygonise then filter on area,
opening operator (negativ buffer of X then positiv buffer of X)
...
Cheers,
Rémi-C
2014-02-10 13:43 GMT+01:00 Richard LEHAUT richard.leh...@crpf.fr:
Hi
I succeeded to correct interior rings, but I've problem wih interior
Hey,
geometry equality can be defined in many ways.
For your duplication problem, it is a simple postgres problem :
you want that for any couple (line, poly), you have at most one result
(given polygon are convex, which they are if they are squares).
so at the end of you computing you just add
Maybe you could try to expressely set it to corresponding
iso-8859-1 = LATIN1
(instead of latin1)
or maybe IBM850.
If you are on windows, you can just donwload previous version to try with
older binaries.
Cheers,
Rémi-C
2014-02-07 Michael Fricker michael.fric...@altusgroup.com:
Hey,
you use far too big coordinates.
Please consider using st translate on your data, then curve offset, then
inverse translate.
This is mandatory (when using so many digits, you artificially increase the
need for precision in numeric computing, which is problematic).
It is true for all
Oups I put a -15 translation in above example, it also works for -16 of
course like required.
Cheers,
Rémi-C
2014-02-03 Rémi Cura remi.c...@gmail.com:
Hey,
you use far too big coordinates.
Please consider using st translate on your data, then curve offset, then
inverse translate
:
On Mon, Feb 03, 2014 at 10:54:13AM +0100, Rémi Cura wrote:
Hey,
you use far too big coordinates.
Please consider using st translate on your data, then curve offset, then
inverse translate.
Speaking of which, this treatment is one of those performed on catching
an exception in binary
It is possible to do by groupping :
the idea is :
you group small polygons by the coordinate of there centroid, rounded (or
snapped ot grid) so that polygon less than XX meter are in the same group.
Example :
SELECT ST_Union(poly.geom )
FROM my_poly AS poly
GROUP BY
You could use convex or concav envelop.
You can also union the geometries (eventually simplifying by neg buffer
then pos buffer)
That's if you want precise boundary.
If you just need bbox ...
Cheers,
Rémi-C
2014-01-27 Hugues François hugues.franc...@irstea.fr
Hello,
On my side, I often use
Last chance :
this function is a a sloghtly modified topogeom constructor that won't stop
when habving an error.
This way you can import everything that works, and then deal with the
unworking ...
2014/1/15 Sandro Santilli s...@keybit.net
On Tue, Jan 14, 2014 at 04:10:45PM +0100, Ludovic
: %
doing nothing',ageom;
RETURN NULL;
END;
END
$BODY$
LANGUAGE plpgsql VOLATILE STRICT
COST 100;
ALTER FUNCTION public.rc_totopogeomwithouterrors(geometry, character
varying, integer, double precision)
OWNER TO postgres;
2014/1/15 Rémi Cura remi.c...@gmail.com
Last chance :
this function
People answered about finding why postgis_topology is slow :
the 2 suggestions were :
_http://www.postgresql.org/docs/9.2/static/auto-explain.html
_
http://blog.guillaume.lelarge.info/index.php/post/2012/03/31/Profiling-PL/pgsql-functions
First one looks classical and harder
Second one contains
Hey,
you seem to mix 2 different things.
_You can do what you want using the regular postgis function
You would write a trigger on your polygon column, os that on update, you
also update other touching polygons (you would need to do polygon breanking
into lines, which may become difficult if you
Oups,
I figured it would be of no use.
I of course works on pure geometry, raster functions only sees postgis
points and ints.
Here few lines of the point table
qgis_id, st_astext(geom) AS geom,z,reflectance
1 POINT Z (2248.0317003 20680.022999 49.02441796875) 49024 8110
2 POINT Z
Trying to exclude possible sources of errors here :
it tried to convert points to 2D, no sucess.
Cheers,
Rémi-C
2013/12/19 Rémi Cura remi.c...@gmail.com
Oups,
I figured it would be of no use.
I of course works on pure geometry, raster functions only sees postgis
points and ints.
Here
Hey, thanks for the answer, and nice catch !
I'll try to simply buffer points,
then try th eother method to count
About performance :
this is not about plpgsql or C.
I did quit the same processing when importing files of 3 millions points to
split into small 1*1*1 m cubes.
There were a lot of
Buffering make it works.
Cheers,
Rémi-C
2013/12/19 Rémi Cura remi.c...@gmail.com
Hey, thanks for the answer, and nice catch !
I'll try to simply buffer points,
then try th eother method to count
About performance :
this is not about plpgsql or C.
I did quit the same processing when
most likely you havn't !
I don't know your os so answering is difficult.
If you have the size of fb free on list,
you can try a
VACUUM FULL ANALYZE
then try again =)
Other wise, you may have to do you computing piece by piece.
Cheers,
Rémi-C
2013/12/19 isla...@infomed.sld.cu
ok i execute
For the record :
when outputting the image to disk then loading it into postgis, image shows
fine (band 1 and band2)
When 'add canvas' on the image in QGIS, a nex layer is created, but empty.
Do you know other way to do it?
Cheers,
Rémi-C
2013/12/19 Rémi Cura remi.c...@gmail.com
QGis 2.0.1
it is a regular sql command :
use pgadmin or psql
vacuum full analyze;
To execute by pieces, there are many possibilities,
one would be to use pgscript.
The idea is to split your big query into smaller queries.
Cheers,
Rémi-C
2013/12/19 isla...@infomed.sld.cu
Rémi Cura remi.c...@gmail.com
Hey,
using this in sql :
SELECT ST_AsText(ST_GeomFromText(
'GEOMETRYCOLLECTIONM(
POINTM(10 10 100),
LINESTRINGM(0 50 100, 100 50 100),
POLYGONM((25 25 100,75 25 100,75 75 100,25 75 100,25 25 100),(45 45 100,55
45 100,55 55 100,45 55 100,45 45 100))
)'
))
It gives you the expected output.
I would
Hey list,
again noob question.
It seems impossible to create an empty raster with less than one scale
(takes integer !).
Now when I want to update the scale, the number of pixel is not updated :
Is it normal/what should I do?
Cheers,
Rémi-C
DROP TABLE IF EXISTS patch_to_raster;
CREATE TABLE
Juste use one or 2 polygon !
you can use limits on your first query,
or add a where condition with gid=XX, I guess you have a gid field (primary
key).
Cheers,
Rémi-C
2013/12/16 J.Alejandro Martinez Linares isla...@infomed.sld.cu
El 16/12/13 14:38, Rémi Cura escribió:
And of course
Hey,
http://trac.osgeo.org/postgis/wiki/UsersWikiPostgreSQLPostGIS
Cheers,
Rémi-C
2013/12/17 Thomas Schweikle tschwei...@gmail.com
Hi!
I am having some problems with postgis. The first one: which postgis
version is related to which postgresql-version?
Would be nice to have some link.
Hey,
I put some slides about postgis topology data model on the wiki :
http://trac.osgeo.org/postgis/wiki/PostgisTopology_Data_Model
Cheers,
Rémi-C
___
postgis-users mailing list
postgis-users@lists.osgeo.org
something like
(ST_Dump(ST_CollectionExtract(the_geom,3))).geom AS the_geom
)
(you will need to generate an unique ID, see
http://trac.osgeo.org/postgis/wiki/UsersWikiQGIS
row_number() OVER() AS id_qgis
;-) )
2013/12/17 isla...@infomed.sld.cu
Rémi Cura remi.c...@gmail.com escribió:
Juste use one
Not complaining or whatever,
but a truly generic rotation function can be made by using a single
ST_Affine call
ST_Affine(T^-1. R . T) ,
where T is the 4-4 translation matrix (id + translation vector on last
column), and R is a rotation matrix (possibly a composition of rotation
around X,Y,Z. It
...@lists.osgeo.org
[mailto:postgis-users-boun...@lists.osgeo.org] *On Behalf Of *Rémi
Cura
*Sent:* Wednesday, December 11, 2013 10:08 AM
*To:* PostGIS Users Discussion
*Subject:* Re: [postgis-users] Old question resurfacing
I would be very interested to know any attempt to visualize 3D
/IGN_spatial_ref_for_PostGIS
Cheers,
Rémi-C
2013/12/12 Sandro Santilli s...@keybit.net
On Thu, Dec 12, 2013 at 05:55:20PM +0100, Rémi Cura wrote:
an example :
insert into spatial_ref_sys values
(110014000,'IGNF',110014000,'GEOCCS[Ile
I don't think that SRID is a valid one.
Unfortunately I
Seems like If you want :
for each little square, get the id of big square overlaping.
If this is simply this, you don't need the intersection function,
but only the intersects function :
You could do something like this
SELECT ss.id, bs.id , ss.geom
FROM smal_square_table AS ss, big_square_table
,dni10C numeric,
dni11C numeric,dni12C numeric,dniannC numeric,the_geom geometry
Rémi Cura remi.c...@gmail.com escribió:
Seems like If you want :
for each little square, get the id of big square overlaping.
If this is simply this, you don't need the intersection function,
but only
],AUTHORITY[IGNF,REG014]],PRIMEM[Greenwich,0.0,AUTHORITY[IGNF,LGO01]],UNIT[metre,1],AXIS[X,OTHER],AXIS[Y,EAST],AXIS[Z,NORTH],AUTHORITY[IGNF,AMST63]]'
,'+init=IGNF:AMST63');
Cheers,
Rémi-C
2013/12/12 Andreas Laggner andreas.lagg...@ti.bund.de
Am 12.12.2013 17:27, schrieb Rémi Cura:
Hey
, just ask
what do you want to a better understanding. please i need it
Rémi Cura remi.c...@gmail.com escribió:
I'm very sorry it is difficult to understand what you mean.
Maybe with a real drawing and link here to it, or to some screenshots?
;-)
Cheers,
Rémi-C
2013/12/12 isla
I would be very interested to know any attempt to visualize 3D point cloud
from data base !
We did the same but our solution is far from perfect.
Bob, is you rporject public/open source, have you any paper/doc about it ?
Cheers,
Rémi-C
2013/12/11 Basques, Bob (CI-StPaul)
waht would you mean by intersection in 3D space?
volumetric intersection, or 2D intersection of colinear geometries?
Rémi-C
2013/12/12 Mateusz Loskot mate...@loskot.net
http://postgis.net/docs/ST_3DIntersects.html
Mateusz Loskot, http://mateusz. loskot.net
(Sent from mobile, apology for
On a more high level:
I'm guessing you are using the recursive cte to walk trough a graph.
There are ways to accelerate this, but at the cost of increase of memory
usage and duplication of data.
It depends a lot on you graph properties (has it a lots of nodes,
connectivity, centralness, etc)
(of
If your radar return is well discretized in time (I'd say max 10^6
points/sec),
pointcloud could do the trick (you can have any number of attributes per
point).
Else, you would have to cut the data into pieces.
Cheers,
Rémi-C
2013/12/10 George Silva georger.si...@gmail.com
I'm glad this might
I had a bug with this so be a little carefull ;-)
Cheers,
Rémi-C
2013/12/8 Stephen Mather step...@smathermather.com
Awesome. Thank you.
On Sun, Dec 8, 2013 at 12:53 AM, Bborie Park dustym...@gmail.com wrote:
SET postgis.backend = geos;
The docs for postgis.gdal_datapath have
Maybe you could try with a more powerfull langage,
like PL-R.
You would probably find some already existing useful brick for what you
want to do.
Cheers,
Rémi-C
2013/12/5 Martin Davis mtncl...@telus.net
Is there an existing function to do this? No.
Is it easy to write such a function? No,
Or you can use pl-R to use R directly inside DB, which has some pros and
cons
Cheers,
Rémi-C
2013/12/3 Lee Hachadoorian lee.hachadooria...@gmail.com
On Tue, Dec 3, 2013 at 11:50 AM, James David Smith
james.david.sm...@gmail.com wrote:
On 3 December 2013 15:55, Lee Hachadoorian
Bonjour,
on aurait gagné du temps avec cette info dès le début.
Ne cherchez pas plus loin l'erreur est là.
D'abord la force d'une base de donnée réside dans les indexes, sinon pas
d'accélération
Si vous avez un seul polygone, l'index ne sert à rien. Pour qu'il soit
utile il vous faudrait par
Both work fine here (9.3).
Cheers,
Rémi-C
2013/12/3 Christophe Vergon christophevergon-gir...@orange.fr
Hello,
When I write :
CREATE TABLE cadastre.p__edigeo_2012_2b_thread1_com_148_EAD0A01 (idobj
serial,ptrobj integer, refedigeo varchar(255))
it works with an implicit conversion from
Hey all,
somebody has a trick to clean all the topology infrastructure in DB?
I mean ensuring that there is no orphan in relation, that layer
references existing tables, and so on?
Thanks,
Rémi-C
___
postgis-users mailing list
wrecking everything ? ;-)
Cheers,
Rémi-C
2013/12/3 Sandro Santilli s...@keybit.net
On Tue, Dec 03, 2013 at 05:55:54PM +0100, Rémi Cura wrote:
For records :
I add to desactivate trigger on layer to delete manually rows
concerning
deleted schema, then reactivate trigger, then delete from
, 2013 at 06:19:13PM +0100, Rémi Cura wrote:
Yep, but the table with the topogeometry column was deleted,
so no way to use classical function.
We could make the DropGeometryColumn function tolerant
to the fact that the column isn't there anymore while
still having it cleanup topology.layer
Hm what you seem to do is
compute a minimal bouding box (not aligned with NS or EW),
then use the greatest side as a reference, translate it so it passes by the
center of this bbox,
then several times offset the shortest side by a small step in the normal
direction, until you are out of the
Salut Laurent,
on peut imaginer plusieurs approches pour ton problème, en fonction de la
taille de tes données, de la précision voulue, etc.
En résumé tu as un fichier de polygons (combien?) qui représentent un
maillage de la France (département, IRIS, région?).
Et un fichier de point Lidar (2
-resource.html
Rémi-C
Le 2 décembre 2013 19:33, Rémi Cura remi.c...@gmail.com a écrit :
Salut Laurent,
on peut imaginer plusieurs approches pour ton problème, en fonction de la
taille de tes données, de la précision voulue, etc.
En résumé tu as un fichier de polygons (combien?) qui
I don't know what's in you table , but your result may be very big
numberofresult = numberof(province ) * numberof(potentialite_tmp )
You can manually test this by using WITH statement with limits :
WITH province AS (
SELECT *
FROM province
LIMIT 2),
potentialite_tmp AS (
SELECT *
FROM
On 25 November 2013 17:25, Rémi Cura remi.c...@gmail.com wrote:
You could use the default ending (which is well defined),
then split the resulting with the line like endcap=flat (easy to build,
translate endpoint by radius and -radius in normal direction)
I still fail to understand why you
About sharp turns : you could use a small buffer then errosion or straight
skeleton (out of the box with sfcgal).
For the flat encapt I suggest you to use st_split afterward.
Cheers,
Rémi-C
2013/11/26 Mike Toews mwto...@gmail.com
On 26 November 2013 06:04, James David Smith
Does offset curve gives the same result?
(seems lile offsetting both side have same behavior)
Maybe you can try several buffers with increasing size?
(default appears wau before 1000)
Also, can you try to simplify your line : each coordinates uses 15 digits,
surely you don't need all of this !
(with copious commentary) at
http://geospatial.commons.gc.cuny.edu/2013/11/21/finding-islands-or-the-converse/
Best,
--Lee
On Thu, Nov 21, 2013 at 7:28 AM, Rémi Cura remi.c...@gmail.com wrote:
Hey ,
Oops,
vocaulary error
because it's making (2 million)^2 comparisons. - it's cartesian
You may try
http://postgis.net/docs/ST_SharedPaths.html
Cheers,
Rémi-C
2013/11/22 Ani Alamo alamo@gmail.com
I tried this function:
intersects(buffer(LINESTRING1, 0.001), buffer(LINESTRING2, 0.001))
Always gives me false result.
I don't know what to do!
2013/11/22 Ani Alamo
else
decompose line B into points
if any point of B in on line A
return true
else return false
(you chekc point on line with st_intersects)
Cheers,
Rémi-C
2013/11/22 Rémi Cura remi.c...@gmail.com
You may try
http://postgis.net/docs/ST_SharedPaths.html
Cheers,
Rémi-C
2013/11/22 Ani
, Rémi Cura remi.c...@gmail.com wrote:
Maybe you can try the stupidest way to go,
anyway you have to do a inner product because you have to consider each
pair of polygons.
You don't have to do inner product. `a LEFT JOIN b ... WHERE b.gid IS
NULL` is a typical unmatched query. By doing self
Hey,
It's difficult to help if you don't provide an example of
non-working-as-intended-query.
The 2 queries above can't run without 'soils.vw_zones' table, so I can't
reproduce your bug.
Can you try to update all your table with
UPDATE soils.vw_zones SET geometry = ST_SetSRID(geom, XXX);
Also
If you are desperate,
I think you can use OGR directly.
Good luck,
Cheers,
Rémi-C
2013/11/21 Sandro Santilli s...@keybit.net
On Thu, Nov 21, 2013 at 12:59:16PM +0100, Paolo Cavallini wrote:
I have hit the issue:
http://trac.osgeo.org/postgis/ticket/393
on a large file. I have PostGIS
Maybe you can try the stupidest way to go,
anyway you have to do a inner product because you have to consider each
pair of polygons.
CREATE TABLE result AS
SELECT DISTINCT ON (a.gid) a.*
FROM table AS a , table AS b
WHERE ST_Intersects(a.geom, b.geom) = TRUE
AND a.gid != b.gid
Now you can
There is a very big difference between :' i tweak one of your function, so
my tweak must be open source' and 'i use one of your function, so my code
must be open source'.
Unfortunately GPL imposes both.
Cheers,
Rémi-C
2013/11/20 Sandro Santilli s...@keybit.net
On Wed, Nov 20, 2013 at
,
Rémi-C
2013/11/20 Sandro Santilli s...@keybit.net
On Wed, Nov 20, 2013 at 04:32:12PM +0100, Rémi Cura wrote:
There is a very big difference between :' i tweak one of your function,
so
my tweak must be open source' and 'i use one of your function, so my code
must be open source
From what I understand of your needs, Postigs topology was designed for
this.
Cheers,
Rémi-C
2013/11/20 Denis Rouzaud denis.rouz...@gmail.com
Hi all,
I am drawing some multipolygons in QGIS and sometimes, I have parts of
them which are adjacent and I'd like to homogenize them to have less
Message-
From: postgis-users-boun...@lists.osgeo.org [mailto:postgis-users-
boun...@lists.osgeo.org] On Behalf Of Rémi Cura
Sent: Monday, November 18, 2013 11:49 AM
To: PostGIS Users Discussion
Subject: Re: [postgis-users] The first release of the PostGIS Add-ons is
out!
Very
at or near point 420114.82424666081
298567.38722311892
st_isvalid
f
f
t
f
f
f
t
t
(8 rows)
I guess that I should try to fix these geometries first before we
attempt to continue any further right?
Thanks
James
On 13 November 2013 18:04, Rémi Cura
It should be finished by now,
can you check you have geom indexes on :
ukmajrdbuffer.geom and all_lines.the_geom
How many geoms do you have in ukmajrdbuffer?
Cheers,
Rémi-C
2013/11/15 Rémi Cura remi.c...@gmail.com
Hey Sandro,
Thanks for this, it is at least twice faster =)
Cheers
= 8
This is because before I was given the data the roads buffer had
already been dissolved unfortunately.
James
On 15 November 2013 13:43, Rémi Cura remi.c...@gmail.com wrote:
Good !
Something is strange with the result,
you get only 190k lines intersecting road buffer
.
Thanks
James
On 15 November 2013 14:01, Rémi Cura remi.c...@gmail.com wrote:
Outch,
you have only 8 roads in GB? =)
This is not going to go fast till you don't have some dozens k's of
roads.
I'm guessing you are not at will to send those roads?
The next step will be (when
as clSN, cutted_lines_EW AS clEW
WHERE ST_Intersects(clSN.lines_cutted, clEW.lines_cutted)=TRUE --no point
ot compute an intersection if lines don't intersect
)
SELECT row_number() over() AS id , point
FROM points ;
On 15 November 2013 15:42, Rémi Cura remi.c...@gmail.com wrote:
Still
On 11 November 2013 14:51, Rémi Cura remi.c...@gmail.com wrote:
Hey,
the whole point on using a sgbds like postgis is using index.
If you have one line you don't use indexes...
So in short, don't make one polygon with a buffer of all the road, but a
table with a line for the buffer
the lines and the only storing where they intersect and are within the
polygons is a great idea. I'm going to give that a go now.
Thanks again,
James
On 13 November 2013 11:41, Rémi Cura remi.c...@gmail.com wrote:
This would be an improvement but still non efficient.
you have 2
ST_Intersection(east_to_west.the_geom, north_to_south.the_geom)
FROM east_to_west, north_to_south, buff_union_roads
WHERE st_within(buff_union_roads.geom,
ST_Intersection(east_to_west.the_geom, north_to_south.the_geom)) =
TRUE);
Thanks
James
On 13 November 2013 13:33, Rémi Cura remi.c
Hey,
the whole point on using a sgbds like postgis is using index.
If you have one line you don't use indexes...
So in short, don't make one polygon with a buffer of all the road, but a
table with a line for the buffer for every road, then do you computation to
create grid of points inside of
Hey Richard,
I may be completely wrong, but maybe you could use ST_DWithin instead of
Buffer+intersect.
It is usually a good boost in perf, but maybe you absolutely need this
shrinking.
In the same way I'm not sure the way this query is written is the best (you
really need a subquerry here?).
201 - 300 of 346 matches
Mail list logo