Re: [MapServer-users] breaking changes in branch-7-6 after commit of experimental changes from branch-8-0

2023-05-30 Thread Oliver Christen
before that commit, the following code was working:

STYLE
  COLOR 255 255 255
  SYMBOL "rectangle3"
  ANGLE -45
  SIZE 8
  GAP 10
END

after that commit, it no longer works, as ANGLE -45 is not accepted
anymore, instead it had to be changed to

STYLE
  COLOR 255 255 255
  SYMBOL "rectangle3"
  ANGLE 315
  SIZE 8
  GAP 10
END

the documentation https://mapserver.org/mapfile/style.html say it must be
of type DOUBLE, but double can be negative and nothing say explicitly in
the documentation that the value can not be negative.

I see this discussion here:
https://github.com/MapServer/MapServer/issues/6463
but no comment if something was done or agreed on about this.
A note should be added to the documentation to make that clear, as it was
done for label size (see below).
I can do a PR if that would be acceptable ?

for label size, I see that a note was added saying the value must be
greater than zero on 26 january 2022, so I guess this is on me for not
noticing it (but im not alone not noticing it so I guess it wasnt clearly
stated/announced either)

Best regards
Oliver






On Tue, May 30, 2023 at 1:27 PM Rahkonen Jukka <
jukka.rahko...@maanmittauslaitos.fi> wrote:

> Hi,
>
>
>
> Maybe the commit could be adjusted in Mapserver 7. Out of curiosity, what
> is the use case for zero sized labels and symbols? To disable them, or?
> About the angle, by the Mapserver 7.0 documentation, doesn’t it feel like
> negative angles were not thought to work ever?
>
>
> “Angle, given in degrees, to rotate the symbol (counter clockwise).
> Default is 0 (no rotation). If you have an attribute that specifies angles
> in a clockwise direction (compass direction), you have to adjust the angle
> attribute values before they reach MapServer (360-ANGLE), as it is not
> possible to use a mathematical expression for ANGLE.”
>
>
>
> -Jukka Rahkonen-
>
>
>
> *Lähettäjä:* MapServer-users  
> *Puolesta
> *Oliver Christen
> *Lähetetty:* tiistai 30. toukokuuta 2023 13.59
> *Vastaanottaja:* mapserver-users@lists.osgeo.org
> *Aihe:* [MapServer-users] breaking changes in branch-7-6 after commit of
> experimental changes from branch-8-0
>
>
>
> Dear Mapserver commiters
>
>
>
> this commit there:
>
> https://github.com/MapServer/MapServer/pull/6818
>
>
>
> broke a lot of things,
>
> like style angles not supporting negative values anymore or label size
> mandatorily greater than zero (also possibly for symbol)
>
>
>
> Im not against changes, but changes should be:
>
> a) documented
>
> b) announced with preferably a deprecation period
>
>
>
> so, please revert that commit or provide adequate change/documentation for
> all the breaking changes.
>
>
>
> Thank you
>
> Best regards
>
> Oliver Christen
>
>
>
> --
>
>
>
>
> <https://www.camptocamp.com/en/news-events/camptocamp-flagship-office-bussigny>
>
> *New address from 1st September 2022*
>
>
>
> *Camptocamp SA*
>
> Route de la Chaux 4
>
> 1030 Bussigny
>
> Suisse
>
___
MapServer-users mailing list
MapServer-users@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/mapserver-users


[MapServer-users] breaking changes in branch-7-6 after commit of experimental changes from branch-8-0

2023-05-30 Thread Oliver Christen
Dear Mapserver commiters

this commit there:
https://github.com/MapServer/MapServer/pull/6818

broke a lot of things,
like style angles not supporting negative values anymore or label size
mandatorily greater than zero (also possibly for symbol)

Im not against changes, but changes should be:
a) documented
b) announced with preferably a deprecation period

so, please revert that commit or provide adequate change/documentation for
all the breaking changes.

Thank you
Best regards
Oliver Christen

-- 

<https://www.camptocamp.com/en/news-events/camptocamp-flagship-office-bussigny>

New address from 1st September 2022

Camptocamp SA

Route de la Chaux 4

1030 Bussigny

Suisse
___
MapServer-users mailing list
MapServer-users@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] how to make symbol rotate with the map?

2014-09-04 Thread Oliver Christen

dear all

I need to make a symbol rotate with the map, i.e. if my symbol has an 
angle of 45° and the map is rotated by, let say, 10°, the symbol angle 
should be 55° to keep it's orientation relative to the map itself.


at the moment the only way I see is to dynamically set in my database 
the angle values to be used for the symbol calculated from the angle of 
the map selected by the user (which is totally dynamic, not a configured 
value), and that's horrendously annoying as I would have to dynamically 
set the data value for the given layer so it use the dynamically 
generated angles values.


is there any other way to make a symbol angle follow the map rotation ?

thanks for reading
best regards
Oliver

--
Camptocamp SA
Oliver Christen
PSE A, EPFL
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] how to make symbol rotate with the map?

2014-09-04 Thread Oliver Christen
a bit easier way to do it is to use mapserver substitution to recover 
the map angle and calculate the final angle value directly in the 
postgres data string :


DATA geom FROM (SELECT ..., (%map_angle%::float+angle)*-1 AS angle
FROM table
WHERE condition)
as foo USING UNIQUE id USING srid=21781

...

METADATA
  ...
  default_map_angle 0
  map_angle_validation_pattern [0-9.\-]* # for map angle
END

CLASS
STYLE
SYMBOL 'bla.png'
SIZE 120.0
ANGLE [angle]
OFFSET 0 60.00
END
END

maybe there is an even easier way to do it?


dear all

I need to make a symbol rotate with the map, i.e. if my symbol has an 
angle of 45° and the map is rotated by, let say, 10°, the symbol angle 
should be 55° to keep it's orientation relative to the map itself.


at the moment the only way I see is to dynamically set in my database 
the angle values to be used for the symbol calculated from the angle 
of the map selected by the user (which is totally dynamic, not a 
configured value), and that's horrendously annoying as I would have to 
dynamically set the data value for the given layer so it use the 
dynamically generated angles values.


is there any other way to make a symbol angle follow the map rotation ?

thanks for reading
best regards
Oliver




--
Camptocamp SA
Oliver Christen
PSE A, EPFL
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] shp2img working, mapserv not

2014-07-04 Thread Oliver Christen

open your .dbf file and look at the paths inside (the file is binary but the 
path are readable), the paths must be something mapserver can find based on 
your mapfile SHAPEPATH, otherwise Mapserver will not succeed to access the 
related images.
I would advice using the -write_absolute_path option (if thats not already the 
case)




Hi all,

thanks in advance for any help on this.  I've been trying to create a layer
of tiled geotiffs.  I made a .shp file index with gdaltindex and am
referencing this in the layer.  When I run shp2img locally I get a correct
.png image, but when I try to call mapserv from a web browser with the url:

http://192.168.1.200/cgi-bin/mapserv?map=/var/www/html/topex/srtm.maplayer=SRTM30_plusmode=map

I get the following error:

msDrawMap(): Image handling error. Failed to draw layer named 'SRTM30_Plus'.
msDrawRasterLayerLow(): Unable to access file. Corrupt, empty or missing
file '/var/www/html/topex/../../data/SRTM/srtm30_plus_geotiff/topoN0E0.tiff'
for layer 'SRTM30_Plus'.

I have the same results trying to look at each of the GeoTIFFs individually.
Here is the mapfile I'm working with:

MAP
   NAME Global_Grid
   SIZE 1200 600
   UNITS dd
   EXTENT -180 -84 180 84
   CONFIG MS_ERRORFILE /tmp/ms_error.txt
   DEBUG 5

   LAYER
  NAME SRTM30_Plus
  STATUS DEFAULT
  TYPE RASTER
  TILEINDEX ../../data/SRTM/srtm_index.shp
   END

END

I suspect it may be a permissions problem with the apache server, but I am
currently at a loss.  Any help would be appreciated.


--
Camptocamp SA
Oliver Christen
PSE A, EPFL
CH-1015 Lausanne

www.camptocamp.com

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] symbol rotation follow map rotation ?

2014-06-30 Thread Oliver Christen
I see in the code a function msRotateVectorSymbol but that function 
doesnt seen to be called from anywhere.

is it something only usable from mapscript ?


dear list

Observed using mapserver 6.0.1, I have noticed that symbol rotation 
doesn't seem to follow map rotation, ie. the symbols keep the same 
absolute angle whatever the map angle is.
I have the feeling this is a current limitation of Mapserver, as it 
doesn't know if the symbol angle is an absolute value (independent of 
the map angle) or relative (map angle should be added to symbol angle).


example here:

no map rotation (signs symbols with angle aligning them with the 
streets):
http://map-preprod.morges.ch/preprod/wsgi/mapserv_proxy?BBOX=527784.2%2C151341.33%2C527831.44%2C151402.27HEIGHT=2438WIDTH=1890map_resolution=254TRANSPARENT=trueVERSION=1.1.1FORMAT=image%2FpngSERVICE=WMSformat_options=dpi%3A254LAYERS=vd_cad_tpr_cs_s_coul_grl%2Cvd_cad_tpr_baths_s_coul_grl%2Cvd_cad_tpr_odl_l_coul_grl%2Cmf_cad_no_lieudit_label_coul%2Cmf_cad_ab_nom_localisation_label_coul_grl%2Cpri_25_a%2Cgeb_25_a%2Ceis_25_l_modif%2Cstr_25_l%2Coft_tpr_swissnames%2Cmf_sig_signal_physiqueREQUEST=GetMapSTYLES=SRS=EPSG%3A21781 



with map rotation (signs symbols have kept their initial angle and are 
not aligned anymore):
http://map-preprod.morges.ch/preprod/wsgi/mapserv_proxy?BBOX=527784.2%2C151341.33%2C527831.44%2C151402.27HEIGHT=2438WIDTH=1890map_resolution=254TRANSPARENT=trueVERSION=1.1.1FORMAT=image%2FpngSERVICE=WMSmap_angle=59.99format_options=dpi%3A254LAYERS=vd_cad_tpr_cs_s_coul_grl%2Cvd_cad_tpr_baths_s_coul_grl%2Cvd_cad_tpr_odl_l_coul_grl%2Cmf_cad_no_lieudit_label_coul%2Cmf_cad_ab_nom_localisation_label_coul_grl%2Cpri_25_a%2Cgeb_25_a%2Ceis_25_l_modif%2Cstr_25_l%2Coft_tpr_swissnames%2Cmf_sig_signal_physiqueREQUEST=GetMapSTYLES=angle=59.99SRS=EPSG%3A21781 



could someone confirm or infirm?

In my opinion I would prefer a relative angle calculation by default, 
but that's just me :)


thanks for reading
kind regards
Oliver




--
Camptocamp SA
Oliver Christen
PSE A, EPFL
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] symbol rotation follow map rotation ?

2014-06-27 Thread Oliver Christen

dear list

Observed using mapserver 6.0.1, I have noticed that symbol rotation 
doesn't seem to follow map rotation, ie. the symbols keep the same 
absolute angle whatever the map angle is.
I have the feeling this is a current limitation of Mapserver, as it 
doesn't know if the symbol angle is an absolute value (independent of 
the map angle) or relative (map angle should be added to symbol angle).


example here:

no map rotation (signs symbols with angle aligning them with the streets):
http://map-preprod.morges.ch/preprod/wsgi/mapserv_proxy?BBOX=527784.2%2C151341.33%2C527831.44%2C151402.27HEIGHT=2438WIDTH=1890map_resolution=254TRANSPARENT=trueVERSION=1.1.1FORMAT=image%2FpngSERVICE=WMSformat_options=dpi%3A254LAYERS=vd_cad_tpr_cs_s_coul_grl%2Cvd_cad_tpr_baths_s_coul_grl%2Cvd_cad_tpr_odl_l_coul_grl%2Cmf_cad_no_lieudit_label_coul%2Cmf_cad_ab_nom_localisation_label_coul_grl%2Cpri_25_a%2Cgeb_25_a%2Ceis_25_l_modif%2Cstr_25_l%2Coft_tpr_swissnames%2Cmf_sig_signal_physiqueREQUEST=GetMapSTYLES=SRS=EPSG%3A21781

with map rotation (signs symbols have kept their initial angle and are 
not aligned anymore):
http://map-preprod.morges.ch/preprod/wsgi/mapserv_proxy?BBOX=527784.2%2C151341.33%2C527831.44%2C151402.27HEIGHT=2438WIDTH=1890map_resolution=254TRANSPARENT=trueVERSION=1.1.1FORMAT=image%2FpngSERVICE=WMSmap_angle=59.99format_options=dpi%3A254LAYERS=vd_cad_tpr_cs_s_coul_grl%2Cvd_cad_tpr_baths_s_coul_grl%2Cvd_cad_tpr_odl_l_coul_grl%2Cmf_cad_no_lieudit_label_coul%2Cmf_cad_ab_nom_localisation_label_coul_grl%2Cpri_25_a%2Cgeb_25_a%2Ceis_25_l_modif%2Cstr_25_l%2Coft_tpr_swissnames%2Cmf_sig_signal_physiqueREQUEST=GetMapSTYLES=angle=59.99SRS=EPSG%3A21781 



could someone confirm or infirm?

In my opinion I would prefer a relative angle calculation by default, 
but that's just me :)


thanks for reading
kind regards
Oliver

--
Camptocamp SA
Oliver Christen
PSE A, EPFL
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] geotiff resampling and transparency - artifact problem

2013-04-26 Thread Oliver Christen

Hello Tomas

as far as I can see the tiff are already in RGBA, from what I can see 
from the gdalinfo (see below)


I tried using OFFSITE, but I couldnt see any difference in the output 
with or without it.


In my opinion something is going wrong in the image treatments when a 
resampling is applied, as if the alpha transparency value for the border 
pixels was not preserved during the processing.

but then that's pure speculation.


Warning 1: TIFFReadDirectory:Unknown field with tag 37724 (0x935c) 
encountered

Driver: GTiff/GeoTIFF
Files: 592116.tif
   592116.tfw
Size is 5020, 5020
Coordinate System is `'
Origin = (2591997.8906868,1117002.1093132)
Pixel Size = (0.200,-0.200)
Metadata:
  TIFFTAG_SOFTWARE=Adobe Photoshop CS2 Windows
  TIFFTAG_DATETIME=2013:04:18 14:52:47
  TIFFTAG_XRESOLUTION=72
  TIFFTAG_YRESOLUTION=72
  TIFFTAG_RESOLUTIONUNIT=2 (pixels/inch)
Image Structure Metadata:
  INTERLEAVE=PIXEL
Corner Coordinates:
Upper Left  ( 2591997.900, 1117002.100)
Lower Left  ( 2591997.900, 1115998.100)
Upper Right ( 2593001.900, 1117002.100)
Lower Right ( 2593001.900, 1115998.100)
Center  ( 2592499.900, 1116500.100)
Band 1 Block=5020x1 Type=Byte, ColorInterp=Red
  Mask Flags: PER_DATASET ALPHA
Band 2 Block=5020x1 Type=Byte, ColorInterp=Green
  Mask Flags: PER_DATASET ALPHA
Band 3 Block=5020x1 Type=Byte, ColorInterp=Blue
  Mask Flags: PER_DATASET ALPHA
Band 4 Block=5020x1 Type=Byte, ColorInterp=Alpha


Hi Oliver,

Are you using mapserver's OFFSITE parameter, and/or do your tiff files 
have a nodata pixel set ?
What happens if you convert your tiff to 32bit RGBA? (Not suggesting 
this is a valid workaround, just trying to narrow down the reasons for 
this).


--
thomas


On 18 April 2013 15:27, Oliver Christen 
oliver.chris...@camptocamp.com 
mailto:oliver.chris...@camptocamp.com wrote:


dear all

Im having some issue when applying PROCESSING RESAMPLE=AVERAGE
or RESAMPLE=BILINEAR on some geotiff with transparency.
Either a clear or dark artifact appears on the border of the
image, as seen on these examples:

http://dev.camptocamp.com/files/mapserver/fuzzy_edge_AVERAGE.png
http://dev.camptocamp.com/files/mapserver/fuzzy_edge_BILINEAR.png

if I specify no PROCESSING at all, the transparency is correct
(but not the image quality which I wanted to improve with the
PROCESSING ):
http://dev.camptocamp.com/files/mapserver/fuzzy_edge_NOPROCESSING.png

at first I though the problem may have been caused by a fuzzy edge
on the transparent edges in the geotiff:
http://dev.camptocamp.com/files/mapserver/fuzzy_edge.png
so I retreated one image to have clean edges:
http://dev.camptocamp.com/files/mapserver/clean_edge.png

but I couldnt see much difference in the result:
http://dev.camptocamp.com/files/mapserver/clean_edge_AVERAGE.png
http://dev.camptocamp.com/files/mapserver/clean_edge_BILINEAR.png

there is no visible artifact for all border multipe of 90°
(horizontal and vertical edges), only the non 90° edges show that
problem.

Im not knowledgeable with image manipulation and resampling with
transaprency, so I dont know if this is normal or not.

Any idea if there is some way to solve this problem ?

thank you for your attention

best regards
Oliver



-- 
Camptocamp SA

Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com http://www.camptocamp.com

+41 21 619 10 23 tel:%2B41%2021%20619%2010%2023 (direct)
+41 21 619 10 10 tel:%2B41%2021%20619%2010%2010 (centrale)
+41 21 619 10 00 tel:%2B41%2021%20619%2010%2000 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
mailto:mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users





--
Camptocamp SA
Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] geotiff resampling and transparency - artifact problem

2013-04-18 Thread Oliver Christen

dear all

Im having some issue when applying PROCESSING RESAMPLE=AVERAGE or 
RESAMPLE=BILINEAR on some geotiff with transparency.
Either a clear or dark artifact appears on the border of the image, as 
seen on these examples:


http://dev.camptocamp.com/files/mapserver/fuzzy_edge_AVERAGE.png
http://dev.camptocamp.com/files/mapserver/fuzzy_edge_BILINEAR.png

if I specify no PROCESSING at all, the transparency is correct (but not 
the image quality which I wanted to improve with the PROCESSING ):

http://dev.camptocamp.com/files/mapserver/fuzzy_edge_NOPROCESSING.png

at first I though the problem may have been caused by a fuzzy edge on 
the transparent edges in the geotiff:

http://dev.camptocamp.com/files/mapserver/fuzzy_edge.png
so I retreated one image to have clean edges:
http://dev.camptocamp.com/files/mapserver/clean_edge.png

but I couldnt see much difference in the result:
http://dev.camptocamp.com/files/mapserver/clean_edge_AVERAGE.png
http://dev.camptocamp.com/files/mapserver/clean_edge_BILINEAR.png

there is no visible artifact for all border multipe of 90° (horizontal 
and vertical edges), only the non 90° edges show that problem.


Im not knowledgeable with image manipulation and resampling with 
transaprency, so I dont know if this is normal or not.


Any idea if there is some way to solve this problem ?

thank you for your attention

best regards
Oliver



--
Camptocamp SA
Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] problem with TRUETYPE and GEOMTRANSFORM 'labelpoly': wrong calculation by mapserver ?

2013-02-22 Thread Oliver Christen

Hello Steve
I have created a test case and opened a issue:
https://github.com/mapserver/mapserver/issues/4595

as stated in the issue, it is possible the problem is caused by the 
TRUETYPE font itself, but I dont have the knowledge to debug the 
processing inside mapserver to find out.


best regards
Oliver


Oliver: Any chance you can create a test case and a ticket?

Steve

___

From: mapserver-users-boun...@lists.osgeo.org 
[mapserver-users-boun...@lists.osgeo.org] on behalf of Oliver Christen 
[oliver.chris...@camptocamp.com]
Sent: Thursday, February 21, 2013 8:40 AM
To: mapserver-users@lists.osgeo.org
Subject: [mapserver-users] problem with TRUETYPE and GEOMTRANSFORM 'labelpoly': 
wrong calculation by mapserver ?

hello dear mailinglist

Im having a strange result when using a TRUETYPE symbol used as LABEL,
with a GEOMTRANSFORM 'labelpoly' to give some background color of my symbol,

the background polygon created by the GEOMTRANSFORM 'labelpoly' is
smaller than my symbol, ending in a not so great result, as seen on this
example:

http://dev.camptocamp.com/files/mapserver/test_train_symbol.png

also, the OUTLINECOLOR show some color inside the symbol, something I
dont understand at all.

if someone could possibly tell me how to adjust the labelpoly size and
explain me the OUTLINECOLOR behavior in this case, I would be very grateful.

here is the CLASS used to render that layer:

CLASS
  LABEL
TYPE TRUETYPE
FONT HPMDRG
SIZE 20
ANTIALIAS TRUE
COLOR  0 51 153 # dark blue
OUTLINECOLOR 255 0 0 # red, just for testing, I dont understand
why there is an outline inside the symbol
STYLE
  GEOMTRANSFORM 'labelpoly'
  COLOR 0 255 255 # background cyan rectangle, not the correct size
END
POSITION cc
  END
  TEXT 1
END

Im using mapserver 6.3.0 on debian

thank you for reading this,
best regards
Oliver

--
Camptocamp SA
Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users





--
Camptocamp SA
Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] problem with TRUETYPE and GEOMTRANSFORM 'labelpoly': wrong calculation by mapserver ?

2013-02-21 Thread Oliver Christen

hello dear mailinglist

Im having a strange result when using a TRUETYPE symbol used as LABEL, 
with a GEOMTRANSFORM 'labelpoly' to give some background color of my symbol,


the background polygon created by the GEOMTRANSFORM 'labelpoly' is 
smaller than my symbol, ending in a not so great result, as seen on this 
example:


http://dev.camptocamp.com/files/mapserver/test_train_symbol.png

also, the OUTLINECOLOR show some color inside the symbol, something I 
dont understand at all.


if someone could possibly tell me how to adjust the labelpoly size and 
explain me the OUTLINECOLOR behavior in this case, I would be very grateful.


here is the CLASS used to render that layer:

  CLASS
LABEL
  TYPE TRUETYPE
  FONT HPMDRG
  SIZE 20
  ANTIALIAS TRUE
  COLOR  0 51 153 # dark blue
  OUTLINECOLOR 255 0 0 # red, just for testing, I dont understand 
why there is an outline inside the symbol

  STYLE
GEOMTRANSFORM 'labelpoly'
COLOR 0 255 255 # background cyan rectangle, not the correct size
  END
  POSITION cc
END
TEXT 1
  END

Im using mapserver 6.3.0 on debian

thank you for reading this,
best regards
Oliver

--
Camptocamp SA
Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] Mapserver with postgis connection behind ELB - timeout problem

2013-02-01 Thread Oliver Christen

dear list

Im using Mapserver with postgis layers,
the postgres server is behind an AWS ELB (amazon load balancer) which 
timout connections after 60 seconds or so.
If I configure my layers with PROCESSING CLOSE_CONNECTION=DEFER to 
increase performances, I often get the following error in Mapserver 
after a period of inactivity which seems to correspond to the timeout 
duration of the ELB:


msPostGISRetrieveVersion(): Query error. Error executing SQL: SELECT 
postgis_version()


when this error occur, I effectively see nothing in Postrgres logs, 
meaning there are no new connection attempt.


Now if I remove the CLOSE_CONNECTION=DEFER I dont have any problem anymore.

Is there a way to configure the timeout duration in Mapserver for 
postgis connection so I could set it to be the same as the ELB one?

Also, what is the internal Mapserver default timeout duration ?

thank you for you attention.
best regards
Oliver

--
Camptocamp SA
Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] wfs_encoding usage and problems

2012-10-12 Thread Oliver Christen
Dear list

I need some precision about the usage of the wfs_encoding parameter.

I have tried to set it at a layer level (LAYER  METADATA 
wfs_encoding someencoding), but Mapserver simply ignored it.

It only work if I set it at the MAP level (MAP  WEB  METADATA 
wfs_encoding someencoding)

is this normal?

Im using Mapserver 6.0.1 on RedHat

thank you for any details you can provide.

best regards
Oliver
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] problem using sld stroke-opacity

2012-09-10 Thread Oliver Christen

dear list

I am trying to use sld to stylize the map in a project and Im having 
some problem with stroke-opacity, I dont see any visible change whatever 
value I set.


the sld is there:
http://dev.camptocamp.com/files/mapserver/sld_line_width.xml

I have tried with the mapserver demo too, same result:
http://demo.mapserver.org/cgi-bin/wms?SERVICE=wmsVERSION=1.1.1REQUEST=GetMapLAYERS=country_boundsFORMAT=pngWIDTH=400HEIGHT=300SRS=EPSG:4326STYLES=BBOX=-180.0,-90.0,180.0,90.0TRANSPARENT=TRUESLD=http://dev.camptocamp.com/files/mapserver/sld_line_width.xml

did I do something wrong or forgot to set something in the sld to have 
it to work correctly?


Im using mapserver 6.0.3

thank you for reading, any help will be greatly appreciated

best regards
Oliver

--
Camptocamp SA
Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] problem using sld stroke-opacity

2012-09-10 Thread Oliver Christen

ah nevemind, I found the related bug entry:
https://github.com/mapserver/mapserver/issues/4132

so it would be ok in the next 6.2 release :)


dear list

I am trying to use sld to stylize the map in a project and Im having 
some problem with stroke-opacity, I dont see any visible change 
whatever value I set.


the sld is there:
http://dev.camptocamp.com/files/mapserver/sld_line_width.xml

I have tried with the mapserver demo too, same result:
http://demo.mapserver.org/cgi-bin/wms?SERVICE=wmsVERSION=1.1.1REQUEST=GetMapLAYERS=country_boundsFORMAT=pngWIDTH=400HEIGHT=300SRS=EPSG:4326STYLES=BBOX=-180.0,-90.0,180.0,90.0TRANSPARENT=TRUESLD=http://dev.camptocamp.com/files/mapserver/sld_line_width.xml 



did I do something wrong or forgot to set something in the sld to have 
it to work correctly?


Im using mapserver 6.0.3

thank you for reading, any help will be greatly appreciated

best regards
Oliver




--
Camptocamp SA
Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] mapserver crash on wfs query, looks like issue 4166, could someone confirm ?

2012-07-18 Thread Oliver Christen
hi Steve

sorry, I forgot to say, yes it is a bbox query:
http://dev.camptocamp.com/files/mapserver/query.xml

Oliver

On Tue, Jul 17, 2012 at 6:59 PM, Lime, Steve D (DNR)
steve.l...@state.mn.us wrote:
 Just a bbox query like the ticket referenced?



 Steve



 From: mapserver-users-boun...@lists.osgeo.org
 [mailto:mapserver-users-boun...@lists.osgeo.org] On Behalf Of Oliver
 Christen
 Sent: Tuesday, July 17, 2012 7:09 AM
 To: mapserver-users@lists.osgeo.org
 Subject: [mapserver-users] mapserver crash on wfs query, looks like issue
 4166, could someone confirm ?



 dear all

 I stumbled upon a Mapserver problem, which looks like the existing bug
 https://github.com/mapserver/mapserver/issues/4166
 but im not exactly sure,
 I would really appreciate if someone could confirm (or not) if that's what
 happening in this case.

 Here is the problem description:

 I have two Shapefile layers.

 When I do a wfs query on both of them, Mapserver crash.
 (Strangely doing a wfs query on each of them, alone, gives a normal result)

 As proposed in the bug description, I replaced the DATA by CONNECTIONTYPE
 OGR/CONNECTION and the problem was solved.

 This problem could be reproduced with Mapserver 6.0.0 and 6.0.3

 I dont really know how to do a WFS query on mapserver using commandline with
 gdb so I only have the raw stacktrace:
 http://dev.camptocamp.com/files/mapserver/raw_backtrace.txt

 thanks in advance for any info
 best regards
 Oliver

 --

 Camptocamp SA

 Oliver Christen

 PSE A

 CH-1015 Lausanne



 www.camptocamp.com



 +41 21 619 10 23 (direct)

 +41 21 619 10 10 (centrale)

 +41 21 619 10 00 (fax)
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] mapserver crash on wfs query, looks like issue 4166, could someone confirm ?

2012-07-17 Thread Oliver Christen

|dear all

I stumbled upon a Mapserver problem, which looks like the existing bug
https://github.com/mapserver/mapserver/issues/4166
but im not exactly sure,
I would really appreciate if someone could confirm (or not) if that's 
what happening in this case.


Here is the problem description:

I have two Shapefile layers.

When I do a wfs query on both of them, Mapserver crash.
(Strangely doing a wfs query on each of them, alone, gives a normal result)

As proposed in the bug description, I replaced the DATA by 
CONNECTIONTYPE OGR/CONNECTION and the problem was solved.


This problem could be reproduced with Mapserver 6.0.0 and 6.0.3

I dont really know how to do a WFS query on mapserver using commandline 
with gdb so I only have the raw stacktrace:

http://dev.camptocamp.com/files/mapserver/raw_backtrace.txt

thanks in advance for any info
best regards
Oliver
|

--
Camptocamp SA
Oliver Christen
PSE A
CH-1015 Lausanne

www.camptocamp.com

+41 21 619 10 23 (direct)
+41 21 619 10 10 (centrale)
+41 21 619 10 00 (fax)

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] GetLegendGraphic legend image generated as line for polygon without color

2012-03-07 Thread Oliver Christen
dear all

Im trying to find a way to tell Mapserver to render the legend icon
for a polygon layer with no color (transparent) (only outlinecolor) as
a polygon legend (rectangle) and not as a line legend (zig-zag line)

see below for my layer definition

the currently generated legend image is:
http://dev.camptocamp.com/files/mapserver/legend.png

I have tried to specify a
COLOR -1 -1 -1
to force Mapserver to handle the legend image as a polygon (rectangle)
without success.

any idea if there is a trick to force Mapserver to render the legend
as a rectangle and not as a zig-zag line ?

thanks in advance
Oliver

 LAYER
   NAME movd_cad_tpr_prf_s
   METADATA
 wms_title SitCdParcellaireLineDDPtt2_
   END
   TYPE POLYGON
   DATA MOVD_CAD/MOVD_CAD_TPR_PRF_S
   STATUS ON

   MAXSCALE 10001

   CLASS
 NAME Prf
 STYLE
   OUTLINECOLOR 255 0 0
   WIDTH 3
 END
   END
 END
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] GetLegendGraphic legend image generated as line for polygon without color

2012-03-07 Thread Oliver Christen
excellent!
this works perfectly
thank you very much

On Wed, Mar 7, 2012 at 1:35 PM, thomas bonfort thomas.bonf...@gmail.com wrote:
 c.f. http://trac.osgeo.org/mapserver/ticket/4104

 here's an ugly hack that *should* work (not tested):

 CLASS
  STYLE
    COLOR 0 0 0
    OPACITY 0
  END
  STYLE
    OUTLINECOLOR 255 0 0
    WIDTH 3
  END
 END

 On Wed, Mar 7, 2012 at 13:26, Oliver Christen
 oliver.chris...@camptocamp.com wrote:
 dear all

 Im trying to find a way to tell Mapserver to render the legend icon
 for a polygon layer with no color (transparent) (only outlinecolor) as
 a polygon legend (rectangle) and not as a line legend (zig-zag line)

 see below for my layer definition

 the currently generated legend image is:
 http://dev.camptocamp.com/files/mapserver/legend.png

 I have tried to specify a
 COLOR -1 -1 -1
 to force Mapserver to handle the legend image as a polygon (rectangle)
 without success.

 any idea if there is a trick to force Mapserver to render the legend
 as a rectangle and not as a zig-zag line ?

 thanks in advance
 Oliver

  LAYER
   NAME movd_cad_tpr_prf_s
   METADATA
     wms_title SitCdParcellaireLineDDPtt2_
   END
   TYPE POLYGON
   DATA MOVD_CAD/MOVD_CAD_TPR_PRF_S
   STATUS ON

   MAXSCALE 10001

   CLASS
     NAME Prf
     STYLE
       OUTLINECOLOR 255 0 0
       WIDTH 3
     END
   END
  END
 ___
 mapserver-users mailing list
 mapserver-users@lists.osgeo.org
 http://lists.osgeo.org/mailman/listinfo/mapserver-users
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] crash mapserver with some specific jpeg output and truetype symbol over raster

2011-11-16 Thread Oliver Christen
thank you Steve
I will do it

best regards
Oliver

On Tue, Nov 15, 2011 at 11:48 PM, Lime, Steve D (DNR)
steve.l...@state.mn.us wrote:
 This is enough to open a bug, can you do that? -Steve

 -Original Message-
 From: mapserver-users-boun...@lists.osgeo.org 
 [mailto:mapserver-users-boun...@lists.osgeo.org] On Behalf Of Oliver Christen
 Sent: Tuesday, November 15, 2011 7:15 AM
 To: mapserver-users@lists.osgeo.org
 Subject: [mapserver-users] crash mapserver with some specific jpeg output and 
 truetype symbol over raster

 hi all

 Im experimenting some weird behavior with some jpeg output on a
 Mapserver 6.0.0

 I have two layers: a raster layer and a watermark using a truetype
 symbol over it.
 When requesting the jpeg output, at the very specific FORMATOPTION
 QUALITY=88, Mapserver crash with a Floating point exception.

 The image is still generated (???), but the error make it impossible to
 server the image with fastcgi, as the Mapserver error trigger an error
 500 in apache.

 I DONT have any problem if:
 1) I request a png image
 2) I change the FORMATOPTION QUALITY=88 to 87 or 89

 I have prepared a minimal testcase which allow to reproduce the problem:

 http://publicshare.camptocamp.com/mapserver/testcase.zip

 which can be triggered with:

 shp2img -m testcase.map -l ch.swisstopo.pixelkarte-farbe-pk200.noscale
 ch.swisstopo.watermark -i jpeg -s 256 256 -e 624800.0 132400.0 625440.0
 133040.0  test.jpg

 or:

 /usr/lib/cgi-bin/mapserv -nh
 QUERY_STRING=map=testcase.mapmode=maplayers=ch.swisstopo.pixelkarte-farbe-pk200.noscale+ch.swisstopo.watermarkmapext=624800.0+132400.0+625440.0+133040.0mapsize=256+256
   test2.jpeg

 I just had a look using gdb (but I dont have enough knowledge to get any
 info from this sadely):

 gdb /usr/lib/cgi-bin/mapserv
 GNU gdb (GDB) 7.0.1-debian
 Copyright (C) 2009 Free Software Foundation, Inc.
 License GPLv3+: GNU GPL version 3 or later
 http://gnu.org/licenses/gpl.html
 This is free software: you are free to change and redistribute it.
 There is NO WARRANTY, to the extent permitted by law. Type show copying
 and show warranty for details.
 This GDB was configured as x86_64-linux-gnu.
 For bug reporting instructions, please see:
 http://www.gnu.org/software/gdb/bugs/...
 Reading symbols from /usr/lib/cgi-bin/mapserv...(no debugging symbols
 found)...done.
 (gdb) run -nh
 QUERY_STRING=map=testcase.mapmode=maplayers=ch.swisstopo.pixelkarte-farbe-pk200.noscale+ch.swisstopo.watermarkmapext=624800.0+132400.0+625440.0+133040.0mapsize=256+256
   test2.jpeg
 Starting program: /usr/lib/cgi-bin/mapserv -nh
 QUERY_STRING=map=testcase.mapmode=maplayers=ch.swisstopo.pixelkarte-farbe-pk200.noscale+ch.swisstopo.watermarkmapext=624800.0+132400.0+625440.0+133040.0mapsize=256+256
   test20.jpeg
 [Thread debugging using libthread_db enabled]

 Program received signal SIGFPE, Arithmetic exception.
 0x004e20dd in ?? ()
 (gdb) backtrace
 #0 0x004e20dd in ?? ()
 #1 0x0051401e in jpeg_stream_term_destination ()
 #2 0x774dac6f in jpeg_finish_compress () from /usr/lib/libjpeg.so.62
 #3 0x005144af in ?? ()
 #4 0x005168ea in ?? ()
 #5 0x00454fbf in ?? ()
 #6 0x0041e1bb in main ()

 I had a look at Mapserver 6.0.1 changelog but I couldnt find a bug which
 may be similar.
 Is this a known issue?
 Could someone tell me if it is enough to open a bug?

 Thanks in advance
 best regards
 Oliver
 ___
 mapserver-users mailing list
 mapserver-users@lists.osgeo.org
 http://lists.osgeo.org/mailman/listinfo/mapserver-users



___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] crash mapserver with some specific jpeg output and truetype symbol over raster

2011-11-16 Thread Oliver Christen
I have opened the bug, I hope I did it right ^_^

http://trac.osgeo.org/mapserver/ticket/4084

On Wed, Nov 16, 2011 at 10:15 AM, Oliver Christen
oliver.chris...@camptocamp.com wrote:
 thank you Steve
 I will do it

 best regards
 Oliver

 On Tue, Nov 15, 2011 at 11:48 PM, Lime, Steve D (DNR)
 steve.l...@state.mn.us wrote:
 This is enough to open a bug, can you do that? -Steve

 -Original Message-
 From: mapserver-users-boun...@lists.osgeo.org 
 [mailto:mapserver-users-boun...@lists.osgeo.org] On Behalf Of Oliver Christen
 Sent: Tuesday, November 15, 2011 7:15 AM
 To: mapserver-users@lists.osgeo.org
 Subject: [mapserver-users] crash mapserver with some specific jpeg output 
 and truetype symbol over raster

 hi all

 Im experimenting some weird behavior with some jpeg output on a
 Mapserver 6.0.0

 I have two layers: a raster layer and a watermark using a truetype
 symbol over it.
 When requesting the jpeg output, at the very specific FORMATOPTION
 QUALITY=88, Mapserver crash with a Floating point exception.

 The image is still generated (???), but the error make it impossible to
 server the image with fastcgi, as the Mapserver error trigger an error
 500 in apache.

 I DONT have any problem if:
 1) I request a png image
 2) I change the FORMATOPTION QUALITY=88 to 87 or 89

 I have prepared a minimal testcase which allow to reproduce the problem:

 http://publicshare.camptocamp.com/mapserver/testcase.zip

 which can be triggered with:

 shp2img -m testcase.map -l ch.swisstopo.pixelkarte-farbe-pk200.noscale
 ch.swisstopo.watermark -i jpeg -s 256 256 -e 624800.0 132400.0 625440.0
 133040.0  test.jpg

 or:

 /usr/lib/cgi-bin/mapserv -nh
 QUERY_STRING=map=testcase.mapmode=maplayers=ch.swisstopo.pixelkarte-farbe-pk200.noscale+ch.swisstopo.watermarkmapext=624800.0+132400.0+625440.0+133040.0mapsize=256+256
   test2.jpeg

 I just had a look using gdb (but I dont have enough knowledge to get any
 info from this sadely):

 gdb /usr/lib/cgi-bin/mapserv
 GNU gdb (GDB) 7.0.1-debian
 Copyright (C) 2009 Free Software Foundation, Inc.
 License GPLv3+: GNU GPL version 3 or later
 http://gnu.org/licenses/gpl.html
 This is free software: you are free to change and redistribute it.
 There is NO WARRANTY, to the extent permitted by law. Type show copying
 and show warranty for details.
 This GDB was configured as x86_64-linux-gnu.
 For bug reporting instructions, please see:
 http://www.gnu.org/software/gdb/bugs/...
 Reading symbols from /usr/lib/cgi-bin/mapserv...(no debugging symbols
 found)...done.
 (gdb) run -nh
 QUERY_STRING=map=testcase.mapmode=maplayers=ch.swisstopo.pixelkarte-farbe-pk200.noscale+ch.swisstopo.watermarkmapext=624800.0+132400.0+625440.0+133040.0mapsize=256+256
   test2.jpeg
 Starting program: /usr/lib/cgi-bin/mapserv -nh
 QUERY_STRING=map=testcase.mapmode=maplayers=ch.swisstopo.pixelkarte-farbe-pk200.noscale+ch.swisstopo.watermarkmapext=624800.0+132400.0+625440.0+133040.0mapsize=256+256
   test20.jpeg
 [Thread debugging using libthread_db enabled]

 Program received signal SIGFPE, Arithmetic exception.
 0x004e20dd in ?? ()
 (gdb) backtrace
 #0 0x004e20dd in ?? ()
 #1 0x0051401e in jpeg_stream_term_destination ()
 #2 0x774dac6f in jpeg_finish_compress () from /usr/lib/libjpeg.so.62
 #3 0x005144af in ?? ()
 #4 0x005168ea in ?? ()
 #5 0x00454fbf in ?? ()
 #6 0x0041e1bb in main ()

 I had a look at Mapserver 6.0.1 changelog but I couldnt find a bug which
 may be similar.
 Is this a known issue?
 Could someone tell me if it is enough to open a bug?

 Thanks in advance
 best regards
 Oliver
 ___
 mapserver-users mailing list
 mapserver-users@lists.osgeo.org
 http://lists.osgeo.org/mailman/listinfo/mapserver-users




___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] crash mapserver with some specific jpeg output and truetype symbol over raster

2011-11-15 Thread Oliver Christen

hi all

Im experimenting some weird behavior with some jpeg output on a 
Mapserver 6.0.0


I have two layers: a raster layer and a watermark using a truetype 
symbol over it.
When requesting the jpeg output, at the very specific FORMATOPTION 
QUALITY=88, Mapserver crash with a Floating point exception.


The image is still generated (???), but the error make it impossible to 
server the image with fastcgi, as the Mapserver error trigger an error 
500 in apache.


I DONT have any problem if:
1) I request a png image
2) I change the FORMATOPTION QUALITY=88 to 87 or 89

I have prepared a minimal testcase which allow to reproduce the problem:

http://publicshare.camptocamp.com/mapserver/testcase.zip

which can be triggered with:

shp2img -m testcase.map -l ch.swisstopo.pixelkarte-farbe-pk200.noscale 
ch.swisstopo.watermark -i jpeg -s 256 256 -e 624800.0 132400.0 625440.0 
133040.0  test.jpg


or:

/usr/lib/cgi-bin/mapserv -nh 
QUERY_STRING=map=testcase.mapmode=maplayers=ch.swisstopo.pixelkarte-farbe-pk200.noscale+ch.swisstopo.watermarkmapext=624800.0+132400.0+625440.0+133040.0mapsize=256+256 
 test2.jpeg


I just had a look using gdb (but I dont have enough knowledge to get any 
info from this sadely):


gdb /usr/lib/cgi-bin/mapserv
GNU gdb (GDB) 7.0.1-debian
Copyright (C) 2009 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later 
http://gnu.org/licenses/gpl.html

This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law. Type show copying
and show warranty for details.
This GDB was configured as x86_64-linux-gnu.
For bug reporting instructions, please see:
http://www.gnu.org/software/gdb/bugs/...
Reading symbols from /usr/lib/cgi-bin/mapserv...(no debugging symbols 
found)...done.
(gdb) run -nh 
QUERY_STRING=map=testcase.mapmode=maplayers=ch.swisstopo.pixelkarte-farbe-pk200.noscale+ch.swisstopo.watermarkmapext=624800.0+132400.0+625440.0+133040.0mapsize=256+256 
 test2.jpeg
Starting program: /usr/lib/cgi-bin/mapserv -nh 
QUERY_STRING=map=testcase.mapmode=maplayers=ch.swisstopo.pixelkarte-farbe-pk200.noscale+ch.swisstopo.watermarkmapext=624800.0+132400.0+625440.0+133040.0mapsize=256+256 
 test20.jpeg

[Thread debugging using libthread_db enabled]

Program received signal SIGFPE, Arithmetic exception.
0x004e20dd in ?? ()
(gdb) backtrace
#0 0x004e20dd in ?? ()
#1 0x0051401e in jpeg_stream_term_destination ()
#2 0x774dac6f in jpeg_finish_compress () from /usr/lib/libjpeg.so.62
#3 0x005144af in ?? ()
#4 0x005168ea in ?? ()
#5 0x00454fbf in ?? ()
#6 0x0041e1bb in main ()

I had a look at Mapserver 6.0.1 changelog but I couldnt find a bug which 
may be similar.

Is this a known issue?
Could someone tell me if it is enough to open a bug?

Thanks in advance
best regards
Oliver
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] wrong tileindex path make mapserver 6 crash

2011-08-30 Thread Oliver Christen

hello
Im just wondering if it was normal for mapserver 6 to crash instead of 
returning an error message if a tileindex path was incorrect for a 
raster layer ?


Im using 6.0.0-1

I have a raster layer using a tileindex (ortho/ortho.*):

LAYER
NAME 'ortho'
TYPE RASTER
STATUS ON
TILEINDEX ortho/ortho
TILEITEM LOCATION
END

at first the TILEINDEX was wrong set to ortho, so effectively pointing 
to the folder instead of the files, which is of course wrong, but I 
would have expected an error message, not a crash.


all I have in my error log is:
mod_fcgid: error reading data, FastCGI server closed connection
Premature end of script headers: mapserv

I tried to turn on the debug with:
CONFIG MS_ERRORFILE /tmp/ms_error.txt
DEBUG 5
CONFIG CPL_DEBUG ON

but ms_error.txt is empty

is there any way to be able to obtain some error message which may help 
knowing where the error is?

would this problem be fixed in 6.0.1?

thanks in advance
best regards
Oliver
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] Mapserver WMS server, IIS and wrong content-type

2011-07-04 Thread Oliver Christen

Hi Jeff

that would make some sens, but would IIS recognise the extension type 
when the url requested is a dynamic wms url ?
ie. would IIS use the filename returned by Mapserver to match the 
extension ?


O.


Hello Oliver,

Quick guess: maybe that IIS instance needs to be configured for that 
MIME type 
(http://technet.microsoft.com/en-us/library/cc786786%28WS.10%29.aspx).


-jeff




___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] Mapserver WMS server, IIS and wrong content-type

2011-07-04 Thread Oliver Christen

Hi Jeff,
I do not have my hand on that server so I cant test it myself, I was 
only relaying the question.

I did read the article but I failed to see how it would solve the problem.
I will relay that info to the server admin and see if it works.

thank you
best regards
Oliver


Hello Oliver,

It can't hurt to try the steps outlined in that link I provided.  It 
would answer the questions you posed in your last email.


-jeff




___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] line width unit ignored after mapserver upgrade

2010-03-01 Thread Oliver Christen

hello

I recently upgraded mapserver from 5.4 to 5.6.1 and all my line style are 
computed in meter (I believe) instead of pixel.


here is the result im getting:
http://dev.camptocamp.com/~ochriste/wms_line.png

here is my layer definition:

LAYER
 NAME SitPrProjetConsLine
 GROUP GrpConstruction
 METADATA
   wms_title SitPrProjetConsLine
 END
 TYPE LINE
 DATA SIT_PR_Projet_cons_line
 STATUS ON

 DEBUG 5

 MAXSCALE 10001

 SIZEUNITS pixels

 CLASS
   STYLE
 COLOR 0 0 255
 WIDTH 1
 MAXWIDTH 1
 MINWIDTH 1
   END
 END
END


here is the query I do:

?LAYERS=SitPrProjetConsLineFORMAT=image%2FpngSERVICE=WMSVERSION=1.1.1REQUEST=GetMapSTYLES=EXCEPTIONS=application%2Fvnd.ogc.se_inimageSRS=EPSG%3A21781BBOX=506263.99910144,143335.99917056,506391.99903232,143463.99910144WIDTH=256HEIGHT=256

I want a line width of 1px
What am I doing wrong ?

thanks in advance
Oliver 


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] line width unit ignored after mapserver upgrade

2010-03-01 Thread Oliver Christen

Hi Bart

that was it!

my map has RESOLUTION 254
so adding a DEFRESOLUTION 254 solved the problem

thank you very much

best regards
Oliver

- Original Message - 


Hi,

is your MAP-RESOLUTION different from 72? If so, set DEFRESOLUTION to your 
value as well.


Best regards,
Bart

On Mar 1, 2010, at 9:32 AM, Oliver Christen wrote:


hello

I recently upgraded mapserver from 5.4 to 5.6.1 and all my line style are 
computed in meter (I believe) instead of pixel.


here is the result im getting:
http://dev.camptocamp.com/~ochriste/wms_line.png

here is my layer definition:

LAYER
NAME SitPrProjetConsLine
GROUP GrpConstruction
METADATA
  wms_title SitPrProjetConsLine
END
TYPE LINE
DATA SIT_PR_Projet_cons_line
STATUS ON

DEBUG 5

MAXSCALE 10001

SIZEUNITS pixels

CLASS
  STYLE
COLOR 0 0 255
WIDTH 1
MAXWIDTH 1
MINWIDTH 1
  END
END
END


here is the query I do:

?LAYERS=SitPrProjetConsLineFORMAT=image%2FpngSERVICE=WMSVERSION=1.1.1REQUEST=GetMapSTYLES=EXCEPTIONS=application%2Fvnd.ogc.se_inimageSRS=EPSG%3A21781BBOX=506263.99910144,143335.99917056,506391.99903232,143463.99910144WIDTH=256HEIGHT=256

I want a line width of 1px
What am I doing wrong ?

thanks in advance
Oliver
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users




___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] Mapserver ignore DATA in layer definition

2010-01-25 Thread Oliver Christen

hi

I have a mapserver 5.2.1 on a server which totally ignore the DATA in a 
postgis layer definition in my mapfile.


I can write anything I want, even gibberish, in the DATA without even geting 
an error.


This is my layer :

LAYER
 NAME plan_points
 STATUS ON
 TYPE POINT
 CONNECTIONTYPE POSTGIS
 CONNECTION dbname=sitn user=plans_intervention 
password=sitn_plans_intervention host=nesitn2
 #DATA 'geom from (select 
geom,oid,id,fk_obj,code,com,sf,op_saisie,rem,rotation,url_lien,owner from 
plans_intervention.plan_points) as foo'

 #DATA geom from (select * from test.plan_points) as foo using unique id
 DATA dfsdfsdfd
 TEMPLATE ttt
 OPACITY 100
 CLASSITEM code
 TOLERANCE 20
 CLASS
   STYLE
 SYMBOL background1
 SIZE 20
 COLOR 255 255 255
 ANGLE [rotation]
   END
   STYLE
 SYMBOL VHC1
 SIZE 20
 COLOR 0 0 255
 ANGLE [rotation]
   END
   TEXT ([com])
   LABEL
 COLOR  0 0 255
 OUTLINECOLOR 255 255 255
 FONT arial
 TYPE truetype
 SIZE 12
 ANGLE [rotation]
 POSITION uc
 PRIORITY 10
 PARTIALS FALSE
 FORCE true
 ANTIALIAS true
   END
 END
END



as you can see I have made different try for the DATA, all are ignored, even 
the DATA dfsdfsdfd one.


the server is running on Windows Server 2008

any idea why the DATA string is ignored ?

this is the full mapserver version info:
MapServer Version MapServer version 5.2.1 OUTPUT=GIF OUTPUT=PNG OUTPUT=JPEG 
OUTPUT=WBMP OUTPUT=PDF OUTPUT=SWF OUTPUT=SVG SUPPORTS=PROJ SUPPORTS=AGG 
SUPPORTS=FREETYPE SUPPORTS=ICONV SUPPORTS=FRIBIDI SUPPORTS=WMS_SERVER 
SUPPORTS=WMS_CLIENT SUPPORTS=WFS_SERVER SUPPORTS=WFS_CLIENT 
SUPPORTS=WCS_SERVER SUPPORTS=SOS_SERVER SUPPORTS=FASTCGI SUPPORTS=THREADS 
SUPPORTS=GEOS SUPPORTS=RGBA_PNG INPUT=JPEG INPUT=POSTGIS INPUT=OGR 
INPUT=GDAL INPUT=SHAPEFILE PHP MapScript Version ($Revision: 7937 $ $Date: 
2008-09-30 10:37:46 -0400 (Tue, 30 Sep 2008) $)


thanks in advance
Oliver 


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] Mapserver ignore DATA in layer definition

2010-01-25 Thread Oliver Christen

hi Lars

yes I checked that,
the layer is effectively used as I get a layer not found error as soon as 
I remove/comment it


Oliver



My first idea: Are you sure that this mapfile is used? Or do you have
two layer with the same name?

I just tried to use DATA kakja as statement end got an internal server
error back. But I tried on current trunk (5.7-dev)

Lars
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users



___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] symbol at start and end of polyline ?

2010-01-19 Thread Oliver Christen

hello

I would like to add a symbol at the start and end of a polyline.

I tried using GEOMTRANSFORM with start and end in a style block in the 
layer but sadely this will add a symbol to all start and end of every vertex 
of the polyline, which is not the result wanted :/


is there any way to do that, beside creating a new point layer ?

thanks in advance
Oliver 


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] symbol at start and end of polyline ?

2010-01-19 Thread Oliver Christen
thx Thomas, but im not sure I understand what you are telling me here.
do line segments have id ?
  - Original Message - 
  From: thomas bonfort 
  To: Oliver Christen 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Tuesday, January 19, 2010 10:30 AM
  Subject: Re: [mapserver-users] symbol at start and end of polyline ?


  hi,

  class
expression ([id]=321)
style
  geomtransform end
  ...
end
  end
  class
style
  ... #all the others
end
  end


  On Tue, Jan 19, 2010 at 10:14, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

hello

I would like to add a symbol at the start and end of a polyline.

I tried using GEOMTRANSFORM with start and end in a style block in the 
layer but sadely this will add a symbol to all start and end of every vertex of 
the polyline, which is not the result wanted :/

is there any way to do that, beside creating a new point layer ?

thanks in advance
Oliver 
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] symbol at start and end of polyline ?

2010-01-19 Thread Oliver Christen
it makes sens, but how do I do that ?
do line segment get automaticaly an id ?
or how to get (or set) the id of the line segment?
  - Original Message - 
  From: thomas bonfort 
  To: Oliver Christen 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Tuesday, January 19, 2010 10:50 AM
  Subject: Re: [mapserver-users] symbol at start and end of polyline ?


  well, if you want only one line segment to have start markers, you must  tell 
mapserver which segment that is.


  On Tue, Jan 19, 2010 at 10:48, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

thx Thomas, but im not sure I understand what you are telling me here.
do line segments have id ?
  - Original Message - 
  From: thomas bonfort 
  To: Oliver Christen 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Tuesday, January 19, 2010 10:30 AM
  Subject: Re: [mapserver-users] symbol at start and end of polyline ?


  hi,

  class
expression ([id]=321)
style
  geomtransform end
  ...
end
  end
  class
style
  ... #all the others
end
  end


  On Tue, Jan 19, 2010 at 10:14, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

hello

I would like to add a symbol at the start and end of a polyline.

I tried using GEOMTRANSFORM with start and end in a style block in 
the layer but sadely this will add a symbol to all start and end of every 
vertex of the polyline, which is not the result wanted :/

is there any way to do that, beside creating a new point layer ?

thanks in advance
Oliver 
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users




___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] symbol at start and end of polyline ?

2010-01-19 Thread Oliver Christen
At first I thought I had some oddity with the data I used for my testing
( see http://dev.camptocamp.com/~ochriste/line_with_style.png with the star 
symbol on every vertex of one line feature)

but after trying with some other set of data, I only have the star symbol on 
start and end of the whole line feature ( and, strangely, also on cross-section 
if the line separate in two at some point ) 
http://dev.camptocamp.com/~ochriste/line_feature_extremities.png
so I thought the problem was solved, but upon zooming on the extrmities of the 
line feature, im again having strange behavior:
http://dev.camptocamp.com/~ochriste/line_feature_extremity_zoomed.png
with multiple star symbols.
the red star in these two image are from the same line feature.

why is there more than 2 stars ?
does the geomtransform 'end' only work for line features consisting of only 2 
points ?
  - Original Message - 
  From: thomas bonfort 
  To: Oliver Christen 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Tuesday, January 19, 2010 2:10 PM
  Subject: Re: [mapserver-users] symbol at start and end of polyline ?


  if you're using postgis, you can use a unique key from your table (i.e. gid)
  if you have a shapefile, you'll have to find an or a combination of 
attributes in the dbf that's unique to identify that particular feature.

  --
  thomas


  On Tue, Jan 19, 2010 at 11:59, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

it makes sens, but how do I do that ?
do line segment get automaticaly an id ?
or how to get (or set) the id of the line segment?
  - Original Message - 
  From: thomas bonfort 
  To: Oliver Christen 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Tuesday, January 19, 2010 10:50 AM
  Subject: Re: [mapserver-users] symbol at start and end of polyline ?


  well, if you want only one line segment to have start markers, you must  
tell mapserver which segment that is.


  On Tue, Jan 19, 2010 at 10:48, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

thx Thomas, but im not sure I understand what you are telling me here.
do line segments have id ?
  - Original Message - 
  From: thomas bonfort 
  To: Oliver Christen 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Tuesday, January 19, 2010 10:30 AM
  Subject: Re: [mapserver-users] symbol at start and end of polyline ?


  hi,

  class
expression ([id]=321)
style
  geomtransform end
  ...
end
  end
  class
style
  ... #all the others
end
  end


  On Tue, Jan 19, 2010 at 10:14, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

hello

I would like to add a symbol at the start and end of a polyline.

I tried using GEOMTRANSFORM with start and end in a style block 
in the layer but sadely this will add a symbol to all start and end of every 
vertex of the polyline, which is not the result wanted :/

is there any way to do that, beside creating a new point layer ?

thanks in advance
Oliver 
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users






___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-14 Thread Oliver Christen
mmm, I tested a bit and it seems the calculation is more complex than that, as 
the ellipse size definition of the symbol has also some effect.

style-size 8, points 6 6 end, gap 20 = spacing of ~20px between dots
style-size 8, points 8 8 end, gap 20 = spacing of ~14px between dots
style-size 10, points 6 6 end, gap 20 = spacing of ~26px between dots

I fear I dont understand at all how this is calculated :(
  - Original Message - 
  From: Oliver Christen 
  To: thomas bonfort 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Wednesday, January 13, 2010 10:08 AM
  Subject: Re: [mapserver-users] symbol circle (ellipse) rendered not round but 
oval


  thank you Thomas

  so if I have a dot of, for example, 8px and I want a space pf 20px between 
the dots, I need a gap of 20/8 =  2.5

  but as far as I can see, Mapserver doesnt like none-integer value for gap.

  what should I do? :(


  2010/1/12 thomas bonfort thomas.bonf...@gmail.com

gap in pixels = symbol-gap * style-size



On Tue, Jan 12, 2010 at 16:52, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

  GAP 20 is really giving unwanted result:

  http://dev.camptocamp.com/~ochriste/tilecache9.png

  though using only one style instead of two seems to solve the oval 
problem (but why ???)

  style
   symbol dots
   size 8

   color 204 102 0
   outlinecolor  102 51 0
  end

  I dont understand how GAP is supposed working:

  with GAP 2, I had ~10px between dots,
  with GAP 3, I had ~18px between dots,
  with GAP 4, I had ~26px between dots

  whats the logic ?
- Original Message - 
From: thomas bonfort 
To: Oliver Christen 
Cc: mapserver-users@lists.osgeo.org 
Sent: Tuesday, January 12, 2010 4:36 PM
Subject: Re: [mapserver-users] symbol circle (ellipse) rendered not 
round but oval


symbol
 name dots
 type ellipse
 points 1 1 end
 gap 20
end

...

style
 symbol dots
 size 8
 color r g b
 outlinecolor  r g b
end





On Tue, Jan 12, 2010 at 16:31, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

  Hi Thomas :)

  thanks for the answer, but sadely it didnt solve my problem:

  here are a few tests I did:

  using GAP 20 gave me really weird result:
  http://dev.camptocamp.com/~ochriste/tilecache6_gap_20.png

  using PATTERN 1 22 END gave me better result, but the dot are still 
slightly oval:
  http://dev.camptocamp.com/~ochriste/tilecache7_pattern.png

  I wondered if the visual effect may have come from the fact im 
pilling up two circle one over the other, but even when trying to display only 
the outer or inner circles, the circle are still slightly off:
  http://dev.camptocamp.com/~ochriste/tilecache4_outercircle.png
  http://dev.camptocamp.com/~ochriste/tilecache5_innercircle.png
  (both at once 
http://dev.camptocamp.com/~ochriste/tilecache3_bothcircle.png)

  any other idea? ^_^
- Original Message - 
From: thomas bonfort 
To: Oliver Christen 
Cc: mapserver-users@lists.osgeo.org 
Sent: Tuesday, January 12, 2010 3:18 PM
Subject: Re: [mapserver-users] symbol circle (ellipse) rendered not 
round but oval


hi Oliver :)

try with GAP 20 instead of STYLE 1 20 END (which is deprecated btw, 
use PATTERN now)

cheers,
thomas


On Tue, Jan 12, 2010 at 15:04, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

  hello

  I have a weird problem with symbology.
  Im rendering a line with dots, but the dots are not round but 
slightly oval, see image http://dev.camptocamp.com/~ochriste/tilecache.png

  Any idea what could be the cause ?

  Here is my symbol definition:

  SYMBOL
   NAME linie-gepunkt10
   TYPE ELLIPSE
   POINTS
1 1
   END
   STYLE
1 20
   END
   FILLED TRUE
  END

  and my layer's class definition:

   CLASS
EXPRESSION 3
MAXSCALEDENOM 21
STYLE
  SYMBOL linie-gepunkt10
  SIZE 10
  COLOR 102 51 0
END
STYLE
  SYMBOL linie-gepunkt10
  SIZE 8
  COLOR 204 102 0
END
   END

  and the map outputformat:

  OUTPUTFORMAT
   NAME png
   DRIVER AGG/PNG
   MIMETYPE image/png
   IMAGEMODE RGBA
   EXTENSION png
   FORMATOPTION

Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-14 Thread Oliver Christen
mmm, then whats the difference between points 1 1 end and points 8 8 end 
?




Oliver Christen wrote:

Thank you Håvard
I see you have changed points 1 1 end by points 8 8 end.
Tell me if im wrong, but these values are only used to render the 
proportions of the elipse but not it's size, is that correct?
so if I understand correctly, using larger values for the elipse 
width/height refine the rendering of the vectorised dot geometry ?


The size of the symbols are determined by the values given
in the style of the class in the layer.  So the magnitude
of the values used in the symbol definition only have
effect within the symbol definition (makes it possible to
use only integers, and makes it easier to do the
arithmetics :-) ).
The magnitude of the values used in the symbol definition
should have no bearing on the way the symbols are rendered
by Mapserver.

Håvard



regards
Oliver


Dear Oliver Christen,

Using Mapserver 5.6.0 and AGG rendering, you can achieve what
you want by using a combination of ellipse size and gap size
in the symbol definition.
There is something strange in the calculations (a gap of
about half the ellipse size seems to produce touching symbols),
so in your case, you could try to use 8 as the diameter of the
ellipses, and 20 + 4 (half the ellipse size) = 24 as the
gap:

symbol
 name dots
 type ellipse
 points 8 8 end
 gap 24
end


Håvard

Oliver Christen wrote:

thank you Thomas

so if I have a dot of, for example, 8px and I want a space pf 20px 
between the dots, I need a gap of 20/8 =  2.5


but as far as I can see, Mapserver doesnt like none-integer value for 
gap.


what should I do? :(

2010/1/12 thomas bonfort thomas.bonf...@gmail.com 
mailto:thomas.bonf...@gmail.com


gap in pixels = symbol-gap * style-size


On Tue, Jan 12, 2010 at 16:52, Oliver Christen
oliver.chris...@camptocamp.com
mailto:oliver.chris...@camptocamp.com wrote:

GAP 20 is really giving unwanted result:
 http://dev.camptocamp.com/~ochriste/tilecache9.png
http://dev.camptocamp.com/%7Eochriste/tilecache9.png
 though using only one style instead of two seems to solve the
oval problem (but why ???)
 style
 symbol dots
 size 8
 color 204 102 0
 outlinecolor  102 51 0
end
 I dont understand how GAP is supposed working:
 with GAP 2, I had ~10px between dots,
with GAP 3, I had ~18px between dots,
with GAP 4, I had ~26px between dots
 whats the logic ?

- Original Message -
*From:* thomas bonfort mailto:thomas.bonf...@gmail.com
*To:* Oliver Christen 
mailto:oliver.chris...@camptocamp.com

*Cc:* mapserver-users@lists.osgeo.org
mailto:mapserver-users@lists.osgeo.org
*Sent:* Tuesday, January 12, 2010 4:36 PM
*Subject:* Re: [mapserver-users] symbol circle (ellipse)
rendered not round but oval

symbol
 name dots
 type ellipse
 points 1 1 end
 gap 20
end

...

style
 symbol dots
 size 8
 color r g b
 outlinecolor  r g b
end




On Tue, Jan 12, 2010 at 16:31, Oliver Christen
oliver.chris...@camptocamp.com
mailto:oliver.chris...@camptocamp.com wrote:

Hi Thomas :)
 thanks for the answer, but sadely it didnt solve my 
problem:

 here are a few tests I did:
 using GAP 20 gave me really weird result:

http://dev.camptocamp.com/~ochriste/tilecache6_gap_20.png

http://dev.camptocamp.com/%7Eochriste/tilecache6_gap_20.png
 using PATTERN 1 22 END gave me better result, but the
dot are still slightly oval:

http://dev.camptocamp.com/~ochriste/tilecache7_pattern.png

http://dev.camptocamp.com/%7Eochriste/tilecache7_pattern.png
 I wondered if the visual effect may have come from the
fact im pilling up two circle one over the other, but
even when trying to display only the outer or inner
circles, the circle are still slightly off:

http://dev.camptocamp.com/~ochriste/tilecache4_outercircle.png

http://dev.camptocamp.com/%7Eochriste/tilecache4_outercircle.png

http://dev.camptocamp.com/~ochriste/tilecache5_innercircle.png

http://dev.camptocamp.com/%7Eochriste/tilecache5_innercircle.png
(both at once

http://dev.camptocamp.com/~ochriste/tilecache3_bothcircle.png

http://dev.camptocamp.com/%7Eochriste/tilecache3_bothcircle.png)
 any other idea? ^_^

- Original Message -
*From:* thomas bonfort
mailto:thomas.bonf...@gmail.com
*To:* Oliver Christen
mailto:oliver.chris...@camptocamp.com

Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-14 Thread Oliver Christen

Thank you for the explanation.

Im still trying to figure out how to calculate exactly the GAP and ellipse 
POINT depending on the symbol's size and the distance between the points 
wanted.
empirically I have come with the formula: distance = (GAP * SIZE * 0.75) / 
POINTS
seems weird but at least I get some way to estimate the different values to 
use for GAP and POINTS



Oliver Christen wrote:
mmm, then whats the difference between points 1 1 end and points 8 8 
end ?


For an ELLIPSE symbol definition without GAP, there is no
difference.
The bounding box of the symbol will be used when rendering
the symbol.  The height of the symbol bounding box on the
map will be the size given in the STYLE of the CLASS in
the LAYER.
I have found this practice very convenient when creating
complex vector symbols.

When you have an ELLIPSE symbol definition with a GAP
specified, the relationship between the values used in
POINTS and GAP is very important.

For the current version of Mapserver and AGG, if you set
SIZE in STYLE to 8, and combine that with the symbol
definition I suggested (points 8 8, gap 24), I think you
would get a reasonable result.

Håvard


Oliver Christen wrote:

Thank you Håvard
I see you have changed points 1 1 end by points 8 8 end.
Tell me if im wrong, but these values are only used to render the 
proportions of the elipse but not it's size, is that correct?
so if I understand correctly, using larger values for the elipse 
width/height refine the rendering of the vectorised dot geometry ?

The size of the symbols are determined by the values given
in the style of the class in the layer.  So the magnitude
of the values used in the symbol definition only have
effect within the symbol definition (makes it possible to
use only integers, and makes it easier to do the
arithmetics :-) ).
The magnitude of the values used in the symbol definition
should have no bearing on the way the symbols are rendered
by Mapserver.

Håvard


regards
Oliver


Dear Oliver Christen,

Using Mapserver 5.6.0 and AGG rendering, you can achieve what
you want by using a combination of ellipse size and gap size
in the symbol definition.
There is something strange in the calculations (a gap of
about half the ellipse size seems to produce touching symbols),
so in your case, you could try to use 8 as the diameter of the
ellipses, and 20 + 4 (half the ellipse size) = 24 as the
gap:

symbol
 name dots
 type ellipse
 points 8 8 end
 gap 24
end


Håvard

Oliver Christen wrote:

thank you Thomas

so if I have a dot of, for example, 8px and I want a space pf 20px 
between the dots, I need a gap of 20/8 =  2.5


but as far as I can see, Mapserver doesnt like none-integer value for 
gap.


what should I do? :(

2010/1/12 thomas bonfort thomas.bonf...@gmail.com 
mailto:thomas.bonf...@gmail.com


gap in pixels = symbol-gap * style-size


On Tue, Jan 12, 2010 at 16:52, Oliver Christen
oliver.chris...@camptocamp.com
mailto:oliver.chris...@camptocamp.com wrote:

GAP 20 is really giving unwanted result:
 http://dev.camptocamp.com/~ochriste/tilecache9.png
http://dev.camptocamp.com/%7Eochriste/tilecache9.png
 though using only one style instead of two seems to solve 
the

oval problem (but why ???)
 style
 symbol dots
 size 8
 color 204 102 0
 outlinecolor  102 51 0
end
 I dont understand how GAP is supposed working:
 with GAP 2, I had ~10px between dots,
with GAP 3, I had ~18px between dots,
with GAP 4, I had ~26px between dots
 whats the logic ?

- Original Message -
*From:* thomas bonfort mailto:thomas.bonf...@gmail.com
*To:* Oliver Christen 
mailto:oliver.chris...@camptocamp.com

*Cc:* mapserver-users@lists.osgeo.org
mailto:mapserver-users@lists.osgeo.org
*Sent:* Tuesday, January 12, 2010 4:36 PM
*Subject:* Re: [mapserver-users] symbol circle (ellipse)
rendered not round but oval

symbol
 name dots
 type ellipse
 points 1 1 end
 gap 20
end

...

style
 symbol dots
 size 8
 color r g b
 outlinecolor  r g b
end




On Tue, Jan 12, 2010 at 16:31, Oliver Christen
oliver.chris...@camptocamp.com
mailto:oliver.chris...@camptocamp.com wrote:

Hi Thomas :)
 thanks for the answer, but sadely it didnt solve my 
problem:

 here are a few tests I did:
 using GAP 20 gave me really weird result:

http://dev.camptocamp.com/~ochriste/tilecache6_gap_20.png

http://dev.camptocamp.com/%7Eochriste/tilecache6_gap_20.png
 using PATTERN 1 22 END gave me better result, but 
the

dot are still slightly oval:

http

Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-14 Thread Oliver Christen

Hi Steve

I followed your suggestion and opened a documentation bug

thanks
Oliver


Hi guys,

I haven't followed this whole thread but I think it should be better 
documented so people can understand how to use this. Please write a doc 
bug and put as much information in the bug about how this works as you 
have been able to discover.


Thanks,
  -Steve W

Oliver Christen wrote:


Thank you Håvard
I see you have changed points 1 1 end by points 8 8 end.
Tell me if im wrong, but these values are only used to render the 
proportions of the elipse but not it's size, is that correct?
so if I understand correctly, using larger values for the elipse 
width/height refine the rendering of the vectorised dot geometry ?


regards
Oliver


Dear Oliver Christen,

Using Mapserver 5.6.0 and AGG rendering, you can achieve what
you want by using a combination of ellipse size and gap size
in the symbol definition.
There is something strange in the calculations (a gap of
about half the ellipse size seems to produce touching symbols),
so in your case, you could try to use 8 as the diameter of the
ellipses, and 20 + 4 (half the ellipse size) = 24 as the
gap:

symbol
 name dots
 type ellipse
 points 8 8 end
 gap 24
end


Håvard

Oliver Christen wrote:

thank you Thomas

so if I have a dot of, for example, 8px and I want a space pf 20px 
between the dots, I need a gap of 20/8 =  2.5


but as far as I can see, Mapserver doesnt like none-integer value for 
gap.


what should I do? :(

2010/1/12 thomas bonfort thomas.bonf...@gmail.com 
mailto:thomas.bonf...@gmail.com


gap in pixels = symbol-gap * style-size


On Tue, Jan 12, 2010 at 16:52, Oliver Christen
oliver.chris...@camptocamp.com
mailto:oliver.chris...@camptocamp.com wrote:

GAP 20 is really giving unwanted result:
 http://dev.camptocamp.com/~ochriste/tilecache9.png
http://dev.camptocamp.com/%7Eochriste/tilecache9.png
 though using only one style instead of two seems to solve the
oval problem (but why ???)
 style
 symbol dots
 size 8
 color 204 102 0
 outlinecolor  102 51 0
end
 I dont understand how GAP is supposed working:
 with GAP 2, I had ~10px between dots,
with GAP 3, I had ~18px between dots,
with GAP 4, I had ~26px between dots
 whats the logic ?

- Original Message -
*From:* thomas bonfort mailto:thomas.bonf...@gmail.com
*To:* Oliver Christen 
mailto:oliver.chris...@camptocamp.com

*Cc:* mapserver-users@lists.osgeo.org
mailto:mapserver-users@lists.osgeo.org
*Sent:* Tuesday, January 12, 2010 4:36 PM
*Subject:* Re: [mapserver-users] symbol circle (ellipse)
rendered not round but oval

symbol
 name dots
 type ellipse
 points 1 1 end
 gap 20
end

...

style
 symbol dots
 size 8
 color r g b
 outlinecolor  r g b
end




On Tue, Jan 12, 2010 at 16:31, Oliver Christen
oliver.chris...@camptocamp.com
mailto:oliver.chris...@camptocamp.com wrote:

Hi Thomas :)
 thanks for the answer, but sadely it didnt solve my 
problem:

 here are a few tests I did:
 using GAP 20 gave me really weird result:

http://dev.camptocamp.com/~ochriste/tilecache6_gap_20.png

http://dev.camptocamp.com/%7Eochriste/tilecache6_gap_20.png
 using PATTERN 1 22 END gave me better result, but the
dot are still slightly oval:

http://dev.camptocamp.com/~ochriste/tilecache7_pattern.png

http://dev.camptocamp.com/%7Eochriste/tilecache7_pattern.png
 I wondered if the visual effect may have come from the
fact im pilling up two circle one over the other, but
even when trying to display only the outer or inner
circles, the circle are still slightly off:

http://dev.camptocamp.com/~ochriste/tilecache4_outercircle.png

http://dev.camptocamp.com/%7Eochriste/tilecache4_outercircle.png

http://dev.camptocamp.com/~ochriste/tilecache5_innercircle.png

http://dev.camptocamp.com/%7Eochriste/tilecache5_innercircle.png
(both at once

http://dev.camptocamp.com/~ochriste/tilecache3_bothcircle.png

http://dev.camptocamp.com/%7Eochriste/tilecache3_bothcircle.png)
 any other idea? ^_^

- Original Message -
*From:* thomas bonfort
mailto:thomas.bonf...@gmail.com
*To:* Oliver Christen
mailto:oliver.chris...@camptocamp.com
*Cc:* mapserver-users@lists.osgeo.org
mailto:mapserver-users@lists.osgeo.org
*Sent:* Tuesday, January 12

Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-14 Thread Oliver Christen
mmm, in my opinion it would be clearer if the GAP didnt depend from 
anything, set the GAP to 10 would make you a spacing of 10px between each 
symbols border, whatever the values of the symbol POINTS.

but that may be a change too important ?

Oliver


Stephen Woodbridge wrote:
So if I want ten pixel ellipses with a 1, 2, or 3 pixels gap how would 
you specify that?


points 10 10 gap 1 (or 2 or 3)?


Yes, if the rendering behaviour had been according to my
preferences,
POINTS 10 10 END
GAP 1 (or 2 or 3)
should do it (if you had specified SIZE 10 in the STYLE).
You specify the gap, and not the centre to centre distance
between the ellipses.

But currently, this is not the behaviour.
Currently (with AGG in 5.6.0) you would have to specify:
POINTS 10 10 END
GAP 6 (or 7 or 8)
to get something close (if you specify SIZE 10 in the STYLE).

Something seems to be wrong when the GAP is close to
half the diameter specified in POINTS - some of the gaps
will be as expected, while others are larger (closer to the
actual number specified in GAP).

Håvard



-Steve W

Havard Tveite wrote:

If Thomas B. agrees, I think it would be reasonable to also
open a bug on the gap calculation.
In my opinion, since the word gap is used, the correct
way to specify a symbol with ellipses that has the same
gap between them as the diameter of the ellipses should be
as follows:

symbol
  name dots
  type ellipse
  points 1 1 end
  gap 1
end

or an equivalent scaled version such as:

symbol
  name dots
  type ellipse
  points 23 23 end
  gap 23
end

The actual size of the ellipses (and the width of the gaps)
should be determined by the SIZE in CLASS- STYLE.  If
SIZE is 10, the ellipses should have a 10 pixel diameter
(and should have a 10 pixel gap between them).

I have been hesitant to file bugs regarding symbology, since
the documentation on symbology is lacking and there is not
complete agreement on how some of the things that has to do
with symbology should work.  There have also been
discussions on changing some behaviour and mechanisms for
symbology.
First we have to agree on how things should work and document
it, then we can file bugs if behaviour is not as expected.

Håvard

Oliver Christen wrote:

Hi Steve

I followed your suggestion and opened a documentation bug

thanks
Oliver


Hi guys,

I haven't followed this whole thread but I think it should be better 
documented so people can understand how to use this. Please write a 
doc bug and put as much information in the bug about how this works as 
you have been able to discover.


Thanks,
  -Steve W

Oliver Christen wrote:

Thank you Håvard
I see you have changed points 1 1 end by points 8 8 end.
Tell me if im wrong, but these values are only used to render the 
proportions of the elipse but not it's size, is that correct?
so if I understand correctly, using larger values for the elipse 
width/height refine the rendering of the vectorised dot geometry ?


regards
Oliver


Dear Oliver Christen,

Using Mapserver 5.6.0 and AGG rendering, you can achieve what
you want by using a combination of ellipse size and gap size
in the symbol definition.
There is something strange in the calculations (a gap of
about half the ellipse size seems to produce touching symbols),
so in your case, you could try to use 8 as the diameter of the
ellipses, and 20 + 4 (half the ellipse size) = 24 as the
gap:

symbol
 name dots
 type ellipse
 points 8 8 end
 gap 24
end


Håvard

Oliver Christen wrote:

thank you Thomas

so if I have a dot of, for example, 8px and I want a space pf 20px 
between the dots, I need a gap of 20/8 =  2.5


but as far as I can see, Mapserver doesnt like none-integer value 
for gap.


what should I do? :(

2010/1/12 thomas bonfort thomas.bonf...@gmail.com 
mailto:thomas.bonf...@gmail.com


gap in pixels = symbol-gap * style-size


On Tue, Jan 12, 2010 at 16:52, Oliver Christen
oliver.chris...@camptocamp.com
mailto:oliver.chris...@camptocamp.com wrote:

GAP 20 is really giving unwanted result:
 http://dev.camptocamp.com/~ochriste/tilecache9.png
http://dev.camptocamp.com/%7Eochriste/tilecache9.png
 though using only one style instead of two seems to solve 
the

oval problem (but why ???)
 style
 symbol dots
 size 8
 color 204 102 0
 outlinecolor  102 51 0
end
 I dont understand how GAP is supposed working:
 with GAP 2, I had ~10px between dots,
with GAP 3, I had ~18px between dots,
with GAP 4, I had ~26px between dots
 whats the logic ?

- Original Message -
*From:* thomas bonfort 
mailto:thomas.bonf...@gmail.com
*To:* Oliver Christen 
mailto:oliver.chris...@camptocamp.com

*Cc:* mapserver-users@lists.osgeo.org
mailto:mapserver-users@lists.osgeo.org
*Sent:* Tuesday, January 12, 2010 4:36 PM
*Subject:* Re: [mapserver-users

Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-13 Thread Oliver Christen
hi David

Im requesting the image through wms indeed.
Though the bbox x/y ratio is exactly the same as the image x/y.
I tried to request the same image directly in cgi, sadely I get exactly the
same problem:

image result in cgi: http://dev.camptocamp.com/~ochriste/cgi.png

(as compared to the wms image http://dev.camptocamp.com/~ochriste/wms.png)

this is the cgi request I used:
/usr/lib/cgi-bin/mapserv -nh
QUERY_STRING=map=/var/www/veloland/private/veloland/veloland/wms/veloland.mapmode=mapLAYERS=MtblandRoutenNationalmapext=556960+220720+557600+221360mapsize=256+256
 cgi.png

I used exactly the same bbox and image size as the wms request:

/wms?FORMAT=image%2FpngLAYERS=MtblandRoutenNationalSERVICE=WMSVERSION=1.1.1REQUEST=GetMapSTYLES=EXCEPTIONS=application%2Fvnd.ogc.se_inimageSRS=EPSG%3A21781BBOX=556960,220720,557600,221360WIDTH=256HEIGHT=256

this is when using STYLE 1 20 END or PATTERN 1 20 END for the symbol

when using GAP  the dots are correctly shaped

so I suspect the problem is within the interpretation of STYLE/PATTERN for
the symbol generation ?


2010/1/12 Fawcett, David (MPCA) david.fawc...@state.mn.us

  Are you requesting your image through a WMS and maybe getting a
 ‘non-square pixel’ effect?



 If you are using WMS, are the symbols round when you do a cgi request in
 mode=map or mode=tile?



 David.



 *From:* mapserver-users-boun...@lists.osgeo.org [mailto:
 mapserver-users-boun...@lists.osgeo.org] *On Behalf Of *Oliver Christen
 *Sent:* Tuesday, January 12, 2010 9:31 AM
 *To:* thomas bonfort

 *Cc:* mapserver-users@lists.osgeo.org
 *Subject:* Re: [mapserver-users] symbol circle (ellipse) rendered not
 round but oval



 Hi Thomas :)



 thanks for the answer, but sadely it didnt solve my problem:



 here are a few tests I did:



 using GAP 20 gave me really weird result:

 http://dev.camptocamp.com/~ochriste/tilecache6_gap_20.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache6_gap_20.png



 using PATTERN 1 22 END gave me better result, but the dot are still
 slightly oval:

 http://dev.camptocamp.com/~ochriste/tilecache7_pattern.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache7_pattern.png



 I wondered if the visual effect may have come from the fact im pilling up
 two circle one over the other, but even when trying to display only the
 outer or inner circles, the circle are still slightly off:

 http://dev.camptocamp.com/~ochriste/tilecache4_outercircle.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache4_outercircle.png

 http://dev.camptocamp.com/~ochriste/tilecache5_innercircle.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache5_innercircle.png

 (both at once
 http://dev.camptocamp.com/~ochriste/tilecache3_bothcircle.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache3_bothcircle.png
 )



 any other idea? ^_^

  - Original Message -

 *From:* thomas bonfort thomas.bonf...@gmail.com

 *To:* Oliver Christen oliver.chris...@camptocamp.com

 *Cc:* mapserver-users@lists.osgeo.org

 *Sent:* Tuesday, January 12, 2010 3:18 PM

 *Subject:* Re: [mapserver-users] symbol circle (ellipse) rendered not
 round but oval



 hi Oliver :)

 try with GAP 20 instead of STYLE 1 20 END (which is deprecated btw, use
 PATTERN now)

 cheers,
 thomas

 On Tue, Jan 12, 2010 at 15:04, Oliver Christen 
 oliver.chris...@camptocamp.com wrote:

 hello

 I have a weird problem with symbology.
 Im rendering a line with dots, but the dots are not round but slightly
 oval, see image 
 http://dev.camptocamp.com/~ochriste/tilecache.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache.png

 Any idea what could be the cause ?

 Here is my symbol definition:

 SYMBOL
  NAME linie-gepunkt10
  TYPE ELLIPSE
  POINTS
   1 1
  END
  STYLE
   1 20
  END
  FILLED TRUE
 END

 and my layer's class definition:

  CLASS
   EXPRESSION 3
   MAXSCALEDENOM 21
   STYLE
 SYMBOL linie-gepunkt10
 SIZE 10
 COLOR 102 51 0
   END
   STYLE
 SYMBOL linie-gepunkt10
 SIZE 8
 COLOR 204 102 0
   END
  END

 and the map outputformat:

 OUTPUTFORMAT
  NAME png
  DRIVER AGG/PNG
  MIMETYPE image/png
  IMAGEMODE RGBA
  EXTENSION png
  FORMATOPTION INTERLACE=OFF
 END
 ___
 mapserver-users mailing list
 mapserver-users@lists.osgeo.org
 http://lists.osgeo.org/mailman/listinfo/mapserver-users




___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-13 Thread Oliver Christen
thank you Thomas

so if I have a dot of, for example, 8px and I want a space pf 20px between
the dots, I need a gap of 20/8 =  2.5

but as far as I can see, Mapserver doesnt like none-integer value for gap.

what should I do? :(

2010/1/12 thomas bonfort thomas.bonf...@gmail.com

 gap in pixels = symbol-gap * style-size


 On Tue, Jan 12, 2010 at 16:52, Oliver Christen 
 oliver.chris...@camptocamp.com wrote:

  GAP 20 is really giving unwanted result:

 http://dev.camptocamp.com/~ochriste/tilecache9.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache9.png

 though using only one style instead of two seems to solve the oval problem
 (but why ???)

 style
  symbol dots
  size 8
  color 204 102 0
  outlinecolor  102 51 0
 end

 I dont understand how GAP is supposed working:

 with GAP 2, I had ~10px between dots,
  with GAP 3, I had ~18px between dots,
  with GAP 4, I had ~26px between dots

 whats the logic ?

 - Original Message -
  *From:* thomas bonfort thomas.bonf...@gmail.com
 *To:* Oliver Christen oliver.chris...@camptocamp.com
 *Cc:* mapserver-users@lists.osgeo.org
  *Sent:* Tuesday, January 12, 2010 4:36 PM
 *Subject:* Re: [mapserver-users] symbol circle (ellipse) rendered not
 round but oval

 symbol
  name dots
  type ellipse
  points 1 1 end
  gap 20
 end

 ...

 style
  symbol dots
  size 8
  color r g b
  outlinecolor  r g b
 end




 On Tue, Jan 12, 2010 at 16:31, Oliver Christen 
 oliver.chris...@camptocamp.com wrote:

  Hi Thomas :)

 thanks for the answer, but sadely it didnt solve my problem:

 here are a few tests I did:

 using GAP 20 gave me really weird result:
 http://dev.camptocamp.com/~ochriste/tilecache6_gap_20.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache6_gap_20.png

 using PATTERN 1 22 END gave me better result, but the dot are still
 slightly oval:
 http://dev.camptocamp.com/~ochriste/tilecache7_pattern.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache7_pattern.png

 I wondered if the visual effect may have come from the fact im pilling up
 two circle one over the other, but even when trying to display only the
 outer or inner circles, the circle are still slightly off:
 http://dev.camptocamp.com/~ochriste/tilecache4_outercircle.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache4_outercircle.png
 http://dev.camptocamp.com/~ochriste/tilecache5_innercircle.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache5_innercircle.png
 (both at once
 http://dev.camptocamp.com/~ochriste/tilecache3_bothcircle.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache3_bothcircle.png
 )

 any other idea? ^_^

 - Original Message -
  *From:* thomas bonfort thomas.bonf...@gmail.com
 *To:* Oliver Christen oliver.chris...@camptocamp.com
 *Cc:* mapserver-users@lists.osgeo.org
 *Sent:* Tuesday, January 12, 2010 3:18 PM
 *Subject:* Re: [mapserver-users] symbol circle (ellipse) rendered not
 round but oval

 hi Oliver :)

 try with GAP 20 instead of STYLE 1 20 END (which is deprecated btw, use
 PATTERN now)

 cheers,
 thomas

 On Tue, Jan 12, 2010 at 15:04, Oliver Christen 
 oliver.chris...@camptocamp.com wrote:

 hello

 I have a weird problem with symbology.
 Im rendering a line with dots, but the dots are not round but slightly
 oval, see image 
 http://dev.camptocamp.com/~ochriste/tilecache.pnghttp://dev.camptocamp.com/%7Eochriste/tilecache.png

 Any idea what could be the cause ?

 Here is my symbol definition:

 SYMBOL
  NAME linie-gepunkt10
  TYPE ELLIPSE
  POINTS
   1 1
  END
  STYLE
   1 20
  END
  FILLED TRUE
 END

 and my layer's class definition:

  CLASS
   EXPRESSION 3
   MAXSCALEDENOM 21
   STYLE
 SYMBOL linie-gepunkt10
 SIZE 10
 COLOR 102 51 0
   END
   STYLE
 SYMBOL linie-gepunkt10
 SIZE 8
 COLOR 204 102 0
   END
  END

 and the map outputformat:

 OUTPUTFORMAT
  NAME png
  DRIVER AGG/PNG
  MIMETYPE image/png
  IMAGEMODE RGBA
  EXTENSION png
  FORMATOPTION INTERLACE=OFF
 END
 ___
 mapserver-users mailing list
 mapserver-users@lists.osgeo.org
 http://lists.osgeo.org/mailman/listinfo/mapserver-users





___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-13 Thread Oliver Christen


Thank you Håvard
I see you have changed points 1 1 end by points 8 8 end.
Tell me if im wrong, but these values are only used to render the 
proportions of the elipse but not it's size, is that correct?
so if I understand correctly, using larger values for the elipse 
width/height refine the rendering of the vectorised dot geometry ?


regards
Oliver


Dear Oliver Christen,

Using Mapserver 5.6.0 and AGG rendering, you can achieve what
you want by using a combination of ellipse size and gap size
in the symbol definition.
There is something strange in the calculations (a gap of
about half the ellipse size seems to produce touching symbols),
so in your case, you could try to use 8 as the diameter of the
ellipses, and 20 + 4 (half the ellipse size) = 24 as the
gap:

symbol
 name dots
 type ellipse
 points 8 8 end
 gap 24
end


Håvard

Oliver Christen wrote:

thank you Thomas

so if I have a dot of, for example, 8px and I want a space pf 20px 
between the dots, I need a gap of 20/8 =  2.5


but as far as I can see, Mapserver doesnt like none-integer value for 
gap.


what should I do? :(

2010/1/12 thomas bonfort thomas.bonf...@gmail.com 
mailto:thomas.bonf...@gmail.com


gap in pixels = symbol-gap * style-size


On Tue, Jan 12, 2010 at 16:52, Oliver Christen
oliver.chris...@camptocamp.com
mailto:oliver.chris...@camptocamp.com wrote:

GAP 20 is really giving unwanted result:
 http://dev.camptocamp.com/~ochriste/tilecache9.png
http://dev.camptocamp.com/%7Eochriste/tilecache9.png
 though using only one style instead of two seems to solve the
oval problem (but why ???)
 style
 symbol dots
 size 8
 color 204 102 0
 outlinecolor  102 51 0
end
 I dont understand how GAP is supposed working:
 with GAP 2, I had ~10px between dots,
with GAP 3, I had ~18px between dots,
with GAP 4, I had ~26px between dots
 whats the logic ?

- Original Message -
*From:* thomas bonfort mailto:thomas.bonf...@gmail.com
*To:* Oliver Christen mailto:oliver.chris...@camptocamp.com
*Cc:* mapserver-users@lists.osgeo.org
mailto:mapserver-users@lists.osgeo.org
*Sent:* Tuesday, January 12, 2010 4:36 PM
*Subject:* Re: [mapserver-users] symbol circle (ellipse)
rendered not round but oval

symbol
 name dots
 type ellipse
 points 1 1 end
 gap 20
end

...

style
 symbol dots
 size 8
 color r g b
 outlinecolor  r g b
end




On Tue, Jan 12, 2010 at 16:31, Oliver Christen
oliver.chris...@camptocamp.com
mailto:oliver.chris...@camptocamp.com wrote:

Hi Thomas :)
 thanks for the answer, but sadely it didnt solve my 
problem:

 here are a few tests I did:
 using GAP 20 gave me really weird result:
http://dev.camptocamp.com/~ochriste/tilecache6_gap_20.png

http://dev.camptocamp.com/%7Eochriste/tilecache6_gap_20.png
 using PATTERN 1 22 END gave me better result, but the
dot are still slightly oval:

http://dev.camptocamp.com/~ochriste/tilecache7_pattern.png

http://dev.camptocamp.com/%7Eochriste/tilecache7_pattern.png
 I wondered if the visual effect may have come from the
fact im pilling up two circle one over the other, but
even when trying to display only the outer or inner
circles, the circle are still slightly off:

http://dev.camptocamp.com/~ochriste/tilecache4_outercircle.png

http://dev.camptocamp.com/%7Eochriste/tilecache4_outercircle.png

http://dev.camptocamp.com/~ochriste/tilecache5_innercircle.png

http://dev.camptocamp.com/%7Eochriste/tilecache5_innercircle.png
(both at once

http://dev.camptocamp.com/~ochriste/tilecache3_bothcircle.png

http://dev.camptocamp.com/%7Eochriste/tilecache3_bothcircle.png)
 any other idea? ^_^

- Original Message -
*From:* thomas bonfort
mailto:thomas.bonf...@gmail.com
*To:* Oliver Christen
mailto:oliver.chris...@camptocamp.com
*Cc:* mapserver-users@lists.osgeo.org
mailto:mapserver-users@lists.osgeo.org
*Sent:* Tuesday, January 12, 2010 3:18 PM
*Subject:* Re: [mapserver-users] symbol circle
(ellipse) rendered not round but oval

hi Oliver :)

try with GAP 20 instead of STYLE 1 20 END (which is
deprecated btw, use PATTERN now)

cheers,
thomas

[mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-12 Thread Oliver Christen

hello

I have a weird problem with symbology.
Im rendering a line with dots, but the dots are not round but slightly oval, 
see image http://dev.camptocamp.com/~ochriste/tilecache.png


Any idea what could be the cause ?

Here is my symbol definition:

SYMBOL
 NAME linie-gepunkt10
 TYPE ELLIPSE
 POINTS
   1 1
 END
 STYLE
   1 20
 END
 FILLED TRUE
END

and my layer's class definition:

 CLASS
   EXPRESSION 3
   MAXSCALEDENOM 21
   STYLE
 SYMBOL linie-gepunkt10
 SIZE 10
 COLOR 102 51 0
   END
   STYLE
 SYMBOL linie-gepunkt10
 SIZE 8
 COLOR 204 102 0
   END
 END

and the map outputformat:

OUTPUTFORMAT
 NAME png
 DRIVER AGG/PNG
 MIMETYPE image/png
 IMAGEMODE RGBA
 EXTENSION png
 FORMATOPTION INTERLACE=OFF
END 


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-12 Thread Oliver Christen
Hi Thomas :)

thanks for the answer, but sadely it didnt solve my problem:

here are a few tests I did:

using GAP 20 gave me really weird result:
http://dev.camptocamp.com/~ochriste/tilecache6_gap_20.png

using PATTERN 1 22 END gave me better result, but the dot are still slightly 
oval:
http://dev.camptocamp.com/~ochriste/tilecache7_pattern.png

I wondered if the visual effect may have come from the fact im pilling up two 
circle one over the other, but even when trying to display only the outer or 
inner circles, the circle are still slightly off:
http://dev.camptocamp.com/~ochriste/tilecache4_outercircle.png
http://dev.camptocamp.com/~ochriste/tilecache5_innercircle.png
(both at once http://dev.camptocamp.com/~ochriste/tilecache3_bothcircle.png)

any other idea? ^_^
  - Original Message - 
  From: thomas bonfort 
  To: Oliver Christen 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Tuesday, January 12, 2010 3:18 PM
  Subject: Re: [mapserver-users] symbol circle (ellipse) rendered not round but 
oval


  hi Oliver :)

  try with GAP 20 instead of STYLE 1 20 END (which is deprecated btw, use 
PATTERN now)

  cheers,
  thomas


  On Tue, Jan 12, 2010 at 15:04, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

hello

I have a weird problem with symbology.
Im rendering a line with dots, but the dots are not round but slightly 
oval, see image http://dev.camptocamp.com/~ochriste/tilecache.png

Any idea what could be the cause ?

Here is my symbol definition:

SYMBOL
 NAME linie-gepunkt10
 TYPE ELLIPSE
 POINTS
  1 1
 END
 STYLE
  1 20
 END
 FILLED TRUE
END

and my layer's class definition:

 CLASS
  EXPRESSION 3
  MAXSCALEDENOM 21
  STYLE
SYMBOL linie-gepunkt10
SIZE 10
COLOR 102 51 0
  END
  STYLE
SYMBOL linie-gepunkt10
SIZE 8
COLOR 204 102 0
  END
 END

and the map outputformat:

OUTPUTFORMAT
 NAME png
 DRIVER AGG/PNG
 MIMETYPE image/png
 IMAGEMODE RGBA
 EXTENSION png
 FORMATOPTION INTERLACE=OFF
END 
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] symbol circle (ellipse) rendered not round but oval

2010-01-12 Thread Oliver Christen
GAP 20 is really giving unwanted result:

http://dev.camptocamp.com/~ochriste/tilecache9.png

though using only one style instead of two seems to solve the oval problem (but 
why ???)

style
 symbol dots
 size 8
 color 204 102 0
 outlinecolor  102 51 0
end

I dont understand how GAP is supposed working:

with GAP 2, I had ~10px between dots,
with GAP 3, I had ~18px between dots,
with GAP 4, I had ~26px between dots

whats the logic ?
  - Original Message - 
  From: thomas bonfort 
  To: Oliver Christen 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Tuesday, January 12, 2010 4:36 PM
  Subject: Re: [mapserver-users] symbol circle (ellipse) rendered not round but 
oval


  symbol
   name dots
   type ellipse
   points 1 1 end
   gap 20
  end

  ...

  style
   symbol dots
   size 8
   color r g b
   outlinecolor  r g b
  end





  On Tue, Jan 12, 2010 at 16:31, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

Hi Thomas :)

thanks for the answer, but sadely it didnt solve my problem:

here are a few tests I did:

using GAP 20 gave me really weird result:
http://dev.camptocamp.com/~ochriste/tilecache6_gap_20.png

using PATTERN 1 22 END gave me better result, but the dot are still 
slightly oval:
http://dev.camptocamp.com/~ochriste/tilecache7_pattern.png

I wondered if the visual effect may have come from the fact im pilling up 
two circle one over the other, but even when trying to display only the outer 
or inner circles, the circle are still slightly off:
http://dev.camptocamp.com/~ochriste/tilecache4_outercircle.png
http://dev.camptocamp.com/~ochriste/tilecache5_innercircle.png
(both at once http://dev.camptocamp.com/~ochriste/tilecache3_bothcircle.png)

any other idea? ^_^
  - Original Message - 
  From: thomas bonfort 
  To: Oliver Christen 
  Cc: mapserver-users@lists.osgeo.org 
  Sent: Tuesday, January 12, 2010 3:18 PM
  Subject: Re: [mapserver-users] symbol circle (ellipse) rendered not round 
but oval


  hi Oliver :)

  try with GAP 20 instead of STYLE 1 20 END (which is deprecated btw, use 
PATTERN now)

  cheers,
  thomas


  On Tue, Jan 12, 2010 at 15:04, Oliver Christen 
oliver.chris...@camptocamp.com wrote:

hello

I have a weird problem with symbology.
Im rendering a line with dots, but the dots are not round but slightly 
oval, see image http://dev.camptocamp.com/~ochriste/tilecache.png

Any idea what could be the cause ?

Here is my symbol definition:

SYMBOL
 NAME linie-gepunkt10
 TYPE ELLIPSE
 POINTS
  1 1
 END
 STYLE
  1 20
 END
 FILLED TRUE
END

and my layer's class definition:

 CLASS
  EXPRESSION 3
  MAXSCALEDENOM 21
  STYLE
SYMBOL linie-gepunkt10
SIZE 10
COLOR 102 51 0
  END
  STYLE
SYMBOL linie-gepunkt10
SIZE 8
COLOR 204 102 0
  END
 END

and the map outputformat:

OUTPUTFORMAT
 NAME png
 DRIVER AGG/PNG
 MIMETYPE image/png
 IMAGEMODE RGBA
 EXTENSION png
 FORMATOPTION INTERLACE=OFF
END 
___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users




___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] reprojected layer = changed min/max scale ?

2009-07-17 Thread Oliver Christen

Hello

This is probably a silly question but I cant find an answer for it.

I have a mapfile with a given EPSG projection (EPSG:27582)
In this mapfile I have a WMS layer with a different projection (EPSG:4326)

when displayed, the max scale of the wms layer is different from the max 
scale value specified in the layer's definition in the mapfile


so I would like to know, when mapserver does the reprojection of the wms 
layer from EPSG:4326 to EPSG:27582, the min/max scale are also modified in 
the process ?


thanks in advance

Oliver 


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


[mapserver-users] Re: phpmapscript whichShapes return code ?

2008-09-09 Thread Oliver Christen

Hi

using mapserver 5.0.2/php 5.2.5

im doing something like this:

  $map = ms_newmapobj(d:/msapps/gmap-ms40/htdocs/gmap75.map);
  $layer = $map-getLayerByName('road');
  $status = $layer-open();
  $status = $layer-whichShapes($map-extent);
  while ($shape = $layer-nextShape())
  {
 echo $shape-index .br\n;
  }
  $layer-close();

when checking the $status value of the $layer-whichShapes($map-extent); 
instructions I get the values 0 or 2.

0 seems to be equal to MS_SUCCESS
but 2 is definitively not equal to MS_FAILURE

I did that:

if ($status === MS_SUCCESS) {
   print success; // 0
}  else if ($status === MS_FAILURE) {
   print failure;  // 1?
} else {
   print unknown; // 2
}

I never get failure, I only get success or unknown

so, what is the meaning of a code 2 value?

thanks in advance
Oliver 


___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users


Re: [mapserver-users] mapscript queryByPoint - tolerancedepends ofresolution ?

2008-07-15 Thread Oliver Christen

Hi

The problem was not in mapserver but came from our script:
when the dpi is changed, the tolerance is globaly adjusted so all the 
related features dont get over/undersized.


But when a value of -1 is used in queryByPoint, the global tolerance 
recalculated result in -3.25~ when passing from 96 to 150 dpi
and mapserver do not know what to do with such value, it only understand -1 
or =0.


The problem was solved by specifying a special case when -1 was used in our 
script for queryByPoint, so the tolerance stay at -1 at 96 and 150 dpi


Could you possible modify a bit the mapscript documentation of  queryByPoint 
?
It say Passing buffer =0 defaults to tolerances set in the map file ... 
but thats not exactly the case.
It would be Passing buffer -1 defaults to tolerances set in the map 
file...


Thank you
Best regards
Oliver

- Original Message - 
From: Steve Lime [EMAIL PROTECTED]
To: Oliver Christen [EMAIL PROTECTED]; 
mapserver-users@lists.osgeo.org

Sent: Wednesday, July 09, 2008 6:39 PM
Subject: Re: [mapserver-users] mapscript queryByPoint - tolerancedepends 
ofresolution ?



Tolerance computations are based on cellsize. Resolution affects only scale 
computations, so it
is not clear to me how one could affect the other. I just went through the 
code to verify. Is there
any chance that layer scale limits are coming into play here and you're not 
querying the layer you

think you are? Does the behavior persist in 5.x?

If you can create a concise test case I'll gladly take a look.

Steve

On 7/9/2008 at 7:41 AM, in message 
[EMAIL PROTECTED],

Oliver Christen [EMAIL PROTECTED] wrote:

Hi

I have a problem when doing a queryByPoint with phpmapscript 4.10.1

By default, the resolution of my map is 96 dpi.
I do a queryByPoint on a layer containing polygonal areas. The point is
right in the middle of one area.
I use a default tolerance (-1) corresponding, if I read correctly the
documentation, to 3 pixels.

The result obtained is the area containing the coordinates I used, it's
absolutly normal and I get what I expect.

Now, if I dynamicaly change the resolution of the map to 150 dpi and then 
do


exactly the same queryByPoint as before,
I get nothing!
Mapserver consider there was nothing at the coordinate I provided.

Then, if I redo the query, at 150 dpi, BUT with a tolerance set to 10
(instead of -1), I get a correct and expected result.

So, somehow, Mapserver seems to make some very weird calculation with the
tolerance and resolution which make it to believe there is nothing when 
then


is something.

have you ever heard of such behaviour?
is that a bug?

any help would be very welcome.

thanks in advance
Oliver

___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users



___
mapserver-users mailing list
mapserver-users@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/mapserver-users