Re: [OpenMap Users] VMAP2Shape

From: Don Dietrick <dietrick_at_bbn.com>
Date: Wed, 10 Nov 2004 19:23:41 -0500

Hi Todd,

I think the eps is the decimal degree distance that the thinning
algorithm uses in determining if two points can be merged. The
threshold is a decimal degree radius, where polygons are eliminated if
all of their points fit within that radius.


On Nov 8, 2004, at 7:05 PM, Webb, Todd wrote:

> Hey Don
>
> Thanks, it worked and I was able to generate the shape files from our
> VMAP Edition 5 CDs. A .dbf, .shp, and .shx file are created for each.
> However, I still haven't been able to accomplish what I set out to,
> so maybe it would help to explain my goal.
>
> I wanted to create an updated version of the VMAP shape files that are
> available on the OpenMap website, and to have everything set up to do
> it again when we receive Edition 6 of VMAP. So I wanted to create 4
> shape files: one with higher resolution area data and one with higher
> resolution edge data, Then one with overview/reference area data, and
> one with overview edge data. I wanted the high-res files to contain
> as much detail as can be packed into 30Mb, and each of the overview
> shape files to be under 2Mb each, much like the existing shape files
> on the website.
>
> The shape files that I generate are ALWAYS the same size: 12Mb. So
> the past few days I have playing with the options, I have used eps
> values of between .0001 to 1.0 and threshhold values of .001 to 1000.0
> and they seem to have very little affect on the resulting file size...
> it is always almost exactly 12Mb. I guess it would help if I had a
> better understanding of these values and the effective ranges of each.
> I also tried different scale values in the properties file, but to no
> effect.

I don't think scale values are used in the properties file.

>
> So I modified VMAP2Shape.java. I added some additional details to the
> output, and added scale as a command line option. It looked like the
> thinning code was commented out, so I re-enabled it and made the
> thinning algorithm run recursively, so it doesn't matter if the list
> contains an OMGraphics or another OMGraphicList. I tried multiple eps
> and threshhold values again, as well as scales of 30,000 to
> 300,000,000 but to no change. I even tried different resolutions
> passed to the drawTile() method. I can see that graphics are actually
> being trimmed, but the resulting file size is still always 12Mb.

Whoops, that was my bad. I commented out that code when modifying it
to use ESRIShapeExport.

Thanks for the update, I'll check it out. I'm not sure why you would
be seeing files of the same size with different eps settings, if
thinning is enabled.

-Don

> I attached my modified VMAP2Shape.java, and below is the command I am
> running and the contents of the property file.
>
> Am I going to be able to accomplish my goal with this tool, if so,
> what am I missing?
>
> Thanks!!
>
> Todd
>
> : Create thinned coastline
> :java -Xmx256m -cp %OPENMAP_HOME% com.bbn.openmap.layer.vpf.VMAP2Shape
> -props VMAPThinned.properties -prefix vmapCoastline -scale 300000000
> -thin 0.1 0.5 VMAP_Coastline3
> :java -Xmx256m -cp %OPENMAP_HOME% com.bbn.openmap.layer.vpf.VMAP2Shape
> -props VMAPThinned.properties -prefix vmapCoastline -thin 0.1 0.9
> VMAP_Coastline
> java -Xmx256m -cp %OPENMAP_HOME% com.bbn.openmap.layer.vpf.VMAP2Shape
> -props VMAPThinned.properties -prefix vmapCoastline -scale 3000000
> -thin 0.1 0.9 VMAP_Coastline2
>
> : Create thinned political areas (continents)
> :java -Xmx256m -cp %OPENMAP_HOME% com.bbn.openmap.layer.vpf.VMAP2Shape
> -props VMAPThinned.properties -prefix vmapContinents -scale 3000000
> -thin 0.2 0.5 VMAP_Continents
>
> : Create thinned political boundaries
> :java -Xmx256m -cp %OPENMAP_HOME% com.bbn.openmap.layer.vpf.VMAP2Shape
> -props VMAPThinned.properties -prefix vmapPolitical -scale 30000
> -thin 0.3 0.5 VMAP_Borders
>
> ##
> ## property file that directs thinning on VMAP data
> ##
>
> # Just political areas
> vmapContinents.coverageType=bnd
> vmapContinents.featureTypes=area
> vmapContinents.edge=polbnda
> vmapContinents.searchByFeature=true
> vmapContinents.vpfPath=C\:/VMAP/V0EUR/VMAPLV0;C\:/VMAP/V0NOA/VMAPLV0;
> C\:/VMAP/V0SAS/VMAPLV0;C\:/VMAP/V0SOA/VMAPLV0
>
> # Just political boundaries
> vmapPolitical.coverageType=bnd
> vmapPolitical.featureTypes=edge
> vmapPolitical.edge=polbndl
> vmapPolitical.searchByFeature=true
> vmapPolitical.vpfPath=C\:/VMAP/V0EUR/VMAPLV0;C\:/VMAP/V0NOA/VMAPLV0;
> C\:/VMAP/V0SAS/VMAPLV0;C\:/VMAP/V0SOA/VMAPLV0
>
> # Just coastline
> vmapCoastline.coverageType=bnd
> vmapCoastline.featureTypes=edge
> vmapCoastline.edge=coastl
> vmapCoastline.searchByFeature=true
> vmapCoastline.vpfPath=C\:/VMAP/V0EUR/VMAPLV0;C\:/VMAP/V0NOA/VMAPLV0;
> C\:/VMAP/V0SAS/VMAPLV0;C\:/VMAP/V0SOA/VMAPLV0
> #vmapCoastline.scale=300000f
>
>
> -----Original Message-----
> From: Don Dietrick [mailto:dietrick_at_bbn.com]
> Sent: Friday, November 05, 2004 7:14 AM
> To: Webb, Todd
> Cc: Openmap-Users (E-mail)
> Subject: Re: [OpenMap Users] VMAP2Shape NullPointerException
>
>
> It seems to work pretty well...
>
> <VMAP2Shape.java>


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Don Dietrick, dietrick_at_bbn.com
BBN Technologies, Cambridge, MA
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

--
[To unsubscribe to this list send an email to "majdart_at_bbn.com"
with the following text in the BODY of the message "unsubscribe openmap-users"]
Received on Wed Nov 10 2004 - 19:27:03 EST

This archive was generated by hypermail 2.3.0 : Tue Mar 28 2017 - 23:25:06 EDT