Re: [Qgis-developer] QGIS inconsistency

[Bernd Vogelgesang]

Hi Steven,

just having a quick look at the data you provided:
- your Munros.shp doesn't have 27700 projection, but some user defined and  
only becomes visible when setting to 27700 manually.
- the geojson and the shape file have quite some offset. (75 whatsoevers,  
meteres, miles ??)


Apart from the fact that there might  be a bug, your data seems to me more  
buggy ;)

Besides that, i never dare to do any spatial analysis within epsg 4326.



Greetz
Bernd

Am 18.10.2013, 21:13 Uhr, schrieb Steven Horner ste...@stevenhorner.com:


Hello,

I have been doing a project to calculate mountain viewsheds, basically
where a mountain summits can be seen from.

I have calculated the viewsheds and polygonized to Vector. So far so  
good,
but then I wanted to show which mountains could be seen from the summit  
of

other mountains I appear to have 2 problems.

1. I ran a spacial query to show how many mountains (Munros) are Within  
my

polgonized viewshed. This returns the Munros. Unfortunately I get a
different number in QGIS 1.8 to QGIS 2.0.1, I don't know which is true

2. I had saved my Viewshed originally as a shapefile with a CRS of 27700
(British National Grid), I saved a version as GeoJSON with CRS of 4326.  
If

I run the same spacial query on this file I get a different number of
Munros again.

So I now have 4 different figures from what I would have expected to be
identical.

Is data lost when a shapefile is saved as a GeoJSON file in a different  
CRS?

and which should I beleive 1.8 or 2.0.1 or neither?

If anyone wants to test this you can download the files here:
Viewshed  
Shapefilehttps://dl.dropboxusercontent.com/u/2657852/AmBodach_shp.zip-

CRS of 27700
Viewshed  
GeoJSONhttps://dl.dropboxusercontent.com/u/2657852/AmBodach_geojson.zip-

Converted fromthe above but with CRS of 4326
Munros Shapefile  
https://dl.dropboxusercontent.com/u/2657852/Munros.zip -

CRS of 27700

Wish I had realised there was something wrong before I had calculated 200
of these.

Thanks,
Steven



--
Bernd Vogelgesang * Siedlerstrasse 2 * 91083 Baiersdorf * +49-9133-825374
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer

[Qgis-developer] QGIS inconsistency

[Steven Horner]

Hello,

I have been doing a project to calculate mountain viewsheds, basically
where a mountain summits can be seen from.

I have calculated the viewsheds and polygonized to Vector. So far so good,
but then I wanted to show which mountains could be seen from the summit of
other mountains I appear to have 2 problems.

1. I ran a spacial query to show how many mountains (Munros) are Within my
polgonized viewshed. This returns the Munros. Unfortunately I get a
different number in QGIS 1.8 to QGIS 2.0.1, I don't know which is true

2. I had saved my Viewshed originally as a shapefile with a CRS of 27700
(British National Grid), I saved a version as GeoJSON with CRS of 4326. If
I run the same spacial query on this file I get a different number of
Munros again.

So I now have 4 different figures from what I would have expected to be
identical.

Is data lost when a shapefile is saved as a GeoJSON file in a different CRS?
and which should I beleive 1.8 or 2.0.1 or neither?

If anyone wants to test this you can download the files here:
Viewshed 
Shapefilehttps://dl.dropboxusercontent.com/u/2657852/AmBodach_shp.zip-
CRS of 27700
Viewshed 
GeoJSONhttps://dl.dropboxusercontent.com/u/2657852/AmBodach_geojson.zip-
Converted fromthe above but with CRS of 4326
Munros Shapefile https://dl.dropboxusercontent.com/u/2657852/Munros.zip -
CRS of 27700

Wish I had realised there was something wrong before I had calculated 200
of these.

Thanks,
Steven
___
Qgis-developer mailing list
Qgis-developer@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/qgis-developer