Hi All,
I just finished using v.digit to retrace the contour lines of a vector file.
The old vector file was created from a raster file. The lines this process
created were thick and broken.
My first question is, how do I remove the old lines so I just have the new
lines that I just traced.
Second question. The book, and v.support, make mention about nodes. I
assume these are a kind of anchor point. Will I need these for the
contours, and if I do, would it be some kind of box drawn around the
contours that the end points would be snapped to?
Third Question. When I go to label the contours: some of the contours go
off the end of the map then come back on at a later point. Do I have to do
anything special, or will v.digit be able to keep track of this.
Last question: I will be receiving contour maps made by the Irish ordinance
survey in .dxf form from my professor. The contours should have been
labelled by the OS. I expect that v.in.dxf should do the job. Are there
any surprises that I should watch for?
Kurt
I also have a question on v.digit. If I make changes in v.digit and then
decide that they are not correct or are really bad then how can I bail out
without saving the changes. I can't seem to find a way short of using
"kill" from another console. That isn't a very good solution.
Hello
On Mon, 24 Mar 2003, Kurt Springs wrote:
Last question: I will be receiving contour maps made by the Irish ordinance
survey in .dxf form from my professor. The contours should have been
labelled by the OS. I expect that v.in.dxf should do the job. Are there
any surprises that I should watch for?
I have used v.in.dxf and v.in.dxf3d to input contour maps supplied by the
Ordnance Survey of Northern Ireland; I don't know if OSi use the same format
(probably not!). But the problem I found with v.in.dxf3d was it attached
the attributes of each line in the dxf file as an attribute of each line in
GRASS. However in GRASS the attributes are all supposed to be integer
numbers greater than 0 and for areas near the sea, where the attribute of
the HWMMT line was 0, v.support just silently deleted all these lines.
If you don't mind going down into the files in the GRASS database, the
solution I found was to use a Perl script to move the attributes
from the attribute file (dig_att directory) to a new categories file
(dig_cats directory), and replace the dig_att file with a new one
containing only integer numbers greater than 0.
I have attached the script below but it is a bit crude as I only had to
use it a few times. It is important to run it after running v.in.dxf and
v.in.dxf3d but before running v.support. It must be run from the dig_att
directory, with the name of the vector file as its only command-line
argument.
As far as I can remember (this was at least a year ago) v.in.dxf3d has a
few bugs, documentation and otherwise. From my notes I see that 'lines='
is not optional and must be supplied, and 'prefix=' is not an option here
so you may have to rename the dig_att file after running v.in.dxf3d if
you used the prefix option with v.in.dxf.
Paul
(attachments)
dxffix.pl (1.38 KB)