[GRASSLIST:2212] Re: Another problem with r.proj

[Eric G. Miller]

On Sun, Jul 29, 2001 at 11:40:43AM -0700, Rich Shepard wrote:
>    I suspect that I know why r.proj failed for me but worked for Eric. If I'm
> right, we have a major deficiency in GRASS -- IMO, of course. :-)
> 
>   The files I was trying to reproject are vector files: public land survey
> system (PLSS) and mining claims. Once I get everything set up properly and
> working, I'll be converting these to rasters for analytical purposes, but I
> also need the vectors for printed output purposes. I never made the
> connection before between the "r." and the vector files.

Fifty lashes with a wet noodle for you!

>   Is this the problem? Is running affine transformations on the arc-nodes of
> a vector map so much different from running them on a grid?

Why would you use an affine transformation for vectors?  v.proj will do
an inverse followed by a foward projection for each vertex (depending on
input/output projections -- e.g. only the inverse or forward is done if
the target or source is lat/lon).  See v.plant for adding extra vertices
at a specified interval (will reduce apparent error in line work that is
projected if the original map has widely spaced vertices).

-- 
Eric G. Miller <egm2 at jps.net>