[GRASS-user] v.clean oddities

[M S]

sorry for the repetitive posting about the same issue, but i'm determined to
get this.  I have been at this for a week, but think I finally found what i
need to do.  hopefully there is a way to automate this effort.  i'm not
getting the results i would anticipate.

I whittled my data down to a subset because running v.net.steiner on 28,000
lines with 208 points took 3 days.  when i used 2,000 lines and 17 pionts i
was able to get some results quicker and try to figure out whats wrong.

i have points (buildings) and lines (streets).  I used v.distance to make
the connecting lines.  i used v.category to add categories to the connecting
lines, because all the cats were zero.  when i would patch them together
like this, the connectors wound up on on layer 0.  not sure if this matters
for network analysis.  when i add categories and patch them together, the
lines are on layer 1 and the points are on layer 1.  all seems good so far.


v.clean.... i am very familiar with vector data editing and operations in
a/i, but i dont understand the outputs from v.clean.  i have used default
values and also specified thresholds.  not all the lines get snapped to new
lines from the connectors to the streets.  in the last run of 17, 1 line did
not intersect/snap to the streets, even with varying thresholds.  i found
that i had to use v.digit to add a verticie and split the line prior to
v.clean (after knowing which lines were not snapping).

Is there a way to automate this?  Does v.clean have to have a verticie in
the threshold radius to snap to the street lines?  Or can it snap nodes to
lines when a verticie is not in the threshold?

Using v.distance, i would expect the connectors to be exactly at the street
lines.  in the event that they are such a fraction of a distance from the
streets, a "1" threshold should get them snapped to the streets.  any reason
why it wouldnt?  it seems that there has to be a verticie in the threshold
and the man pages illude to this as well

thanks,
Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/grass-user/attachments/20060701/9951031c/attachment.html