At the end of v.proj processing of the public_forests map three warnings
were presented:
WARNING: Number of incorrect boundaries: 39934
WARNING: Number of centroids outside area: 528
WARNING: Number of duplicate centroids: 107
The v.clean module has 13 tools and an action threshold for use. I want to
learn which tools and use sequence I should apply to clean this map. Also,
how do I determine what threshold to apply?
TIA,
Rich
On Wed, Jun 27, 2018 at 4:43 PM, Rich Shepard <rshepard@appl-ecosys.com> wrote:
At the end of v.proj processing of the public_forests map three warnings
were presented:WARNING: Number of incorrect boundaries: 39934
WARNING: Number of centroids outside area: 528
WARNING: Number of duplicate centroids: 107The v.clean module has 13 tools and an action threshold for use. I want to
learn which tools and use sequence I should apply to clean this map.
You may want to take a look here:
https://grasswiki.osgeo.org/wiki/Vector_topology_cleaning
Edits are always welcome in the Wiki.
Also, how do I determine what threshold to apply?
Normally the module gives you a suggestion.
Markus
On Wed, 27 Jun 2018, Markus Neteler wrote:
You may want to take a look here:
https://grasswiki.osgeo.org/wiki/Vector_topology_cleaning
Normally the module gives you a suggestion.
Markus,
Thanks for the URL. The focus is on v.in.ogr rather than v.proj so I'll
try to translate from the former to the latter.
Best regards,
Rich
On Wed, Jun 27, 2018 at 8:59 PM, Rich Shepard <rshepard@appl-ecosys.com> wrote:
On Wed, 27 Jun 2018, Markus Neteler wrote:
You may want to take a look here:
https://grasswiki.osgeo.org/wiki/Vector_topology_cleaningNormally the module gives you a suggestion.
Markus,
Thanks for the URL. The focus is on v.in.ogr rather than v.proj so I’ll
try to translate from the former to the latter.
In your case, the focus is also on v.in.ogr rather than v.proj. Try to get a reasonably clean vector with v.in.ogr (+ v.clean) before reprojecting the vector.
Markus M
Best regards,
Rich
grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user
On Wed, 27 Jun 2018, Markus Metz wrote:
In your case, the focus is also on v.in.ogr rather than v.proj. Try to get
a reasonably clean vector with v.in.ogr (+ v.clean) before reprojecting
the vector.
Markus,
Yes, I'll re-import it and play with the threshold values.
Thanks,
Rich
On Wed, 27 Jun 2018, Markus Neteler wrote:
You may want to take a look here:
https://grasswiki.osgeo.org/wiki/Vector_topology_cleaning
Markus,
I find this sentence ambiguous: "For import, try to find a snapping
threshold for v.in.ogr that produces an error-free output (during import the
module suggests a threshold)." I've not seen a suggested threshold when I
run v.in.ogr. When I press the [Enter] key it takes off and does its thing.
Does 'suggest a threshold' mean that v.in.ogr picks one and uses it without
user knowledge or interaction?
Regards,
Rich
On Thu, 28 Jun 2018, Rich Shepard wrote:
I find this sentence ambiguous: "For import, try to find a snapping
threshold for v.in.ogr that produces an error-free output (during import
the module suggests a threshold)." I've not seen a suggested threshold
when I run v.in.ogr. When I press the [Enter] key it takes off and does
its thing. Does 'suggest a threshold' mean that v.in.ogr picks one and
uses it without user knowledge or interaction?
The answer to another thread answered this one, too.
Rich
On Wed, 27 Jun 2018, Markus Neteler wrote:
You may want to take a look here:
https://grasswiki.osgeo.org/wiki/Vector_topology_cleaning
Markus,
After repeating v.in.ogr with larger snap thresholds until it told me it
did as well as it could and I should now use v.clean, is there a tool that
tells me when a polygonal map is cleaned as well as it can be?
Regards,
Rich
On Fri, Jun 29, 2018 at 5:49 PM, Rich Shepard <rshepard@appl-ecosys.com> wrote:
On Wed, 27 Jun 2018, Markus Neteler wrote:
You may want to take a look here:
https://grasswiki.osgeo.org/wiki/Vector_topology_cleaningMarkus,
After repeating v.in.ogr with larger snap thresholds until it told me it
did as well as it could and I should now use v.clean, is there a tool that
tells me when a polygonal map is cleaned as well as it can be?
After import with v.in.ogr, overlapping parts have a category value set in layer 2. If you want to get rid off all overlapping parts, check with v.category op=report if any centroids still have a category value in layer 2. Sometimes these areas can be cleaned away with v.clean tool=rmarea, sometimes manual editing with the vector digitizer is needed.
Markus M
Regards,
Rich
grass-user mailing list
grass-user@lists.osgeo.org
https://lists.osgeo.org/mailman/listinfo/grass-user
On Fri, 29 Jun 2018, Markus Metz wrote:
After import with v.in.ogr, overlapping parts have a category value set in
layer 2. If you want to get rid off all overlapping parts, check with
v.category op=report if any centroids still have a category value in layer
2. Sometimes these areas can be cleaned away with v.clean tool=rmarea,
sometimes manual editing with the vector digitizer is needed.
Markus,
I have seen that overlapping parts are assigned cat 2, but did not know
what to do about it.
The warnings left after importing with v.in.ogr are:
WARNING: Number of incorrect boundaries: 39934
WARNING: Number of centroids outside area: 528
WARNING: Number of duplicate centroids: 107
and I've looked at the v.clean tools (and the wiki page) trying to find the
appropriate ones. I tried 'v.clean -c tool=break type=line' which runs
rmdupl after breaking and rebuilding topology, but v.proj reports the above
warnings. I'll check v.category first, then read the 'See also' references
at the bottom of that wiki page.
Thanks again,
Rich
On Fri, 29 Jun 2018, Markus Metz wrote:
After import with v.in.ogr, overlapping parts have a category value set in
layer 2. If you want to get rid off all overlapping parts, check with
v.category op=report if any centroids still have a category value in layer
2. Sometimes these areas can be cleaned away with v.clean tool=rmarea,
sometimes manual editing with the vector digitizer is needed.
Markus,
Followed your advice and there are still the same warnings with the same
values. Is there a way for me to determine where in this large map of
publicly-owned forests in the state the 107 overlapping polygons are
located?
Regards,
Rich
Rich Shepard wrote
On Fri, 29 Jun 2018, Markus Metz wrote:
After import with v.in.ogr, overlapping parts have a category value set
in
layer 2. If you want to get rid off all overlapping parts, check with
v.category op=report if any centroids still have a category value in
layer
2. Sometimes these areas can be cleaned away with v.clean tool=rmarea,
sometimes manual editing with the vector digitizer is needed.Markus,
Followed your advice and there are still the same warnings with the
same
values. Is there a way for me to determine where in this large map of
publicly-owned forests in the state the 107 overlapping polygons are
located?Regards,
Rich
_______________________________________________
grass-user mailing list
grass-user@.osgeo
d.vect layer=2 .... ?
the manuals are a good source where to look.
-----
best regards
Helmut
--
Sent from: http://osgeo-org.1560.x6.nabble.com/Grass-Users-f3884509.html
On Fri, 29 Jun 2018, Helmut Kudrnovsky wrote:
d.vect layer=2 .... ?
Helmut,
I did this then ran v.clean. The number of duplicate centroids dropped
from 107 to 4, but refreshing the display showed the same areas as when
there were more duplicates. When time permits I might try looking at each
small portion of the map and try to find the four duplicates, but for now
it's not worth the time investment.
Ideally, the overlaps and duplicates would/could be displayed in a
contrasting color for easy identification.
Best regards,
Rich
Ideally, the overlaps and duplicates would/could be displayed in a
contrasting color for easy identification.
citing the d.vect manual:
color=name
Feature color
Either a standard color name, R:G:B triplet, or "none"
Default: black
fill_color=name
Area fill color
Either a standard color name, R:G:B triplet, or "none"
Default: 200:200:200
?
-----
best regards
Helmut
--
Sent from: http://osgeo-org.1560.x6.nabble.com/Grass-Users-f3884509.html
On Fri, 29 Jun 2018, Helmut Kudrnovsky wrote:
citing the d.vect manual:
color=name
Feature color
Either a standard color name, R:G:B triplet, or "none"
Default: blackfill_color=name
Area fill color
Either a standard color name, R:G:B triplet, or "none"
Default: 200:200:200
I've not specified colors by layers. I'll look into doing this.
Regards,
Rich