Brian,<div><br></div><div>Thanks for running the clean command on the themes!</div><div><br></div><div>For the upcoming 1.3.1 release of Natural Earth our format is now "ne_" prefix on each of the 10m, 50m, and 110m scale set themes.</div>
<div><br></div><div>_Nathaniel</div><div><br></div><div><br><br><div class="gmail_quote">On Fri, Feb 25, 2011 at 01:00, Brian Hamlin <span dir="ltr"><<a href="mailto:maplabs@light42.com">maplabs@light42.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">Hamish-<br>
<br>
I am just finishing a set of tables cleaned via st_MakeValid() in PostGIS 2.0b<br>
Conveniently, I renamed the tables as c10m_this_that<br>
so as to no longer need the quoting..<br>
<br>
I wonder how many people use the .shp files..<br>
or how we should name those..<br><font color="#888888">
<br>
-Brian</font><div><div></div><div class="h5"><br>
<br>
<br>
On Feb 24, 2011, at 8:21 PM, Hamish wrote:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Nathaniel wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I'm not seeing the any geometry errors in ArcGIS (the horror,<br>
the horror) on the 10m lakes theme fresh off the website.<br>
</blockquote>
<br>
Arc is not a topological GIS so it misses (or ignores) many<br>
errors in the data which cause problems when the data is used<br>
in other ways which demand topological correctness.<br>
([river] network analysis, etc)<br>
I understand that Natural Earth is mainly focused on cartographic<br>
needs, but ...<br>
<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
The 1.3 version that was released a couple weeks ago should have<br>
resolved most the obvious topology errors on the themes that<br>
were ID'd to me before. If you want to run a custom repair<br>
geometry / topology script on them as they're imported to the<br>
LiveDemo system, sounds fine to me.<br>
</blockquote>
<br>
Loading a fresh download of 10m-lakes.zip into GRASS GIS (which<br>
is fully topological) exposes a number of errors in the data.<br>
Probably most are due to overlapping areas; small sliver areas<br>
resulting from those after "flattening" are appear below as<br>
areas without centroids (no attributes).<br>
<br>
GRASS65> v.in.ogr dsn=natearth/10m_lakes.shp out=ne_10m_lakes<br>
[...]<br>
Building topology for vector map <ne_10m_lakes_tmp>...<br>
Building areas...<br>
100%<br>
1708 areas built<br>
1472 isles built<br>
Attaching islands...<br>
100%<br>
Number of nodes: 2627<br>
Number of primitives: 4691<br>
Number of points: 0<br>
Number of lines: 0<br>
Number of boundaries: 4691<br>
Number of centroids: 0<br>
Number of areas: 1708<br>
Number of isles: 1472<br>
Number of areas without centroid: 1708<br>
-----------------------------------------------------<br>
Find centroids for layer: 10m_lakes<br>
-----------------------------------------------------<br>
Write centroids:<br>
100%<br>
WARNING: 82 areas represent more (overlapping) features, because polygons<br>
overlap in input layer(s). Such areas are linked to more than 1<br>
row in attribute table. The number of features for those areas is<br>
stored as category in layer 2<br>
-----------------------------------------------------<br>
1358 input polygons<br>
Total area: 1.26155E+12 (1708 areas)<br>
Overlapping area: 2.02361E+11 (82 areas)<br>
Area without category: 2.39549E+10 (240 areas)<br>
-----------------------------------------------------<br>
Building topology for vector map <ne_10m_lakes>...<br>
Registering primitives...<br>
3489 primitives registered<br>
138259 vertices registered<br>
Building areas...<br>
100%<br>
1708 areas built<br>
1472 isles built<br>
Attaching islands...<br>
100%<br>
Attaching centroids...<br>
100%<br>
Number of nodes: 3253<br>
Number of primitives: 3489<br>
Number of points: 0<br>
Number of lines: 0<br>
Number of boundaries: 2021<br>
Number of centroids: 1468<br>
Number of areas: 1708<br>
Number of isles: 1472<br>
Number of centroids outside area: 3<br>
Number of duplicate centroids: 5<br>
Number of areas without centroid: 248<br>
<br>
#extract problems to another layer:<br>
GRASS65> v.extract in=ne_10m_lakes out=ne_10m_lakes_problems layer=2<br>
<br>
displaying that extract shows problems are all in the northern<br>
hemisphere, the Great Lakes of N.America being the biggest<br>
pile of red ink. (Lakes Ontario and Erie look ok though)<br>
<br>
a map layer created with "v.clean error=problem_areas_map" can<br>
help quickly pinpoint where the problems are.<br>
<br>
e.g. an easy place to start looking are the finger lakes within<br>
this bounding box, all of which show errors:<br>
<br>
north: 43N<br>
south: 42:18N<br>
west: 77:24W<br>
east: 76:12W<br>
<br>
(south of Rochester-Syracuse, NY)<br>
<br>
<br>
best,<br>
Hamish<br>
<br>
<br>
<br>
<br>
<br>
</blockquote>
<br>
</div></div></blockquote></div><br></div>