<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Tahoma
}
--></style>
</head>
<body class='hmmessage'><div dir='ltr'>
In your experience, it there a method to predict where failure<div>might occur, or some characteristics of offending geometries...</div><div><br></div><div>I continue to run samples, some large, some small in hopes</div><div>of isolating. Right now running a large sample ie. edge 41330 splitted at face 0</div><div>64551200 or so ms into the process. Tends to slow other processes down</div><div>on my box.</div><div><br></div><div>I was just wondering if through the multitudinous process of st_split, addface, addedgemodface,</div><div>if a progression of small remainders from the splitting (or others) get carried forward</div><div>that would fault st_createtopogeo as it gets around to updating {left,right}_face fields and the like.</div><div><br></div><div><pre class="fragment" style="font-family: monospace, fixed; font-size: 14px; border-top-width: 1px; border-right-width: 1px; border-bottom-width: 1px; border-left-width: 1px; border-top-style: solid; border-right-style: solid; border-bottom-style: solid; border-left-style: solid; border-top-color: rgb(204, 204, 204); border-right-color: rgb(204, 204, 204); border-bottom-color: rgb(204, 204, 204); border-left-color: rgb(204, 204, 204); background-color: rgb(255, 255, 255); margin-top: 4px; margin-bottom: 4px; margin-left: 2px; margin-right: 8px; padding-left: 6px; padding-right: 6px; padding-top: 4px; padding-bottom: 4px; ">01771 <span class="comment" style="color: rgb(128, 0, 0); ">/*******************************************************************************</span>
<a name="l01772" style="color: rgb(26, 65, 168); "></a>01772 <span class="comment" style="color: rgb(128, 0, 0); "> * GEOS-dependent extra functions on LWGEOM</span>
<a name="l01773" style="color: rgb(26, 65, 168); "></a>01773 <span class="comment" style="color: rgb(128, 0, 0); "> ******************************************************************************/</span>
<a name="l01774" style="color: rgb(26, 65, 168); "></a>01774
<a name="l01782" style="color: rgb(26, 65, 168); "></a>01782 <a class="code" href="http://postgis.refractions.net/documentation/postgis-doxygen/d4/d00/structLWGEOM.html" style="color: rgb(0, 0, 255); text-decoration: none; ">LWGEOM</a>* <a class="code" href="http://postgis.refractions.net/documentation/postgis-doxygen/da/de7/liblwgeom_8h_9e7442e41dfb29854c0762ef858b97bf.html#9e7442e41dfb29854c0762ef858b97bf" style="color: rgb(0, 0, 255); text-decoration: none; ">lwgeom_buildarea</a>(<span class="keyword" style="color: rgb(0, 128, 0); ">const</span> <a class="code" href="http://postgis.refractions.net/documentation/postgis-doxygen/d4/d00/structLWGEOM.html" style="color: rgb(0, 0, 255); text-decoration: none; ">LWGEOM</a> *geom) ;</pre></div><div><pre class="fragment" style="font-family: monospace, fixed; font-size: 14px; border-top-width: 1px; border-right-width: 1px; border-bottom-width: 1px; border-left-width: 1px; border-top-style: solid; border-right-style: solid; border-bottom-style: solid; border-left-style: solid; border-top-color: rgb(204, 204, 204); border-right-color: rgb(204, 204, 204); border-bottom-color: rgb(204, 204, 204); border-left-color: rgb(204, 204, 204); background-color: rgb(255, 255, 255); margin-top: 4px; margin-bottom: 4px; margin-left: 2px; margin-right: 8px; padding-left: 6px; padding-right: 6px; padding-top: 4px; padding-bottom: 4px; ">01791 <span class="comment" style="color: rgb(128, 0, 0); "> * Split polygon by line, line by line, line by point.</span>
<a name="l01792" style="color: rgb(26, 65, 168); "></a>01792 <span class="comment" style="color: rgb(128, 0, 0); "> *</span>
<a name="l01793" style="color: rgb(26, 65, 168); "></a>01793 <span class="comment" style="color: rgb(128, 0, 0); "> * Returns all components as a collection.</span>
<a name="l01794" style="color: rgb(26, 65, 168); "></a>01794 <span class="comment" style="color: rgb(128, 0, 0); "> * First element of the collection is always the part which</span>
<a name="l01795" style="color: rgb(26, 65, 168); "></a>01795 <span class="comment" style="color: rgb(128, 0, 0); "> * remains after the cut, while the second element is the</span>
<a name="l01796" style="color: rgb(26, 65, 168); "></a>01796 <span class="comment" style="color: rgb(128, 0, 0); "> * part which has been cut out. We arbitrarely take the part</span>
<a name="l01797" style="color: rgb(26, 65, 168); "></a>01797 <span class="comment" style="color: rgb(128, 0, 0); "> * on the *right* of cut lines as the part which has been cut out.</span>
<a name="l01798" style="color: rgb(26, 65, 168); "></a>01798 <span class="comment" style="color: rgb(128, 0, 0); "> * For a line cut by a point the part which remains is the one</span>
<a name="l01799" style="color: rgb(26, 65, 168); "></a>01799 <span class="comment" style="color: rgb(128, 0, 0); "> * from start of the line to the cut point.</span></pre></div><div>Again, I note the 'arbitrarily take the part on the right' and wonder if there is some carried remainder that</div><div>messes up face boundary edges down the line as these are run after repeated splitting and addedgefacemod operations</div><div>have been registered in what I assume to be a temp table before finally being tested for their true merit as topologies.</div><div><br></div><div>Anyway, my computer continues to search. If I was smarter, I would construct a sieve scein and run this 100 polygons</div><div>at a time, tossing successful topological populations and creating tables for the failures, this process is looking for</div><div>what fails.</div><div>Chris</div><div><pre class="fragment" style="font-family: monospace, fixed; font-size: 14px; border-top-width: 1px; border-right-width: 1px; border-bottom-width: 1px; border-left-width: 1px; border-top-style: solid; border-right-style: solid; border-bottom-style: solid; border-left-style: solid; border-top-color: rgb(204, 204, 204); border-right-color: rgb(204, 204, 204); border-bottom-color: rgb(204, 204, 204); border-left-color: rgb(204, 204, 204); background-color: rgb(255, 255, 255); margin-top: 4px; margin-bottom: 4px; margin-left: 2px; margin-right: 8px; padding-left: 6px; padding-right: 6px; padding-top: 4px; padding-bottom: 4px; "><span class="Apple-style-span" style="font-family: Tahoma; font-size: 13px; white-space: normal; ">> Date: Mon, 5 Dec 2011 11:06:23 +0100</span></pre><div>> From: strk@keybit.net<br>> To: postgis-users@postgis.refractions.net<br>> Subject: Re: [postgis-users] Preparing for Topology - St_CreateTopoGeo<br>> <br>> On Sun, Dec 04, 2011 at 06:55:56PM -0500, Chris English wrote:<br>> > <br>> > Looking for triggers town by town<br>> <br>> Sounds like a good plan.<br>> <br>> --strk; <br>> <br>> ,------o-. <br>> | __/ | Thank you for PostGIS-2.0 Topology !<br>> | / 2.0 | http://www.pledgebank.com/postgistopology<br>> `-o------'<br>> <br>> _______________________________________________<br>> postgis-users mailing list<br>> postgis-users@postgis.refractions.net<br>> http://postgis.refractions.net/mailman/listinfo/postgis-users<br></div></div> </div></body>
</html>