<div dir="ltr">Hi Bruce,<div><br></div><div>An interesting read. I think the test of any model is how well it handles the exceptions. In the case of my own projects (where I use my recommended structure of course) I have cases where I have to draw up a section of cave which spans more than one trip in one drawing (typically difficult junctions where approach passages were surveyed on different trips). So here I have to add a .th2 file at a higher level than the trip and have to draw the scrap referencing the station names with more than just the number. The tip to specify the namespace in the scrap header is how I handle this where I can keep each scrap to an area belonging to a single trip (I still might want to draw multiple scraps in the same map editor file, hence the need to place the .th2 file outside the trip). But in some cases to get things to output how I want, there is no way to avoid having stations from multiple trips in the same scrap. So here each station name needs to include the namespace. So in reality the model I recommend only works in idea cases, and needs to be merged with parts of some of your other models to deal with these exceptions. The key thing I try to avoid as far as possible is having to add namespaces to each station (as it means you can no longer just click on the stations in a sketch imported from PocketTopo and have the station numbers set automatically). I usually end up clicking on all the station points to give them automatically assigned numbers, and then add the namespace using search and replace in a plain text editor.</div><div><br></div><div>It would be good to add an analysis of the pros and cons of each structure to your document, covering how they handle the imperfect structure of real world data.</div><div><br></div><div>Footleg</div></div><br><div class="gmail_quote"><div dir="ltr">On Thu, Jul 6, 2017 at 12:38 PM Vasily Vl. Suhachev via Therion <<a href="mailto:therion@speleo.sk">therion@speleo.sk</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF">
<p>Hello Bruce and All<br>
</p>
<p>I had to draw a few large caves and I want to share my
experience:<br>
<br>
1) I always keep scanned survey notebooks. This is an artifact
obtained in a cave and is the source of truth.<br>
<br>
2) I store the entire centerline in a separate directory with a
split into files on a trip basis. I think this is more correct
because the centerline is a direct reflection of the survey
notebooks. In the case of an error, this allows you to quickly
navigate to notebooks. Entire cave centerline assembled in index
file inside centerline directory.<br>
<br>
3) I do not tie maps tightly to survey trips because it's usually
more convenient to draw a map that covers a cave part (passage,
grotto) rather than surveying trip.</p>
<p>I call such dedicated part of the cave as `subsystem`. Subsystem
maps are stored in separate subsystem directories. In each
directory there are files `thconfig`, * .th2 and a file `<a href="http://maps.th" target="_blank">maps.th</a>`
in which the definition of maps and joins is stored.<br>
</p>
<p>I store the definition of maps of the entire cave in the top
directory in the `<a href="http://maps.th" target="_blank">maps.th</a>` file in which I connect the` <a href="http://maps.th" target="_blank">maps.th</a>`
files from the subsystems' directories. I do not use therion
namespaces for maps because the number of maps are much smaller
than the surveys. Uniqueness of names of maps I support manually.</p>
<p>Folder structure example:<br>
</p>
<tt>cave</tt><tt><br>
</tt><tt><tt> +-- survey</tt><tt> </tt></tt><tt><tt><tt>#
scanned notebooks</tt><tt><br>
</tt></tt></tt><tt><tt><tt><tt> +-- trip1<br>
+-- 1.jpg<br>
</tt></tt></tt></tt><tt><tt><tt><tt><tt> +-- 2.jpg<br>
</tt></tt></tt></tt></tt><tt><tt><tt><tt><tt><tt><tt><tt><tt>
+-- 3.jpg<br>
</tt></tt></tt></tt></tt></tt></tt></tt></tt><tt><tt><tt><tt><tt>
+-- trip2<br>
</tt></tt></tt></tt></tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt>
+-- trip3<br>
</tt></tt></tt></tt></tt></tt></tt></tt></tt></tt></tt></tt></tt></tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt><tt>
+-- trip4<br>
</tt></tt></tt></tt></tt></tt></tt></tt></tt></tt></tt></tt></tt></tt>
+-- cl</tt><tt> # centerline<br>
+-- <a href="http://cave.th" target="_blank">cave.th</a> # index file with equates<br>
+-- <a href="http://trip1.th" target="_blank">trip1.th</a><br>
</tt><tt><tt> +-- <a href="http://trip2.th" target="_blank">trip2.th</a><br>
</tt></tt><tt><tt><tt><tt> +-- <a href="http://trip3.th" target="_blank">trip3.th</a><br>
</tt></tt></tt></tt><tt><tt><tt><tt> +-- <a href="http://trip4.th" target="_blank">trip4.th</a><br>
</tt></tt> +-- model</tt><tt><br>
</tt></tt><tt><tt><tt> +-- thconfig</tt><tt><br>
</tt><tt>
</tt></tt> +-- map</tt><tt><br>
</tt><tt> +-- p # plan</tt><tt><br>
</tt><tt> +-- subsystem1</tt><tt><br>
</tt><tt> +-- 1.th2</tt><tt><br>
</tt><tt> +-- 2.th2</tt><tt><br>
</tt><tt>
+-- <a href="http://maps.th" target="_blank">maps.th</a></tt><tt><br>
</tt><tt> +-- thconfig</tt><tt><br>
</tt><tt>
+-- subsystem2</tt><tt><br>
</tt><tt>
</tt><tt><tt> +-- 3.th2</tt><tt><br>
</tt><tt> +-- <a href="http://maps.th" target="_blank">maps.th</a></tt><tt><br>
</tt><tt> +-- thconfig</tt><tt><br>
</tt><tt>
</tt> +-- <a href="http://maps.th" target="_blank">maps.th</a></tt><tt><br>
</tt><tt>
+-- thconfig</tt><tt><br>
</tt><tt>
</tt><tt> +-- rr # extended elevation</tt><tt><br>
</tt><tt>
</tt><tt> +-- r240 # elevation</tt><tt> at
240<br>
</tt> <br>
<br>
<div class="m_1172764276928653763moz-cite-prefix">06.07.2017 03:08, Bruce Mutton via
Therion пишет:<br>
</div></div><div text="#000000" bgcolor="#FFFFFF">
<blockquote type="cite">
<div class="m_1172764276928653763WordSection1">
<p class="MsoNormal">A little while ago I explored the effects
of different ways of relating scraps and maps to surveys, and
to the files that contain them.<u></u><u></u></p>
<p class="MsoNormal">I came up with the attached text, which I
might get around to turning into a wiki page sometime. There
are links to some examples already available via the wiki.<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">Before I wiki-ise it, I’d like some
feedback, or criticism, as appropriate! Is it helpful? (I
think so)<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">Footleg, I plagiarised your page, just a
little.<u></u><u></u></p>
<p class="MsoNormal">Also, if Danilo Magnani or Will Urbanski
are still listening, or if someone knows how to contact them,
I’d like permission to put on the wiki the project samples
they posted to this forum some years ago. Those are named
Complesso - Buc di V, and DDC – Radio Collar Cave.<u></u><u></u></p>
<p class="MsoNormal"><u></u> <u></u></p>
<p class="MsoNormal">Bruce<u></u><u></u></p>
</div>
<br>
<fieldset class="m_1172764276928653763mimeAttachmentHeader"></fieldset>
<br>
<pre>_______________________________________________
Therion mailing list
<a class="m_1172764276928653763moz-txt-link-abbreviated" href="mailto:Therion@speleo.sk" target="_blank">Therion@speleo.sk</a>
<a class="m_1172764276928653763moz-txt-link-freetext" href="https://mailman.speleo.sk/listinfo/therion" target="_blank">https://mailman.speleo.sk/listinfo/therion</a>
</pre>
</blockquote>
<br>
</div><div text="#000000" bgcolor="#FFFFFF"><pre class="m_1172764276928653763moz-signature" cols="72">--
WBR, Vasily</pre>
</div>
_______________________________________________<br>
Therion mailing list<br>
<a href="mailto:Therion@speleo.sk" target="_blank">Therion@speleo.sk</a><br>
<a href="https://mailman.speleo.sk/listinfo/therion" rel="noreferrer" target="_blank">https://mailman.speleo.sk/listinfo/therion</a><br>
</blockquote></div>