Re: [Therion] Revisiting Breaking extended elevations on specific stations
last weekend i completed the survey of a cave with some parallel close pits connected through several "windows". the last survey joined the previous surveys at seven stations. needless to say, i was not able to tell therion how to break the centerline for the extended elevation. Eventually i removed a few "equates" (the loops closed quite well anyways) extended elevation requires a spanning tree of the centerline graph. "extend" is meant to tell therion how to break loops "removing" edges/legs. I suspect that this semantics of "extend stop" is not sufficient because if the user moves the data around (which is acceptable because in therion data can be reordered) the result could be different. I believe that we need a way the user tell therion how to construct the spanning tree without ambiguities. The only solution i can see is telling which edge to remove. The drawback is that each station appears only once (at one node) in the tree, and there cannot be two stations "7" as in Tarquin's problem. back to the problem i experienced: what about an option to "equate" to tell therion that it must not be used for the extended elevation ? marco
`> needless to say, i was not able to tell therion how to break the centerline for the extended elevation. Eventually i removed a few "equates"
(the loops closed quite well anyways)
what about an option to "equate" to tell therion that it must not be used for the extended elevation ? marco
Thanks Marco Something important I had not considered. I wonder what it does to tarquin's example. Of course, there are two types of equate, although Therion rightly does not distinguish. It was only important for very old software about 30 years ago. The first type of equate, is that which is necessary to join one survey to the next. Without this, you only have individual surveys that are not connected in any way. It would make no sense for Therion's extend to ignore these. The second type of equate, is the second, third etc to a particular survey, which cause the two centrelines to form loops. Ah, and there is possibly a third distinction. Those defined explicitly with 'equate 1@survey2 43@survey1' and those that just happen to traverse the same station more than once within a particular survey trip. I imagine you mean that a proposed 'extend ignore-equates' would be telling Therion to honour only those equates necessary to join the extended survey network, and ignore all other equates? I guess 'extend ignore-equates' could honour namespaces, and be started at particular stations/legs and stopped at particular stations legs? Maybe... extend [equates (default) | ignore-equates ] [ [station1] [[station2]] ] This could be used where the extended network is defined 'in-line' with the survey data, and 'in blocks' separated from the survey data. Bruce
Something important I had not considered. I wonder what it does to tarquin's example.
I don't even begin to know how to cope with the situation where there is a separate branch from 7 to 12 in my example, where you want 2-10-9-8-7-12 to be shown as one branch in the extended elevation, and 1-2-3-4-5-6-7-11 in the other branch, with two station 7s. At that point, I think the only way to cope with it is to say: 7 7a 0 0 0 7a 8 1 2 3 ... 7a 12 1 2 3 extend ignore 7-7a Manipulate the data to add an extra station, allowing a break. I also don't think this discussion of "extend stop 8 7" can cope with it either. I sense a large spanner in the works... Tarquin
participants (3)
-
Bruce Mutton -
Marco Corvi -
Tarquin Wilton-Jones