[Therion] extend

[Bruce Mutton]

After thinking in a more structured way, and after Evaristo’s clear explanation and request, I have gone back to my original line of thinking; that extend (left right vertical ignore) for splays has no direct meaning.  Perhaps it is more correct to say; it is not generally meaningful to apply extend control statements to splays.  Extend controls applied to legs will, however, affect where the splays are projected.  And in most cases the splay projection can be calculated explicitly without need of further user input.

 

I will try to explain my proposal with a word picture.  I would draw a diagram, but my skills don’t extend that far.

Plan Projection

Consider looking down on the cave, and projecting onto a horizontal plane, the shape and position of the stations, legs and splays. This is a fixed shape, once loops are closed.  It would not be appropriate for user control to move the projected position of splays.  I don’t think there would be any dispute here.

Elevation projection

This is the same as for a plan projection, except that the whole cave is viewed from the side, projected onto a vertical plane, looking say north-east, or any other compass direction.  Again, user control of projected splay positions would not be appropriate.

Extended Elevation (projection)

Here I am including both the concepts that Evaristo introduced, of Extended Profile and Idealised Profile. To some degree they are the same thing, once artistic input has been applied.

1.	We could think of one survey leg that has been extended left or right as been projected in a vertical plane, that passes through the leg.  The stations at each end of the leg are of course in that plane as well.
2.	Now the splays at each station are projected onto that plane, in the same way as described for plan projections and elevation projections above.  This means that for this leg, and the splays attached to its stations, there is only one possible projected position and shape for its splays, in its vertical plane.  Splays that are not perfectly aligned with the plane will be foreshortened.
3.	Now the legs vertical plane is rotated or flattened out onto the ‘page’, left or right as we specify, and we have one leg for our extended elevation.  No further user control is desirable at this point.
4.	If there is another leg at the end of the leg we are considering, we have to think about which splays are associated with which of the legs.  
5.	If there are only two legs from each station, this is easy, if the splay projected position is not between the stations, then the splay is projected in the other legs vertical plane.  If there are multiple legs intersecting, say at a passage junction, then it is more complicated, but a rational computation could be devised.  There may also be some justification for a user control that allows the user to assign which survey leg each splay is associated with. 
ie user assigns extend left or right to legs, user optionally assigns legs to splays.  This is not the same as extending splays left or right, as if we did that, the splay lengths would be too long (not projected into the legs vertical plane).
6.	Therion now follows this process for all legs that make up the cave and we have an extended elevation.  And as described above, splays are not directly manipulated by extend left, right or ignore.
7.	But there are some difficulties.  Vertical legs that are extended left or right will be a problem.  And any leg that is ‘extend vertical’ will be a problem.  In both cases there is no unique vertical plane in which to project the splays.  In each case, if there is only one vertical leg at a time ie there is a non-vertical leg at each end, then the non-vertical legs planes can be used to project the splays at each end.  No problem.
8.	Where there are multiple (near) vertical legs, either because you are surveying down a shaft, or because the user has extended multiple legs vertically, there is still a problem.  I can think of two expedient solutions:
-Allow splays to be extended left and right – This will have the effect of making these passages seem larger, as the splay lengths will not be ‘projected’, or
-Have Therion refer back to a bearing associated with the vertical leg (entered in the centreline data) to decide the orientation of the vertical plane for splay projections, and then continue exactly as described above.  Although additional control would be required…
                extend vertical -splay-left
                extend vertical -spay-right
…in order for Therion to know which way around to project the splays.
I prefer the second option I think.

 

Does that make sense?

 

Aside from my explanation above, I have a selfish reason for preferring to minimise the ability to explicitly control splay extensions.  It is because I suspect that splay control can only be specified if the extend control statements are in-line with the survey centreline data.  It is hard to imagine how one could efficiently identify particular splays if the control is not in-line, as in my preference <https://therion.speleo.sk/wiki/extend#an_extended_elevation_paradigm>  to accommodate the option of multiple extended elevations for one set of data.

 

Bruce

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.speleo.sk/pipermail/therion/attachments/20180610/2dd32c14/attachment.htm>