Interactive extended elevations

[Gavin Lowe]

Phil Underwood wrote:

> I'm currently working on yet more bells and whistles for chasm.
> In this case, a slightly more intelligent extended elevation processor.
> 
> If I recall correctly, there was a discussion on this a while back; this
> would be my implementation 

I produced a partial implementation of extended elevations about a year
ago.  However, it used the pre-loop-closure data, rather than the closed
data, and I haven't had time since to sort it out.

My implementation used extra commands in the survex file; doing it
graphically could certainly be advantageous.

I can let you have the code if you think it would be useful.

> A selectable starting point.
> Process consequent legs from left to right.
> Individual legs can be *breakable* and/or *reversed*
> Specification of several continuous legs, which will be extended as a whole 

Discussion of projecting a series of legs (rather than straightening them
out) snipped.  This is a useful feature, but a bit of a side point.

> If a loop contains no breakable legs, then it will be broken at a
> junction, which is unspecified.

I think a warning should also be given in this case, so that the user knows
(s)he should choose a more intelligent break point.  BTW, IMO it normally
looks best if you don't break at a junction.

> If a loop contains one *breakable* leg, then it is broken at that leg. If it
> contains more than one, it will be broken at one of them (unspecified).

In the latter case, I think a warning should be given again.

> On encountering a *reversed* leg, the direction of unfolding is reversed.

Do you mean just that leg is reversed, or from then on?  I hope the latter:
you really want to be able to reverse a section of survey easily.

> The graphics interface will allow selection of the starting point,
> modification of the properties of legs(breakable/reversible), and display
> of legs according to their properties (breakable & broken, not breakable
> but broken anyway, reversed)

You also want to be able to specify that certain legs should be omitted
from the extended elevation, for clarity.

> The specification of which legs are to be breakable or reversed is done
> after processing (ie there would no added survex commands)
> 
> Having rambled on at length about this, does anyone have any better
> ideas/improvements? Can anyone see any potential problems with this?

One thing that wasn't clear from the description is how the EE information
is subsequently stored.  I hope you're planning to store it in a way that
it can be retrieved, and such that it can still be applied to surveys once
extra data has been added.

Gavin