highlighting selected stations Re: Color by...

[Warren Family]

 > Knowing what sorts of things people want to achieve, for what purposes
 > would help design something that might stand the test of time.

Taking the opportunity to further respond here ... :-)

In earlier versions of aven, hovering over a name in the survey 
structure on the RHS of the GUI would cause all the corresponding 
stations to light up, as it were.  This got changed at some point so now 
instead a loop is thrown around the selected stations.  I can see why 
this change is useful in many circumstances, but it actually broke some 
important use-cases for me.  The basic problem with using a loop is that 
if you zoom in, you completely lose the critical information on which 
survey stations have been highlighted.  This is a critical issue if you 
want to see in detail where a particular part of a cave system lies in 
relation to another part, or (in my case, in the Easegill data set) 
figuring out exactly where the different parts join up.

Another use-case is that it is sometimes very illuminating to easily see 
the stations in a given cave passage.  For example in the Easegill data 
set, Wretched Rabbit very clearly lights up as a single inlet series, 
into the main drain, in the middle of the complex network of passages at 
that end of the system.  It's virtually impossible to see this any other 
way.

The current option of hiding / showing cave levels doesn't address this 
need (or, I didn't figure out how to make it work), since you still want 
to see the rest of the cave system in relation to the selected part. 
Being able to color by level as in the last email would help, but as a 
solution it seems a bit over-complicated.

What would be great would be to have both options available: 
highlighting the selected stations, or throwing a loop around the 
selected stations.  Presumably this wouldn't take much to implement 
since it must just amount to restoring the original highlighting code, 
and adding a suitable pair menu options.  At one point I did try to do 
this myself but I lacked sufficient familiarity with the aven code to be 
able to easily figure it out.

Patrick


On 15/02/2025 00:12, Wookey wrote:
> On 2025-02-14 18:27 -0500, Richard Knapp wrote:
>> Is there a way to group section of a large project so the colors be consistent: Main cave as one color, first branch another, etc….?
> 
> Not a good way no.
> 
> The only way I can think of to do it currently is to put fake dates on
> surveys (matching dates for areas you want coloured the same) and
> 'colour by date'. Which isn't really something one wants to do, as
> real dates matter.
> 
> Or if you re-arranged the survey structure so that things you wanted
> the same colour were the same survey. and then 'colour by
> srurvey'. But that doesn't sound at all satisfactory either.
> 
> If the 'colour by survey' mode could perhaps be adjusted to show
> higher level survey groupings then the current 'every survey is a
> different colour no matter how small', that might be a sufficieint
> mechanism for many cases, but possibly what's wanted is an entirely
> independent grouping, which is just used for display purposes.
> 
> There are plenty of occasions where this sort of thing would be
> useful, but it's not obvious what the best way to do it would be.
> 
> One could have *group (or whatever name) to define a display style (we
> already used 'style' for the survey method). So that would go in the
> data, and if it propogated withthe survey structure by default, and
> begin/ends, you might not have to change too many places.
> 
> Or one could write a separate 'mapping' a bit like the espec files
> that say what surveys to put in what styles, and also what those
> styles look like. The you load a survey and a style to get that
> 'view'. This allows more than one view for a given dataset.
> 
> Or it could be a point-and-click sort of thing in aven (but then you
> need a way to record it or you'll be doing it over and over again,
> which would get boring fast).
> 
> Would any of these work for people?
> 
> Knowing what sorts of things people want to achieve, for what purposes
> would help design something that might stand the test of time.
> 
> For our group being able to 'colour by cave' would be very
> helpful/useful. And that fairly directly matches the survey structure,
> but not at the top level (kataster area), not the bottom level
> (individual surveys). So some mechanism to say what 'level' of the
> survey grouping the colouring should traverse might be sufficient. The
> issue is that people have all sorts of different structures, and
> 'cave' might be one level down in one part of the dataset, but two
> levels doewn in another. It's not an easy thing to generalise, but
> some experiments at least might not need much code.
> 
> Wookey
> 
>