search

iho-ohi / S-164-Sub-Group

Development and tracking of IHO S-164 Test datasets and user manual documentation
15 stars 2 forks source link

Encoding of restricted areas #80

Open DavidGrant-NIWC opened 8 months ago

DavidGrant-NIWC commented 8 months ago

When two restricted areas share a geometry and the only attribute difference is the restriction then they should be encoded as a single area with multiple values of restriction. For instance, the two restricted areas below should be a single restricted area with two restrictions: Anchoring Prohibited, and Fishing Prohibited. Encoding the areas individually results in multiple outline patterns since the drawing priorities are equal:

Fishing Prohibited Anchoring Prohibited
image image
kusala9 commented 8 months ago

Now that restricted area is a single feature there shouldn't be a need to encode them with shared geometry, they should be a single REstrictedArea with multiple restriction attributes. This is probably a hangover from the separated RestrictedAreas so needs re-encoding. This is in the SW cell, I will see if we can find others and re-encode this one.

MikusRL commented 8 months ago

There were a case with S-57 ECDIS that no more than 7 restrictions could be encoded on one restriction feature, and we had some cases in PRIMAR where the restrictions were more and it crashed an ECDIS. So the solution was to "duplicate" the restriction object and split the restrictions between them.

I can see also usecases where the restrictions could use the same geometry, but need different restrictions and different according to the different restrictions also information either in "INFORM" or as attached TXT.

So I think if this is already considered, and almost developed, then it should be developed til the end that it works.

Also in simplified view the dashes for different feature coherent areas could come together using the same dash pattern. Then again the alignment of the dashes would be a very good for the user solution. Else the dashes are seen as one continuous line at some points and can be ambiguous to interpret at first look on ECDIS screen.

Or is this only about the split objects from one initial resare object in S-57? Then please ignore my comment :)

DavidGrant-NIWC commented 8 months ago

@MikusRL I originally noticed this because it looks like a portrayal error. One would expect line suppression to prevent the conflict between the two symbolized lines, but upon closer inspection it is apparent that this is an encoding issue.

I would expect the mariner to prefer that if there is different information to be provided it is all included on the same restricted area feature. This seems to match the current modeling, where information is bound to RestrictedArea with a 0:∞ multiplicity.