You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Now in spec the first byte in twkb is used like this:
bit 1: Is ID included in twkb?
bit 2-4: serialization method
bit 5-8: precision, number of decimals in coordinates
I don't think the possibility to use different serialization methods will ever be used. I think the varInt method will be the one and only. So, for that reason I think those bits instead should be used for optional features, the same way as the ID? bit.
The other 2 optional features I can think of is:
If totally leaving the idea of different serialization methods we still have one bit left for future optional features.
Then comes the order of those 3 bits with optional features.
Now the ID? bit is the first bit.
I think it wold make sense to have the order
size?
bbox?
id?
unused bit
5-8) precision, number of decimals.
because that is the order they would apear later on in the geometry for practical resons.
But I don't know if there is any implementation out there that will break if moving the ID-bit.
I don't think any implementation of twkb is released yet so I guess this is still the right time to changing the design of the initial byte.
The text was updated successfully, but these errors were encountered:
Now in spec the first byte in twkb is used like this:
bit 1: Is ID included in twkb?
bit 2-4: serialization method
bit 5-8: precision, number of decimals in coordinates
I don't think the possibility to use different serialization methods will ever be used. I think the varInt method will be the one and only. So, for that reason I think those bits instead should be used for optional features, the same way as the ID? bit.
The other 2 optional features I can think of is:
If totally leaving the idea of different serialization methods we still have one bit left for future optional features.
Then comes the order of those 3 bits with optional features.
Now the ID? bit is the first bit.
I think it wold make sense to have the order
5-8) precision, number of decimals.
because that is the order they would apear later on in the geometry for practical resons.
But I don't know if there is any implementation out there that will break if moving the ID-bit.
I don't think any implementation of twkb is released yet so I guess this is still the right time to changing the design of the initial byte.
The text was updated successfully, but these errors were encountered: