[css-backgrounds] Order of values for border shorthands is unclear

[vwkd]

The definitions of the order of values for the shorthands border-color, border-style and border-width seem a bit unclear. For all three the spec states

If it has four values, they set top, right, bottom and left in that order. If left is missing, it is the same as right; if bottom is missing, it is the same as top; if right is missing, it is the same as top.

So what happens with three or two values, e.g. border-width: a b c or border-width: a b? How does it know which side I left out?

By trial and error I know the order is the same as for the margin and padding shorthands. The definitions of margin and padding are much clearer though. The spec states

If there is only one component value, it applies to all sides. If there are two values, the top and bottom [..] are set to the first value and the right and left [..] are set to the second. If there are three values, the top is set to the first value, the left and right are set to the second, and the bottom is set to the third.

I would suggest taking over this part of the definition from margin / padding for the border properties as well.

[Loirooriol]

The full syntax is like border-width: top right bottom left.
If you only provide 3 values (border-width: top right bottom), then left is missing.
If you only provide 2 values (border-width: top right), then bottom and left are missing.
If you only provide 1 value (border-width: top), then right, bottom and left are missing.

So you know which sides are missing because you match greedily, sides can only be omitted at the end.

[vwkd]

Ok thanks, I understand now. The sentence assumed that for each left out value the ones following it were also left out. So a somewhat clearer formulation would be

If left is missing, it is the same as right; if in addition to that bottom is missing, it is the same as top; if in addition to that right is missing, it is the same as top.

Still the margin / padding definitions seem the clearest.

[fantasai]

Merged your pull request, thanks! Looks very clean. :) Closing out the issue, but filed #5687 as a follow-up.