Code review request: has_key('dotted.key') about coil HOT 4 OPEN

Why don't we fix the __contains__ operation instead? has_key() is more of a 
legacy API that has been replaced by the 'in' operator.

I didn't want to change the default behavior of iterating over a struct.Struct; 
i.e., I did not want to force an iteration over all keys in a struct 
(recursively), just the immediate keys of that struct.

>>> x = coil.text.fromString('a: { b: c d: { e: True } }')
>>> for key in x:
...     print key
a
>>> x.has_key('a.b')
True

I suppose __contains__ works, I hadn't looked into that, but I wanted to keep 
iteration distinct from whether the (fully-dotted) key exists in the struct.

Making iteration "for key in x" slightly different from comtains "if key in x" 
slightly different is slightly odd, but I'd prefer that than making the legacy 
contains api "x.has_key" different from the new api. Or perhaps an entirely new 
method would be better, "x.has_child" or some such where the recursiveness of 
the lookup is explicit. I think I'm leaning to the entirely new method at the 
moment.

how about adding a recursive=False kwarg to has_key (or write a new 
coil-specific fn)?

It seems to me that the correct way to iterate and test keys in coil is the 
same way as in dicts.  If you want a coil-specific way to test for existence, 
make a coil-specific function?