[Python-3000] AST access (WAS: Adaptation vs. Generic Functions)

[Phillip J. Eby]

At 07:39 PM 4/10/2006 -0400, Blake Winton wrote:
>Phillip J. Eby wrote:
>>I tried doing something like this when I was writing RuleDispatch, and 
>>gave up in disgust because there's no sane way to implement "and" and 
>>"or" operations with this approach.  The bitwise operators (&, |, and ~) 
>>bind too tightly to be used with comparison expressions, so doing 
>>something like "x>y & z" would mean "x>(y&z)" instead of "(x>y)&z".
>
>How would symbols, or ast access have helped you with that problem? Would 
>you really go around manually mangling the parse tree to move the "&" 
>higher in the branches somehow?

No - I'd have used 'and', 'or' and 'not': they're what I wanted to use in 
the first place.


>What do you think about the oft-maligned curly brace?

I think it should remain the syntax for dictionary literals, and not get 
involved here.


>And it should be possible to tell the difference between it and a dict 
>with a bit of a lookahead...  (Really it's the conflict with dict 
>initialization that I expect to kill this proposal.)

For both implementation and readability reasons, Python syntax should 
require no more than 1 token of lookahead - by which I mean you should be 
able to parse it in purely token-by-token fashion without ever peeking ahead.