Implement the current set of rules for precedence partial ordering #1096
Conversation
In the process, switch to an unambiguous grammar, using the precedence-climbing method for operator precedence (suitably modified to handle a partial precedence order) rather than Bison %precedence / %prec.
used in the real grammar.
|
@geoffromer I've reworked the presentation here and moved most of the description to README.md in an attempt to address your comments. What do you think? |
| @@ -15,3 +15,134 @@ techniques can be applied to other kinds of AST nodes as needed. See the | |||
| handling of the `UNIMPL_EXAMPLE` token for an example of how this is done, and | |||
| see [`unimplemented_example_test.cpp`](unimplemented_example_test.cpp) for an | |||
| example of how to test it. | |||
|
|
|||
| ## Precedence and associativity | |||
There was a problem hiding this comment.
Non-blocking comment: reading this section, I still have a hard time working out what rules I should follow when adding/modifying grammar rules, or conversely how to read grammar rules to understand the structure they describe. I think part of the issue is that the "what" (as opposed to "why") part is split between the idealised rules on lines 45-63 and the modifications to that ideal on lines 96-102.
However, I'm not sure that trying to wordsmith this is a good use of time right now; we'll probably have a better idea of how to clarify this once we've worked with it for a while. I think all the necessary information is here (modulo my question below), and that's probably enough for now.
…1096) In the process, switch to an unambiguous grammar, using the precedence-climbing method for operator precedence (suitably modified to handle a partial precedence order) rather than Bison %precedence / %prec.
In the process, switch to an unambiguous grammar, using the
precedence-climbing method for operator precedence (suitably modified to
handle a partial precedence order) rather than Bison
%precedence/%prec.