#4430: Better support for resolving infix expressions in template haskell
---------------------------------+------------------------------------------
Reporter: reinerp | Owner: igloo
Type: feature request | Status: patch
Priority: normal | Milestone: 7.4.1
Component: Template Haskell | Version: 6.12.3
Keywords: | Testcase:
Blockedby: | Difficulty:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: None/Unknown
---------------------------------+------------------------------------------
Comment(by simonpj):
Re bullet 3, I see now. (Please put that in the big comment, or whatever
we agree.)
Perhaps we should just say that
{{{
InfixE Nothing plus (UInfixE 1 plus 2) -----> (+ (1 + 2))
}}}
There is only one (legal) meaning for what the data structre represents,
so let's just pick that. That's what I thought the idea was, and it's why
I was confused. Why generate a possibly bogus term if there's an obvious
legal meaning?
After all we have that
{{{
InfixE 0 * (UInfixE 1 + 2) -----> (0 * (1 + 2))
}}}
(that is, we parenthesise the arguments of `InfixE` regardless of their
form) so it seems entirely consistent for sections too.
Re last bullet:
* I think it would be worth documenting the left-biased form that GHC's
renamer expects. There is a bit in `RnTypes.lhs` line 280 or so, but it
lacks a `Note [blah]` heading -- you could add that and expand the note.
* I was thinking of a comment to say "this is why `cvtOpAppP` has to
reassociate the tree. Documenting the implementation (to convince oneself
that it does indeed do so) is a separate matter. Your comment on that
looks fine; except we need to know that the return value of `cvtl`
satisfies the invariant, since that supplies the initial arg to
`cvtOpAppP`.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4430#comment:16>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
