#5375: Regression in newName
---------------------------------+------------------------------------------
Reporter: reinerp | Owner:
Type: bug | Status: new
Priority: highest | Milestone: 7.2.1
Component: Template Haskell | Version: 7.3
Keywords: | Testcase:
Blockedby: | Difficulty:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: GHC rejects valid program
---------------------------------+------------------------------------------
Comment(by simonpj):
I think that's the wrong approach. It places a global burden on every
binding enclosing the occurrence of "x".
But the problem is not with those bindings; rather, it's with the
''occurrence'' of x. The quasiquoter should not use `mkName` to generate
a string-name for `x` -- as you yourself pointed out in #4429. Rather it
should use the proper `Name` of `x`! Just like `'x` in Haskell code
So how can a quasiquoter get hold of the `Name` for x? That's a whole
separate question, which we have open in #4429. If we had the
`lookupValue` described there, the quasiquoter could use it to look x up.
Now, I grant that #4429 is languishing unloved, but I think that's the
right fix because it adds a simple-to-understand facility that may be
independently useful, rather than making a regrettably-complex situation
more complex still.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5375#comment:9>
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
