#4429: Ability to specify the namespace in mkName
---------------------------------+------------------------------------------
Reporter: reinerp | Owner:
Type: feature request | Status: new
Priority: normal | Milestone:
Component: Template Haskell | Version: 6.12.3
Keywords: | Testcase:
Blockedby: | Difficulty:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: None/Unknown
---------------------------------+------------------------------------------
Old description:
> Given
>
> {{{
> data Foo
> data Bar = Foo
> }}}
>
> If we do {{{reify (mkName "Foo")}}} then we get the information about
> "{{{Foo}}} the type", and not about "{{{Foo}}} the constructor".
>
> (This is problematic, say, for a quasiquoter
>
> {{{ [qq| ... Foo ... |] }}}
>
> because the quasiquoter is forced to use {{{mkName "Foo"}}} as the
> {{{Name}}} for reify -- the forms {{{'Foo}}} and {{{''Foo}}} are
> unavailable to it.)
>
> I would like a way around this problem. It seems like it would be enough
> to communicate the namespace to {{{mkName}}}, so that the ambiguity no
> longer exists.
New description:
Given
{{{
data Foo
data Bar = Foo
}}}
If we do {{{reify (mkName "Foo")}}} then we get the information about
"{{{Foo}}} the type", and not about "{{{Foo}}} the constructor".
(This is problematic, say, for a quasiquoter
{{{
[qq| ... Foo ... |]
}}}
because the quasiquoter is forced to use {{{mkName "Foo"}}} as the
{{{Name}}} for reify -- the forms {{{'Foo}}} and {{{''Foo}}} are
unavailable to it.)
I would like a way around this problem. It seems like it would be enough
to communicate the namespace to {{{mkName}}}, so that the ambiguity no
longer exists.
--
Comment(by simonpj):
I think `mkName` is the wrong thing for you here. Fundamentally, you
want to get the `TH.Name` of the data type called "Foo" that is currently
in scope, yes? You could give that `Name` to `reify`, or you could use it
in a type. Suppose we had
{{{
lookupType :: String -> Q Name
lookupValue :: String -> Q Name
}}}
that were like `mkName` except that (a) they are monadic, and (a) they
expect the string to be in scope. The would be the precise monadic
equivalents of `'Foo` and `''Foo`. Would that do the job?
Anyone else have comments?
Simon
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4429#comment:1>
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
