Typed splices and type checking
Hello, I've run into another misunderstanding with Template Haskell and typed splices. For example, I was hoping to use typed splices to generate earlier errors from Printf. Here's the world's least handy printf:
class Printf t where printf :: String -> Q (TExp String) -> Q (TExp t)
instance Printf String where printf s t | '%' `notElem` s = [|| $$t ++ s ||] | otherwise = fail ("Unexpected format %" ++ [c]) where (_, _:c:_) = break ('%' ==) s
instance Printf t => Printf (Char -> t) where printf s t | c /= 'c' = fail ("Unexpected format %" ++ [c] ++ " for character") | otherwise = [|| \c -> $$(printf s'' [|| $$t ++ s' ++ [c] ||]) ||] where (s', '%':c:s'') = break ('%' ==) s
Now, consider these two definitions:
f :: Char -> String f = $$(printf "foo %c" [||""||])
h :: Char -> String h y = $$(printf "foo %c" [||""||]) y
I would have expected these to be equivalent, from a type checking perspective. However, while the first types fine, the second generates the error message:
Printing.hs:14:12: No instance for (Printf t0) arising from a use of `printf' The type variable `t0' is ambiguous Note: there are several potential instances: instance Printf t => Printf (Char -> t) -- Defined at Printf.hs:20:10 instance Printf String -- Defined at Printf.hs:9:10 In the expression: printf "foo %c" [|| "" ||] In the Template Haskell splice $$(printf "foo %c" [|| "" ||]) In the expression: $$(printf "foo %c" [|| "" ||])
Should I have anticipated this? Ought the interaction of typed splices and eta-expansion be problematic? /g -- The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336.
On Fri, Mar 27, 2015 at 10:30 AM, J. Garrett Morris wrote: Should I have anticipated this? Ought the interaction of typed splices
and eta-expansion be problematic? In the specific case of printf, which uses typeclasses to play dirty tricks
that interact "interestingly" with eta expansion, it's not particularly
surprising. I would suggest using wrappers if possible.
--
brandon s allbery kf8nh sine nomine associates
allbery.b@gmail.com ballbery@sinenomine.net
unix, openafs, kerberos, infrastructure, xmonad http://sinenomine.net
Good question! See https://ghc.haskell.org/trac/ghc/ticket/10271. Simon | -----Original Message----- | From: Glasgow-haskell-users [mailto:glasgow-haskell-users- | bounces@haskell.org] On Behalf Of J. Garrett Morris | Sent: 27 March 2015 14:30 | To: GHC users | Subject: Typed splices and type checking | | Hello, | | I've run into another misunderstanding with Template Haskell and typed | splices. For example, I was hoping to use typed splices to generate | earlier errors from Printf. Here's the world's least handy printf: | | > class Printf t | > where printf :: String -> Q (TExp String) -> Q (TExp t) | > | > instance Printf String | > where printf s t | '%' `notElem` s = [|| $$t ++ s ||] | > | otherwise = fail ("Unexpected format %" | > ++ [c]) | > where (_, _:c:_) = break ('%' ==) s | > | > instance Printf t => Printf (Char -> t) | > where printf s t | > | c /= 'c' = fail ("Unexpected format %" ++ [c] ++ | > " for character") | > | otherwise = [|| \c -> $$(printf s'' | > [|| $$t ++ s' ++ [c] ||]) | > ||] | > where (s', '%':c:s'') = break ('%' ==) s | | Now, consider these two definitions: | | > f :: Char -> String | > f = $$(printf "foo %c" [||""||]) | | > h :: Char -> String | > h y = $$(printf "foo %c" [||""||]) y | | I would have expected these to be equivalent, from a type checking | perspective. However, while the first types fine, the second generates | the error message: | | > Printing.hs:14:12: | > No instance for (Printf t0) arising from a use of `printf' | > The type variable `t0' is ambiguous | > Note: there are several potential instances: | > instance Printf t => Printf (Char -> t) | > -- Defined at Printf.hs:20:10 | > instance Printf String -- Defined at Printf.hs:9:10 | > In the expression: printf "foo %c" [|| "" ||] | > In the Template Haskell splice | > $$(printf "foo %c" [|| "" ||]) | > In the expression: $$(printf "foo %c" [|| "" ||]) | | Should I have anticipated this? Ought the interaction of typed splices | and eta-expansion be problematic? | | /g | | -- | The University of Edinburgh is a charitable body, registered in | Scotland, with registration number SC005336. | | _______________________________________________ | Glasgow-haskell-users mailing list | Glasgow-haskell-users@haskell.org | http://mail.haskell.org/cgi-bin/mailman/listinfo/glasgow-haskell-users
participants (3)
-
Brandon Allbery -
J. Garrett Morris -
Simon Peyton Jones