[GHC] #5592: Incorrect "redundant import" warning
GHC
cvs-ghc at haskell.org
Sat Oct 29 02:26:37 CEST 2011
#5592: Incorrect "redundant import" warning
---------------------------------+------------------------------------------
Reporter: ryantrinkle | Owner:
Type: bug | Status: new
Priority: normal | Component: Compiler
Version: 7.2.1 | Keywords:
Testcase: | Blockedby:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: Incorrect warning at compile-time
---------------------------------+------------------------------------------
Record fields are incorrectly deemed "redundant" by GHC in at least some
circumstances. This causes a confusing warning when -Wall is on.
Given the following module,
{{{
module MyRecord where
data MyRecord = MyRecord { field :: String } deriving (Show)
}}}
the following program's meaning will change depending on whether or not
"field" is imported:
{{{
{-# LANGUAGE RecordWildCards #-}
import qualified MyRecord as MR (MyRecord(MyRecord))
main = do
let field = "Hello, world!"
rec = MR.MyRecord {..}
print rec
}}}
The program as written will crash, because the record wildcard will not
use the 'field' variable, since the 'field' record label is not imported.
A warning is correctly issued regarding the missing field.
However, if the user adds an import for 'field',
{{{
import qualified MyRecord as MR (MyRecord(MyRecord, field))
}}}
GHC will issue a warning:
{{{
Main.hs:3:1:
Warning: The import of `MR.field'
from module `MyRecord' is redundant
}}}
However, adding the import of 'field' causes the record wildcard to have
access to the 'field' label, so the program does not crash.
Since the import changes the meaning of the program, it shouldn't be
considered redundant.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5592>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
More information about the Glasgow-haskell-bugs
mailing list