Hi Ian, It almost certainly was caused by that patch, but the test works OK for me!
I think I can see what causes this: the output of this new test depends on the locale of the machine. My locale is en_GB.UTF-8 and and trying to write a surrogate byte to stdout in this encoding fails with an exception. Perhaps on your machine you have a Latin 1 locale which succeeds in writing the surrogate byte and hence gets different output. I'll have the test explicitly set the locale of stdout to latin1 to fix this. Max On 24 September 2011 21:46, Ian Lynagh <[email protected]> wrote: > > Hi Max, > > I'm seeing a test failure when validating: > > Actual stdout output differs from expected: > --- ./lib/IO/decodingerror002.stdout 2011-09-24 12:36:55.000000000 +0100 > +++ ./lib/IO/decodingerror002.run.stdout 2011-09-24 13:43:09.000000000 > +0100 > @@ -1,4 +1,4 @@ > Left decodingerror002.in: hGetChar: invalid argument (invalid byte sequence) > Left decodingerror002.in: hGetChar: end of file > �Left decodingerror002.in: hGetChar: end of file > -Left <stdout>: hPutChar: invalid argument (invalid character) > +�Left decodingerror002.in: hGetChar: end of file > *** unexpected failure for decodingerror002(normal) > > I wonder if this was caused by this patch?: > > commit 901edcb2bb342e7943400afe2ea6772998ecbf95 > Author: Max Bolingbroke <[email protected]> > Date: Fri Sep 23 23:24:36 2011 +0900 > > Fix #5436 by using 'recover' on handle EOF > > > Thanks > Ian > > > _______________________________________________ > Cvs-ghc mailing list > [email protected] > http://www.haskell.org/mailman/listinfo/cvs-ghc > _______________________________________________ Cvs-ghc mailing list [email protected] http://www.haskell.org/mailman/listinfo/cvs-ghc
