RE: runghc, ".hs" suffix

I'm working on this.  It's not trivial because GHC assumes all over the
place that the suffix on a file determines how it should be compiled,
but I've now implemented something similar to gcc's -x flag.

Cheers,
        Simon

On 14 May 2005 20:04, Frederik Eaton wrote:

> (moving to a separate thread)
> 
> Also, how hard would it be to make it so that runghc doesn't require
> script file names to end with ".hs"? Some people like to keep their
> executable names language-neutral.
> 
> Frederik
> 
> On Wed, May 11, 2005 at 10:55:03AM +0100, Simon Marlow wrote:
>> On 11 May 2005 07:37, Frederik Eaton wrote:
>> 
>>> It looks like runghc is exiting with status 0 if there is a
>>> problem. Shouldn't this be non-zero? 
>>> 
>>> $ runghc -V
>>> The Glorious Glasgow Haskell Compilation System, version 6.4 $
>>> runghc /dev/null; echo $? Could not find module `/dev/null':
>>>   use -v to see a list of the files searched for
>>>   (one of the roots of the dependency analysis)
>>> 
>>> <interactive>:1:76:
>>>     Failed to load interface for `Main':
>>>         Could not find module `Main':
>>>           it is not a module in the current program, or in any known
>>> package. 0
>> 
>> It turns out I fixed this bug after 6.4, but later the fix was
>> accidentally reverted.  So thanks for reporting this again, it'll
>> work properly in 6.4.1. 
>> 
>> Cheers,
>>      Simon
>> 
> 
> _______________________________________________
> Glasgow-haskell-bugs mailing list
> Glasgow-haskell-bugs@xxxxxxxxxxx
> http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs