On Sat, Sep 25, 2010 at 2:54 PM, Ian Lynagh
Thanks for your work on this. Please see http://www.haskell.org/haskellwiki/Library_submissions for the process to propose changes to the core libraries.
Yes, I know the process. However, consider my mail more as a question if the current use of throw is always equivalent to using throwIO. Take the current onException as an example: onException io what = io `catch` \e -> do _ <- what throw (e :: SomeException) is this always equivalent to: onException io what = io `catch` \e -> do _ <- what throwIO (e :: SomeException) ? Thinking about this some more I can't seem to find a situation where the two are different. So my patch can safely be ignored and I apologize for the noise. However I do wonder why onException is using throw instead of the recommended throwIO. Has this to do with performance since they're both primitive operations: throw :: Exception e => e -> a throw e = raise# (toException e) throwIO :: Exception e => e -> IO a throwIO e = IO (raiseIO# (toException e)) ? Thanks, Bas