igloo:
> Sun Oct 5 05:11:50 PDT 2008 Ian Lynagh <[EMAIL PROTECTED]>
> * Make sure that we hClose file handles when we are done with them
>
> M ./Data/ByteString.hs -1 +2
> M ./Data/ByteString/Lazy.hs -1 +1
This stuff worries me, ad hoc fixing of bugs already closed upstream.
And without tests, (when there are already tests in the upstream repo for this
very behaviour).
Wed Dec 19 22:06:13 PST 2007 Don Stewart <[EMAIL PROTECTED]>
* For lazy IO operations, be sure to hClose the resource on EOF
Wed Dec 19 22:07:36 PST 2007 Don Stewart <[EMAIL PROTECTED]>
* add test for lazy hclose working
Meanwhile, there's twelve months of other patches you're not
merging into the GHC fork.
I worry that the concerns of the GHC release process don't align with the
concerns of library writers. We want to ship the best Haskell code, versus
shipping whatever builds in the GHC tree. There seems to be unclear direction
about what we're trying to do here.
Duncan's already sent the win32 dep removal patch already. Now we just need
GHCi to not use bytestring to do its little bit of IO.
Then you can update to bytestring 0.9.1.3 released yesterday, specifically for
the release,
http://hackage.haskell.org/cgi-bin/hackage-scripts/package/bytestring-0.9.1.3
-- Don
_______________________________________________
Cvs-libraries mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-libraries
