On behalf of the maintainers team I'm happy to announce that bytestring-0.11.4.0 (http://hackage.haskell.org/package/bytestring-0.11.4.0) is finally released. Highlights from the changelog ( https://github.com/haskell/bytestring/blob/0.11.4.0/Changelog.md): * A serious bug in `instance Lift ShortByteString` has been fixed. * `Builder` literals no longer break `warp` versions older than 3.3.22 by demanding too large of a buffer. * `unsafeIndex` is once again exported from Data.ShortByteString.Internal. * Many operations on `ShortByteString` have been made much more efficient. Many people contributed their time and effort to make this release happen. Thanks to all of them! Just to name a few, in no particular order: * Alexander Biehl * Xia Li-yao * Sergey Vinokurov * Viktor Dukhovni * Julian Ospald * Andreas Klebinger * Simon Jakobi * Andrew Lelechenko * every contributor to the older bytestring releases Happy Haskelling! ~-Matthew Craven (Hopefully the send is successful this time!)
Hi, I tried to get a reference to the Lift instance bug of ShortByteString, but I couldn't. All I see is the switch to ByteArray: https://github.com/haskell/bytestring/commit/91a727c2e52b13e4ad259293ed00245... What was the bug? Cheers On 1/13/23 05:49, Matthew Craven wrote:
On behalf of the maintainers team I'm happy to announce that bytestring-0.11.4.0 (http://hackage.haskell.org/package/bytestring-0.11.4.0 http://hackage.haskell.org/package/bytestring-0.11.4.0) is finally released. Highlights from the changelog (https://github.com/haskell/bytestring/blob/0.11.4.0/Changelog.md https://github.com/haskell/bytestring/blob/0.11.4.0/Changelog.md):
* A serious bug in `instance Lift ShortByteString` has been fixed. * `Builder` literals no longer break `warp` versions older than 3.3.22 by demanding too large of a buffer. * `unsafeIndex` is once again exported from Data.ShortByteString.Internal. * Many operations on `ShortByteString` have been made much more efficient.
Many people contributed their time and effort to make this release happen. Thanks to all of them! Just to name a few, in no particular order:
* Alexander Biehl * Xia Li-yao * Sergey Vinokurov * Viktor Dukhovni * Julian Ospald * Andreas Klebinger * Simon Jakobi * Andrew Lelechenko * every contributor to the older bytestring releases
Happy Haskelling! ~-Matthew Craven
(Hopefully the send is successful this time!)
_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries
This was the first item in the main changelog, which provided a link. See:
* https://github.com/haskell/bytestring/pull/542
* https://github.com/haskell/aeson/issues/967
* https://gitlab.haskell.org/ghc/ghc/-/issues/22204
In short, if Lift was used to produce two ShortByteStrings that have
the same length, then GHC would sometimes use the same buffer for both
when the code actually gets run. It's pretty bad.
Thanks for asking.
On Thu, Jan 12, 2023 at 11:29 PM Julian Ospald
Hi,
I tried to get a reference to the Lift instance bug of ShortByteString, but I couldn't.
All I see is the switch to ByteArray: https://github.com/haskell/bytestring/commit/91a727c2e52b13e4ad259293ed00245...
What was the bug?
Cheers
On 1/13/23 05:49, Matthew Craven wrote:
On behalf of the maintainers team I'm happy to announce that bytestring-0.11.4.0 (http://hackage.haskell.org/package/bytestring-0.11.4.0 http://hackage.haskell.org/package/bytestring-0.11.4.0) is finally released. Highlights from the changelog (https://github.com/haskell/bytestring/blob/0.11.4.0/Changelog.md https://github.com/haskell/bytestring/blob/0.11.4.0/Changelog.md):
* A serious bug in `instance Lift ShortByteString` has been fixed. * `Builder` literals no longer break `warp` versions older than 3.3.22 by demanding too large of a buffer. * `unsafeIndex` is once again exported from Data.ShortByteString.Internal. * Many operations on `ShortByteString` have been made much more efficient.
Many people contributed their time and effort to make this release happen. Thanks to all of them! Just to name a few, in no particular order:
* Alexander Biehl * Xia Li-yao * Sergey Vinokurov * Viktor Dukhovni * Julian Ospald * Andreas Klebinger * Simon Jakobi * Andrew Lelechenko * every contributor to the older bytestring releases
Happy Haskelling! ~-Matthew Craven
(Hopefully the send is successful this time!)
_______________________________________________ Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries
Libraries mailing list Libraries@haskell.org http://mail.haskell.org/cgi-bin/mailman/listinfo/libraries
participants (2)
-
Julian Ospald -
Matthew Craven