- Haskell-Cafe - Haskell.org

[ANNOUNCE] GHC 9.6.2 is now available
by Ben Gamari 23 May '23

23 May '23

The GHC developers are happy to announce the availability of GHC 9.6.2. Binary distributions, source distributions, and documentation are available at https://downloads.haskell.org/ghc/9.6.2 This release is primarily a bug-fix release addressing a few issues found in 9.6.2. These include: * a number of simplifier and specialisation issues (#22761, #22549) * A bug resulting in crashes of programs using the new ``listThreads#`` primop (#23071). * A compiler crash triggered by certain uses of quantified constraints (#23171) * Various bugs in the Javascript backend have been fixed (#23399, #23360, #23346) * A missing write barrier in the non-moving collector's handling of selector thunks, resulting in undefined behavior (#22930). * The non-moving garbage collector's treatment of weak pointers has been revamped which should allow more reliable finalization of ``Weak#`` closures (#22327) * The non-moving garbage collector now bounds the amount of marking it will do during the post-marking stop-the-world phase, greatly reducing tail latencies in some programs (#22929) A full accounting of changes can be found in the [release notes]. As some of the fixed issues do affect correctness users are encouraged to upgrade promptly. We would like to thank Microsoft Azure, GitHub, IOG, the Zw3rk stake pool, Well-Typed, Tweag I/O, Serokell, Equinix, SimSpace, Haskell Foundation, and other anonymous contributors whose on-going financial and in-kind support has facilitated GHC maintenance and release management over the years. Finally, this release would not have been possible without the hundreds of open-source contributors whose work comprise this release. As always, do give this release a try and open a [ticket] if you see anything amiss. Happy compiling, ~ Ben [ticket]: https://gitlab.haskell.org/ghc/ghc/-/issues/new [release notes]: https://downloads.haskell.org/~ghc/9.6.2/docs/users_guide/9.6.2-notes.html

1 0

[ANN] NASA's Ogma 1.0.9
by Ivan Perez 22 May '23

22 May '23

Hi everyone! I'm thrilled to announce the release of Ogma 1.0.9! Ogma is a NASA tool that facilitates integrating runtime monitors or runtime verification applications into other systems. Use cases supported by Ogma include producing Robot Operating System (ROS 2) packages [3], NASA Core Flight System (cFS) applications [4], and components for FPrime [1] (the software framework used for the Mars Helicopter), as well as generating monitors from requirements specified in natural language [5]. Ogma is also one of the solutions recommended for monitoring in Space ROS applications [2]. Ogma leverages existing Haskell work, like the Copilot language [6] (also funded by NASA) and BNFC [7]. For more details, including videos of monitors being generated and flown in simulators, see: https://github.com/nasa/ogma * What's changed This release improves the user experience and fixes a number of bugs in the code and documentation. A new flag `--target-file-name` allows users to specify the root for C monitoring files generated by some of the backends, which makes for a simpler process. For details about the release, see: https://github.com/nasa/ogma/releases/tag/v1.0.9 * Releases Ogma is released as a collection of packages in Hackage. The entry point is https://hackage.haskell.org/package/ogma-cli. * Code The github repo is located at: https://github.com/nasa/ogma. * What's coming The next release is planned for Jul 21st, 2023. Ogma is currently undergoing the qualification process necessary for NASA Class D Software. Apart from the changes required by that process, we also have the following in our roadmap: - Extend ROS 2 monitors with further information about sources of violations. - Add tests to generated code. - Simplify NASA cFS monitor generation process. - Simplify the architecture. More specifically, we now have improved the compilation process by using a higher-kinded data structure, allowing us to incorporate different kinds of information to syntax trees used during both parsing and code generation. This feature has been prototyped, and is currently undergoing testing. We hope that you are as excited as we are and that our work demonstrates that, with the right support, Haskell can reach farther than we ever thought possible. Happy Haskelling! Ivan [1] https://github.com/nasa/fprime [2] https://space.ros.org/ [3] https://www.ros.org/ [4] https://github.com/nasa/cFS [5] https://github.com/NASA-SW-VnV/fret [6] https://github.com/Copilot-Language/copilot [7] https://github.com/BNFC/bnfc

1 0

Summer of Haskell 2023 Project Selections
by Aaron Allen 21 May '23

21 May '23

This year, Google did not select Haskell as an open-source project for the Google Summer of Code program. In line with tradition, the Haskell community stepped up, with multiple individuals, companies, and non-profit organizations funding nine significant open-source contributions to the Haskell ecosystem: - Cabal file support for HLS - Implement Resolution Methods in HLS - Goto Definition for Third-Party Libraries in HLS - Teaching Weeder About Type Class Instances - Standardize GHC’s Error Dump in JSON Format - Maximally Decoupling Haddock and GHC - Representing Pattern - Improving Calligraphy - Structured Errors and Error Codes for cabal-install For more details about the accepted projects, the participants, and the mentors, see this announcement: https://summer.haskell.org/news/2023-05-14-summer-of-haskell-2023-project-s… We would like to thank the sponsors who made this possible: - The Haskell Foundation - Kadena - Jane Street - Holmusk - Gershom Bazerman - Edward Kmett - MLabs - Flipstone

1 0

Re: [Haskell-cafe] Stackage question
by Henning Thielemann 19 May '23

19 May '23

On Fri, 19 May 2023, Henning Thielemann wrote: > On Fri, 19 May 2023, Dominick Samperi wrote: > >> The page recommends adding the package to Stackage Nightly, and I guess I >> could try the third option, resolver: nightly-2023-05-19, but this seems >> odd. Why would I release a package with a particular hard-coded date for >> nightly? Shouldn't it just be "nightly" or something like this? > > You must add your package to the build-constraints.yaml file in the > repository at: > https://github.com/fpco/stackage/ I see, you have done this. Then I think you must wait a little bit until your package is built.

1 0

Stackage question
by Dominick Samperi 19 May '23

19 May '23

Hello, Can somebody explain how to resolve a problem with the resolver, specifically, my package (mathlist) appears on the Stackage page, but with the message "This package is not in any snapshots." I tried resolver: lts-20.21 (in stack.yaml), then changed this to resolver: ghc-9.4.5, but this doesn't make any difference. The page recommends adding the package to Stackage Nightly, and I guess I could try the third option, resolver: nightly-2023-05-19, but this seems odd. Why would I release a package with a particular hard-coded date for nightly? Shouldn't it just be "nightly" or something like this? Thanks, Dominick

1 0

[Haskell] [ANNOUNCE] Haskell Language Server 2.0.0.0 released
by Zubin Duggal 19 May '23

19 May '23

Binaries for this release are available at https://downloads.haskell.org/~hls/haskell-language-server-2.0.0.0/. These binaries can be installed using [GHCup](https://www.haskell.org/ghcup/) or the Haskell VSCode extension. This is a fairly small release, marking the start of a new versioning where HLS and all the included packages in the distribution (plugins, ghcide etc.) will be versioned in lockstep. # Changelog - New versioning scheme for all packages distributed as part of HLS, versioning them in lockstep for each release. - Binaries for GHC 9.4.5 - Keep instance lenses stable even if parsed results are unavailable (#3545) - Keep stale lenses for module name (#3570) - Keep type lenses stable (#3558) Happy editing! - Zubin

2 1

Help with package/module naming
by David Feuer 15 May '23

15 May '23

I'm currently putting the finishing touches on the first version of a new priority queue package. It implements amortized-optimal priority queues and is intended to compete with the heaps package with a simpler, more compact, and hopefully faster implementation†. The main types are key-only and key-value priority queues, which I'm naming based on the conventions of the pqueue package: data MinQueue k data MinPQueue k a data MaxQueue k data MaxPQueue k a Additionally, the implementation uses underlying types with constant time insertion, logarithmic-time union, logarithmic time peekMin, and logarithmic time deleteMin: -- Key-value priority queues that are strict in both the key and the -- value. data MinPQueue k a -- Similar to the above, but associating -- two values with each key. The @a@ -- value is stored strictly and the @b@ value -- is stored lazily. data MinPQueue2 k a b So I have a few questions: 1. What should I name this package? 2. What should I name the modules containing the bootstrapped queues? 3. What should I name the modules containing the underlying oddly-strict queues? I don't know how many people will want to use them, but I figure they're probably just what's needed from time to time and it's better to let them loose than keep them under wraps. Finally, there is the question of how to expose internals. In addition to the queues above, with are all reasonable things for public consumption, there are modules implementing those, as well as a couple modules implementing helper types and such. I'm really no good at figuring out how to "lay out" modules in a package, so any advice would be greatly appreciated. † The heaps package uses bootstrapped skew binomial heaps (Brodal and Okasaki), achieving worst case constant-time union and peekMin and worst case logarithmic time deleteMin. The package I'm working on uses the same bootstrapping technique, but with a space-saving modification of Louis Wasserman's realization of Okasaki lazy binomial heap as the base priority queue. It achieves the same bounds, but union is only constant time in the (persistently) amortized sense; it degrades to logarithmic time in the worst case. In practice, this theoretical degradation likely doesn't matter, both because it's small and because GHC's garbage collection imposes occasional delays regardless.

2 3

QuickCheck update
by David Feuer 15 May '23

15 May '23

The latest QuickCheck release was uploaded to Hackage on 2020-11-14. There have been multiple commits to the master branch since then. konsumlamm opened an issue on 2023-03-10* requesting a new release. You responded positively, but there has been no visible progress towards that goal. Does QuickCheck need additional maintainers? * https://github.com/nick8325/quickcheck/issues/351

1 0

Is anybody on 9.6.1 having problems profiling on a Mac?
by George Colpitts 15 May '23

15 May '23

Is anybody on 9.6.1 having problems profiling on a Mac? As I reported in https://gitlab.haskell.org/ghc/ghc/-/issues/23211 ghc -prof -fprof-auto hello.hs where the file hello.hs consists of the line: main = print "hello" gets the error: Could not find module ‘Prelude’ Perhaps you haven't installed the "p_dyn" libraries for package ‘base-4.18.0.0’? It seems to me like a regression of the closed bug https://gitlab.haskell.org/ghc/ghc/-/issues/20707. Thanks George

3 10

ICFP 2023: Call for Tutorial, Panel, and Discussion Proposals
by ICFP Publicity 15 May '23

15 May '23

ICFP 2023 CALL FOR TUTORIAL, PANEL, AND DISCUSSION PROPOSALS 28th ACM SIGPLAN International Conference on Functional Programming September 4 - 9, 2023 Seattle, WA, USA https://icfp23.sigplan.org/ The 28th ACM SIGPLAN International Conference on Functional Programming will be held in Seattle, WA, USA on September 4 - 9, 2023, with the option of virtual participation. ICFP provides a forum for researchers and developers to hear about the latest work on the design, implementations, principles, and uses of functional programming. Proposals are invited for tutorials, lasting approximately 3 hours each, to be presented during ICFP and its co-located workshops and other events. The tutorials may target an audience who is interested in commercial uses of functional programming, but we also welcome tutorials whose primary audience is researchers rather than practitioners. Tutorials may focus either on a concrete technology or on a theoretical or mathematical tool. Ideally, tutorials will have a concrete result, such as "Learn to do X with Y" rather than "Learn language Y". Just like last year, following the success of the #ShutDownPL event, we are also inviting proposals for panels and discussions on topics of broader interest to the PL community. Tutorials, panels, and discussions may occur before or after ICFP, co-located with the associated workshops, on September 4 or September 8-9. ---------------------------------------------------------------------- Submission details Deadline for submission: May 23rd, 2023 Notification of acceptance: May 26th, 2023 Prospective organizers of tutorials are invited to submit a completed tutorial proposal form in plain text format to the ICFP 2023 workshop co-chairs (Arthur Azevedo de Amorim and Yannick Forster), via email to icfp-workshops-2023 AT googlegroups.com by May 23rd, 2023. Please note that this is a firm deadline. The proposal form is available at: http://www.icfpconference.org/icfp2023-files/icfp23-panel-form.txt http://www.icfpconference.org/icfp2023-files/icfp23-tutorials-form.txt ---------------------------------------------------------------------- Selection committee The proposals will be evaluated by a committee comprising the following members of the ICFP 2023 organizing committee, together with the members of the SIGPLAN executive committee. Workshop Co-Chair: Arthur Azevedo de Amorim (Boston University) Workshop Co-Chair: Yannick Forster (Inria Nantes) General Chair: Nikhil Swamy (Microsoft Research) Program Chair: Sam Lindley (University of Edinburgh) ---------------------------------------------------------------------- Further information Any queries should be addressed to the workshop co-chairs (Arthur Azevedo de Amorim and Yannick Forster), via email to icfp-workshops-2023 AT googlegroups.com

1 0