February 2024 - Haskell-Cafe

Best way to implement "scoped exceptions"?
by Tom Ellis 27 May '25

27 May '25

I'd like to implement "scoped exceptions"[1], that is, a combinator of type withScopedException :: ((forall a. e -> IO a) -> IO r) -> IO (Either e r) so I can use exceptions in a convenient way, without having to set up fragile "action at a distance" based on the type of the thing I'm throwing and catching, for example: scopedExceptionExample :: IO (Either String (Either Int Void)) scopedExceptionExample = do withScopedException $ \throw1 -> withScopedException $ \throw2 -> if (1 :: Int) < 0 then throw1 "Hello" else throw2 1234 -- ghci> scopedExceptionExample -- Right (Left 1234) Here's a hacky way of doing it, based on tagging each exception with a unique value, and then filtering when handling: data MyException where MyException :: e -> Data.Unique.Unique -> MyException instance Show MyException where show _ = "<MyException>" instance Exception MyException withScopedException :: ((forall a. e -> IO a) -> IO r) -> IO (Either e r) wiathScopedException f = do fresh <- Data.Unique.newUnique flip tryJust (f (\e -> throwIO (MyException e fresh))) $ \case MyException e tag -> -- unsafeCoerce is very unpleasant if tag == fresh then Just (unsafeCoerce e) else Nothing This is the approach taken by the effectful library[2]. But is there a better way? Can I persuade GHC's RTS to work like this directly? Tom [1] I don't know if these already have a name [2] https://hackage.haskell.org/package/effectful-core-2.2.1.0/docs/src/Effectf…

5 7

S-REPLS 14 at Jane Street London on March 8th: Call for participation
by Richard Eisenberg 29 Feb '24

29 Feb '24

============================================================================ S-REPLS 14 South of England Regional Programming Language Seminar series Jane Street London, 2 & A Half Devonshire Square, Premier Place, London EC2M 4UJ Fri 8 March 2024 ============================================================================ S-REPLS is a series of informal meetings for people in the South of England interested in the principles, design and implementation of programming languages. Events take place approximately thrice yearly, organised by members of the community. Attendance at S-REPLS 14 is free and lunch will be provided. Due to security concerns and planning needs all attendees need to register. Please RSVP here: https://docs.google.com/forms/d/e/1FAIpQLSeODsd6ZXNi58hTBBP1oMOLkjgkZK40QdK… <https://docs.google.com/forms/d/e/1FAIpQLSeODsd6ZXNi58hTBBP1oMOLkjgkZK40QdK…> Schedule below. Talk abstracts and further details can be found on the S-REPLS 14 website: https://www.cl.cam.ac.uk/events/s-repls14/ <https://www.cl.cam.ac.uk/events/s-repls14/> 0900 Arrival and breakfast 1000 Keynote: Graded types and Algebraic Effects Dominic Orchard (University of Kent and University of Cambridge) 1100 Break 1130 Automatically Finding Upper and Lower Time Complexity Bounds for Parallel Evaluation Carsten Fuhs (Birkbeck, University of London) 1200 Somewhat Dynamic Build Systems Neil Mitchell (Meta) 1230 Lunch 1330 The Dafny Programming Language and Static Verifier Stefan Zetzsche (Amazon) 1400 Implementing a Dependently Typed Language Modelling Algebraic Equivalences Zhiyi Liu (University of Cambridge) 1430 Programming with First-class Constructor Contexts Anton Lorenzen (University of Edinburgh) 1500 Break 1530 Finding cheaper straightline instruction sequences more cheaply Maria A Schett 1600 Starlark: between configuration and programming language Stiopa Koltsov (Meta) 1630 Ill-Typed programs don’t evaluate Charlie Walpole (University of Bristol)

1 0

Is it possible to type a function to accept only literal values?
by Daneel Yaitskov 27 Feb '24

27 Feb '24

Hi List, Recently I was experimenting with a GCC intrinsic functions for x86 SMID instructions to write a fast CSV parser and noticed that some arguments are required to be literal values (known at compile time), which makes live hard, e.g. bit shift operation expects number of bits to shift in such a way. Nonetheless the intel manual defines function prototypes as normal: __m128i _mm_bsrli_si128 (__m128i a, int imm8) Using a variable as second argument in _mm_bsrli_si128 produces unclear, but compilation error with exact line number. Definitely some magic happens here, because it is not possible to express this constraint type system of C language. Haskell is known as one of the best languages in type acrobatics, but could it beat C here? -- Best regards, Daniil Iaitskov

3 2

[ANNOUNCE] Agda 2.6.4.2
by Andreas Abel 26 Feb '24

26 Feb '24

Dear all, The Agda Team is pleased to announce the release of Agda 2.6.4.2. Agda 2.6.4.2 is a bugfix release over Agda 2.6.4.1: - Fix an inconsistency in Cubical Agda related to catch-all clauses. - Fix some internal errors. - Fix an issue with `opaque`. - Fix building with cabal flag `-f debug-serialisation`. # GHC supported versions Agda 2.6.4.2 has been tested with GHC 9.8.1, GHC 9.6.4, 9.4.8, 9.2.8, 9.0.2, 8.10.7, 8.8.4 and 8.6.5 on Linux, macOS and Windows. # Installation Agda 2.6.4.2 can be installed using cabal-install or stack: 1. Getting the tarball $ cabal update $ cabal get Agda-2.6.4.2 $ cd Agda-2.6.4.2 2. a. Using cabal-install $ cabal install -f +optimise-heavily -f +enable-cluster-counting 2. b. Using stack $ stack --stack-yaml stack-a.b.c.yaml install --flag Agda:optimise-heavily --flag Agda:enable-cluster-counting replacing `a.b.c` with your version of GHC. The flags mean: - optimise-heavily: Turn on extra optimisation for a faster Agda. Takes large resources during compilation of Agda. - enable-cluster-counting: Enable unicode clusters for alignment in the LaTeX backend. Requires the ICU lib to be installed and known to pkg-config. These flags can be dropped from the install if causing trouble. # Standard library You can use standard library v1.7.3, v2.0 or the `master` branch of the standard library with Agda 2.6.4.2. This branch is available at https://github.com/agda/agda-stdlib/ # Fixed issues over Agda 2.6.4.1 https://hackage.haskell.org/package/Agda-2.6.4.2/candidate/changelog Enjoy Agda 2.6.4.2! Andreas, on behalf of the Agda Team -- Andreas Abel <>< Du bist der geliebte Mensch. Department of Computer Science and Engineering Chalmers and Gothenburg University, Sweden andreas.abel(a)gu.se http://www.cse.chalmers.se/~abela/

1 1

[ANN] Darcs 2.18.1 release
by Ganesh Sittampalam 25 Feb '24

25 Feb '24

Hi, On behalf of the Darcs team, I would like to announce the release of Darcs 2.18.1 [1]. The full release notes [2] are appended below. Regards, Ganesh [1] https://hackage.haskell.org/package/darcs-2.18.1 [2] https://hackage.haskell.org/package/darcs-2.18.1/changelog Darcs 2.18.1, 25 Feb 2024 * Supports GHC 9.8 and the most recent version of other dependencies at the time of release, with the exception of the tls 2.0 package, which has been held back because of problems connecting to hub.darcs.net (see https://bugs.darcs.net/issue2715) * Substantial rewrite of the 'darcs test' command. The most important user visible change is that a test script can now return an exit code of 125 to reflect an untestable/skipped state (as with with "git bisect run"). This in turn means that a group of patches can be found to be responsible for a failure rather than just a single one. By default, Darcs will try to minimise such a group by reordering patches to remove irrelevant ones from the initial group found from the patch ordering in the repository. This behaviour can be disabled with --no-shrink-failure. * Remove support for downloading via curl This is no longer particularly useful as we now use modern, maintained Haskell libraries for native HTTP downloading, and substantially simplifies this area of the code. * Patch index: Significant performance improvement The patch index is used in commands like annotate and log. A couple of performance improvements were made that should speed up using the patch index. * Progress reporting Progress reports are now provided during more long-running operations, including updating the "index" (a cache that speeds up detecting changes in the working directory), and during merge operations. They also behave better on Windows and when outputting long lines. * Other changes/fixes: * Use hardlinks more often to share files between repositories/caches. * Support --leave-test-dir for all commands that support --test * Avoid extraneous "repo:." entries in _darcs/prefs/sources [issue2672] * Add 'darcs clean' command as an alias for 'darcs revert -l' * 'darcs rebase unsuspend': add more patch editing options * Fix stale lock files after Ctrl-C * External merge tools: preserve output, and fail if tool does * Properly reference renamed files in external merge [issue189] * Mark conflicts properly if tag pulled at the same time [issue2682] * Remove the useless optimize pristine subcommand * 'darcs convert': honour the --compress and --diff-algorithm options * Fix contrib/darcs-shell [issue2646] * Fix 'darcs pull --dont-allow-conflicts' with external-merge [issue1819] * Problems with local pristine files now tell user to run 'darcs repair' [issue1981] * Fix various problems with symlinks, including on Windows * Add --no-prefs-templates option when creating a repository * Allow 'darcs rebase unsuspend' when there are non-conflicting unrecorded changes * Handle broken pending patches in 'darcs check' and 'darcs repair' * Improve error reporting when remote _darcs/format doesn't exist * 'darcs optimize reorder': add --deep/--shallow options * 'darcs optimize compress/uncompress': also handle pristine files * 'darcs optimize cache': don't work with lists of darcs repos Instead just the global cache is cleaned by checking hard-link counts * Skip the pager when $DARCS_PAGER / $PAGER are set to the empty string * 'darcs convert export': allow relative paths for --read-marks and --write-marks * Fix 'darcs amend --unrecord' to move unrecorded changes to pending [issue2697] * Don't treat cancelling an operation as failure [issue2074] * Fix cloning of ssh repo when using Ctrl-C to stop getting patches [issue2701] * Don't report invalid regexes as a bug in darcs [issue2702] * Add short option -n for --dry-run * 'darcs diff': support --look-for-moves and --look-for-adds * Fix buffering problem with 'darcs diff' [issue2704] * 'darcs obliterate' and 'darcs rebase': offer to revert any conflicting unrecorded changes * Stop displaying context lines in various interactive scenarios: it didn't work properly and would require a lot of work to fix. * Improve conflict display for Rebase and V3 patches * Increase size limit to 100K for environment variables like DARCS_PATCHES_XML, and warn when it is exceeded * 'darcs rebase unsuspend': improve the display of dropped dependencies * 'darcs amend --ask-deps': also provide a way to remove dependencies * 'darcs push': support --reorder-patches option * Remove the --remote-repo option * Don't display hints about using --set-default * Conflict resolution: unrecorded changes will suppress conflict marking of appropriate changes [issue2708] * Explain what a "clean tag" is in help for tag command * Fix problem with naming patch bundles after patches that contain characters incompatible with the current locale [issue2716]

2 1

Package Takeover: Hexpat
by Pavan Rikhi 25 Feb '24

25 Feb '24

Hi, as per the "Taking over a package" wiki page, I am sending this message as I would like to takeover basic maintenance of the hexpat package: https://hackage.haskell.org/package/hexpat I have attempted to reach the original author from multiple of their publicly available addresses 3 weeks ago & have gotten no response. The last commit to the package was over 7 years ago. I have submit multiple PRs to fix compilations to their repository with no response: https://github.com/the-real-blackh/hexpat/pull/13 https://github.com/the-real-blackh/hexpat/pull/14 The hackage trustees have applied some of these, but co-maintainership by me would mean less work for them, and eventually code changes will be required to keep the package compiling: https://github.com/haskell-infra/hackage-trustees/issues/363 https://github.com/haskell-infra/hackage-trustees/issues/394 Pavan Rikhi

1 0

[ANNOUNCE] GHC 9.8.2 is now available
by Zubin Duggal 23 Feb '24

23 Feb '24

The GHC developers are happy to announce the availability of GHC 9.8.2. Binary distributions, source distributions, and documentation are available on the [release page](https://www.haskell.org/ghc/download_ghc_9_8_2.html). GHC Blog Post: https://www.haskell.org/ghc/blog/20240223-ghc-9.8.2-released.html This release is primarily a bugfix release addressing many issues found in the 9.8 series. These include: * A fix for a bug where certain warnings flags were not recognised (#24071) * Fixes for bugs in the renamer and typechecker (#24084, #24134, #24279, #24083) * Fixes for bugs in the simplifier and code generator (#24160, #24242, #23628, #23659, #24160, #23862, #24295, #24370) * Fixes for some memory leaks in GHCi (#24107, #24118) * Improvements to error messages (#21097, #16996, #11050, #24196, #24275, #23768, #23784, #23778) * A fix for a recompilation checking bug where GHC may miss changes in transitive dependencies when deciding to relink a program (#23724). * And many more fixes 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. Enjoy! -Zubin [ticket]: https://gitlab.haskell.org/ghc/ghc/-/issues/new [release notes]: https://downloads.haskell.org/~ghc/9.8.2/docs/users_guide/9.8.2-notes.html

1 0

what monad is this?
by Johannes Waldmann 21 Feb '24

21 Feb '24

Dear Cafe, I was surprised to see ghc accept this code ghci> do let {x = False}; not x True because I initially thought: there's a `do`, so there must be some monad, but which is it? some implicit Identity monad perhaps? But the type is indeed plain `Bool`, and the explanation is (I think): there is a `do` but there is no monad, since the translation according to ( https://www.haskell.org/onlinereport/haskell2010/haskellch3.html#x8-470003.… ) does never produce `(>>=)`. So, all is fine, nothing to see here. Perhaps keep in mind as an edge case, useful to confuse students. Actually, to motivate them to read the language standard ... - J.W.

5 4

Invitation Guest Speaker @ ACFTI Cybersecurity stream
by Andrew Zayine 08 Feb '24

08 Feb '24

On behalf of the Association of Cyber Forensics and Threat Investigators (ACFTI), I am pleased to invite you to the new Cybersecurity stream lecture/seminar series. The presentation is a maximum of 1 hour in length, with an audience of about 60+, made up of undergraduate and postgraduate students plus cybersecurity students from developing countries. Our goal is to shine a spotlight on the broad array of new advances in cybersecurity science and operations currently adopted in the industry. This session will be conducted online. It will be fantastic to have any hands-on topics related to cyber forensics. Your discussion on this topic will be a great addition to our event. Expressions of interest to present from anyone doing research or applying cybersecurity techniques to practical or theoretical applications related to the interactions between cyber forensics and threat investigations can be sent as a summary of your work (c.200 words) to acfti (at) acfti (dot) org by February 15, 2024 Thank you in advance for your consideration, and we are very much looking forward to hearing from you. To get more news about our events, please join our low-traffic announcement group @ https://groups.google.com/g/acfti ________________________________________________________ Association of Cyber Forensics and Threat Investigators https://www.acfti.org Twitter: @acfti

5 4

Haskell Ecosystem Workshop @ Zurihac
by Jose Calderon 08 Feb '24

08 Feb '24

Hello everyone, Much like last year, the Haskell Foundation is putting on a technical workshop co-located with Zurihac. Last year the focus was on GHC, this year we are changing the scope of the workshop in order to include tools that ‘fill out’ the Haskell and GHC development ecosystem, including: * Haskell Language Server * Cabal * Haskell Ecosystem Security * GHC The event will be held during the 2 days prior to Zurihac , 6-7 June, 2024. Please take this into account when planning your trip to Zurihac. The event's aim is to help folks interested in contributing to these tools get up to speed on the context, design, and workflow necessary in order to make contributions to these projects. We're asking our speakers to have 'good first issues' ready for volunteers to work on, and Zurihac itself will be a great time to dive in whilst still having access to the various maintainers. Additionally, the talks will be recorded as a form of documentation for these tools. We have a few speakers committed, but are still working on the overall program. Once we have the program finalized, we will open registration. We expect to have all speakers and talks finalized within the next few weeks. Just as last year, the event will have a modest registration fee in order to cover its costs and pay for the speakers to attend. If you have any suggestions on what tools/topics you would like to see represented, I’m happy to hear them. While this year’s program is mostly filled, it can help us know what to plan for future events like this. I hope to see you there!

1 0