- Haskell-Cafe - Haskell.org

ANN: stack-1.6.1
by Emanuel Borsboom 07 Dec '17

07 Dec '17

See https://haskellstack.org <https://haskellstack.org/> for installation and upgrade instructions. Note: we are releasing a bit earlier than planned due to #3624 <https://github.com/commercialhaskell/stack/issues/3624>. As such, not all the binaries have been built yet, but the commonly used 64-bit binaries of Linux static, macOS, and Windows are available. In addition, the Windows installer and binary has not been codesigned (we are awaiting validation of a new codesign certificate), and you may see a "Windows Defender SmartScreen prevented an unrecognized app from starting" warning when you try to run them. If so, click on More info, and then click on the Run anyway button that appears. Changes since v1.5.1 Major changes: Complete overhaul of how snapshots are defined, the packages and extra-deps fields, and a number of related items. For full details, please see the writeup on these changes <https://www.fpcomplete.com/blog/2017/07/stacks-new-extensible-snapshots>. PR #3249 <https://github.com/commercialhaskell/stack/pull/3249>, see the PR description for a number of related issues. Upgraded to version 2.0 of the Cabal library. Behavior changes: The --install-ghc flag is now on by default. For example, if you run stack build in a directory requiring a GHC that you do not currently have, Stack will automatically download and install that GHC. You can explicitly set install-ghc: false or pass the flag --no-install-ghc to regain the previous behavior. stack ghci no longer loads modules grouped by package. This is always an improvement for plain ghci - it makes loading faster and less noisy. For intero, this has the side-effect that it will no longer load multiple packages that depend on TH loading relative paths. TH relative paths will still work when loading a single package into intero. See #3309 <https://github.com/commercialhaskell/stack/issues/3309> Setting GHC options for a package via ghc-options: in your stack.yaml will promote it to a local package, providing for more consistency with flags and better reproducibility. See: #849 <https://github.com/commercialhaskell/stack/issues/849> The package-indices setting with Hackage no longer works with the 00-index.tar.gz tarball, but must use the 01-index.tar.gz file to allow revised packages to be found. Options passsed via --ghci-options are now passed to the end of the invocation of ghci, instead of the middle. This allows using +RTS without an accompanying -RTS. When auto-detecting --ghc-build, tinfo6 is now preferred over standard if both versions of libtinfo are installed Addition of stack build --copy-compiler-tool, to allow tools like intero to be installed globally for a particular compiler. #2643 <https://github.com/commercialhaskell/stack/issues/2643> Stack will ask before saving hackage credentials to file. This new prompt can be avoided by using the save-hackage-creds setting. Please see #2159 <https://github.com/commercialhaskell/stack/issues/2159>. The GHCRTS environment variable will no longer be passed through to every program stack runs. Instead, it will only be passed through commands like exec, runghc, script, ghci, etc. See #3444 <https://github.com/commercialhaskell/stack/issues/3444>. ghc-options: for specific packages will now come after the options specified for all packages / particular sets of packages. See #3573 <https://github.com/commercialhaskell/stack/issues/3573>. The pvp-bounds feature is no longer fully functional, due to some issues with the Cabal library's printer. See #3550 <https://github.com/commercialhaskell/stack/issues/3550>. Other enhancements: The with-hpack configuration option specifies an Hpack executable to use instead of the Hpack bundled with Stack. Please see #3179 <https://github.com/commercialhaskell/stack/issues/3179>. It's now possible to skip tests and benchmarks using --skip flag GitSHA1 is now StaticSHA256 and is implemented using the StaticSize 64 ByteString for improved performance. See #3006 <https://github.com/commercialhaskell/stack/issues/3006> Dependencies via HTTP(S) archives have been generalized to allow local file path archives, as well as to support setting a cryptographic hash (SHA256) of the contents for better reproducibility. Allow specifying --git-branch when upgrading When running stack upgrade from a file which is different from the default executable path (e.g., on POSIX systems, ~/.local/bin/stack), it will now additionally copy the new executable over the currently running stack executable. If permission is denied (such as in /usr/local/bin/stack), the user will be prompted to try again using sudo. This is intended to assist with the user experience when the PATH environment variable has not been properly configured, see #3232 <https://github.com/commercialhaskell/stack/issues/3232>. stack setup for ghcjs will now install alex and happy if they are not present. See #3109 <https://github.com/commercialhaskell/stack/issues/3232>. Added stack ghci --only-main flag, to skip loading / importing all but main modules. See the ghci documentation page for further info. Allow GHC's colored output to show through. GHC colors output starting with version 8.2.1, for older GHC this does nothing. Sometimes GHC's heuristics would work fine even before this change, for example in stack ghci, but this override's GHC's heuristics when they're broken by our collecting and processing GHC's output. Extended the ghc-options field to support $locals, $targets, and $everything. See: #3329 <https://github.com/commercialhaskell/stack/issues/3329> Better error message for case that stack ghci file targets are combined with invalid package targets. See: #3342 <https://github.com/commercialhaskell/stack/issues/3342> For profiling now uses -fprof-auto -fprof-cafs instead of the deprecated -auto-all -caf-all. See: #3360 <https://github.com/commercialhaskell/stack/issues/3360> Better descriptions are now available for stack upgrade --help. See: #3070 <https://github.com/commercialhaskell/stack/issues/3070> When using Nix, nix-shell now depends always on gcc to prevent build errors when using the FFI. As ghc depends on gcc anyway, this doesn't increase the dependency footprint. --cwd DIR can now be passed to stack exec in order to execute the program in a different directory. See: #3264 <https://github.com/commercialhaskell/stack/issues/3264> Plan construction will detect if you add an executable-only package as a library dependency, resulting in much clearer error messages. See: #2195 <https://github.com/commercialhaskell/stack/issues/2195>. Addition of --ghc-options to stack script to pass options directly to GHC. See: #3454 <https://github.com/commercialhaskell/stack/issues/3454> Add hpack package.yaml to build Stack itself Add ignore-revision-mismatch setting. See: #3520 <https://github.com/commercialhaskell/stack/issues/3520>. Log when each individual test suite finishes. See: #3552 <https://github.com/commercialhaskell/stack/issues/3552>. Avoid spurious rebuilds when using --file-watch by not watching files for executable, test and benchmark components that aren't a target. See: #3483 <https://github.com/commercialhaskell/stack/issues/3483>. Stack will now try to detect the width of the running terminal (only on POSIX for the moment) and use that to better display output messages. Work is ongoing, so some messages will not be optimal yet. The terminal width can be overriden with the new --terminal-width command-line option (this works even on non-POSIX). Passing non local packages as targets to stack ghci will now cause them to be used as -package args along with package hiding. Detect when user changed .cabal file instead of package.yaml. This was implemented upstream in hpack. See #3383 <https://github.com/commercialhaskell/stack/issues/3383>. Automatically run autoreconf -i as necessary when a configure script is missing. See #3534 <https://github.com/commercialhaskell/stack/issues/3534> GHC bindists can now be identified by their SHA256 checksum in addition to their SHA1 checksum, allowing for more security in download. For filesystem setup-info paths, it's no longer assumed that the directory is writable, instead a temp dir is used. See #3188 <https://github.com/commercialhaskell/stack/issues/3188>. Bug fixes: stack hoogle correctly generates Hoogle databases. See: #3362 <https://github.com/commercialhaskell/stack/issues/3362> stack --docker-help is now clearer about --docker implying system-ghc: true, rather than both --docker and --no-docker. stack haddock now includes package names for all modules in the Haddock index page. See: #2886 <https://github.com/commercialhaskell/stack/issues/2886> Fixed an issue where Stack wouldn't detect missing Docker images properly with newer Docker versions. #3171 <https://github.com/commercialhaskell/stack/pull/3171> Previously, cabal files with just test-suite could cause build to fail (#2862 <https://github.com/commercialhaskell/stack/issues/2862>) If an invalid snapshot file has been detected (usually due to mismatched hashes), Stack will delete the downloaded file and recommend either retrying or filing an issue upstream. See #3319 <https://github.com/commercialhaskell/stack/issues/3319>. Modified the flag parser within Stack to match the behavior of Cabal's flag parser, which allows multiple sequential dashes. See #3345 <https://github.com/commercialhaskell/stack/issues/3345> Now clears the hackage index cache if it is older than the downloaded index. Fixes potential issue if stack was interrupted when updating index. See #3033 <https://github.com/commercialhaskell/stack/issues/3033> The Stack install script now respects the -d option. See #3366 <https://github.com/commercialhaskell/stack/pull/3366>. stack script can now handle relative paths to source files. See #3372 <https://github.com/commercialhaskell/stack/issues/3372>. Fixes explanation of why a target is needed by the build plan, when the target is an extra dependency from the commandline. See #3378 <https://github.com/commercialhaskell/stack/issues/3378>. Previously, if you delete a yaml file from ~/.stack/build-plan, it would trust the etag and not re-download. Fixed in this version. Invoking stack --docker in parallel now correctly locks the sqlite database. See #3400 <https://github.com/commercialhaskell/stack/issues/3400>. docs.haskellstack.org RTD documentation search is replaced by the mkdocs search. Please see #3376 <https://github.com/commercialhaskell/stack/issues/3376>. stack clean now works with nix. See #3468 <https://github.com/commercialhaskell/stack/issues/3376>. stack build --only-dependencies no longer builds local project packages that are depended on. See #3476 <https://github.com/commercialhaskell/stack/issues/3476>. Properly handle relative paths stored in the precompiled cache files. See #3431 <https://github.com/commercialhaskell/stack/issues/3431>. In some cases, Cabal does not realize that it needs to reconfigure, and must be told to do so automatically. This would manifest as a "shadowed dependency" error message. We now force a reconfigure whenever a dependency is built, even if the package ID remained the same. See #2781 <https://github.com/commercialhaskell/stack/issues/2781>. When --pvp-bounds is enabled for sdist or upload, internal dependencies could cause errors when uploaded to hackage. This is fixed, see #3290 <https://github.com/commercialhaskell/stack/issues/3290> Fixes a bug where nonexistent hackage versions would cause stack to suggest the same package name, without giving version info. See #3562 <https://github.com/commercialhaskell/stack/issues/3562> Fixes a bug that has existed since 1.5.0, where stack setup --upgrade-cabal would say that Cabal is already the latest version, when it wasn't. Ensure that an extra-dep from a local directory is not treated as a $locals for GHC options purposes. See #3574 <https://github.com/commercialhaskell/stack/issues/3574>. Building all executables only happens once instead of every time. See #3229 <https://github.com/commercialhaskell/stack/issues/3229> for more info. Thanks to all our contributors for this release: Aaron McDaid Adam McCullough Alexey Zabelin Andy Ashley Towns Chris Done Chris Martin d-dorazio Deni Bertovic Dmitry Ivanov Echo Nolan Emanuel Borsboom Felix Yan Filippo Vitale Gábor Lipták Ivan Lazar Miljenovic Joshua Simmons Judah Jacobson Khan Thompson Lizao Li Luke Murphy Martin Kolinek Mathieu Boespflug Matt Audesse Matthias Heinzel Michael Sloan Michael Snoyman mrkkrp Neil Mitchell Oleg Grenrus OvermindDL1 Paolo G. Giarrusso Rafe Reuben D'Netto Roman Cheplyaka Samuli Thomasson Schlueter Scott Fleischman Shea Levy Simon Jakobi Tom Sydney Kerckhove tswelsh Walter Franzini

1 0

[Announce] ZuriHac 2018: Registration now open
by Jasper Van der Jeugt 06 Dec '17

06 Dec '17

We are happy to announce that registration for ZuriHac 2018 is now open. Participation is free but limited to 300 attendees. You can register at: http://zurihac.info/register This year, the Haskell Hackathon will take place Friday June 8th to Sunday the 10th. It will be hosted at the Hochschule Rapperswil right besides the beautiful lake Zurich, like last year. The Zurich Haskell Hackathon is a free (as in beer), international, grassroots collaborative coding festival whose goal is to expand the community and to build and improve Haskell libraries, tools, and infrastructure. This is already the 7th Haskell Hackathon in Zurich! This year, we will enjoy keynotes from: - Niki Vazou - Edward Kmett - Stephen Diehl More keynote speakers will be announced. This event is open to any experience level, from beginners to gurus. This year, Julie Moronuki, co-author of Haskell Programming from first principles [1], has kindly agreed to teach a beginners course in one of the classrooms we have available. Additionally, there will be mentors on site whom you can directly approach during the whole event with any Haskell-related question you have. This is a great opportunity to meet your fellow Haskellers in real life, find new contributors for your project, improve existing libraries and tools or even start new ones! More information about ZuriHac can be found on our website [2]. We would also like to thank our sponsors Adjoint [3], Digital Asset [4], HSR [5] for supporting this great event! Looking forward to see you there, the Zurich HaskellerZ meetup group [1]: https://www.goodreads.com/book/show/25587599-haskell-programming [2]: https://zurihac.info/ [3]: https://www.adjoint.io/ [4]: https://digitalasset.com/careers.html [5]: https://www.hsr.ch/

2 1

How to match on such type
by Baa 05 Dec '17

05 Dec '17

Hello, All! I have type: infixr 9 ||| data a ||| b = A a|B b deriving (Eq, Data, Show) and usually I wraps into it values which are instances of my class IsTag: class IsTag a where anyTag :: a So, I need to check if some value wrapping by `a|||b` is equal to `anyTag`, i.e.: A (B (A x)) == (anyTag::TypeOf_x) ==> True This function must be generic, ie, it can not know anything about concreate TypeOf_x, only: `a|||b` and `IsTag`. How to do it??? I added `Data` to `a|||b` and even to values which I "wrap" with `a|||b` (I assumed to use `gmap*` and Co), but this does not help me. Is it even possible?? == Best regadrs, Paul

2 3

Taking over a maintainership on hsyslog-udp
by Kostiantyn Rybnikov 04 Dec '17

04 Dec '17

I'd like to take over a maintainership of the hsyslog-udp package [0]. I wrote Jon few days ago and got no response, so I'm acting as per the "taking over" instructions [1] Thanks [0]: http://hackage.haskell.org/package/hsyslog-udp [1]: https://wiki.haskell.org/Taking_over_a_package

2 2

10 PhD studentships in Nottingham
by Graham Hutton 04 Dec '17

04 Dec '17

Dear all, The School of Computer Science in Nottingham is advertising 10 fully-funded PhD studentships. Applicants in the area of the Functional Programming Lab (https://tinyurl.com/fp-notts) are encouraged! If you are interested in applying, please contact a potential supervisor prior to submitting your application: Thorsten Altenkirch - constructive logic, proof assistants, homotopy type theory, category theory, lambda calculus. Venanzio Capretta - type theory, mathematical logic, corecursive structures, proof assistants, dependently-typed programming. Graham Hutton - program calculation and verification, category theory, recursion operators, coinductive types. Henrik Nilsson - functional reactive programming, modelling and simulation, domain-specific languages, probabilistic languages. Best wishes, Graham +---------------------------------------------------------------+ 10 Fully-Funded PhD Studentships School of Computer Science University of Nottingham, UK https://tinyurl.com/10-phds-2018 Applications are invited for up to ten fully-funded PhD studentships in the School of Computer Science at the University of Nottingham, starting on 1 October 2018. The topics for the studentships are open, but should relate to the interests of one of the School's research groups: Agents Lab; Automated Scheduling, Optimisation and Planning; Computer Vision Lab; Functional Programming Lab; Intelligent Modelling and Analysis; Mixed Reality Lab; Data Driven Algorithms, Systems and Design and Uncertainty in Data and Decision Making. The studentships are for three years and include a stipend of £14,553 per year and tuition fees, and are available to students of any nationality. Applicants are normally expected to have a first-class Masters or Bachelors degree in Computer Science or a related discipline, and must obtain the support of a potential supervisor in the School prior to submitting their application. Initial contact with supervisors should be made at least two weeks prior to the closing date for applications. Informal enquiries may be addressed to SS-PGR-JC(a)nottingham.ac.uk To apply, please submit the following items by email to: Christine.Fletcher(a)nottingham.ac.uk: (1) a brief covering letter that describes your reasons for wishing to pursue a PhD, your proposed research area and topic, and the name of a potential supervisor; (2) a copy of your CV, including your actual or expected degree class(es), and results of all University examinations; (3) an example of your technical writing, such as a project report or dissertation; (4) contact details for two academic referees. Closing date for applications: 19th January 2018 +---------------------------------------------------------------+ This message and any attachment are intended solely for the addressee and may contain confidential information. If you have received this message in error, please send it back to me, and immediately delete it. Please do not use, copy or disclose the information contained in this message or in any attachment. Any views or opinions expressed by the author of this email do not necessarily reflect the views of the University of Nottingham. This message has been checked for viruses but the contents of an attachment may still contain software viruses which could damage your computer system, you are advised to perform your own checks. Email communications with the University of Nottingham may be monitored as permitted by UK legislation.

1 0

Call for Participation: BOB 2018 (February 23, Berlin)
by Matthias Fischmann 04 Dec '17

04 Dec '17

BOB 2018 Conference “What happens if we simply use what’s best?” February 23, 2018, Berlin http://bobkonf.de/2018/ Program: http://bobkonf.de/2018/en/program.html Registration: http://bobkonf.de/2018/en/registration.html ================================================================ BOB is the conference for developers, architects and decision-makers to explore technologies beyond the mainstream in software development, and to find the best tools available to software developers today. Our goal is for all participants of BOB to return home with new insights that enable them to improve their own software development experiences. The program features 14 talks and 8 tutorials on current topics: http://bobkonf.de/2018/en/program.html The subject range of talks includes functional programming, verticalization, formal methods, and data analytics. The tutorials feature introductions to Haskell, Clojure, Livecoding, terminal programming, Liquid Haskell, functional reactive programming, and domain-driven design. Leif Andersen will give the keynote talk. Registration is open online: http://bobkonf.de/2018/en/registration.html NOTE: The early-bird rates expire on January 22, 2018! BOB cooperates with the :clojured conference on the following day. There is a registration discount available for participants of both events. http://www.clojured.de/

1 0

ghc-8.2.2 downloads: missing checksums
by Jeremy Henty 03 Dec '17

03 Dec '17

The SHA1SUMS and SHA256SUMS files in https://downloads.haskell.org/~ghc/8.2.2/ do not have checksums for ghc-8.2.2-x86_64-unknown-linux.tar.xz . Regards, Jeremy Henty

2 2

out-commented code vs. case with redundant pattern matches
by Henning Thielemann 30 Nov '17

30 Nov '17

Occasionally I have multiple implementations of the same task and want to choose one quickly but statically. I do not want to out-comment unused branches because they shall still be type checked. So far I used this scheme: case 0::Int of 0 -> putStrLn "A" 1 -> putStrLn "B" _ -> putStrLn "C" With ghc-8.0.2 and ghc-8.2.2 I get these warnings: RedundantCase.hs:4:7: warning: [-Woverlapping-patterns] Pattern match is redundant In a case alternative: 0 -> ... RedundantCase.hs:5:7: warning: [-Woverlapping-patterns] Pattern match is redundant In a case alternative: 1 -> ... I thought that "redundant" means that the first two cases overlap with '_'. But if I replace '_' by '2' I get not only the non-exhaustive patterns warning but an additional redundancy warning on pattern '2'. Is there a nice way to tell GHC that the unused branches are intended, without generally disabling overlapping patterns warning? I mean, this one does not provoke any warnings: if True then putStrLn "X" else putStrLn "Y" but is limited to two branches.

4 5

Re: [Haskell-cafe] out-commented code vs. case with redundant pattern matches
by Olaf Klinke 29 Nov '17

29 Nov '17

Henning, how about putting the branches into their own variables. Admittedly this pollutes the namespace but is much more readable IMHO. workflow = new_workflow where new_workflow = putStrLn "A" old_workflow = putStrLn "B" I suppose GHC complains neither about unused where clauses nor about unused top-level bindings. At least with GHCi 7.8.3 the above code produces no warnings. Olaf

2 1

Re: [Haskell-cafe] out-commented code vs. case with redundant pattern matches
by Doug McIlroy 29 Nov '17

29 Nov '17

Surely GHC is too clever by half in asserting that just 1 of two impossible branches is redundant in the code > case 0::Int of > 0 -> putStrLn "A" > 1 -> putStrLn "B" > _ -> putStrLn "C" The assertion suggests that GHC code optimization would leave some useless vestige of the third case, The long-range solution to Thielmann's quandry is a bug report. Doug

1 0