- Haskell-Cafe - Haskell.org

Re: [Haskell-cafe] [Haskell] Branch Streaming (was GHC is a monopoly compiler)
by Tony Day 01 Oct '16

01 Oct '16

I've actually looked down that particular rabbit hole. Here's a link to a monoid for pipes that represents a branch: https://github.com/tonyday567/pipes-extended/blob/master/src/Pipes/Monoid.h… For flavour, the computation involves hoisting and lifting back and forth over several layers, as the various meta-physical wires cross. That was years ago, but I haven't seen anything since. It looks like Pipes is the wrong abstraction point to stream with branching. I wonder what is? On Wed, Sep 28, 2016 at 9:59 PM, Carter Schonwald < carter.schonwald(a)gmail.com> wrote: > Nope :) > > And I think we are still learning how to do stream. Have have streaming > in a line down pat, but streaming with a DAG or general graphs wendontnhave > yet ;) > > > On Wednesday, September 28, 2016, Tom Ellis <tom-lists-haskell-cafe-2013@ > jaguarpaw.co.uk> wrote: > >> On Mon, Sep 26, 2016 at 08:56:55PM -0500, Christopher Allen wrote: >> > Streaming is a good example here. It's not "obvious" how to do many >> > things in Haskell that are "obvious" in other languages. Partly >> > because industry hasn't used languages like Haskell much, partly >> > because the canvas we work with permits more structure. >> >> Is it obvious how to do streaming in other languages? >> _______________________________________________ >> Haskell-Cafe mailing list >> To (un)subscribe, modify options or view archives go to: >> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe >> Only members subscribed via the mailman list are allowed to post. > > > _______________________________________________ > Haskell-Cafe mailing list > To (un)subscribe, modify options or view archives go to: > http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe > Only members subscribed via the mailman list are allowed to post. >

3 3

Question regarding Candidate Package builds
by Guru Devanla 01 Oct '16

01 Oct '16

Is there any information on what kinds of checks happen on candidate packages that are uploaded. There was a cabal error on a published build. So, I have fixed that and have uploaded a candidate package. I was hoping the same build/checks happen on the candidate package, but I have not seen that happen. This link on the HOME page is very old and does not give any definitive information on what is currently working: https://github.com/haskell/hackage-server/issues/41#issuecomment-25005344 Thanks Guru

1 0

Library pretty print values in Tabular format
by Gurudev Devanla 01 Oct '16

01 Oct '16

Hello All, I am happy to publish a new library on Hackage that lets one print values contained in lists, maps, vectors in a tabular format. This is motivated from the output I am used to seeing while using the Pandas library in Python. This is my first Haskell library and I look forward to feedback and suggestions on improving the library. https://hackage.haskell.org/package/pptable https://github.com/gdevanla/pptable#readme One thing, which I right away do not like about the library is the constraints where the values will have to derive from both Data.Data and GHC.Generic.Generics. Thank you and I look forward to some feedback. Regards Guru P.S: I spent a lot of time trying to make Hackage produce the haddock links. Any pointers to that would be helpful. And "stack sdist" does not package the documentation automatically.

4 3

Re: [Haskell-cafe] Library pretty print values in Tabular format
by Ivan Lazar Miljenovic 30 Sep '16

30 Sep '16

Actually sending this to the list (no idea where the googlegroups bit came from). On 30 September 2016 at 23:49, Ivan Lazar Miljenovic <ivan.miljenovic(a)gmail.com> wrote: > On 30 September 2016 at 22:06, Gurudev Devanla > <gurudev.devanla(a)gmail.com> wrote: >> Hello All, >> >> I am happy to publish a new library on Hackage that lets one print values >> contained in lists, maps, vectors in a tabular format. This is motivated >> from the output I am used to seeing while using the Pandas library in >> Python. This is my first Haskell library and I look forward to feedback and >> suggestions on improving the library. >> >> https://hackage.haskell.org/package/pptable >> https://github.com/gdevanla/pptable#readme >> >> One thing, which I right away do not like about the library is the >> constraints where the values will have to derive from both Data.Data and >> GHC.Generic.Generics. > > You could provide a type-class for values as well; this would also > help avoid the usage of get_type == "Bool" that you have. > >> Thank you and I look forward to some feedback. >> >> Regards >> Guru >> >> P.S: I spent a lot of time trying to make Hackage produce the haddock >> links. Any pointers to that would be helpful. And "stack sdist" does not >> package the documentation automatically. > > Hackage should automatically generate the documentation after a period > of time; they are generated server-side not client-side (hence why > "stack sdist" didn't generate them). > >> >> >> _______________________________________________ >> Haskell-Cafe mailing list >> To (un)subscribe, modify options or view archives go to: >> http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe >> Only members subscribed via the mailman list are allowed to post. > > > > -- > Ivan Lazar Miljenovic > Ivan.Miljenovic(a)gmail.com > http://IvanMiljenovic.wordpress.com -- Ivan Lazar Miljenovic Ivan.Miljenovic(a)gmail.com http://IvanMiljenovic.wordpress.com

1 0

Alternative: you can fool many people some time, and some people many time, but...
by Corentin Dupont 30 Sep '16

30 Sep '16

Hi guys, I'm playing with the mysterious "some" and "many" from Control.Applicative. If I try: many $ Just 1 It just loops, I understand why: http://stackoverflow.com/questions/18108608/what-are-alterna tives-some-and-many-useful-for It seems that some and many are usually used in a context where something is consumed, and can be depleted, so the loop ends. But why doesn't this terminates? take 3 $ many $ Just 1 It's a recursive call, but the construction of the result should be lazy...

4 4

PLDI 2017 - Call for Workshops and Tutorials
by Tobias Grosser 30 Sep '16

30 Sep '16

*Call for Workshops and Tutorials* ============================= 2017 ACM Conference on Programming Language Design and Implementation (PLDI) June 19-23, 2017 in Barcelona, Spain http://conf.researchr.org/home/pldi-2017 PLDI 2017 will host co-located workshops and tutorials for which it calls for proposals. This year, PLDI is a part of a large cluster of co-located conferences including ECOOP, Curry On, LCTES, ISMM, DEBS and others. Take your chance in addressing the diverse audience of the premier forum in programming language design and implementation by proposing your event. PLDI welcomes prominent events focusing on programming language design theory and practice, for which it can provide guidance in ublishing results in the ACM Digital Library. * Submission guidelines * ===================== http://conf.researchr.org/track/pldi-2017/pldi-2017-workshops-and-tutorials Please submit proposals in plain text to Workshops and Tutorials Chair Aaron Smith <aaron.smith(a)microsoft.com>. A proposal should provide: * Name of the workshop/tutorial. * Duration of the workshop/tutorial. * Organizers: names, affiliation, contact information, brief (100 words) biography. * A short description (150-200 words) of the topic. * Event format: workshop/tutorial; type of submissions if any; review process; results dissemination; references to previous events. * Expected attendance and target audience within PLDI community. * Important dates * ================ Published Workshops proposals due 28 Nov 2016 Acceptance Notification 9 Dec 2016 Unpublished Workshops and Tutorials proposals due 30 Jan 2017 Workshop and Tutorials held (tentatively) 18, 22, 23 Jun 2017 *Organizing Committee* ==================== http://conf.researchr.org/committee/pldi-2017/pldi-2017-organizing-committee General Chair: Albert Cohen, INRIA, France Program Chair: Martin Vechev, ETH Zurich, Switzerland Workshops & Tutorials Chair: Aaron Smith, University of Edinburgh Organizing Chair: Fernando Orejas, Universitat Politècnica de Catalunya, Spain Publicity Chairs: Adrian Sampson, Cornell, USA Tobias Grosser, ETH Zurich, Switzerland Web Chair: Oleksandr Zinenko, INRIA, France Financial Chair: Louis-Noël Pouchet, Ohio State University, USA Student Travel Chair: Ronald Mak, San Jose State University, USA

1 0

Writing insertionSort using foldr and a helper function
by jorgemal1960＠gmail.com 30 Sep '16

30 Sep '16

I am posting the same question again because I had not subscribed to the list and I received a message saying it was automatically rejected. I have the following function that takes an element and a list and inserts the element into the list at the first position where it is less than or equal to the next element. So, if the list is sorted, then the list will remain sorted. myInsert :: Ord a => a -> [a] -> [a] myInsert x [] = [x] myInsert x (y:ys) = if x < y then x:y:ys else y:myInsert x ys Now, I have to use the above function myInsert and foldr to implement another function called insertionSort. I have been able to do it without using foldr as follows and it works just fine: insertionSort :: Ord a => [a] -> [a] insertionSort [] = [] insertionSort [x] = [x] insertionSort (x:xs) = myInsert x (insertionSort xs) I have worked for 2 days to use foldr without success, for example: insertionSort :: Ord a => [a] -> [a] insertionSort [] = [] insertionSort [x] = [x] insertionSort (x:xs) = foldr (myInsert) x (insertionSort xs) But it does not even compile, I get the following error which refers to last instruction at position "x" in "foldr (myInsert) x (insertionSort xs)": Couldn't match expected type ‘[a]’ with actual type ‘a’ I will very much appreciate your feedback in order to solve my issue. Best regards.

5 4

Take over word24 package on hackage
by 韩冬(基础平台部) 29 Sep '16

29 Sep '16

Dear hackage administrators: I'd like to take over the maintenance of word24 package on hackage, here is the conversation between me and current maintainer. My hackage account is winterland, currently my package mysql-haskell rely on word24. If there's anything you need to know, please contact me. Thanks! Winterland SEP 29/ 2016 ________________________________ From: 韩冬(基础平台部) Sent: Tuesday, August 30, 2016 5:10 PM To: John Lato Subject: Re: Query about word24 package on package How is the stackage thing going? I hesitate to add it myself because i’m not the maintainer. my hackage account is winterland, If you decide to leave the maintaince to me just drop me a message, thanks! Winter On Aug 27, 2016, at 03:57, John Lato <jwlato(a)gmail.com<mailto:jwlato@gmail.com>> wrote: Hi sorry for the delay, for some reason I've had really inconsistent messages from GitHub. If you want to do maintenance that's awesome I'll see if I can get it set up today. On Thu, Aug 25, 2016, 10:46 PM 韩冬(基础平台部) <handongwinter(a)didichuxing.com<mailto:handongwinter@didichuxing.com>> wrote: Hi, John Really really sorry to bother you by email, recently i’m writing a mysql driver in haskell, i use your word24 package to get a proper Word24 type, so i’d like to ask you to add it to stackage, i open a github issue here: https://github.com/JohnLato/word24/issues/1 , but since you didn’t reply, so i drop this email, hope you can reply ASAP. If you permit, i can do the version bump and package upload maintaince in future. Thanks for your code and time! Winter

2 2

Haskell opportunity at MemCachier
by Amit Levy 29 Sep '16

29 Sep '16

Hi friends! David and I are looking for a software engineer to join MemCachier and help work with us on our automation, analytics monitoring and dashboard, server management and core infrastructure. The work is mostly in Haskell, with the core infrastructure in Go as well as bits in pieces in whatever else is needed. There's lots of opportunity for open source contributions too! I've included the job description below (also at https://www.memcachier.com/job) If you're interested, send a resume and cover letter to jobs(a)memcachier.com or contact David or me. ============================================= MemCachier is the largest provider of memcache-as-service to PaaS applications. Our home-grown, fully distributed and multi-tenant implementation of memcache serves billions of requests each day, powering a range of applications from scrappy startups to fortune 500 companies. As the newest addition to our team, you’ll be working on improving fail over automation, analytics, monitoring, making MemCachier faster and more reliable, and well as contributing to the ecosystem of open source memcache clients in all languages. You’ll also share in on-call duties. Our design is failure resilient, so incidents are rare, but we do share in carrying around a pager. We use tools and languages we genuinely like, with a high skew towards type-safe languages. Our memcache implementation is built in Go and most of the rest of our infrastructure in Haskell. We value well designed software and compile-time guarantees. We are a small team and, like our service, we are distributed—residing on two coasts of the United States and in Europe (or elsewhere, depending on the season). While we are collaborative, this means the job requires being comfortable with a high-degree of independence. In return, you’ll be able to work wherever you are and won’t be bound to a particular schedule most of the time. We encourage functional programmers of all stripes, but we particularly encourage people from groups that are underrepresented in software. Finally, the qualifications below are not strict requirements, but rather guidelines for the kinds of skills we’re looking for. ## Qualifications * B.S. in Computer Science or equivalent experience. Advanced degrees encouraged (the current team either has, or is working on, PhDs), but we know there are plenty of other ways to come by the same knowledge. * Proficiency with distributed systems concepts. * Proficiency in either Go or Haskell, or a closely related language like ML, Rust or a desire and ability to learn. * Good English written communication skills—you don’t to be a native speaker, but you must be proficient. * Experience with web technologies (HTTP, web frameworks, etc). * Experience working independently or in remote teams. ## Responsibilities * Share in on-call duties. Our design is failure resilient, so incidents are rare, but we do share in carrying around a pager. * Contribute throughout the full stack, from provisioning services and core infrastructure, to the analytics dashboard and customer portal. * Participate in discussions about software and product design. ## Benefits * Self-Direction: You'll have a high-degree of autonomy to work on what you think is most interesting and valuable. * Open Source: We try to contribute as much of our work as possible to the open source ecosystem. * Full-Stack: We write software in many languages (Haskell, Go, Ruby, JavaScript) across the stack (front-end, middleware, distributed backends, monitoring). * Hard Problems: We have plenty of unsolved distributed systems problems and lots of data. * Flexible Hours: Work where and when is best for you. To apply, e-mail jobs(a)memcachier.com with a resume, cover letter and other supporting materials.

1 0

Re: [Haskell-cafe] Batteries included (Was: GHC is a monopoly compiler)
by Alex Belanger 29 Sep '16

29 Sep '16

Just a reminder, I think Haskell (the language) is being conflated by `base` (the library). `base` has exactly the minimum required for GHC to bootstrap and compile itself. If you guys are interested in an `other-base`, then do it and list that as a dependency in your cabal file. You choose your dependencies by relevance, but it doesn't affect what Haskell is or capable of. The whole point of having even `base` be a library (and extensions) is what allow Haskell to be such a good research language. No need to discuss morals on a non-existant problem. Go ahead and make a different base. Companies do this all the time. Alex. On Sep 28, 2016 3:22 AM, "Peter" <voldermort(a)hotmail.com> wrote: What are the advantages of batteries included? 1) It's easier for newcomers to find the appropriate library for common tasks. 2) Consistency between different programs, as they will mostly use the same standard libraries for common tasks. Do these require the batteries to be installed with the compiler? Do they require the batteries to be installed as a monolithic unit at all? An "official" wiki page pointing to the recommended packages for common tasks would seem to achieve the purpose almost as well, but without the accompanying problems. As for sets of specific library versions that are known to work well together, Stackage has already solved that. -- View this message in context: http://haskell.1045720.n5. nabble.com/Re-Batteries-included-Was-GHC-is-a-monopoly-compiler- tp5843366p5843416.html Sent from the Haskell - Haskell-Cafe mailing list archive at Nabble.com. _______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.

2 1