- Haskell-Cafe - Haskell.org

Re: [Haskell-cafe] naturally, length :: a -> Int
by Olaf Klinke 05 Mar '21

05 Mar '21

> Date: Thu, 4 Mar 2021 11:36:14 +0100 > From: Ben Franksen <ben.franksen(a)online.de> > To: haskell-cafe(a)haskell.org > Subject: Re: [Haskell-cafe] naturally, length :: a -> Int > Message-ID: <s1qd6u$l43$1(a)ciao.gmane.io> > Content-Type: text/plain; charset=utf-8 > > Am 03.03.21 um 16:17 schrieb Olaf Klinke: > > Which bugs can be caught at compile-time by having length return > > natural numbers? Regarding space, Int instead of Word only wastes the > > sign bit, doesn't it? > > > > I like the idea that Johannes Waldmann and Jaro Reinders brought up: > > Why is length :: Foldable a => a -> Int so convenient? Short answer: > > Because of "affine" things like `availableSpace - length xs'. > > > > There are indeed two types involved here, as the foundation package > > points out: relative offsets (like a tangent space?) and absolute > > counts. Think of NominalDiffTime versus UTCTime, or the two > > interpretations of vectors as points/movements in space. > > > > In this light one could regard the current length as > > "the relative offset of the end of the list" which can readily be > > subtracted from another relative offset. In mathematical terms: Int the > > free group over the monoid of cardinal lengths. > > Hm, interesting point. > > If we do embrace that viewpoint, then I'd say we should go all the way > and interpret indices modulo (non-negative) structure size! This makes > (safe) indexing total (for non-empty structures) and allows things like > xs !! (-1) == last xs as in Perl and some other languages. Unsafe > indexing (as in the vector package) could remain as is for performance > critical code. > > Cheers > Ben I like Haskell particularly for not being like Matlab, where virtually any well-bracketed indexing syntax does produce a result, but not necessarily what you intended or expected. Besides, for some structures length is expensive but index is not so much, so I wonder whether one can modulo-index into a lazy container without evaluating its entire spine. I just wanted to understand/justify why Int is so convenient, and the Time-DiffTime analogy sprang to mind. Following the free group route, one immediately sees that there is an embedding from cardinalities to offsets, but no inverse. We could, of course, bring other categories into the picture such as ordered sets. Then there is an embedding-projection pair between Int and Natural, where all Ints < 0 are mapped to 0. The base functions such as take and drop adhere to this interpretation. Olaf

2 1

Serious bug with Natural in GHC 9.0.1
by Sylvain Henry 05 Mar '21

05 Mar '21

Hi Café, I've made a mistake while implementing constant folding for Natural numbers in GHC 9.0.1 that hasn't been caught before the release. It's a rather serious bug as any number converted into `Natural` with `fromIntegral` can be truncated to Word range if a bogus rule is triggered (see #19345 and #19411 for more details). It is already fixed in HEAD and the fix will be present in GHC 9.0.2 which should be released soon. In the meantime you can check if the rule is triggered with `-ddump-rule-firings`. If it shows that the following rule has fired, then your code may be badly modified: |Rule fired: fromIntegral/Int->Natural (GHC.Real)| (the issue is that despite its name the rule isn't correctly constrained to convert Ints into Natural, but any numeric type...). Sorry about that! Sylvain #19345: https://gitlab.haskell.org/ghc/ghc/-/issues/19345 #19411: https://gitlab.haskell.org/ghc/ghc/-/issues/19411

2 1

naturally, length :: a -> Int
by Johannes Waldmann 04 Mar '21

04 Mar '21

Dear Cafe - let's fight Int. In a lot of code I write (for research, teaching, production), a lot of numbers are in fact natural numbers - because I am counting (e.g., number of elements in a collection) or pointing into what I just counted (e.g., Data.Set.elemAt). But Haskell's "most natural" type is - Int. At least that's my impression from current usage in libraries and teaching examples (see functions length, replicate, ... ). I've developed a horror of reading and writing "length something :: Int". As I will be teaching an introductory class on FP soon, I am looking for ways to handle this. One option is presented in Bird/Gibbons: Algorithm Design with Haskell, which has "type Nat = Int" on page 6. Actually, that's the very first code example that appears in the book! The footnote there refers to a remark in Data.Word that has since been dropped? Now there is Numeric.Natural, (cf. https://mail.haskell.org/pipermail/libraries/2014-November/024203.html ) and numeric literals are already overloaded, so we can just use that? Well, yes and no - we still have to write a type conversion each time we use some library function that insists on Int, or, write cumbersome "genericLength". Also (but that is the lesser concern), Numeric.Natural is for arbitrary-length numbers (corresponding to Integer), while I might sometimes know that numbers are representable in a machine word, and I don't want to pay for the implied pattern match (data Natural = NatS Word | NatJ BigNat). I am willing to pay for underflow checks (on subtraction), and for overflow checks (if there's an upper bound). If I want to live risky (no checks) I could use Data.Word. So, what do you do? (Yes, Peano numerals ...) Or, how do we convince ourselves (and our students) of keeping the current status? NB: here is another practical argument for Natural (from production code, an auto-grading system for student homework) We use datatype-generic parsers heavily. For data Foo = Foo { ... :: Nat ... }, bad input will be rejected by the parser already, and I need no extra checks later on. Less code! I sometimes even use "Pos(itive)". That's all fine and dandy - until I call some library function... - J.W.

8 8

aeson: instance FromJSON IntMap
by Olaf Klinke 04 Mar '21

04 Mar '21

Dear Cafe, why is Aeson's instance FromJSON IntMap so different from the instance of Map? Apparently the Map instance expects an Object (the dictionary) while the IntMap instance [1] is implemented using IntMap.fromList, so it expects an array. I can of course decode the JSON as (Map Int) and then use IntMap.fromAscList . Map.assocs but how efficient is that? And it also relies on Map.assocs producing the list in the order expected by fromAscList which (unlikely) could be broken in a future version of containers. Thanks Olaf [1] https://hackage.haskell.org/package/aeson-1.5.6.0/docs/src/Data.Aeson.Types…

3 2

International Conference on Applied Data Science and Intelligence - (ADSI 2021) Prague
by matteo＠confscience.com 03 Mar '21

03 Mar '21

Call for papers ************************************************* International Conference on Applied Data Science and Intelligence - (ADSI 2021) Prague- Czech Republic, October 14-15, 2021 https://confscience.com/adsi/ Submission deadline: April 1, 2021 All papers accepted in ADSI 2021 will be published in Springer CCIS (Communications in Computer and Information Science). CCIS is abstracted/indexed in Scopus, SCImago, EI-Compendex, Mathematical Reviews, DBLP, Google Scholar, and Thomson Reuters Conference Proceedings Citation (Former ISI Proceedings) *************************************************************************** IMPORTANT DATES: - Paper Submission: April 1, 2021 - Acceptance Notification: July 1, 2021 - Final Manuscript Due: September 1, 2021 *************************************************************************** The ADSI 2021 conference will be held in Conjunction with: International Conference on Recent Theories and Applications in Transportation and Mobility (RTATM 2021) International Conference on Informatics Revolution for Smarter Healthcare (IRSH 2021) *************************************************************************** TOPICS: Authors are invited to submit their original papers to address the topics of the conference, including but not limited to: FUNDAMENTALS AND THEORIES - Theoretical Models - Spatial and temporal multi-models - Multi-dimensional data - Data acquisition and pre-processing - Data inference - Data Classification and Taxonomy - Data Metrics - New approaches for collaboration and competition - Self-organization, self-healing, fault-tolerance approaches - Spatial reasoning - Context awareness - Intelligent mobility - New approaches to supervised and unsupervised learning - New approaches for security, privacy, trust, and ethics in data science - Real-time data analytics - Multi-Agent Systems for data science - Distributed data analytics - Data authenticity - New theories and approaches for Deep learning - New approaches for Business Intelligence - Fuzzy logic - Decision trees - Support vector machines - Evolutionary computation - Statistical methods - Collaborative filtering - Data engineering - Content mining - Indexing schemes - Information retrieval - Metadata use and management INTELLIGENT DATA PROCESSING AND ANALYTICS - Multi-level data processing - Data analytics optimization - Smart data mining - Machine Learning - Deep Learning - Bio-Inspired Computing - Secure data analytics - Privacy in data analytics - Trust in Big Data - Business intelligence - Visualization Analytics - Intelligence as a Service (IaaS) - Data Science as a Service (DSaaS) - Natural Language Processing - Signal Processing - Simulation and Modeling - Data-Intensive Computing SYSTEMS AND INFRASTRUCTURES - Data storage infrastructure - Data warehouses - Data Query and Indexing Technologies - Software Defined Infrastructures - Software Defined Networks (SDN) - Distributed data systems - Smart grid computing - Intelligent data management - Big Data computing - Smart data networking - Internet of Things - Cyber Physical Systems - Blockchain - Fog and Edge intelligence - Parallel Computing systems - Open Source systems for data science - Embedded intelligence - Embedded data science - In-Memory computing - Intelligent drones - Internet of Drones - Real-time data acquisition systems APPLICATIONS - Intelligent Hazard management - Intelligent data science in healthcare - Intelligent data science in farming - Intelligent data science in Oil and Gas - Smart logistics - Intelligent data science in transportation - Intelligent data science in surveillance - Xtech (Fintech, Agritech, etc.) - Intelligent drones - Digital transformation - Bioinformatics - Marketing - Social Science - E-learning and E-services *************************************************************************** OUTSTANDING PAPERS: Based on the peer review scores as well as the presentations at the conference, the authors of outstanding papers will be invited to extend their works for a potential publication in journals special issues with high impact factors. *************************************************************************** PAPER SUBMISSION: Papers must be submitted electronically as PDF files via easychair (https://easychair.org/conferences/?conf=adsi2021) All papers will be peer reviewed. Length of Full papers: 12-15 pages long (written in the LNCS/CCIS one-column page format, 400 words per page) Length of Short papers: less than 12 pages For more information, please refer to the conference website: https://confscience.com/adsi/ *************************************************************************** CONTACT For more information, please send an email to info-adsi(a)confscience.com

5 5

International Conference on Informatics Revolution for Smarter Healthcare (IRSH 2021) Prague
by matteo＠confscience.com 03 Mar '21

03 Mar '21

Call for papers ************************************************* International Conference on Informatics Revolution for Smarter Healthcare (IRSH 2021) Prague- Czech Republic, October 14-15, 2021 https://confscience.com/irsh/ Submission deadline: April 1, 2021 All papers accepted in IRSH 2021 will be published in Springer CCIS (Communications in Computer and Information Science). CCIS is abstracted/indexed in Scopus, SCImago, EI-Compendex, Mathematical Reviews, DBLP, Google Scholar, and Thomson Reuters Conference Proceedings Citation (Former ISI Proceedings) *************************************************************************** IMPORTANT DATES: - Paper Submission: April 1, 2021 - Acceptance Notification: July 1, 2021 - Final Manuscript Due: September 1, 2021 *************************************************************************** The IRSH 2021 conference will be held in Conjunction with: International Conference on Applied Data Science and Intelligence (ADSI 2021) International Conference on Recent Theories and Applications in Transportation and Mobility - (RTATM 2021) *************************************************************************** TOPICS: Authors are invited to submit their original papers to address the topics of the conference, including but not limited to: FUNDAMENTALS AND THEORIES - Interoperability and Data Integration - Confidentiality and Data Security - Data protection - Data Sharing - Security, Privacy, and Trust - Emergent healthcare standards - Emergent healthcare architectures - ICT, Ageing and Disability - Physiological and behavioural modelling - Pandemic and disease modeling - Usability and user experience of medical devices - Human behaviour - Clinical investigation regulatory frameworks - Integrated healthcare approaches - eHealth data standards and interoperability (e.g. HL7/FHIR) - Databases and data warehousing - Big Data and Open Data for healthcare - Design and Development of Methodologies for Healthcare - Emergent Communication Technologies - Real-time interaction theories - Emergent Technologies for Ambient Assisted Living - User Interface Design for healthcare - Sustainability - New approaches for accuracy and effectiveness - Data mining and bioinformatics - Enhanced living environments - Analysis and evaluation of healthcare systems INTELLIGENT HEALTHCARE - Pattern recognition and Machine - Learning for healthcare - Cognitive Informatics - Big Data in Healthcare - Wellbeing Informatics - Data Mining and Data Analytics - Data Visualization - Smart environments - Smart Ambient Assisted Living - Intelligent healthcare solutions - Agent-based solutions for healthcare - Collaboration systems - Intelligent Electronic Health Records - Internet of Things for healthcare - Cyber-Physical Systems for healthcare - Ambient Computing and Reasoning - Context Awareness - Smart devices for eldercare - Autonomy and active ageing - Emergent technologies for intelligent Computer Vision - Service production and delivery - Gamification - Multi-modal interaction - Computer-aided detection and diagnosis - Crowdsourcing for smarted healthcare SERVICES, SYSTEMS, AND INFRASTRUCTURES - Emergent healthcare services - Pervasive health systems and services - Remote healthcare management - Emergent healthcare infrastructure - Industry Revolution 4.0 for healthcare - eHealth - Electronic health records - Assistive technologies - Disease surveillance and patient monitoring systems - Prevention and detection systems - Home monitoring - Healthcare management systems - ICT-based therapeutic systems - ICT-based rehabilitation technologies - Wearable health informatics - Emergent technologies for data analytics - Ambient Assisted Leaving (AAL) - Decision Support Systems - Emergent Technologies for Remote AAL Monitoring - Emergent Technologies and Accessibility - 5G for healthcare - Healthcare supply chain and logistics - Wireless Body Networks - Telemedicine and mobile telemedicine - Mobile Systems - Software Defined infrastructures - Patient empowerment systems - Smart technology for remote patient visits - Biosensors - Medical devices APPLICATIONS - eHealth applications - Application of health informatics in clinical cases - Mobile technologies for healthcare applications - Software Systems in healthcare - Social networking and healthcare - Case Studies - Personalization and patient experience - AR and VR applications - Patient billing - Accounting systems - Personnel and payroll - Materials management - Voice recognition systems - Asset management solutions - Disease management - Feedback integration - Clinical software - Crowd-computing applications - Future directions - Drone-based solutions - Software Defined Networks for healthcare *************************************************************************** OUTSTANDING PAPERS: Based on the peer review scores as well as the presentations at the conference, the authors of outstanding papers will be invited to extend their works for a potential publication in journals special issues with high impact factors. *************************************************************************** PAPER SUBMISSION: Papers must be submitted electronically as PDF files via easychair (https://easychair.org/conferences/?conf=irsh2021) All papers will be peer reviewed. Length of Full papers: 12-15 pages long (written in the LNCS/CCIS one-column page format, 400 words per page) Length of Short papers: less than 12 pages For more information, please refer to the conference website: https://confscience.com/irsh/ *************************************************************************** CONTACT For more information, please send an email to info-irsh(a)confscience.com

1 0

Re: [Haskell-cafe] naturally, length :: a -> Int
by Johannes Waldmann 02 Mar '21

02 Mar '21

Dear Cafe, thanks for ideas and arguments. > [Ben Franksen] I would support a move Int->Word in all libraries > where that makes sense. I find it hard to think of a library where that does NOT make sense... The following is somewhat related but actually orthogonal: I mentioned that numbers often arise as cardinalities (so they are natural, by definition) and serve as pointers. The next step is then: pointer type safety. Vincent Hanquez' Foundation https://hackage.haskell.org/package/foundation-0.0.25 defines class (IsList c, Item c ~ Element c) => Collection c where length :: c -> CountOf (Element c) class ( ... Collection c) => Sequential c where take :: CountOf (Element c) -> c -> c with newtype CountOf ty (and there's also newtype Offset ty) I tend to agree. The phantom type argument for CountOf would catch errors like adding the number of students to the number of exercises (in my application). (Well and the next step after that would be to have the size of the collection in the (dependent) type as well.) Current implementation is newtype CountOf ty = CountOf Int and the author adds (and that was also the point I was making) "Int is a terrible backing type which is hard to get away from, considering that GHC/Haskell are mostly using this for [counting and] offset. Trying to bring some sanity by a lightweight wrapping." NB: I found out about this from Michael Snoyman's (very much entertaining, and too much true) https://www.snoyman.com/blog/2020/12/haskell-bad-parts-3/ Best regards, J.W.

2 1

Puzzling instance method definitions in Data.Sequence.Internal
by Viktor Dukhovni 02 Mar '21

02 Mar '21

In `containers`, Data.Sequence.Internal defines a Foldable instance for `Seq` with method definitions for `length` and `null` that I'm struggling to understand: https://github.com/haskell/containers/blob/master/containers/src/Data/Seque… #if MIN_VERSION_base(4,8,0) length = length {-# INLINE length #-} null = null {-# INLINE null #-} #endif further down the file there are also explicit top-level definitions for these functions: https://github.com/haskell/containers/blob/master/containers/src/Data/Seque… -- | $ O(1) $. Is this the empty sequence? null :: Seq a -> Bool null (Seq EmptyT) = True null _ = False -- | $ O(1) $. The number of elements in the sequence. length :: Seq a -> Int length (Seq xs) = size xs So the intent seems clear, but I don't understand how the instance method definitions are valid. The imports of Both Prelude and Data.Foldable don't include `length` or `null`: https://github.com/haskell/containers/blob/master/containers/src/Data/Seque… So the RHS definitions should plausibly resolve to the top-level functions, but when I try to define a similar type class instance GHC tells me: ‘method’ is not a (visible) method of class ‘Class’ and in any case I'd expect `length = length` to yield an infinite loop, regardless of any other bindings in the outer scope. How is Data.Sequence.Internal getting away with this??? -- Viktor.

2 2

DatatypeContexts / alternative
by CASANOVA Juan 01 Mar '21

01 Mar '21

Hello again, Haskell Cafe, Here again with another doubt on how am I supposed to. This one should be fairly easier, I believe. In short, I would like to have a functor type that can only be applied to certain type parameters. Why? Well, something like this: module DatatypeContextsExample where import Data.Map import Data.Bifunctor data Foo t = Foo (Map t t) instance Functor Foo where fmap f (Foo m) = Foo (fromList (fmap (bimap f f) (toList m))) This does not compile, because I am using toList and fromList, which require (Ord t). But I cannot really have Foo be a functor in this way without it. The thing is, every time I am going to use it, t is actually going to be Ord. But how do I tell Haskell to have this constraint? DatatypeContexts seems to be the extension that allows this: data Ord t => Foo t = Foo (Map t t) But this has two issues. First, DatatypeContexts is deprecated. Second, more importantly, it still does not type check! It produces the exact same error, saying that Ord t could not be inferred. It should be inferred from the very fact that I am using the type Foo, but GHC does not seem to want to understand this. I could also try with GADTs: data Foo t where Foo :: Ord t => Map t t -> Foo t But this still gives the same error on the fmap definition, stating it cannot use Foo without Ord. But I am pattern-matching against it, shouldn't that be enough to know that (Ord t) is going to hold? I don't understand it. Is there anything simple I am missing? Any other simple way to achieve this? Any fundamental problem with what I am trying to do? Thanks as usual, Juan. The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. Is e buidheann carthannais a th' ann an Oilthigh Dhùn Èideann, clàraichte an Alba, àireamh clàraidh SC005336.

5 14

Re: [Haskell-cafe] Not too extensible records?
by Anthony Clayden 01 Mar '21

01 Mar '21

Hi Ignat, as an example of what an extensible records system might look like in Haskell, you could play with Hugs.Trex -- https://www.haskell.org/hugs/pages/hugsman/exts.html#sect7.2 To match a few of its features to your post: > suppose I have two functions: one is a layout algorithm that assigns to nodes some spatial positions, and the second is a topological algorithm that discerns roots and leaves. These two functions are conceptually independent and ... r1 = (label = blah1) r2 = (topology = blah2 | r1) r3 = (v2 = blah3 :: Double | r2) The `( ... | r1)` means extend `r1` with some extra field or fields comma-separated. This is an 'extensible' 'anonymous' records system -- 'anonymous' in that you don't pre-declare your record types via `data`. You can name your record types, either with a `type` decl or with a `newtype` that provides a constructor for known-in-advance types. > may be applied in any order, Yes you can extend an existing record with arbitrary fields, the result is always a 'flat' tuple with the labels-value pairs forming a set: two records are same type providing they have the same set of labels and the same field types at each label, irrespective of how the labels got added. > I would like to make sure their fields are not mixed. > I need to be able to add more and more fields to labels, but in such a way that it is impossible to assign an unsuitable field. There's a so-called 'lacks' constraint you can put on record operations (written with backslash): foo :: r1\topology => Rec r1 -> a -> Rec (edges :: a | r1) is the type of a function that will extend record type `r1` lacking label 'topology' with a field `edges` at type `a`. You can put record types with lacks constraints on instances: instance (r2\v2) => Foo (Rec (label :: String | r2)) (Rec (v2 :: Double, label :: String | r2)) where ... holds for any record that includes label `label` at `String`, and possibly other labels in `r2`, but lacks `v2`. (I'm presuming there's a FunDep on Foo so the second instance parameter is the return type.) > a huge number of `HasThis` and `HasThat` instances That instance head is in effect a `Has` `label` constraint. There's also a shorthand `#label` that generates a function to extract the value at `label` **from any record with a `label`** -- that is, Hugs infers the `Has` constraint for you. > it seems to me that some smart technology is needed to make this manageable. I'm not seeing Trex as 'smart'. It follows conventional relational algebra semantics; was developed ~1996; last release of Hugs was 2006. I'd have thought GHC would come up with something comparable by now. As Olaf's and Henning's replies show, you could fake it with "a lot of boilerplate" and/or some dubious type trickery. AntC

1 0