- Haskell-Cafe - Haskell.org

Final Call for the early round of Papers for the Haskell Symposium 2021
by Hage, J. (Jurriaan) 16 Mar '21

16 Mar '21

Dear all, This is the final call for the early round of papers for the upcoming Haskell Symposium. Please forward to anyone that you believe might be interested. The deadline for this early round is March 19. Note that in May there is still the regular round of submissions. Apologies for receiving multiple copies of this announcement. Best regards, Jurriaan Hage Chair ================================================================================ ACM SIGPLAN CALL FOR SUBMISSIONS Haskell Symposium 2021 ** virtual ** Thu 26 -- Fri 27 August, 2021 http://www.haskell.org/haskell-symposium/2021/ ================================================================================ The ACM SIGPLAN Haskell Symposium 2021 will be co-located with the 2021 International Conference on Functional Programming (ICFP). Due to COVID-19 it will take place **virtually** this year. Like last year, we will be using a lightweight double-blind reviewing process. See further information below. Different from last year is that we offer a new submission category: the tutorial. Details can be found below. The Haskell Symposium presents original research on Haskell, discusses practical experience and future development of the language, and promotes other forms of declarative programming. Topics of interest include: * Language design, with a focus on possible extensions and modifications of Haskell as well as critical discussions of the status quo; * Theory, such as formal semantics of the present language or future extensions, type systems, effects, metatheory, and foundations for program analysis and transformation; * Implementations, including program analysis and transformation, static and dynamic compilation for sequential, parallel, and distributed architectures, memory management, as well as foreign function and component interfaces; * Libraries, that demonstrate new ideas or techniques for functional programming in Haskell; * Tools, such as profilers, tracers, debuggers, preprocessors, and testing tools; * Applications, to scientific and symbolic computing, databases, multimedia, telecommunication, the web, and so forth; * Functional Pearls, being elegant and instructive programming examples; * Experience Reports, to document general practice and experience in education, industry, or other contexts; * Tutorials, to document how to use a particular language feature, programming technique, tool or library within the Haskell ecosystem; * System Demonstrations, based on running software rather than novel research results. Regular papers should explain their research contributions in both general and technical terms, identifying what has been accomplished, explaining why it is significant, and relating it to previous work, and to other languages where appropriate. Experience reports and functional pearls need not necessarily report original academic research results. For example, they may instead report reusable programming idioms, elegant ways to approach a problem, or practical experience that will be useful to other users, implementers, or researchers. The key criterion for such a paper is that it makes a contribution from which other Haskellers can benefit. It is not enough simply to describe a standard solution to a standard programming problem, or report on experience where you used Haskell in the standard way and achieved the result you were expecting. A new submission category for this year's Haskell Symposium is the tutorial. Like with the experience report and the functional pearl, the key criterion for such a paper is that it makes a contribution from which other Haskellers can benefit. What distinguishes a tutorial is that its focus is on explaining an aspect of the Haskell language and/or ecosystem in a way that is generally useful to a Haskell audience. Tutorials for many such topics can be found online; the distinction here is that by writing it up for formal review it will be vetted by experts and formally published. System demonstrations should summarize the system capabilities that would be demonstrated. The proposals will be judged on whether the ensuing session is likely to be important and interesting to the Haskell community at large, whether on grounds academic or industrial, theoretical or practical, technical, social or artistic. Please contact the program chair with any questions about the relevance of a proposal. If your contribution is not a research paper, please mark the title of your experience report, functional pearl, tutorial or system demonstration as such, by supplying a subtitle (Experience Report, Functional Pearl, Tutorial Paper, System Demonstration). Submission Details ================== Early and Regular Track ----------------------- The Haskell Symposium uses a two-track submission process so that some papers can gain early feedback. Strong papers submitted to the early track are accepted outright, and the others will be given their reviews and invited to resubmit to the regular track. Papers accepted via the early and regular tracks are considered of equal value and will not be distinguished in the proceedings. Although all papers may be submitted to the early track, authors of functional pearls and experience reports are particularly encouraged to use this mechanism. The success of these papers depends heavily on the way they are presented, and submitting early will give the program committee a chance to provide feedback and help draw out the key ideas. Formatting ---------- Submitted papers should be in portable document format (PDF), formatted using the ACM SIGPLAN style guidelines. Authors should use the `acmart` format, with the `sigplan` sub-format for ACM proceedings. For details, see: http://www.sigplan.org/Resources/Author/#acmart-format It is recommended to use the `review` option when submitting a paper; this option enables line numbers for easy reference in reviews. Functional pearls, experience reports, tutorials and demo proposals should be labelled clearly as such. Lightweight Double-blind Reviewing ---------------------------------- Haskell Symposium 2021 will use a lightweight double-blind reviewing process. To facilitate this, submitted papers must adhere to two rules: 1. Author names and institutions must be omitted, and 2. References to authors' own related work should be in the third person (e.g., not "We build on our previous work" but rather "We build on the work of "). The purpose of this process is to help the reviewers come to an initial judgment about the paper without bias, not to make it impossible for them to discover the authors if they were to try. Nothing should be done in the name of anonymity that weakens the submission or makes the job of reviewing the paper more difficult (e.g., important background references should not be omitted or anonymized). In addition, authors should feel free to disseminate their ideas or draft versions of their paper as they normally would. For instance, authors may post drafts of their papers on the web or give talks on their research ideas. A reviewer will learn the identity of the author(s) of a paper after a review is submitted. Page Limits ----------- The length of submissions should not exceed the following limits: Regular paper: 12 pages Functional pearl: 12 pages Tutorial: 12 pages Experience report: 6 pages Demo proposal: 2 pages There is no requirement that all pages are used. For example, a functional pearl may be much shorter than 12 pages. In all cases, the list of references is not counted against these page limits. Deadlines --------- Early track: Submission deadline: 19 March 2021 (Fri) Notification: 23 April 2021 (Fri) Regular track and demos: Submission deadline: 21 May 2021 (Fri) Notification: 23 June 2021 (Wed) Deadlines are valid anywhere on Earth. Submission ---------- Submissions must adhere to SIGPLAN's republication policy (http://sigplan.org/Resources/Policies/Republication/) and authors should be aware of ACM's policies on plagiarism (https://www.acm.org/publications/policies/plagiarism) Program Committee members are allowed to submit papers, but their papers will be held to a higher standard. The paper submission deadline and length limitations are firm. There will be no extensions, and papers violating the length limitations will be summarily rejected. Papers should be submitted through HotCRP at: https://haskell21.hotcrp.com/ Improved versions of a paper may be submitted at any point before the submission deadline using the same web interface. Supplementary material: Authors have the option to attach supplementary material to a submission, on the understanding that reviewers may choose not to look at it. This supplementary material should not be submitted as part of the main document; instead, it should be uploaded as a separate PDF document or tarball. Supplementary material should be uploaded at submission time, not by providing a URL in the paper that points to an external repository. Authors can distinguish between anonymized and non-anonymized supplementary material. Anonymized supplementary material will be visible to reviewers immediately; non-anonymized supplementary material will be revealed to reviewers only after they have submitted their review of the paper and learned the identity of the author(s). Resubmitted Papers: authors who submit a revised version of a paper that has previously been rejected by another conference have the option to attach an annotated copy of the reviews of their previous submission(s), explaining how they have addressed these previous reviews in the present submission. If a reviewer identifies him/herself as a reviewer of this previous submission and wishes to see how his/her comments have been addressed, the conference chair will communicate to this reviewer the annotated copy of his/her previous review. Otherwise, no reviewer will read the annotated copies of the previous reviews. Proceedings =========== Accepted papers will be included in the ACM Digital Library. Their authors will be required to choose one of the following options: - Author retains copyright of the work and grants ACM a non-exclusive permission-to-publish license (and, optionally, licenses the work with a Creative Commons license); - Author retains copyright of the work and grants ACM an exclusive permission-to-publish license; - Author transfers copyright of the work to ACM. For more information, please see ACM Copyright Policy (http://www.acm.org/publications/policies/copyright-policy) and ACM Author Rights (http://authors.acm.org/main.html) Accepted proposals for system demonstrations will be posted on the symposium website but not formally published in the proceedings. Publication date: The official publication date of accepted papers is the date the proceedings are made available in the ACM Digital Library. This date may be up to two weeks prior to the first day of the conference. The official publication date affects the deadline for any patent filings related to published work. Artifacts ========= Authors of accepted papers are encouraged to make auxiliary material (artifacts like source code, test data, etc.) available with their paper. They can opt to have these artifacts published alongside their paper in the ACM Digital Library (copyright of artifacts remains with the authors). If an accepted paper's artifacts are made permanently available for retrieval in a publicly accessible archival repository like the ACM Digital Library, that paper qualifies for an Artifacts Available badge (https://www.acm.org/publications/policies/artifact-review-badging#available) Applications for such a badge can be made after paper acceptance and will be reviewed by the PC chair. Program Committee ================= Edwin Brady University of St Andrews Koen Claessen Chalmers University of Technology Dominique Devriese Vrije Universiteit Brussel Andy Gill University of Kansas Jurriaan Hage (chair) Universiteit Utrecht Zhenjiang Hu Peking University Ranjit Jhala University of California Patricia Johann Appalachian State University Yukiyoshi Kameyama University of Tsukuba George Karachalias Tweag Ralf Laemmel University of Koblenz-Landau Daan Leijen Microsoft Research Ben Lippmeier Ghost Locomotion Neil Mitchell Facebook Alberto Pardo Universidad de la Republica, Uruguay Matt Roberts Macquarie University Janis Voigtlaender University of Duisburg-Essen Nicolas Wu Imperial College London If you have questions, please contact the chair at: j.hage(a)uu.nl ================================================================================

1 0

Re: [Haskell-cafe] Future of package cryptonite
by Kazu Yamamoto 16 Mar '21

16 Mar '21

Hi Ximin, Cc: haskell-cafe According to the mail archive[1], your messages are not delivered to haskell-cafe. So, I'm citing your response below. [1] https://mail.haskell.org/pipermail/haskell-cafe/2021-March/ --Kazu From: Ximin Luo <infinity0(a)pwned.gg> Subject: Re: [Haskell-cafe] Future of package cryptonite > Sven Panne: >> Am Mo., 15. März 2021 um 01:17 Uhr schrieb Kazu Yamamoto <kazu(a)iij.ad.jp <mailto:kazu@iij.ad.jp>>: >> >> We should also take care of "memory", "foundation" and "basement". >> No action is taken for over a month to >> https://github.com/haskell-foundation/foundation/pull/549 <https://github.com/haskell-foundation/foundation/pull/549> >> >> >> Somehow I'm getting more and more allergic to all these hasty unwarranted "I want to take over package XY" requests. The last release of cryptonite was less than 7 weeks ago, a PR for foundation was not merged within 5 weeks, etc. etc. For god's sake: If you are in such a hurry, just fork locally! Or even better: Give the maintainers a huge pile of $$$, most of them are doing their stuff in their spare time, so you can't expect SLAs where you would have to spend 5 digit sums as a company. Or you can fork visibly on e.g. GitHub under a different package name and let other people decide which variant to take. >> > > The cryptonite bugfix release in January is the *only* activity on the package for many months. As I understand from Kazu, Vincent has personally decided to quit, he just has not publicly announced it (nor made any other communications). > > So this bugfix release seems more of a "I guess I should really do the minimum to stop possible harm", rather than performing expected maintainer duties & making progress on the overall Haskell ecosystem. > >> "Taking over" a package can almost be seen as robbery from the point of view of the original author, and it is actively discouraging people to make their code Open Source. We should be much, much more sensitive in the Haskell community, I haven't seen such things in other language ecosystems. >> > > Haskell as an ecosystem is much smaller than other language ecosystems, and maintainers of key libraries like cryptonite going AWOL is unfortunately more common. Given that, it is expected to see more requests like this. > > Implicit in all of these requests is that, if the original maintainer restarts their normal maintainer activities, they can immediately re-exercise their maintainership activities and previous access. > > Package namespaces are a shared resource and its consequences affect everyone, and so "ownership" is not as simple as "who got it first". That is why we have a package takeover process in the first place. [1] One extreme case outside of Haskell would be the left-pad npm example. Nobody is taking away copyrights, licenses, or source code away from anyone; nobody has the power to do that. > > [1] https://wiki.haskell.org/Taking_over_a_package > >> Having said that, I think that a few projects are blocked by stack issues before they can support GHC 9.0. It would be great if things would be released more in lock-step, I dream of a world where a new GHC comes out in sync with cabal, stack, Stackage, Haskell language server etc. all supporting the new compiler. Other language ecosystems are lightyears ahead regarding this... :-/ >> >> Cheers, >> S. > > -- > GPG: ed25519/56034877E1F87C35 > https://github.com/infinity0/pubkeys.git

1 0

SML vs Haskell types
by Galaxy Being 15 Mar '21

15 Mar '21

Hello, This is my first post here, and it's an odd one, I'll admit. Basically, I'm trying to translate the material in *The Little MLer *to Haskell, the TLMLer being an excellent types workout. So in SML I have this datatype 'a shish = Bottom of 'a | Onion of 'a shish | Lamb of 'a shish | Tomato of 'a shish and this datatype rod = Dagger | Fork | Sword and then this SML function fun is_veggie (Bottom (x)) = true | is_veggie (Onion (x)) = is_veggie (x) | is_veggie (Lamb (x)) = false | is_veggie (Tomato (x)) = is_veggie (x) which has no problem handling tis is_veggie (Onion(Tomato(Bottom(Dagger)))) Now, in Haskell I've translated this (with minor alterations) to data Shish a = Holder a | Onion a (Shish a) | Lamb a (Shish a) | Tomato a (Shish a) data Rod = Dagger | Fork | Sword However, in Haskell these two expressions are different things entirely meal4 = Tomato Dagger (Onion Fork (Lamb Spear (Holder Fork))) meal5 = (Tomato (Onion (Lamb (Holder Fork)))) Here's my attempt at handling meal4 with a Haskell isVeggie isVeggie (Holder (sh)) = True isVeggie (Onion sh (sk)) = isVeggie sk isVeggie (Tomato sh (sk)) = isVeggie sk isVeggie (Lamb sh (sk)) = False This works for meal4, but not for meal5. And yet in the SML world their is_veggie handles (Onion(Tomato(Bottom(Dagger)))) just fine. TLMLer says Onion (Tomato (Bottom (Dagger))) belongs to the type rod shish, while in Haskell Onion (Tomato (Holder (Dagger))) is a bizarre nested beast due to the fact that the data constructor variable of Onion is Tomato (Holder (Dagger)) etc. etc. Can a single Haskell version of isVeggie handle both meal4 and meal5? No? I thought so. But then how would a separate Haskell version of isVeggie handle meal5 -- or is it just too weird? Also, but not critical, how could the Haskell isVeggie be done with guards, i.e., just like a consed list is matched on (x:xs) in the recursion case? I can see that 1:2:3:[] and Onion (Tomato (Bottom (Dagger))) are both conses, but the latter I don't know how to break out into head and tail for a guard case where the individual food items were not mentioned explicitly. IOW, this doesn't work isVeggieKebab :: Shish -> Bool isVeggieKebab Holder (sk) = True isVeggieKebab (shkb (sk)) | (shkb == Onion) || (shkb == Tomato) = isVeggieKebab sk | otherwise = False I'm feeling some doom and gloom about this project. Right at the start this seems to be an insurmountable difference between SML and Haskell type systems. Or I simply don't understand something fundamental here. LB

4 5

Re: [Haskell-cafe] Future of package cryptonite
by Kazu Yamamoto 15 Mar '21

15 Mar '21

Hello, We should also take care of "memory", "foundation" and "basement". No action is taken for over a month to https://github.com/haskell-foundation/foundation/pull/549 I cannot switch my main GHC version to 9.0 at this moment. --Kazu > Hello Haskell-cafe, > > I'm writing to discuss the future of the package cryptonite, including scenarios where it would be acceptable to take over the package - by me & Kazu (in CC), who already maintains hs-tls, and anyone else who is qualified and willing. > > The current maintainer Vincent (in CC) has been unresponsive to several attempts to contact him by email over the past several months, and is not responsive to PRs [1] on github. He is not entirely dormant, he has made a bugfix release in January, however otherwise he does not communicate. > > Based on [2] and some other hearsay, I understand that Vincent has "quit Haskell". The other maintainer, Oliver Chéron has also explicitly said that he has quit Haskell [3]. > > We have also attempted to contact other members of the haskell-crypto github group, however only one of them responded to say they don't feel they can make decisions for cryptonite. > > Under what time frame does the list think it would be acceptable to take over the package? And additionally, are there any other interested volunteers to co-maintain? (I do not want to be the main maintainer, due to more direct interests in other topics.) > > Best, > Ximin > > [1] https://github.com/haskell-crypto/cryptonite > [2] https://github.com/haskell-crypto/cryptonite/issues/330#issuecomment-670801… > [3] https://github.com/haskell-crypto/cryptonite/pull/323#issuecomment-701885924 > > -- > GPG: ed25519/56034877E1F87C35 > https://github.com/infinity0/pubkeys.git

5 4

Re: [Haskell-cafe] Future of package cryptonite
by Andrew Lelechenko 15 Mar '21

15 Mar '21

> Under what time frame does the list think it would be acceptable to take over the package? And additionally, are there any other interested volunteers to co-maintain? (I do not want to be the main maintainer, due to more direct interests in other topics.) If you do not want to be the main maintainer and do not have much interest in cryptography, then it sounds wrong to initiate a takeover. There is a procedure for non-maintainer uploads via Hackage Trustees, please file an issue at https://github.com/haskell-infra/hackage-trustees/issues <https://github.com/haskell-infra/hackage-trustees/issues>, linking a patched repository. > I cannot switch my main GHC version to 9.0 at this moment. It would be nice to audit that `cryptonite` is not vulnerable to https://mail.haskell.org/pipermail/haskell-cafe/2021-March/133540.html <https://mail.haskell.org/pipermail/haskell-cafe/2021-March/133540.html> Best regards, Andrew

1 0

Re: [Haskell-cafe] Property checks and PostgreSQL?
by Anthony Clayden 15 Mar '21

15 Mar '21

> PostgreSQL data base, Haskell back end, PureScript front end. And there is a fault somewhere in this stack: sometimes some items present in the data base do not make it to the front end. A small fortune in programmer hours has been spent ... Speaking purely from a SQL-and-stuff point of view, "programmer hours spent" sounds to me like a problem with nulls. Are any of your columns nullable? Are you sure they should be nullable? In particular are any of the columns in the keys nullable? Or any of the columns on which you're joining tables? (Perhaps a proper value in one table, but a null in the table it's supposed to equal?) How are you mapping between Haskell field values to/from nullable columns in the SQL? Is that mapping fool-proof? Is it consistent across different tables using what should be the same value? And in general: SQL nulls are an abomination; don't do that; do normalise your database schema to avoid nulls. AntC (35 years an SQL programmer, never trusted it)

1 0

Let's fight Int, Pt. 2
by Johannes Waldmann 14 Mar '21

14 Mar '21

Dear Cafe, it's bad to use Int where you really mean Nat, because this creates extra work (defensive programming, or mental gymnastics to prove that you don't need it) Another Int anti-feature is arithmetics with silent wrap-around - with respect to an unspecified modulus, as highlighted by a recent example: https://gitlab.haskell.org/ghc/ghc/-/issues/19500 What are the possibilities for a "checked Int"? (also, "checked Nat") A historical precedent is set by Ada http://archive.adaic.com/standards/83lrm/html/lrm-03-05.html#3.5.4 "The exception NUMERIC_ERROR is raised by the execution of an operation ... that cannot deliver the correct result" Another point in the design space is Rust: "Integer operators will panic when they overflow when compiled in debug mode." https://doc.rust-lang.org/reference/expressions/operator-expr.html?highligh… The Haskell standard https://www.haskell.org/onlinereport/haskell2010/haskellch6.html#x13-135000… says that "The results of exceptional conditions (such as overflow or underflow) on the fixed-precision numeric types are undefined;" so any program that relies on Int's wrap-around, is already broken. GHC's base library specifies "modulo 2^width" semantics for signed (Data.Int) and unsigned (Data.Word) integer types. Finally, about "fighting" - I guess it's more about strategies to cope with the present situation. - J.W.

2 1

Is Data.Pool not thread-safe or have I done something silly?
by Cody Gman 14 Mar '21

14 Mar '21

I wrote some code with a Data.Pool that has 1 stripe, 1 max resource, and then basically did `Async.replicateConcurrently_ . withResource $ \res -> f res`. I expect withResource to block in each of those threads until the first thread spawned is done with that Resource and releases. To be clear, I get output like: ```shell /tmp $ stack exec --resolver lts-17.5 --package resource-pool --package stm --package hslogger --package time --package concurrent-output -- ghc -O2 -threaded -fwarn-unused-imports testpool.hs && ./testpool acquire 0 - .81428 acquire 1 - .81431 acquire 2 - .81438 acquire 3 - .81440 acquire 4 - .81448 ThreadId 8: processing 1 - .82460 ThreadId 17: processing 4 - .82461 ThreadId 11: processing 2 - .82464 ThreadId 14: processing 3 - .82464 ThreadId 5: processing 0 - .82465 anything else? release 4 - .14427 release 3 - .14430 release 2 - .14431 release 1 - .14431 release 0 - .14432 anything else? anything else? ``` I expect output like: ```shell /tmp $ stack exec --resolver lts-17.5 --package resource-pool --package stm --package hslogger --package time --package concurrent-output -- ghc -O2 -threaded -fwarn-unused-imports testpool.hs && ./testpool [1 of 1] Compiling Main ( testpool.hs, testpool.o ) Linking testpool ... acquire 0 ThreadId 5: processing 0 - .01129 release 0 acquire 1 -- I thought this would have blocked until 0 was released ThreadId 8: processing 1 - .01120 release 1 acquire 2 ThreadId 11: processing 2 - .01123 release 2 acquire 3 ThreadId 14: processing 3 - .01129 release 3 acquire 4 ThreadId 17: processing 4 - .01129 release 4 anything else? anything else? anything else? ``` Here is the code: ``` #!/usr/bin/env stack -- stack script --resolver lts-17.5 --package resource-pool --package stm --package hslogger --package time --package concurrent-output import Control.Concurrent import Control.Concurrent.Async import Control.Concurrent.STM import Data.Pool import Data.Time import System.Console.Concurrent import System.Clock main :: IO () main = do counter <- newTVarIO 0 let acquire = do k <- atomically $ do k <- readTVar counter writeTVar counter (k + 1) return k now <- getTime Monotonic outputConcurrent ("acquire " ++ show k ++ " - "++ show now ++ "\n") return k release k = do now <- getTime Monotonic outputConcurrent ("acquire " ++ show k ++ " - "++ show now ++ "\n") withConcurrentOutput $ do -- create a pool that only allows 1 resource pool <- createPool acquire release 1 500 1 replicateConcurrently_ 5 $ do useResourceFor (seconds 10) pool -- Why do you need these to see the release messages? putStrLn "anything else?" >> threadDelay (seconds 5) putStrLn "anything else?" >> threadDelay (seconds 5) putStrLn "anything else?" >> threadDelay (seconds 5) useResourceFor waitSeconds pool = withResource pool $ \i -> do threadDelay waitSeconds tid <- myThreadId now <- getTime Monotonic outputConcurrent $ show tid <> ": " <> " processing " <> show i <> " - " <> show now <> "\n" seconds = (* 1000000) ``` I'm hoping someone could help explain what I did wrong or confirm it's a bug. Thanks, Cody

2 1

Re: [Haskell-cafe] naturally, length :: a -> Int
by Olaf Klinke 14 Mar '21

14 Mar '21

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. While we're at it: Can there be a Fractional type permitting only positive numbers, as the positive real numbers are closed under division? Can there be a type of rationals between 0 and 1 (which is closed under multiplication) where (/x) is the right adjoint to (*x)? Olaf

4 6

Property checks and PostgreSQL?
by Ignat Insarov 14 Mar '21

14 Mar '21

Note: I posted this [elsewhere] — I hope it is permissible to also post here. There seems to be no single right venue for Haskell questions at this time. [elsewhere]: https://discourse.haskell.org/t/property-checks-and-postgresql/2078 So, I have a three layer stack: PostgreSQL data base, Haskell back end, PureScript front end. And there is a fault somewhere in this stack: sometimes some items present in the data base do not make it to the front end. A small fortune in programmer hours has been spent searching for the fault but there is no progress. No one can figure out even roughly where the fault is. Our last hope is to property check the whole thing, from end to end, verifying that any things put into the data base can be retrieved by the front end. We have in place a fixture that creates a new PostgreSQL cluster, runs the back end on it and makes some scripted HTTP requests. We use it to run unit tests, such like _«if I make this `PUT` request and then that `GET` request to the same end point, I should get back the same thing as I put in»_. In principle it would not be a problem to make a property check out of this property. Practically, tearing down the cluster, building a new pristine one and initializing the data base anew takes seconds, so this way even a trivial property check would take minutes to run. Theoretically, we can carefully reset the data base back to the initial state after each run of the property check. Given that only a small amount of data is modified, it would take imperceptibly little time. But it is hard to know what exactly should be reset, and it is easy to get it wrong if done by hand! One trick we do use is SQL transactions. We take a single connexion, start a transaction, perform a check and then roll the transaction back. Unfortunately, even this is not completely safe: it looks as though sequences used to draw primary keys from are not reset to their previous values! _(Or maybe we are doing something wrong — I am not really a PostgreSQL guru so I am not too sure of myself.)_ But even ignoring this problem _(it is irrelevant to most checks)_, there is another, more severe problem: transactions guarantee that uncommitted data is only visible inside the same connexion. So, there is no way to request it from the HTTP API. This trick is suitable for fuzzing the data base – back end layer, but nothing beyond that. Another thing I heard being done is to selectively drop and re-create exactly the tables affected by a run of a property check. This seems challenging to automate though. How can I know in a general way if a table has been touched? And then, how can I re-create exactly the affected tables given that the data base initialization code is an opaque blob? I wonder if this problem is already solved. In any case, any advice is welcome!

5 9