- Haskell-Cafe - Haskell.org

help w/ improving custom Stream for parsec
by Maurizio Vitale 09 Apr '15

09 Apr '15

I need a custom stream that supports insertion of include files and expansions of macros. I also want to be able to give nice error messages (think of clang macro-expansion backtrace), so I cannot use the standard trick of concatenating included files and expanded macros to the current input with setInput/getInput (I think I can't maybe there's a way of keeping a more complex "position" and since the use in producing an error backtrac is rare, it migth be worth exploring; if anybody has ideas here, I'm listening) Assuming I need a more compelx stream, this is what I have (Macro and File both have a string argument, but it will be more compicated, a list of expansions for Macro for instance). Is there a better way for doing this? What are the performance implications with backtracking? I'll be benchmarking it, but if people see obvious problems, let me know. Thanks a lot, Maurizio {-# LANGUAGE FlexibleInstances #-} {-# LANGUAGE FlexibleContexts #-} {-# LANGUAGE InstanceSigs #-} {-# LANGUAGE MultiParamTypeClasses #-} module Parsing where import Text.Parsec type Parser s m = ParsecT s () m data VStream = File String | Macro String deriving Show newtype StreamStack = StreamStack [VStream] deriving Show instance (Monad m) ⇒ Stream VStream m Char where uncons ∷ VStream -> m (Maybe (Char, VStream)) uncons (File (a:as)) = return $ Just (a, File as) uncons (File []) = return Nothing uncons (Macro (a:as)) = return $ Just (a, File as) uncons (Macro []) = return Nothing instance (Monad m) => Stream StreamStack m Char where uncons (StreamStack []) = return Nothing uncons (StreamStack (s:ss)) = case uncons s of Nothing → uncons $ StreamStack ss Just Nothing → uncons $ StreamStack ss Just (Just (c, File s')) → return $ Just (c, StreamStack (File s': ss)) Just (Just (c, Macro s')) → return $ Just (c, StreamStack (Macro s':ss))

2 5

VACANCIES : 3x PhD position in Fuctional Programming
by Johan Jeuring 09 Apr '15

09 Apr '15

The research group of Software Technology is part of the Software Systems division of in the department of Information and Computer Science at the Utrecht University. We focus our research on functional programming, compiler construction, tools for learning and teaching (serious games, intelligent tutoring systems), program analysis, validation, and verification. Financed by the Technology Foundation STW, the EU, and Utrecht University we currently have job openings for: ** 3x PhD researcher (PhD student) in Functional Programming ** We are looking for PhD students to develop functional programming techniques related to parsing, rewriting, property-based testing, dependently typed programming, or program analysis, and to apply these techniques in several applications, such as distributed systems, applied games, dialogue management systems, or assessment tools. Besides research, the successful candidate will be expected to help supervise MSc students and assist courses. We prefer candidates to start no later than September 2015. --------------------------------- What we are looking for --------------------------------- The candidate should have an MSc in Computer Science with good grades, be highly motivated to pursue a PhD, and speak and write English well. Knowledge of functional programming, such as Haskell or ML is essential. --------------------------------- What we offer --------------------------------- The candidate is offered a full-time salaried position for four years. The salary is supplemented with a holiday bonus of 8% and an end-of-year bonus of 8,3% per year. In addition we offer: a pension scheme, partially paid parental leave, flexible employment conditions. Conditions are based on the Collective Labour Agreement Dutch Universities. The research group will provide the candidate with necessary support on all aspects of the project. More information is available on the website: Terms and employment: http://bit.ly/1elqpM7 Salary starts at € 2,083.- and increases to € 2,664.- gross per month in the fourth year of the appointment. Utrecht is a great place to live, having been ranked as one of the happiest places in the world, according to BBC travel. Living in Utrecht: http://bitly.com/HdbL0X --------------------------------- In order to apply --------------------------------- To apply please attach a letter of motivation, a curriculum vitae, and (email) addresses of two referees. Make sure to also include a transcript of the courses you have followed (at bachelor and master level), with the grades you obtained, and to include a sample of your scientific writing, such as your master thesis. It is possible to apply for this position if you are close to obtaining your Master's. In that case include a letter of your supervisor with an estimate of your progress, and do not forget to include at least a sample of your technical writing skills. Applications are accepted until the positions are filled. Send your application via email to Johan Jeuring: J.T.Jeuring(a)uu.nl --------------- Contact person --------------- For further information you can direct your inquiries to: Johan Jeuring phone: +31 (0)30 253 4115/ (0) 6 40010053 e-mail: J.T.Jeuring(a)uu.nl website: http://www.staff.science.uu.nl/~jeuri101/

1 0

Re: [Haskell-cafe] indexed writer monad
by Doug McIlroy 09 Apr '15

09 Apr '15

I infer from the following dialog that ghci nontransparently caches knowledge about the contents of the library. This is doubtless helpful for the flurry of imports that a single "load" or interactive "import" may cause. But the time saved by caching from one such top-level action to the next is probably imperceptible. The hours wasted in the present instance and others like it (the very quick response on cafe suggests that it's not unusual) probably outweigh all the milliseconds that the cache normally saves. Non-transparent caching is inherently evil. While the risk may be justifiable in some cases (e.g. stdio buffers in C), it is very questionable at top level in ghci. Doesn't this warrant a ticket? Doug >> I install the indexed package by using cabal >> >> Cabal install indexed >> >> https://hackage.haskell.org/package/indexed >> >> Seemed to do something. >> >> Then I want to use that package... >> >>> import Control.Monad.Indexed >> >> Boom...unknown... > > Possibly silly question but... did you restart `ghci` after installing indexed? Genius Maybe this Haskell lark isnt as hard as I thought....

1 0

Generalized Newtype Deriving not allowed in Safe Haskell
by Omari Norman 09 Apr '15

09 Apr '15

When compiling code with Generalized Newtype Deriving and the -fwarn-unsafe flag, I get -XGeneralizedNewtypeDeriving is not allowed in Safe Haskell This happens both in GHC 7.8 and GHC 7.10. I thought I remembered reading somewhere that GNTD is now part of the safe language? The GHC manual used to state that GNTD is not allowed in Safe Haskell: https://downloads.haskell.org/~ghc/7.6.3/docs/html/users_guide/safe-haskell… But this language on GNTD not being part of the safe language was removed in the 7.8 manual: https://downloads.haskell.org/~ghc/7.8.2/docs/html/users_guide/safe-haskell… The GHC release notes don't say anything about this one way or the other. Thoughts?

2 1

Re: [Haskell-cafe] Distributed and persistent events bus in Haskell
by timmy tofu 08 Apr '15

08 Apr '15

We're working on something with 0MQ, if you (or anyone else reading) go that route and want to compare notes, hit me off-list. On Apr 8, 2015 8:01 AM, <haskell-cafe-request(a)haskell.org> wrote: > Send Haskell-Cafe mailing list submissions to > haskell-cafe(a)haskell.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe > or, via email, send a message with subject or body 'help' to > haskell-cafe-request(a)haskell.org > > You can reach the person managing the list at > haskell-cafe-owner(a)haskell.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Haskell-Cafe digest..." > > > Today's Topics: > > 1. Distributed and persistent events bus in Haskell (Arnaud Bailly) > 2. Re: Distributed and persistent events bus in Haskell > (Kyle Marek-Spartz) > 3. Re: Distributed and persistent events bus in Haskell > (Arnaud Bailly) > 4. Re: MinGHC for GHC 7.10.1 available (Aaron Contorer) > 5. status of deb.haskell.org? (Jeremy) > 6. help w/ improving custom Stream for parsec (Maurizio Vitale) > 7. Anyone interested in taking over network-uri? (Johan Tibell) > 8. Re: Anyone interested in taking over network-uri? (Alexey Shmalko) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 7 Apr 2015 16:50:49 +0200 > From: Arnaud Bailly <arnaud.oqube(a)gmail.com> > To: Haskell Cafe <haskell-cafe(a)haskell.org> > Subject: [Haskell-cafe] Distributed and persistent events bus in > Haskell > Message-ID: > < > CAL4zPaqpKXdOibD3_HECt%2B0V0ZniH0kKpA_ESQ30P7g9Ly_5cg(a)mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > Hello, > > I am implementing an application using event sourcing as primary storage > for data, which implies I need a way to durably and reliably store streams > of events on stable storage. I also need to be able to have an event > distribution system on top of that persistent storage so that components > can subscribe to stored events. > > So far I have implemented a simple store, e.g. a flat file, which reuses > the format of Apache Kafka (just in case...). Not very robust nor > sophisticated but can work for moderate loads. Now I am looking for the > event distribution part in the hope of being able to reuse some distributed > event bus system that might exist somewhere and not having to roll my own. > > I have had a look couple of months ago at Vaultaire, Marquise and friends, > but I am not sure they are really suited to my use case: They seem to be > geared toward very high workload and throughput, like log or huge data > streams analysis. > > Thanks for any pointer you might share, > -- > Arnaud Bailly > > twitter: abailly > skype: arnaud-bailly > linkedin: http://fr.linkedin.com/in/arnaudbailly/ >

1 0

ANN: hledger 0.25
by Simon Michael 08 Apr '15

08 Apr '15

I'm pleased to announce hledger and hledger-web 0.25. This release brings GHC 7.10 compatibility, terminal width awareness, useful averages and totals columns, and a more robust hledger-web add form. Full release notes: http://hledger.org/release-notes#hledger-0.25 . Release contributors: Simon Michael, Julien Moutinho. hledger (http://hledger.org) is a command-line tool and haskell library for tracking financial transactions, which are stored in a human-readable plain text format. It can also read CSV or timelog files. It provides useful reports, and can also help you record new transactions interactively. Add-on commands include hledger-web (a web interface), hledger-irr (for calculating internal rate of return) and hledger-interest (for generating interest transactions). hledger is inspired by and largely compatible with Ledger, and can be used with some Ledger files. Installation: cabal update; cabal install hledger [hledger-web] or see http://hledger.org/download and http://hledger.org/installing <http://hledger.org/installing> for more options. Best! -Simon

1 0

MinGHC for GHC 7.10.1 available
by Michael Snoyman 07 Apr '15

07 Apr '15

I've just uploaded a new release of MinGHC, including GHC 7.10.1 and cabal-install 1.22.2.0. This release can be downloaded from: https://s3.amazonaws.com/download.fpcomplete.com/minghc/minghc-7.10.1-i386.… In the process, I also needed to upload a cabal-install binary for Windows, which is available at: https://s3.amazonaws.com/download.fpcomplete.com/minghc/cabal-install-1.22.… I've tested this distribution, but only lightly. Feedback from others would be useful :)

2 1

Distributed and persistent events bus in Haskell
by Arnaud Bailly 07 Apr '15

07 Apr '15

Hello, I am implementing an application using event sourcing as primary storage for data, which implies I need a way to durably and reliably store streams of events on stable storage. I also need to be able to have an event distribution system on top of that persistent storage so that components can subscribe to stored events. So far I have implemented a simple store, e.g. a flat file, which reuses the format of Apache Kafka (just in case...). Not very robust nor sophisticated but can work for moderate loads. Now I am looking for the event distribution part in the hope of being able to reuse some distributed event bus system that might exist somewhere and not having to roll my own. I have had a look couple of months ago at Vaultaire, Marquise and friends, but I am not sure they are really suited to my use case: They seem to be geared toward very high workload and throughput, like log or huge data streams analysis. Thanks for any pointer you might share, -- Arnaud Bailly twitter: abailly skype: arnaud-bailly linkedin: http://fr.linkedin.com/in/arnaudbailly/

2 2

FFI Ptr Question
by Michael Jones 07 Apr '15

07 Apr '15

I am having trouble figuring out how to pass callbacks to C. Given the definition below it is not clear how to define a function and pass it. I tried: getNumberOfRows :: CInt getNumberOfRows = 1 table <- wxcGridTableCreate getNumberOfRows… but the compiler barks at me: src/MainGui.hs:577:32: Not in scope: data constructor ‘Ptr’ even through I do the imports: import Foreign import Foreign.C There is also a _obj as the first value of wxcGridTableCreate and the callbacks, and I am not sure how to make a this* for it. I would have throughout that wxcGridTableCreate would return a Ptr. The goal is to pass haskell functions into wxcGridTableCreate so the grid can call back to haskell. Can someone show an example of just this one function, how to wrap a Ptr around it, and what to do with _obj? DEFINITION wxcGridTableCreate :: Ptr a -> Ptr b -> Ptr c -> Ptr d -> Ptr e -> Ptr f -> Ptr g -> Ptr h -> Ptr i -> Ptr j -> Ptr k -> Ptr l -> Ptr m -> Ptr n -> Ptr o -> Ptr p -> Ptr q -> IO (WXCGridTable ()) wxcGridTableCreate _obj _EifGetNumberRows _EifGetNumberCols _EifGetValue _EifSetValue _EifIsEmptyCell _EifClear _EifInsertRows _EifAppendRows _EifDeleteRows _EifInsertCols _EifAppendCols _EifDeleteCols _EifSetRowLabelValue _EifSetColLabelValue _EifGetRowLabelValue _EifGetColLabelValue = withObjectResult $ wx_ELJGridTable_Create _obj _EifGetNumberRows _EifGetNumberCols _EifGetValue _EifSetValue _EifIsEmptyCell _EifClear _EifInsertRows _EifAppendRows _EifDeleteRows _EifInsertCols _EifAppendCols _EifDeleteCols _EifSetRowLabelValue _EifSetColLabelValue _EifGetRowLabelValue _EifGetColLabelValue foreign import ccall "ELJGridTable_Create" wx_ELJGridTable_Create :: Ptr a -> Ptr b -> Ptr c -> Ptr d -> Ptr e -> Ptr f -> Ptr g -> Ptr h -> Ptr i -> Ptr j -> Ptr k -> Ptr l -> Ptr m -> Ptr n -> Ptr o -> Ptr p -> Ptr q -> IO (Ptr (TWXCGridTable ())) typedef int _cdecl (*TGridGetInt)(void* _obj); typedef int _cdecl (*TGridIsEmpty)(void* _obj, int row, int col); typedef void* _cdecl (*TGridGetValue)(void* _obj, int row, int col); typedef void _cdecl (*TGridSetValue)(void* _obj, int row, int col, void* val); typedef void _cdecl (*TGridClear)(void* _obj); typedef int _cdecl (*TGridModify)(void* _obj, int pos, int num); typedef int _cdecl (*TGridMultiModify)(void* _obj, int num); typedef void _cdecl (*TGridSetLabel)(void* _obj, int idx, void* val); typedef void* _cdecl (*TGridGetLabel)(void* _obj, int idx); EWXWEXPORT(void*,ELJGridTable_Create)(void* self,void* _EifGetNumberRows,void* _EifGetNumberCols,void* _EifGetValue,void* _EifSetValue,void* _EifIsEmptyCell,void* _EifClear,void* _EifInsertRows,void* _EifAppendRows,void* _EifDeleteRows,void* _EifInsertCols,void* _EifAppendCols,void* _EifDeleteCols,void* _EifSetRowLabelValue,void* _EifSetColLabelValue,void* _EifGetRowLabelValue,void* _EifGetColLabelValue) { return (void*)new ELJGridTable (self,

3 6

[haskell.org Google Summer of Code] Call for Mentors
by Edward Kmett 07 Apr '15

07 Apr '15

We have had a rather large pool of potential students apply for this year's Google Summer of Code, but, ultimately, Google won't let us ask for a slot unless we have a potential mentor assigned in advance. On top of that, one thing we try to do with each project is wherever possible, assign both a primary and a backup mentor, so the available mentoring pool is drawn a little thin. Many hands make for light work, though: If you've mentored or thought about mentoring in years past, I'd encourage you to sign up on google-melange for the Google Summer of Code at: https://www.google-melange.com/gsoc/homepage/google/gsoc2015 and request a connection to haskell.org as a Mentor. Once you've done this you can help us vote on proposals, and should something seem appropriate to you, you can flag yourself as available as a potential mentor or backup mentor for one (or more) of the projects. We have a couple of weeks left to rate proposals and request slots, but it'd be good to make as much progress as we can this week. If you have any questions, feel free to reach out to me, or to Shachaf Ben-Kiki or Gershom Bazerman who have been helping out with organizational issues this year. We also have a #haskell-gsoc channel on irc.freenode.net if you have questions about what is involved. Thank you for your time and consideration, -Edward Kmett

1 0