[Haskell-cafe] Proposal for adding finallyError to MonadError
Hi, Café! I have a proposal. I offer to add a new function to the MonadError type class: finallyError :: MonadError e m = m a - m b - m a This is a generalization of the standard finally function that has the following signature: finally: IO a - IO b - IO a Like function finally, the finallyError function runs the first computation and then runs the second computation regardless of whether an error might arise in the first computation. The first computation defines the result. Introducing semantics of a possible cancellation of the m computation, the finallyError function accepts an additional meaning. The second computation must be performed even in case of cancelling the computation, while function catchError must be ignored. It literally means that the finallyError function cannot be expressed in terms of the catchError function. This idea is essentially inspired by the F# async workflow that does allow cancelling the Async computation. I implemented a similar monad Simulation.Aivika.Cont.Cont for my simulation library Aivika [1]. My monad Cont is an extension of the standard monad with the same name. It uses continuations (namely, three different continuations within one computation). It supports the exception handling and it allows cancelling the computation (the liftIO function should also call finallyError to catch the possible cancellation event if my Cont was a generic monad transformer). In the simulation the cancellation of the computation (actually, of the discontinuous process) is a typical use case. Therefore it should be executed as fast as possible. If I raised a specialized IOException for the cancellation event, then it would be a more costly operation. This is a practical point why the finallyError function cannot be expressed in terms of catchError. Another reason is that the catchError function can eat the cancellation exception, which would be absolutely unacceptable. This is a theoretical reason for introducing finallyError. The finallyError function is very useful for guaranteed releasing the external resources (such as the modeled entities in Aivika) in case of raising an exception and even in case of cancelling the active computation, which is important for simulation, for example. Regarding the standard instances of MonadError, the finallyError function seems to be simple for implementing. It is difficult mainly for the Cont-based monads. So, in Aivika it is more difficult than catchError for my Cont monad. The actual name of the function is not important but here it looks consistent with other functions of the MonadError type class. Finally, the finallyError function can be actually defined in an additional type class, which would extend MonadError like that how the ArrowChoice type class extends Arrow. Kind regards, David Sorokin [1] https://github.com/dsorokin/aivika ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] Dual-licensing the package on Hackage
Hi, Cafe! Probably, it was asked before but I could not find an answer with help of Google. I have a library which is hosted on Hackage. The library is licensed under BSD3. It is a very specialized library for a small target group. Now I'm going to relicense it and release a new version already under the dual-license: GPLv3 and commercial. In most cases GPL will be sufficient as this is not a library in common sense. Can I specify the GPL license in the .cabal file, or should I write OtherLicense? I'm going to add the information about dual-licensing in the description section of the .cabal file, though. Thanks, David ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Dual-licensing the package on Hackage
Thanks Thu, I agree with you. Just I don't know what to write in the license field of the .cabal file: GPL or OtherLicense. The both choices seem correct to me and misleading at the same time. Cheers, David 30.07.2013, в 12:53, Vo Minh Thu написал(а): 2013/7/30 David Sorokin david.soro...@gmail.com: Hi, Cafe! Probably, it was asked before but I could not find an answer with help of Google. I have a library which is hosted on Hackage. The library is licensed under BSD3. It is a very specialized library for a small target group. Now I'm going to relicense it and release a new version already under the dual-license: GPLv3 and commercial. In most cases GPL will be sufficient as this is not a library in common sense. Can I specify the GPL license in the .cabal file, or should I write OtherLicense? I'm going to add the information about dual-licensing in the description section of the .cabal file, though. Although you can indeed license your software under different licences, in the case of your question it doesn't seem to be a concern with Hackage: The license displayed on Hackage is the one for the corresponding .cabal file (or at least I think it is). So you issue your new version with the changed license, the new version is available with the new license, the old versions are still available with the old license. Everything is fine. Now about the dual licensing. It seems it is again not a problem with Hackage: you are not granting through Hackage such a commercial license. I guess you provide it upon request (for some money). I.e. when I download your library from Hackage, I receive it under the terms of the BSD (or GPL) license you have chosen, not under a commercial license that I would have to receive through other means. Otherwise the semantic of the license field on Hackage would mean the library is available under such and such licenses, which are not granted to you when you download the library on Hackage. Only when you download the package you can actually find the licensing terms (e.g. in the LICENSE file). But this seems unlikely to me. Cheers, Thu ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Dual-licensing the package on Hackage
I am inclined to use value OtherLicense but state in the description that the package is available either under GPL or a commercial license. The latter must be requested to me. Then there would be no required additional steps to use the package under GPL. Only the LICENSE file must be appropriate. Probably, I will need two files LICENSE and LICENSE-GPLv3. In the former I will have add my copyright and write in a simple form that the license is dual and everyone is free to use the library under GPLv3 (which is the main use case) according the terms provided in the corresponded second file. Thanks, David On 30.07.2013 13:57, Vo Minh Thu wrote: Unless you want to provide multiple open source licenses, I don't see the point: Anybody that needs a commercial license (and has some money) will simply ask for such a commercial license when seeing that the code is available under GPL. Another reason it is pointless is that you will certainly not want to list all the commercial licenses you have used/will use with different clients (there are virtually infinite commercial licenses that you can invent as needs arise: per seat, per core, per year, and so on depending on the clients/projects). I.e. you don't need to state upfront that commercial licences exist (although I understand that you think it is better to advertise your willingness to provide such commercial license, but a comment is enough, the fact is that license is not provided through Hackage). 2013/7/30 Krzysztof Skrzętnicki gte...@gmail.com: Perhaps it would be best if .cabal allowed to have more than one license listed. Another solution would be to use custom field, for example: License: GPL x-Other-License: Commercial, see License-Commercial.txt All best, Krzysztof Skrzętnicki On Tue, Jul 30, 2013 at 11:44 AM, David Sorokin david.soro...@gmail.com wrote: Thanks Thu, I agree with you. Just I don't know what to write in the license field of the .cabal file: GPL or OtherLicense. The both choices seem correct to me and misleading at the same time. Cheers, David 30.07.2013, в 12:53, Vo Minh Thu написал(а): 2013/7/30 David Sorokin david.soro...@gmail.com: Hi, Cafe! Probably, it was asked before but I could not find an answer with help of Google. I have a library which is hosted on Hackage. The library is licensed under BSD3. It is a very specialized library for a small target group. Now I'm going to relicense it and release a new version already under the dual-license: GPLv3 and commercial. In most cases GPL will be sufficient as this is not a library in common sense. Can I specify the GPL license in the .cabal file, or should I write OtherLicense? I'm going to add the information about dual-licensing in the description section of the .cabal file, though. Although you can indeed license your software under different licences, in the case of your question it doesn't seem to be a concern with Hackage: The license displayed on Hackage is the one for the corresponding .cabal file (or at least I think it is). So you issue your new version with the changed license, the new version is available with the new license, the old versions are still available with the old license. Everything is fine. Now about the dual licensing. It seems it is again not a problem with Hackage: you are not granting through Hackage such a commercial license. I guess you provide it upon request (for some money). I.e. when I download your library from Hackage, I receive it under the terms of the BSD (or GPL) license you have chosen, not under a commercial license that I would have to receive through other means. Otherwise the semantic of the license field on Hackage would mean the library is available under such and such licenses, which are not granted to you when you download the library on Hackage. Only when you download the package you can actually find the licensing terms (e.g. in the LICENSE file). But this seems unlikely to me. Cheers, Thu ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Dual-licensing the package on Hackage
This is already another question what license to use :) On 30.07.2013 14:56, Vo Minh Thu wrote: Again I haven't seen a reason to do what you propose: virtually every single GPL library author would gladly accept money for their work to be used in a closed source setting, no need to use OtherLicense to reach that effect. On the other hand, you will stop people interested in open source to look further into your project if they see OtherLicense instead of a well-known open-source license. 2013/7/30 David Sorokin david.soro...@gmail.com: I am inclined to use value OtherLicense but state in the description that the package is available either under GPL or a commercial license. The latter must be requested to me. Then there would be no required additional steps to use the package under GPL. Only the LICENSE file must be appropriate. Probably, I will need two files LICENSE and LICENSE-GPLv3. In the former I will have add my copyright and write in a simple form that the license is dual and everyone is free to use the library under GPLv3 (which is the main use case) according the terms provided in the corresponded second file. Thanks, David On 30.07.2013 13:57, Vo Minh Thu wrote: Unless you want to provide multiple open source licenses, I don't see the point: Anybody that needs a commercial license (and has some money) will simply ask for such a commercial license when seeing that the code is available under GPL. Another reason it is pointless is that you will certainly not want to list all the commercial licenses you have used/will use with different clients (there are virtually infinite commercial licenses that you can invent as needs arise: per seat, per core, per year, and so on depending on the clients/projects). I.e. you don't need to state upfront that commercial licences exist (although I understand that you think it is better to advertise your willingness to provide such commercial license, but a comment is enough, the fact is that license is not provided through Hackage). 2013/7/30 Krzysztof Skrzętnicki gte...@gmail.com: Perhaps it would be best if .cabal allowed to have more than one license listed. Another solution would be to use custom field, for example: License: GPL x-Other-License: Commercial, see License-Commercial.txt All best, Krzysztof Skrzętnicki On Tue, Jul 30, 2013 at 11:44 AM, David Sorokin david.soro...@gmail.com wrote: Thanks Thu, I agree with you. Just I don't know what to write in the license field of the .cabal file: GPL or OtherLicense. The both choices seem correct to me and misleading at the same time. Cheers, David 30.07.2013, в 12:53, Vo Minh Thu написал(а): 2013/7/30 David Sorokin david.soro...@gmail.com: Hi, Cafe! Probably, it was asked before but I could not find an answer with help of Google. I have a library which is hosted on Hackage. The library is licensed under BSD3. It is a very specialized library for a small target group. Now I'm going to relicense it and release a new version already under the dual-license: GPLv3 and commercial. In most cases GPL will be sufficient as this is not a library in common sense. Can I specify the GPL license in the .cabal file, or should I write OtherLicense? I'm going to add the information about dual-licensing in the description section of the .cabal file, though. Although you can indeed license your software under different licences, in the case of your question it doesn't seem to be a concern with Hackage: The license displayed on Hackage is the one for the corresponding .cabal file (or at least I think it is). So you issue your new version with the changed license, the new version is available with the new license, the old versions are still available with the old license. Everything is fine. Now about the dual licensing. It seems it is again not a problem with Hackage: you are not granting through Hackage such a commercial license. I guess you provide it upon request (for some money). I.e. when I download your library from Hackage, I receive it under the terms of the BSD (or GPL) license you have chosen, not under a commercial license that I would have to receive through other means. Otherwise the semantic of the license field on Hackage would mean the library is available under such and such licenses, which are not granted to you when you download the library on Hackage. Only when you download the package you can actually find the licensing terms (e.g. in the LICENSE file). But this seems unlikely to me. Cheers, Thu ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] catching IO errors in a monad transformer stack
Hi!
The exception handling is a difficult thing. It is usually simple enough but
sometimes it can be very difficult, especially when using continuations within
the monadic computation. To feel it, I often remember how the exceptions are
handled in the F# async workflow (the sources are open), but their approach
should be slightly adopted for Haskell what I did in one my simulation library
(as far as I understand, the IO exception cannot arise in a pure value;
therefore IOException should be caught in another place, namely in the liftIO
function).
I'm not sure whether there is a common pattern for handling the exceptions (the
mentioned MonadCatchIO instance contains a warning regarding ContT). Therefore
it is reasonable to allow the programmer himself/herself to define these
handlers through the type class.
Thanks,
David
19.07.2013, в 3:23, Alberto G. Corona написал(а):
Hi Eric:
The pattern may be the MonadCatchIO class:
http://hackage.haskell.org/package/MonadCatchIO-transformers
2013/7/18 Eric Rasmussen ericrasmus...@gmail.com
Hello,
I am writing a small application that uses a monad transformer stack, and I'm
looking for advice on the best way to handle IO errors. Ideally I'd like to
be able to perform an action (such as readFile file_that_does_not_exist),
catch the IOError, and then convert it to a string error in MonadError.
Here's an example of what I'm doing now:
{-# LANGUAGE FlexibleContexts #-}
import Control.Monad.Error
import Control.Monad.State
import System.IO.Error (tryIOError)
catcher :: (MonadIO m, MonadError String m) = IO a - m a
catcher action = do
result - liftIO $ tryIOError action
case result of
Left e - throwError (show e)
Right r - return r
This does work as expected, but I get the nagging feeling that I'm missing an
underlying pattern here. I have tried catch, catchError, and several others,
but (unless I misused them) they don't actually help here. The tryIOError
function from System.IO.Error is the most helpful, but I still have to
manually inspect the result to throwError or return to my underlying monad.
Since this has come up for me a few times now, I welcome any advice or
suggestions on alternative approaches or whether this functionality already
exists somewhere.
Thanks!
Eric
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
--
Alberto.
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
___
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe
[Haskell-cafe] ANNOUNCE: Aivika v0.6, Aivika Experiment v0.3, Aivika Experiment Chart v0.3
Hi! I’m glad to announce new versions of the Aivika [1] simulation library and its additional packages Aivika Experiment [2] and Aivika Experiment Chart [3]. The library is divided to decrease the dependency on GTK, although all three were tested on Linux, Windows and OS X. They allow developing the wide range of simulation models (SD, DES, AM), simulating them and saving the results in form of charts, tables and so on, automating many routine procedures. The architecture is quite flexible. This is a result of my 4-years work since 2009. I finally find the library a quite mature. Among many languages I tried, Haskell was the most suitable. Many simulation concepts can be naturally expressed in terms of Haskell. Many technical details are hidden either under the façade of monads, or similar structures. So, the ordinary differential equations can be elegantly defined using the recursive do-notation. But the discontinuous processes are well expressed using the continuations, if more precisely, three continuations in one computation. I prepared a 55-page PDF document An Introduction to Aivika Simulation Library [4]. For demonstration, it contains some charts prepared with help of Aivika Experiment Chart. There are the deviation charts and histograms. Thanks, David [1] http://hackage.haskell.org/package/aivika [2] http://hackage.haskell.org/package/aivika-experiment [3] http://hackage.haskell.org/package/aivika-experiment-chart [4] https://github.com/dsorokin/aivika/blob/master/doc/aivika.pdf ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Modeling the performance of a non-trivial software app
On 05/11/2011 07:55 AM, C K Kashyap wrote: Hi, I was wondering if it would be a good idea to model a software app's performance using Haskell. The idea is a little abstract in my mind right now. I'll try and illustrate it with an example - Let's say, I want to model a web app - to, model it, I could think of the following entities - 1. One or more clients and as many connections (essentially bandwidth of each connection) 2. A Load balancer ( and bandwidths to the webservers) 3. Bunch of webservers (and bandwidths to the database server) 4. A database server Using the model, I could generate performance characteristics and figure out if database is the bottleneck or not ... if so, what level of sharding would be useful etc Is there already a wheel I am trying to re-invent? Has anyone attempted this? Regards, kashyap ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe Sometimes it is possible to write the corresponded model. Then the model can be simulated to receive the performance characteristics. Much depends on that how precise the model is. The keywords are Discrete Event Simulation (DES) and Theory of Queue. It may require some maths. I wrote a small library called Aivika[1]. Perhaps it might be helpful. David [1]http://hackage.haskell.org/package/aivika ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
Re: [Haskell-cafe] Computational Physics in Haskell
31.03.2011 08:57, Mihai Maruseac пишет: On Thu, Mar 31, 2011 at 1:50 AM, KCkc1...@gmail.com wrote: I'd also like to know of any Haskell programs for theoretical/computational physics. H! Maybe converting such programs to Haskell. On Wed, Mar 30, 2011 at 11:23 AM, Azeem -ul-Hasanaze...@live.com wrote: I started learning Haskell a little while ago. Although I am a novice I am still in love with it. I am physics major and primarily interested in Theoretical Physics and would like to use Haskell in this area. So, I just know to what has been done in this area, are there any libraries for simulating physical process in Haskell etc. Don't know if this is what you're looking for but I found this pages: [1], [2], [3]. Maybe one of them will contain what you're looking for. [1]: http://hackage.haskell.org/packages/archive/pkg-list.html#cat:physics [2]: http://hackage.haskell.org/packages/archive/pkg-list.html#cat:scientific%20simulation [3]: http://hackage.haskell.org/packages/archive/pkg-list.html#cat:simulation I think that Aivika[3] doesn't suit this task, although it can integrate the ordinary differential equations but they are relatively slow as the focus was mainly on the hybrid simulation and DES. As far as I understand, the fields are quite different. David ___ Haskell-Cafe mailing list Haskell-Cafe@haskell.org http://www.haskell.org/mailman/listinfo/haskell-cafe
