December 2005 - Haskell

Re: ECT and SearchPath
by S. Alexander Jacobson 13 Dec '05

13 Dec '05

On Tue, 13 Dec 2005, Sven Moritz Hallberg wrote: > is a good scheme. For it to _really_ work, however, we should form a > registry of "officially assigned" module names (i.e. ones without a > disambiguation suffix) and also one for the author and/or organization > names to be used as suffixes. I for example am known by the nickname > Pesco, or (more recently) by the mnemonic SM at domain KHJK.ORG. My solution to this problem in SearchPath is module maps. SearchPath lets you use multiple module maps simultaneously. So you can for example simultaneously use a global module map as haskell.org, an industry module map at yourassociate.org, an internal company map located on your_intranet, and a personal map located on your computer. Conflicts are handled by the order in which you invoke the maps. So a map is official to the extent your community shares it. Different entities can also compete to provide maps of differing qualities e.g. is the map audited for security issues? does the map point to the repository head of all modules or a static version? etc. Clay Shirky famously observed that in namespace construction there is a pretty direct tradeoff between have names be memorable, global, and non-political. I think shared module maps represents a good compromise. >> I just >> read your ECT article which seems to start with the same assumption and >> takes the theory to its logical conclusion. > > I take that as a compliment. ;) Thank you. It was intended that way! :-) > The main issue that springs to my mind is authenticity control. How do I > know the module I'm downloading is the one I want? Note that this > question is always there, even in the local case, but it is usually not > seen as a problem until module sources are distributed over the Net. I think https URLs solve this problem. You choose module maps from authors you trust and access those maps via https. These "friendly" module maps then point to "friendly" modules that you can also access via https. Since, you choose which CAs you trust to sign server keys, you are safe from MITM attacks. Note, I thought about supporting pgp but there is no obvious way to sign a CVS/SVN/darcs repository and what I like about ECT is that you can point to such repositories rather than specific checked out versions in the comfort that things won't break. I LOVE that in combination with ECT, you get notified that you are using a deprecated version on the next compile without having to check all your imports manually. FYI, the main thing that I would really like, but don't have and can't implement, is for :r in ghci to invoke SearchPath to hunt for missing modules rather than forcing you to quit out and restart ghci every time you add an import of an undownloaded module in your code. Using searcpath make thing very smooth, but this feature would make it completely so. -Alex- ______________________________________________________________ S. Alexander Jacobson tel:917-770-6565 http://alexjacobson.com

1 0

RE: [Haskell] TArray?
by Simon Marlow 13 Dec '05

13 Dec '05

On 13 December 2005 13:11, Sebastian Sylvan wrote: > On 12/13/05, Tomasz Zielonka <tomasz.zielonka(a)gmail.com> wrote: >> On Tue, Dec 13, 2005 at 01:32:46PM +0100, Sebastian Sylvan wrote: >>> Are there plans to include a mutable array for use in the STM monad, >>> or a good reason for why this is not needed? Also, an unboxed >>> version would be nice. >> >> Why not use an (Array idx (TVar a)) ? > > Ah.. Good thinking! > This should work for in the STM monad for anything you would IOArray > for in the IO monad, if you write proper instances for MArray of > course... > > It's a bit inefficient, an extra indirection (compared to a primitive > implementation) but that shouldn't matter too much for the general > case. And the TVars themselves introduces some extra overhead, I would > suppose (which could be per-array rather than per-element, without > losing the property that only writes to the same element causes > retries). > It gets even trickier if you want an unboxed mutable array in the STM > monad. > > But still, it's useful and I think a mutable array for the STM monad > should be in the libraries. It could be implemented as an Array at > first, and perhaps later switch to a more efficient representation. In the past I have used arrays of TVars, as Thomasz suggested. It would indeed be better to have a primitive STM array, the only problem with this is the extra complexity. One simplifying assumption is that it should consider changes at the level of the whole array, rather than per-element (otherwise you'd use an array of TVars). Cheers, Simon

4 4

ETAPS 2007 - Call for Satellite Events
by joost.visser＠di.uminho.pt 13 Dec '05

13 Dec '05

------------------------------------------------------------------- ETAPS 2007: CALL FOR SATELLITE EVENT PROPOSALS deadline: January 20, 2006 ------------------------------------------------------------------- ETAPS 2007 European Joint Conferences on Theory And Practice of Software March 24 - April 1, 2007 Braga, Portugal http://www.di.uminho.pt/etaps07/ -- ABOUT ETAPS -- The European Joint Conferences on Theory and Practice of Software (ETAPS) is the primary European forum for academic and industrial researchers working on topics relating to Software Science. ETAPS is an annual event which takes place in Europe each spring since 1998. The tenth meeting, ETAPS 2007, will take place March 24 till April 1 2007 in Braga, Portugal, hosted by the University of Minho. The main conferences of ETAPS are: - FOSSACS: Foundations of Software Science and Computation Structures - FASE: Fundamental Approaches to Software Engineering - ESOP: European Symposium on Programming - CC: International Conference on Compiler Construction - TACAS: Tools and Algorithms for the Construction and Analysis of Systems These conferences will take place March 26-30, 2007. -- SATELLITE EVENTS -- The ETAPS 2007 Organizing Committee invites proposals for Satellite Events (workshops, tutorials, etc.) that will complement the main ETAPS conferences. They should fall within the scope of ETAPS. This encompasses all aspects of the system development process, including specification, design, implementation, analysis and improvement, as well as the languages, methodologies and tools which support these activities, covering a spectrum from practically-motivated theory to soundly-based practice. Satellite Events provide an opportunity to discuss and report on emerging research approaches and practical experience relevant to theory and practice of software. ETAPS 2007 Satellite Events will be held immediately before and after the main conferences, on March 24-25 and March 31, and April 1, 2007. -- SUBMISSION OF SATELLITE EVENT PROPOSALS -- Researchers and practitioners wishing to organize Satellite Events are invited to submit proposals in ASCII, PDF or Postscript format by e-mail to the Satellite Events Co-chairs, Luis Barbosa and Joost Visser, mailto:joost.visser@di.uminho.pt. A proposal should not exceed two pages and should include: - Satellite Event name / acronym - the names and contact information of the organizers - the preferred period: 24-25 March or 31 March-1 April - the duration of the workshop: one-day or two-day event - 120-word description of the workshop topic for later use in publicity material - a brief explanation of the workshop topic and its relevance to ETAPS - a schedule for paper submission, notification of acceptance and final versions (the latter no later than the early registration deadline of February 12, 2007) - expected number of participants - any other relevant information, like event format, invited speakers, publication policy, demo sessions, special space requirements, etc. The proposals will be evaluated by the ETAPS 2007 organizing committee on the basis of their assessed benefit for prospective participants to ETAPS 2007. The titles and brief information about accepted Satellite Events will be included in the ETAPS 2007 web site, call for papers and call for participation. Satellite Events organizers will be responsible for - producing the event's call for papers and call for participations - publicising the event through specialist mailing lists etc. to complement publicity for ETAPS as a whole - hosting and maintaining a web site for the event - reviewing and making acceptance decisions on submitted papers - producing the event proceedings, if any; facilities for printing will be made available by the ETAPS organizers - scheduling workshop activities in consultation with the local organizers Prospective organizers may wish to consult the web pages of previous satellite events as examples: ETAPS 2006: http://www.complang.tuwien.ac.at/etaps06/ ETAPS 2005: http://www.etaps05.inf.ed.ac.uk/ ETAPS 2004: http://www.lsi.upc.es/etaps04/ ETAPS 2003: http://www.mimuw.edu.pl/etaps03/ -- IMPORTANT DATES -- Satellite Event Proposals Deadline: 20 January 2006 Notification of acceptance: 6 February 2006 -- FURTHER INFORMATION AND ENQUIRIES -- http://www.di.uminho.pt/etaps07/ etaps07(a)di.uminho.pt Luis Barbosa Joost Visser ETAPS 2007 Satellite Events Co-chairs lsb(a)di.uminho.pt, joost.visser(a)di.uminho.pt -- THE HOST CITY -- Braga, capital of the Minho province, is an ancient city in the heart of the green and fertile region known as the Costa Verde. The region is known for its attractiveness in terms of climate, gastronomy, prices, and culture. The region is served by the Oporto international airport, providing direct flights to many major European cities. Braga is known for its barroque churches and splendid 18th century houses. The old city is solemn and antique, but animated with commercial activity and academic life. http://www.cm-braga.com.pt/turismo/turismo.asp --

1 0

HWN this week
by John Goerzen 13 Dec '05

13 Dec '05

Hi everyone, We had a fire at my workplace last night. It is unlikely that I will be able to get HWN out this week. (Sorry about missing it last week as well.) We'll just call the next one the Special Holiday Edition, eh? -- John

1 0

TArray?
by Sebastian Sylvan 13 Dec '05

13 Dec '05

Are there plans to include a mutable array for use in the STM monad, or a good reason for why this is not needed? Also, an unboxed version would be nice. A workaround is IOArray with unsafePerformIO wrapped in the STM monad, but it would be preferable if multiple threads could write to the same array without causing conflicts unless they write to the same index. /S -- Sebastian Sylvan +46(0)736-818655 UIN: 44640862

2 2

Type problem
by Emil Axelsson 13 Dec '05

13 Dec '05

Hello all, Could I please get some guidance with this? I'm working on implementing a simple relational language in Haskell. I'm trying to construct a data type that can represent values and patterns for a small set of supported types. See code below. HasX is a class of types that have an unconstrained value (xVal). Number is a typical member of that class. Val is my value/pattern data type. P represents a primitive value and T2 is used to make structure. X represents the unconstrained value or a wildcard pattern. It can only be used for types in HasX. The problem is the commented line in the value function. I want to use the xVal method to get the value for X. This is only allowed if I add the constraint (HasX a => ). But I don't want value to have that constraint, since then I cannot run it on pairs. Furthermore, it should be safe without the constraint. ex2 shows that we cannot use X to construct values that are not in HasX. Is this just a limitation of the current GATDs, or is it unreasonable of me to expect this to work? Is there any workaround, such as coercing the type of the value function? -- / Emil ******************************** {-# OPTIONS -fglasgow-exts #-} class HasX a where xVal :: a data Number = XN -- Unconstrained | N Int -- Constrained instance HasX Number where xVal = XN data Val a where P :: a -> Val a -- Primitive T2 :: (Val a1, Val a2) -> Val (a1,a2) X :: HasX a => Val a -- Unconstrained value :: Val a -> a -- value X = xVal value (P a) = a value (T2 (a1,a2)) = (value a1, value a2) ex1 :: Val (Number,(Number,Number)) ex1 = T2 (P (N 3), T2 (X, P (N 5))) -- ex2 :: Val (Number,Number) -- ex2 = X

2 4

existential type synonyms
by John Meacham 13 Dec '05

13 Dec '05

I was wondering if this would make sense allow extensential type synonyms, so if you had something like the following, > data Type a = .... you could declare a synonym such as > type AnyType = exists a . Type a so you can create functions such as > areSame :: AnyType -> AnyType -> Bool which would expand to > areSame :: forall a b . Type a -> Type b -> Bool this is opposed to the currently allowed in ghc > type AllTypes = forall a . Type a > oddFunc :: AllTypes -> AllTypes -> Bool which expands to the rank 2 type > oddFunc :: (forall a . Type a) -> (forall b . Type b) -> Bool which means something quite different. if we use 'exists' for existential types, this might be another useful use of said name. John -- John Meacham - ⑆repetae.net⑆john⑈

2 1

Re: ECT and SearchPath
by Sven Moritz Hallberg 12 Dec '05

12 Dec '05

[cc'ing to the Haskell mailing list] S. Alexander Jacobson schrieb: > A while back, I had been arguing on the Haskell mailing list that module > meanings were not and should not be qualified by package names. It cannot be argued that introducing a package system does nothing in terms of disambiguating modules except adding a new top level to the existing hierarchy. Then it must be argued whether this is a good idea or not. My current standpoint is that packages make sense as distributional units (at least in some cases - haven't thought this through), but give no real solution to the problem of managing the module namespace. In the latter respect, I think the only solution is to explicitly partition the namespace by author or organization - the same effect as the Java style of embedding DNS domain names in the module namespace. I find the Java way very ugly, though. But as you surely know, current Haskell practice already is to place a module into the hierarchy in some seemingly appropriate place and suffix it with, say, the author's name or some other characteristic identifier to avoid clashes. I tend to think this is a good scheme. For it to _really_ work, however, we should form a registry of "officially assigned" module names (i.e. ones without a disambiguation suffix) and also one for the author and/or organization names to be used as suffixes. I for example am known by the nickname Pesco, or (more recently) by the mnemonic SM at domain KHJK.ORG. So I could register - Pesco as a (personal) disambiguation suffix or - KHJK as an (organizational) suffix. The latter case would put the management of the KHJK subnamespace under organizational control and I could, in this example, use KHJK.SM for my personal use, or just KHJK (if the organization agrees). > I just > read your ECT article which seems to start with the same assumption and > takes the theory to its logical conclusion. I take that as a compliment. ;) Thank you. > On my end, I decided to implement a tool that shows why packages are not > a good model for versioning and implemented SearchPath > (http://searchpath.org) SearchPath takes a map of module locations on > the Internet and does automatic import chasing for your particular > application. Cool, nice work! I've actually thought about building a system just like this. > In your ECT article you say: > > In the long run, I envision a world where I can stumble across some > module on the Net, immediately import it in my code, and, on the > stroke of a single command, have my system locate, fetch, compile, and > install that module, dropping me at an interactive prompt with the new > import available! All of the above without having to worry about > whether or how to bundle that module with my code, prompted by > questions of its availability and compatibility on my users' hosts. > > I think SearchPath helps get us there. I am soliciting any thoughts you > may have to improve SearchPath and better integrate it with people > implementing ECT in their modules. The main issue that springs to my mind is authenticity control. How do I know the module I'm downloading is the one I want? Note that this question is always there, even in the local case, but it is usually not seen as a problem until module sources are distributed over the Net. The issue of course boils down to establishing trust. Oh, I just remember Raph Levien of advogato.org complaining about his trust metrics not being used! Maybe that would be an idea... - Module authors PGP-sign their code. No problem. - Module importers check their imports for trusted signatures. Hm. - To trust a signature means to a) trust that it belongs to the author it claims to belong to, b) trust that author to produce "friendly" code. Here's where trust metrics could be employed. The PGP web of trust already does it for (a) and (b) would be to do by us. Does the following make sense? Or is there an elegant way to make this not rely on the central registry for authority? - Module importers also check their imports against the registry. - The registry records for each "block" of the module namespace, the author/organization - identified by key - responsible for that block. I don't have time just now to think through these things. But many thanks for your mail, and keep going. I'm sure we can devise some brilliant way to solve all these problems -- in theory ;) . Regards, Sven Moritz

1 0

RE: [Haskell] Re: comment vs varsym
by Simon Marlow 12 Dec '05

12 Dec '05

On 09 December 2005 05:47, Taral wrote: > On 12/8/05, Taral <taralx(a)gmail.com> wrote: >> or b) comment, because the maximal munch rule says so (it matches >> the newline): >> >> comment -> dashes [any<symbol> {any}] newline >> varsym -> (symbol {symbol | :})<reservedop | dashes> > > Aha. Maximal munch only applies to non-whitespace, so Hugs is right. > Yay. This looks like a mistake in the syntax. I think GHC is following the letter of the spec, but I also think that the intended behaviour was to admit --: as a varsym, i.e. Hugs' behaviour. The syntax allows --: to be a comment, because the class of characters "symbol" does not include :. In contrast, the syntax does *not* admit varsyms beginning with '--' as comments. I'm not aware of maximal munch being applied only to non-whitespace; does it say so anywhere in the report? Cheers, Simon

2 1

comment vs varsym
by Taral 10 Dec '05

10 Dec '05

The first token in "--:\n" is... a) varsym, because varsym always wins according to section 2.3: An ordinary comment begins with a sequence of two or more consecutive dashes (e.g. --) and extends to the following newline. The sequence of dashes must not form part of a legal lexeme. For example, "-->" or "|--" do not begin a comment, because both of these are legal lexemes; however "--foo" does start a comment. or b) comment, because the maximal munch rule says so (it matches the newline): comment -> dashes [any<symbol> {any}] newline varsym -> (symbol {symbol | :})<reservedop | dashes> GHC says comment. Hugs says varsym. What's the answer? -- Taral <taralx(a)gmail.com> "Computer science is no more about computers than astronomy is about telescopes." -- Edsger Dijkstra

2 2