- Libraries - Haskell.org

Small SMTP client library
by Massimo Zaniboni 14 Dec '03

14 Dec '03

Sorry for my bad english. If you find this limited module useful then you can insert in your project. It is very limited but it could be useful to send email message using insecure plain/login authentication. I have no time to extend it. Regards, Massimo Zaniboni.

1 0

Abstract Data Structures
by Robert Will 10 Dec '03

10 Dec '03

hi, Abstract Data Types plus Data Structures gives Abstract Data Structures. Divide by Conquer is a technique that every Computer Science Undergraduate learns: to sort a listm sort each of its halves then merge. This is what I call "Computational Divide and Conquer" and it is an important algorithm design strategy. Algorithm Design and Programming were once considerd the same and they were the heart of Computer Science. Nowadays, we get our algorithms from libraries and are instead faced with the complexity of programming applications. It is essential that we divide our programs into smaller parts and specify the contracts between those parts. This is what I call "Informational Divide and Conquer". (The words "informational" vs "calculational" go back to Dijkstra who said that a refinement step is good if it makes either "computational or informational progress". The former meaning decrease a loop or recursion variant, the latter meaning to get simpler assertions. I believe that the Art or Science or whatnot of Programming will make a leap forward, when it learns to profit from Dijkstras heritage.) -- Abstraction in Programming firmly rests on the separation -- of Specification and Implementation. Implementations in functional -- languages have always been on a "higher level", many of them -- serving as their own specifications. (Just take a look into -- Haskell's Prelude and try to specify one of the functions.) Due to -- take higher-level of built-in abstraction, they have resisted for a -- long time the use of programmer-defined abstraction, that is, the -- separation of Specification and Implementation. (Which is by the -- way also the basis of well-used polymorphism.) The main data -- structure of the Haskell programmer is still the same as the one -- from the first functional language in 1958. This file might -- convince the reader that a good functional language should have -- more to offer. http://www.stud.tu-ilmenau.de/~robertw/dessy/fun/AlgebraBasic.hs I have always considered it an advantage that almost all of the Prelude functions are there own specifications. If you had a question, just go ask the code. This approach, however, doesn't scale. If we want to include other data structures than functional stacks into a library (or how did you call that structure with the basic axioms head (x : xs) = x tail (x : xs) = xs ?) we must also include abstraction. I don't know why I write such philosphical stuff, actually I only wanted to annouce my little project and ask for comments: http://www.stud.tu-ilmenau.de/~robertw/dessy/fun/ Anyway, it doesn't hurt to remind every now and then. I would especially like interested people to have a look at AlgebraBasic.hs which I think is quite evolved and even already contains the words "standard proposal". Robert

1 0

"Time Warp"
by Graham Klyne 05 Dec '03

05 Dec '03

I spotted this and thought of some recent(ish) discussions... [[ # "Time Warp" New Scientist (11/22/03) Vol. 180, No. 2422, P. 30; Battersby, Stephen An unusual practice of modern timekeeping is the insertion of a "leap second" into the calendar, which occurs less than once a year on average; this is done to make up for the discrepancy between international atomic time and solar time, which is changing as the Earth's rate of rotation gradually slows. Since the inception of leap seconds 30 years ago, coordinated universal time has lagged in fits and starts behind universal atomic time, and the situation has been complicated by the rollout of various technologies that have their own internal time frames. "With [these technologies] becoming more interactive, they need to be synchronized," states the Naval Research Laboratory's Ron Beard. Technologies ranging from GPS to air traffic control systems to the Internet to military systems are not designed with leap seconds in mind, which makes leap second insertion the responsibility of software engineers. This task is difficult enough, but complicating matters is the unpredictable nature of the Earth's slowing spin, a phenomenon attributed to as-yet-indeterminate internal forces; in fact, the planet's rotation has recently sped up, the result being no leap seconds for the last five years. Affected by this is Motorola's Oncore GPS receiver, which will add an extra day to the calendar just after midnight on the morning of Nov. 28, 2003. This extra day will only last for a second, but Motorola GPS product manager Dave Huntingford is worried that this could affect Nov. 28 cell phone bills and lead to lost revenues; Global Timing Services consultant William Klepczynski is even more concerned that such glitches, when combined with worker stress, could lead to disastrous errors in air traffic control. Beard and others think leap seconds should be jettisoned in favor of a continuous timekeeping system called international time, although astronomers fervently oppose this because of their high reliance on software that automates the alignment of their telescopes. ]] -- http://www.acm.org/technews/articles/2003-5/1203w.html#item17 ------------ Graham Klyne For email: http://www.ninebynine.org/#Contact

2 1

"(±¤°í)±â°¡¸·Èù ºÎ¾÷Á¤º¸-½ºÆÄÀÌ´õ(ÀÏ¸í:µ·°Å¹Ì)¹«·áºÐ¾ç@"
by ��з�� 05 Dec '03

05 Dec '03

1 0

Have amazing s\ex up to 20 times per day mfdshi
by Cvs-ghc 03 Dec '03

03 Dec '03

1 0

Data.Graph vs. FGL
by Abraham Egnor 03 Dec '03

03 Dec '03

I'm doing a little cleanup on FGL (http://web.engr.oregonstate.edu/~erwig/fgl/haskell/) Part of that includes renaming modules to fit the hierarchical scheme; my plan was to call the base Data.Graph. Lo and behold, however, Data.Graph already exists! Could someone enlighten me on the reason what is now Data.Graph was chosen over FGL for inclusion into the base library? Also, given that Data.Graph already exists, what would be a good place for FGL? Data.Graph.FGL seems iffy for two reasons - it implies that it's a subset of Data.Graph, and it still uses an acronym instead of a descriptive name. Thanks much! Abe

3 2

RE: How to best add logging/debugging code?
by Simon Marlow 28 Nov '03

28 Nov '03

> Hmm. This suggestion has the potential to confuse the casual user > even more. Consider > > debug s x = trace (s++": "++show x++"\n") x > f x = debug ("entered f") (... g y ...) > g y = debug ("entered g") (...) > main = ... f ... > > With the current definition of 'trace', although the formatting is > ugly, the user (correctly) sees that the evaluation of f is entered > before g. > > entered f: entered g: val_g_y > val_f_x The current implementation of Debug.Trace does indeed do this, but the behaviour can be altered by changing the buffering mode of stderr. Try it: import System.IO import Debug.Trace main = do hSetBuffering stderr (BlockBuffering Nothing) trace ("a" ++ trace "b" "") (return ()) AFAIK, it's been this way for a long time. Cheers, Simon

1 0

Re: How to best add logging/debugging code?
by Ross Paterson 28 Nov '03

28 Nov '03

On Thu, Nov 27, 2003 at 02:19:53PM +0000, Malcolm Wallace wrote: > Apart from changing the evaluation order of expressions, trace has > other drawbacks, noted I think by Lennart(?) but I can't remember > exactly where. One issue is this: > > Consider an expression where the printable argument to 'trace' causes > the evaluation of another expression which itself is defined using > 'trace'. The nested 'trace' outputs will appear interspersed with > each other. In recursive definitions this can become quite painful > to disentangle. Perhaps trace should be redefined: trace msg v = foldr seq (old_trace msg v) msg

2 1

"(±¤°í)´ëÇÑ¹Î±¹-ÀÎÅÍ³ÝÀÇ ÁöÃàÀ» Èçµé°ÔµÉ ¸Å¿ì ³î¶ó¿î»ç½Ç ÇÑ°¡Áö¸¦ ÀüÇÕ´Ï´Ù.@"
by ��ģ�� 23 Nov '03

23 Nov '03

1 0

RE: Treating POSIX signals as exceptions?
by Simon Marlow 17 Nov '03

17 Nov '03

> On Tue, Nov 11, 2003 at 09:39:05AM -0000, Simon Marlow wrote: > > Hmm, there is clearly a choice between having a signal spawn a new > > thread or raise an asynchronous exception, and it's not at all clear > > which is best. > > > > But, since we can implement either using the other one, it doesn't > > really matter too much which one is primitive. > > While either could be implemented in terms of the other, the choice of > which to use for the default handlers will change the behavior of the > program dramatically, simply because it determines whether > bracket cleans up in the presense of signals. So, essentially you're saying that BY DEFAULT, the behaviour should be that certain signals raise an exception (in the main thread only)? It's not clear to me that this should be the default behaviour. If it is, you'd probably want a way to turn it off - the withSignals API doesn't provide a way to turn it off, apart from 'block'. I'd be much happier to just provide the tools for now; if it turns out later that virtually everyone is beginning their programs with 'withSignal sigINT...', then we can possibly make it the default later. > In darcs, I really have no interest in dealing with signals, > I just want to > make sure my temporary files are cleaned up even if the user > hits ^C. If > the default signal handler uses exceptions, this will happen for free. > Currently, my only choice for the former purpose is to use > something like > withSignal. > > I guess the problem is that as far as I can tell, there is no way to > implement a "correct" bracket along with the existing POSIX signal > implementation, since there's no way to find out what the > current handler > is, which means bracket can't install its own handler without > messing up the current handler. Hmm, I don't think I understand this. bracket is just fine as it is; it's just the signal handler which needs to raise an exception, right? Cheers, Simon

2 1