From apfelmus at quantentunnel.de Fri May 1 04:07:37 2009 From: apfelmus at quantentunnel.de (Heinrich Apfelmus) Date: Fri May 1 03:53:23 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: <49F97E3F.5050400@theingots.org> References: <49F8D08E.208@theingots.org> <20090429223325.GD32272@kira.casa> <49F8D776.9000004@theingots.org> <20090430011525.GA12598@kira.casa> <49F964EC.4060308@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> Message-ID: Daniel Carrera wrote: > Thomas Davie wrote: >>> http://wxhaskell.sourceforge.net/images/controls-macosx2.png >> >> My link was on leopard actually, but that still looks significantly >> different to the natively designed UI. > > Significantly different? They are practically identical. IMHO, they are quite different, and it shows. :) The first difference is the spacing. In the wxHaskell version, the enclosing tab is glued to the window border left and right, while there is a considerable margin in the native version. The inner boxes have spacious margins, too. The second is the capitalization in the native version. That gives it a totally different look! I even thought that they were using different fonts at first. Thirdly, the textlogs have different fonts. Again, the wxHaskell one looks non-standard. Overall, I think it's quite visible that the wxHaskell version is not native. What I don't understand is why, though; after all, most of the above is specified in Apple's Human Interface Guidelines, it's just a matter of implementing them. Regards, apfelmus -- http://apfelmus.nfshost.com From daniel.carrera at theingots.org Fri May 1 05:26:44 2009 From: daniel.carrera at theingots.org (Daniel Carrera) Date: Fri May 1 05:12:25 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: References: <49F8D08E.208@theingots.org> <20090429223325.GD32272@kira.casa> <49F8D776.9000004@theingots.org> <20090430011525.GA12598@kira.casa> <49F964EC.4060308@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> Message-ID: <49FAC054.5030808@theingots.org> Heinrich Apfelmus wrote: > The first difference is the spacing. In the wxHaskell version, the > enclosing tab is glued to the window border left and right, while there > is a considerable margin in the native version. The inner boxes have > spacious margins, too. > > The second is the capitalization in the native version. That gives it a > totally different look! I even thought that they were using different > fonts at first. I didn't notice any of these except for the tab margin until you pointed it out. The tab margin left and right is the only thing I noticed. Taking a second look, I also notice the margin *below* the tab box. > Thirdly, the textlogs have different fonts. Again, the wxHaskell one > looks non-standard. Honestly, I can't tell. They are both very similar, sans fonts. Also, the font could be explained by the fact that these screen shots are taken on different computers (different OS versions, different settings). I just went to the wxWidgets website. On that site, the Mac OS screen shots show the buttons capitalized. In any case, I think these differences are minor. More importantly, I'm not sure that a different cross-platform toolkit would do a better job. Daniel. From bugfact at gmail.com Fri May 1 10:08:07 2009 From: bugfact at gmail.com (Peter Verswyvelen) Date: Fri May 1 09:53:47 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: <49FAC054.5030808@theingots.org> References: <49F8D08E.208@theingots.org> <20090430011525.GA12598@kira.casa> <49F964EC.4060308@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> <49FAC054.5030808@theingots.org> Message-ID: I thought wxWidgets was actually one of the only toolkits that actually *used* the native GUI libraries??? See e.g. http://wiki.wxwidgets.org/WxWidgets_Compared_To_Other_Toolkits So I don't understand how come the look and feel is different then... On Fri, May 1, 2009 at 11:26 AM, Daniel Carrera < daniel.carrera@theingots.org> wrote: > Heinrich Apfelmus wrote: > >> The first difference is the spacing. In the wxHaskell version, the >> enclosing tab is glued to the window border left and right, while there >> is a considerable margin in the native version. The inner boxes have >> spacious margins, too. >> >> The second is the capitalization in the native version. That gives it a >> totally different look! I even thought that they were using different >> fonts at first. >> > > I didn't notice any of these except for the tab margin until you pointed > it out. The tab margin left and right is the only thing I noticed. > Taking a second look, I also notice the margin *below* the tab box. > > > Thirdly, the textlogs have different fonts. Again, the wxHaskell one >> looks non-standard. >> > > Honestly, I can't tell. They are both very similar, sans fonts. Also, the > font could be explained by the fact that these screen shots are taken on > different computers (different OS versions, different settings). > > I just went to the wxWidgets website. On that site, the Mac OS screen > shots show the buttons capitalized. In any case, I think these differences > are minor. More importantly, I'm not sure that a different cross-platform > toolkit would do a better job. > > Daniel. > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090501/defc7f19/attachment.htm From stevech1097 at yahoo.com.au Fri May 1 10:19:35 2009 From: stevech1097 at yahoo.com.au (Steve) Date: Fri May 1 10:02:42 2009 Subject: [Haskell-beginners] gcd In-Reply-To: <20090430021128.57C723241F6@www.haskell.org> References: <20090430021128.57C723241F6@www.haskell.org> Message-ID: <1241187575.5168.1.camel@host.localdomain> I had a look at the gcd definition in GHC 6.10.1 ghc-6.10.1/libraries/base/GHC/Real.lhs -- | @'gcd' x y@ is the greatest (positive) integer that divides both @x@ -- and @y@; for example @'gcd' (-3) 6@ = @3@, @'gcd' (-3) (-6)@ = @3@, -- @'gcd' 0 4@ = @4@. @'gcd' 0 0@ raises a runtime error. gcd :: (Integral a) => a -> a -> a gcd 0 0 = error "Prelude.gcd: gcd 0 0 is undefined" gcd x y = gcd' (abs x) (abs y) where gcd' a 0 = a gcd' a b = gcd' b (a `rem` b) Why is gcd 0 0 undefined? http://en.wikipedia.org/wiki/Greatest_common_divisor says: "It is useful to define gcd(0, 0) = 0 and lcm(0, 0) = 0 because then the natural numbers become a complete distributive lattice with gcd as meet and lcm as join operation. This extension of the definition is also compatible with the generalization for commutative rings given below." An added advantage, for haskell, of defining gcd 0 0 = 0 is that gcd would change from being a partial function to a total function. Regards, Steve From apfelmus at quantentunnel.de Fri May 1 11:25:53 2009 From: apfelmus at quantentunnel.de (Heinrich Apfelmus) Date: Fri May 1 11:11:33 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: References: <49F8D08E.208@theingots.org> <20090430011525.GA12598@kira.casa> <49F964EC.4060308@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> <49FAC054.5030808@theingots.org> Message-ID: Peter Verswyvelen wrote: > I thought wxWidgets was actually one of the only toolkits that actually > *used* the native GUI libraries??? > See e.g. http://wiki.wxwidgets.org/WxWidgets_Compared_To_Other_Toolkits > > So I don't understand how come the look and feel is different then... Using native GUI libraries is necessary, but not sufficient for achieving the same look and feel. :) Other elements are layout and spacing, http://developer.apple.decenturl.com/human-interface-guidelines-layout http://developer.apple.decenturl.com/hi-guidelines-windows drag & drop, http://developer.apple.decenturl.com/hi-guidelines-dragndrop writing style and fonts http://developer.apple.decenturl.com/human-interface-guidelines-text etc, etc. Daniel Carrera wrote: >> Thirdly, the textlogs have different fonts. Again, the wxHaskell one >> looks non-standard. >> > Honestly, I can't tell. They are both very similar, sans fonts. Also, the > font could be explained by the fact that these screen shots are taken on > different computers (different OS versions, different settings). There is a default set of fonts http://developer.apple.decenturl.com/human-interface-guidelines-text The "mini system font" is the typography of choice for text boxes, but it looks the "application font" is acceptable as well. > In any case, I think these differences are minor. I don't. :) > More importantly, I'm not sure that a different cross-platform > toolkit would do a better job. True, that. It's just that this doesn't mean that wxWidgets does a very good job. Regards, apfelmus -- http://apfelmus.nfshost.com From tom.davie at gmail.com Fri May 1 11:55:02 2009 From: tom.davie at gmail.com (Thomas Davie) Date: Fri May 1 11:40:43 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: References: <49F8D08E.208@theingots.org> <20090430011525.GA12598@kira.casa> <49F964EC.4060308@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> <49FAC054.5030808@theingots.org> Message-ID: <1D248BD4-7A46-4B14-820C-45B4557F0A84@gmail.com> This is one of many reasons why I don't believe any of these cross- platform toolkits will ever be good enough to make truly native-like UIs ? The bottom line is that tab views are different between Mac OS and other OSes, they have much more padding round them on Mac OS. This means that any layout you create involving them will never work on both platforms. Bob On 1 May 2009, at 17:25, Heinrich Apfelmus wrote: > Peter Verswyvelen wrote: >> I thought wxWidgets was actually one of the only toolkits that >> actually >> *used* the native GUI libraries??? >> See e.g. http://wiki.wxwidgets.org/WxWidgets_Compared_To_Other_Toolkits >> >> So I don't understand how come the look and feel is different then... > > Using native GUI libraries is necessary, but not sufficient for > achieving the same look and feel. :) > > Other elements are layout and spacing, > > http://developer.apple.decenturl.com/human-interface-guidelines- > layout > http://developer.apple.decenturl.com/hi-guidelines-windows > > drag & drop, > > http://developer.apple.decenturl.com/hi-guidelines-dragndrop > > writing style and fonts > > http://developer.apple.decenturl.com/human-interface-guidelines-text > > etc, etc. > > > Daniel Carrera wrote: >>> Thirdly, the textlogs have different fonts. Again, the wxHaskell one >>> looks non-standard. >>> >> Honestly, I can't tell. They are both very similar, sans fonts. >> Also, the >> font could be explained by the fact that these screen shots are >> taken on >> different computers (different OS versions, different settings). > > There is a default set of fonts > > http://developer.apple.decenturl.com/human-interface-guidelines-text > > The "mini system font" is the typography of choice for text boxes, but > it looks the "application font" is acceptable as well. > >> In any case, I think these differences are minor. > > I don't. :) > >> More importantly, I'm not sure that a different cross-platform >> toolkit would do a better job. > > True, that. It's just that this doesn't mean that wxWidgets does a > very > good job. > > > Regards, > apfelmus > > -- > http://apfelmus.nfshost.com > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners From daniel.carrera at theingots.org Fri May 1 16:16:43 2009 From: daniel.carrera at theingots.org (Daniel Carrera) Date: Fri May 1 16:02:21 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: References: <49F8D08E.208@theingots.org> <20090430011525.GA12598@kira.casa> <49F964EC.4060308@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> <49FAC054.5030808@theingots.org> Message-ID: <49FB58AB.4090503@theingots.org> Peter Verswyvelen wrote: > I thought wxWidgets was actually one of the only toolkits that actually > *used* the native GUI libraries??? It does, and while I'm no expert on the subject, my understanding is that wxWidgets does the best job at matching the native UI. So far it seems that the things wx doesn't get right are small. The most significant difference I've seen so far is capitalization, and I'm not sure that this is an issue with wxWidgets per se, because the other screen shots on the wxWidgets site have the correct capitalization. Daniel. From daniel.carrera at theingots.org Fri May 1 16:20:50 2009 From: daniel.carrera at theingots.org (Daniel Carrera) Date: Fri May 1 16:06:29 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: References: <49F8D08E.208@theingots.org> <20090430011525.GA12598@kira.casa> <49F964EC.4060308@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> <49FAC054.5030808@theingots.org> Message-ID: <49FB59A2.9010901@theingots.org> Heinrich Apfelmus wrote: >> More importantly, I'm not sure that a different cross-platform >> toolkit would do a better job. > > True, that. It's just that this doesn't mean that wxWidgets does a very > good job. Well, "very good job" is a really fuzzy term. Based on the very limited set of screen shots that I have seen, I think WX does well enough for anyone but a pedant (and you disagree, I know). Anyways, to bring back some context into the discussion. We were talking about whether WX is better than X11+Gtk on Mac. I am confident that it is. Daniel. From daniel.carrera at theingots.org Fri May 1 16:28:12 2009 From: daniel.carrera at theingots.org (Daniel Carrera) Date: Fri May 1 16:13:50 2009 Subject: [Haskell-beginners] gcd In-Reply-To: <1241187575.5168.1.camel@host.localdomain> References: <20090430021128.57C723241F6@www.haskell.org> <1241187575.5168.1.camel@host.localdomain> Message-ID: <49FB5B5C.1090501@theingots.org> I'm just another Haskell beginner, but what Steve says makes sense. Perhaps something like this: gcd a b | b == 0 = a | b < 0 = gcd (-b) a | otherwise = gcd b (a `rem` b) Daniel. Steve wrote: > I had a look at the gcd definition in GHC 6.10.1 > ghc-6.10.1/libraries/base/GHC/Real.lhs > > -- | @'gcd' x y@ is the greatest (positive) integer that divides both > @x@ > -- and @y@; for example @'gcd' (-3) 6@ = @3@, @'gcd' (-3) (-6)@ = @3@, > -- @'gcd' 0 4@ = @4@. @'gcd' 0 0@ raises a runtime error. > gcd :: (Integral a) => a -> a -> a > gcd 0 0 = error "Prelude.gcd: gcd 0 0 is undefined" > gcd x y = gcd' (abs x) (abs y) > where gcd' a 0 = a > gcd' a b = gcd' b (a `rem` b) > > Why is gcd 0 0 undefined? > > http://en.wikipedia.org/wiki/Greatest_common_divisor says: > "It is useful to define gcd(0, 0) = 0 and lcm(0, 0) = 0 because then the > natural numbers become a complete distributive lattice with gcd as meet > and lcm as join operation. This extension of the definition is also > compatible with the generalization for commutative rings given below." > > An added advantage, for haskell, of defining gcd 0 0 = 0 is that gcd > would change from being a partial function to a total function. > > Regards, > Steve > > > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > From daniel.is.fischer at web.de Fri May 1 16:44:52 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Fri May 1 16:30:54 2009 Subject: [Haskell-beginners] gcd In-Reply-To: <49FB5B5C.1090501@theingots.org> References: <20090430021128.57C723241F6@www.haskell.org> <1241187575.5168.1.camel@host.localdomain> <49FB5B5C.1090501@theingots.org> Message-ID: <200905012244.52462.daniel.is.fischer@web.de> Am Freitag 01 Mai 2009 22:28:12 schrieb Daniel Carrera: > I'm just another Haskell beginner, but what Steve says makes sense. Yes, it's something I've often been annoyed by. It's a wart (rather an error) in the Haskell report http://haskell.org/onlinereport/basic.html#sect6.4.2 where it's specified that gcd 0 0 shall raise a runtime error. I hope that gets fixed in the next standard. > Perhaps something like this: > > gcd a b | b == 0 = a > > | b < 0 = gcd (-b) a > | otherwise = gcd b (a `rem` b) > > Daniel. No, that would check for negative numbers in each iteration. Just remove the line gcd 0 0 = error "Prelude.gcd: gcd 0 0 is undefined" and everything's fine. > > Steve wrote: > > I had a look at the gcd definition in GHC 6.10.1 > > ghc-6.10.1/libraries/base/GHC/Real.lhs > > > > -- | @'gcd' x y@ is the greatest (positive) integer that divides both > > @x@ > > -- and @y@; for example @'gcd' (-3) 6@ = @3@, @'gcd' (-3) (-6)@ = @3@, > > -- @'gcd' 0 4@ = @4@. @'gcd' 0 0@ raises a runtime error. > > gcd :: (Integral a) => a -> a -> a > > gcd 0 0 = error "Prelude.gcd: gcd 0 0 is undefined" > > gcd x y = gcd' (abs x) (abs y) > > where gcd' a 0 = a > > gcd' a b = gcd' b (a `rem` b) > > > > Why is gcd 0 0 undefined? > > > > http://en.wikipedia.org/wiki/Greatest_common_divisor says: > > "It is useful to define gcd(0, 0) = 0 and lcm(0, 0) = 0 because then the > > natural numbers become a complete distributive lattice with gcd as meet > > and lcm as join operation. This extension of the definition is also > > compatible with the generalization for commutative rings given below." > > > > An added advantage, for haskell, of defining gcd 0 0 = 0 is that gcd > > would change from being a partial function to a total function. > > > > Regards, > > Steve From bugfact at gmail.com Fri May 1 16:46:39 2009 From: bugfact at gmail.com (Peter Verswyvelen) Date: Fri May 1 16:32:17 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: <1D248BD4-7A46-4B14-820C-45B4557F0A84@gmail.com> References: <49F8D08E.208@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> <49FAC054.5030808@theingots.org> <1D248BD4-7A46-4B14-820C-45B4557F0A84@gmail.com> Message-ID: Well, cross-platform toolkits could work if they don't provide widgets as first class objects :-) So never allow specifying the GUI widgets and layout directly, only indirectly. So you basically provide just the model and some hints/constraints and let the GUI toolkit automatically generate native look&feel widgets to present and edit the model. This means the GUI toolkit needs to understand and implement the guidelines on each platform (how many GUI developers actually follow these guidelines 100%?) If you provide constraints that are conflicting between GUI platforms, you have to either relax the constraints or provide different widgets for each platform. All nice theoretically, but not done yet? On Fri, May 1, 2009 at 5:55 PM, Thomas Davie wrote: > This is one of many reasons why I don't believe any of these cross-platform > toolkits will ever be good enough to make truly native-like UIs ? The bottom > line is that tab views are different between Mac OS and other OSes, they > have much more padding round them on Mac OS. This means that any layout you > create involving them will never work on both platforms. > > Bob > > > On 1 May 2009, at 17:25, Heinrich Apfelmus wrote: > > Peter Verswyvelen wrote: >> >>> I thought wxWidgets was actually one of the only toolkits that actually >>> *used* the native GUI libraries??? >>> See e.g. http://wiki.wxwidgets.org/WxWidgets_Compared_To_Other_Toolkits >>> >>> So I don't understand how come the look and feel is different then... >>> >> >> Using native GUI libraries is necessary, but not sufficient for >> achieving the same look and feel. :) >> >> Other elements are layout and spacing, >> >> http://developer.apple.decenturl.com/human-interface-guidelines-layout >> http://developer.apple.decenturl.com/hi-guidelines-windows >> >> drag & drop, >> >> http://developer.apple.decenturl.com/hi-guidelines-dragndrop >> >> writing style and fonts >> >> http://developer.apple.decenturl.com/human-interface-guidelines-text >> >> etc, etc. >> >> >> Daniel Carrera wrote: >> >>> Thirdly, the textlogs have different fonts. Again, the wxHaskell one >>>> looks non-standard. >>>> >>>> Honestly, I can't tell. They are both very similar, sans fonts. Also, >>> the >>> font could be explained by the fact that these screen shots are taken on >>> different computers (different OS versions, different settings). >>> >> >> There is a default set of fonts >> >> http://developer.apple.decenturl.com/human-interface-guidelines-text >> >> The "mini system font" is the typography of choice for text boxes, but >> it looks the "application font" is acceptable as well. >> >> In any case, I think these differences are minor. >>> >> >> I don't. :) >> >> More importantly, I'm not sure that a different cross-platform >>> toolkit would do a better job. >>> >> >> True, that. It's just that this doesn't mean that wxWidgets does a very >> good job. >> >> >> Regards, >> apfelmus >> >> -- >> http://apfelmus.nfshost.com >> >> _______________________________________________ >> Beginners mailing list >> Beginners@haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090501/9e66d037/attachment.htm From stevech1097 at yahoo.com.au Fri May 1 22:08:16 2009 From: stevech1097 at yahoo.com.au (Steve) Date: Fri May 1 21:51:17 2009 Subject: [Haskell-beginners] gcd In-Reply-To: References: <20090430021128.57C723241F6@www.haskell.org> <1241187575.5168.1.camel@host.localdomain> Message-ID: <1241230096.3721.6.camel@host.localdomain> On Fri, 2009-05-01 at 16:19 +0200, Peter Verswyvelen wrote: > I suggest you forward this to the Haskell Cafe, since it's not really > a beginners question, it's a really good question :) Thanks for the advice, I've reposted the question to the Haskell Cafe. Steve > On Fri, May 1, 2009 at 4:19 PM, Steve > wrote: > I had a look at the gcd definition in GHC 6.10.1 > ghc-6.10.1/libraries/base/GHC/Real.lhs > > -- | @'gcd' x y@ is the greatest (positive) integer that > divides both > @x@ > -- and @y@; for example @'gcd' (-3) 6@ = @3@, @'gcd' (-3) > (-6)@ = @3@, > -- @'gcd' 0 4@ = @4@. @'gcd' 0 0@ raises a runtime error. > gcd :: (Integral a) => a -> a -> a > gcd 0 0 = error "Prelude.gcd: gcd 0 0 is undefined" > gcd x y = gcd' (abs x) (abs y) > where gcd' a 0 = a > gcd' a b = gcd' b (a `rem` b) > > Why is gcd 0 0 undefined? > > http://en.wikipedia.org/wiki/Greatest_common_divisor says: > "It is useful to define gcd(0, 0) = 0 and lcm(0, 0) = 0 > because then the > natural numbers become a complete distributive lattice with > gcd as meet > and lcm as join operation. This extension of the definition is > also > compatible with the generalization for commutative rings given > below." > > An added advantage, for haskell, of defining gcd 0 0 = 0 is > that gcd > would change from being a partial function to a total > function. > > Regards, > Steve > > > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > > From daniel.is.fischer at web.de Fri May 1 22:13:47 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Fri May 1 22:00:51 2009 Subject: [Haskell-beginners] gcd In-Reply-To: <1241230096.3721.6.camel@host.localdomain> References: <20090430021128.57C723241F6@www.haskell.org> <1241230096.3721.6.camel@host.localdomain> Message-ID: <200905020413.48032.daniel.is.fischer@web.de> Am Samstag 02 Mai 2009 04:08:16 schrieb Steve: > On Fri, 2009-05-01 at 16:19 +0200, Peter Verswyvelen wrote: > > I suggest you forward this to the Haskell Cafe, since it's not really > > a beginners question, it's a really good question :) I know that wasn't meant as it was written, but ouch. From apfelmus at quantentunnel.de Sat May 2 03:39:03 2009 From: apfelmus at quantentunnel.de (Heinrich Apfelmus) Date: Sat May 2 03:24:54 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: <49FB59A2.9010901@theingots.org> References: <49F8D08E.208@theingots.org> <20090430011525.GA12598@kira.casa> <49F964EC.4060308@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> <49FAC054.5030808@theingots.org> <49FB59A2.9010901@theingots.org> Message-ID: Daniel Carrera wrote: > Anyways, to bring back some context into the discussion. We were talking > about whether WX is better than X11+Gtk on Mac. I am confident that it is. I concur. What has put me off of cross-platform GUIs so far is probably not even the appearance, but rather that the programs I've seen don't work correctly: weird behavior, drawing bugs when updating the screen, sluggishness, and crashes. The Gtk+X11 ones (Gimp, Inkscape) usually work better than those who try to use a native toolkit. Hopefully, Haskell easily avoids these kinds of problems. :) Regards, apfelmus -- http://apfelmus.nfshost.com From tom.davie at gmail.com Sat May 2 04:22:32 2009 From: tom.davie at gmail.com (Thomas Davie) Date: Sat May 2 04:10:22 2009 Subject: [Haskell-beginners] Re: Selecting a GUI toolkit In-Reply-To: <49FB59A2.9010901@theingots.org> References: <49F8D08E.208@theingots.org> <20090430011525.GA12598@kira.casa> <49F964EC.4060308@theingots.org> <49F96B88.6060709@theingots.org> <49F97E3F.5050400@theingots.org> <49FAC054.5030808@theingots.org> <49FB59A2.9010901@theingots.org> Message-ID: On 1 May 2009, at 22:20, Daniel Carrera wrote: > Heinrich Apfelmus wrote: >>> More importantly, I'm not sure that a different cross-platform >>> toolkit would do a better job. >> True, that. It's just that this doesn't mean that wxWidgets does a >> very >> good job. > > Well, "very good job" is a really fuzzy term. Based on the very > limited set of screen shots that I have seen, I think WX does well > enough for anyone but a pedant (and you disagree, I know). Unfortunately for you, many Mac users are pedants - they expect everything to be exactly in place, perfectly. Bob From keithshep at gmail.com Sat May 2 21:27:18 2009 From: keithshep at gmail.com (Keith Sheppard) Date: Sat May 2 21:15:24 2009 Subject: [Haskell-beginners] Parsec tutorial Message-ID: <92e42b740905021827g2c8f6732hca32b2fe78264e05@mail.gmail.com> Hi Patrick, This isn't what you asked for but it may be useful to you. I've created a parsec based SQL parser for my "TxtSushi" command-line program. It's still pretty early in development but it does parse a subset of the SELECT grammar. The project's page is http://keithsheppard.name/txt-sushi/index.html Keith From daniel.carrera at theingots.org Sun May 3 12:06:33 2009 From: daniel.carrera at theingots.org (Daniel Carrera) Date: Sun May 3 11:52:37 2009 Subject: [Haskell-beginners] Haskell vs Clean (speed) Message-ID: <49FDC109.4030302@theingots.org> Hi, I can't subscribe to Haskell-Cafe, so I'll post here. I found something interesting. "General wisdom" is that Clean (or OCaml) is faster than Haskell. The claim is often followed by a link to the Debian shootout. But on closer inspection, I question this conclusion. The Debian shoot out actually has four sets of benchmarks: 1) Intel 32-bit one core. 2) Intel 32-bit quad-core. 3) Intel 64-bit one core. 4) Intel 64-bit quad-core. It turns out that Clean is only faster for (1). For the others, Haskell is faster. Here I compare Haskell, Clean, OCaml, Lisp SBCL, C# Mono and Fortran because they are all in the same ball mark: 32-bit sing core [1]: Lisp, Fortran, Clean, Haskell, C# Mono. 32-bit quad-core [2]: Haskell, C# Mono, Lisp, Clean, Fortran. 64-bit sing core [3]: Fortran, OCaml, Haskell, Clean, C# Mono, Lisp. 64-bit quad-core [4]: Haskell, OCaml, Lisp, C# Mono, Fortran, Clean. Notes: * The order is "fast language first". * There are no results for OCaml for 32-bit. * "Lisp" is "List SCBL" whatever that is. Tentative conclusions: 1) Haskell makes very good use of multiple cores. It smokes Clean. 2) For single core, they are neck and neck. Whether Clean is faster depends non the architecture. What do you think? Daniel. P.S. [1]http://shootout.alioth.debian.org/u32/benchmark.php?test=all&lang=all&d=data&ghc=on&ocaml=on&sbcl=on&csharp=on&ifc=on&clean=on&calc=calculate&box=1 [2]http://shootout.alioth.debian.org/u32q/benchmark.php?test=all&lang=all&d=data&ghc=on&ocaml=on&sbcl=on&csharp=on&ifc=on&clean=on&calc=calculate&box=1 [3]http://shootout.alioth.debian.org/u64/benchmark.php?test=all&lang=all&d=data&ghc=on&ocaml=on&sbcl=on&csharp=on&ifc=on&clean=on&calc=calculate&box=1 [4]http://shootout.alioth.debian.org/u64q/benchmark.php?test=all&lang=all&d=data&ghc=on&ocaml=on&sbcl=on&csharp=on&ifc=on&clean=on&calc=calculate&box=1 From magnus at therning.org Sun May 3 16:11:52 2009 From: magnus at therning.org (Magnus Therning) Date: Sun May 3 15:57:37 2009 Subject: [Haskell-beginners] Haskell vs Clean (speed) In-Reply-To: <49FDC109.4030302@theingots.org> References: <49FDC109.4030302@theingots.org> Message-ID: <49FDFA88.6060303@therning.org> Daniel Carrera wrote: [..] > * "Lisp" is "List SCBL" whatever that is. That should probably be SBCL, as in Steel Bank Common Lisp. No? /M -- Magnus Therning (OpenPGP: 0xAB4DFBA4) magnus?therning?org Jabber: magnus?therning?org http://therning.org/magnus identi.ca|twitter: magthe -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 197 bytes Desc: OpenPGP digital signature Url : http://www.haskell.org/pipermail/beginners/attachments/20090503/b7f423e4/signature.bin From joel.neely at gmail.com Mon May 4 11:34:09 2009 From: joel.neely at gmail.com (Joel Neely) Date: Mon May 4 11:29:14 2009 Subject: [Haskell-beginners] Polymorphism question from an OO-speaking newbie Message-ID: <5e0b968a0905040834t56bf8c33v1e73d5f30f4d8c8f@mail.gmail.com> Short version: Is it possible/reasonable to define a single function that accepts a single string or a list of strings (in which case it maps the single-string "flavor" over the list)? Longer version: Thus far I know that Haskell allows me to define a function on a single string, then define another function that maps the first one across a list of strings, as in: *Main> let quote1 s = "\"" ++ s ++ "\"" *Main> "Quux said " ++ quote1 "foo" ++ " loudly" "Quux said \"foo\" loudly" *Main> let quote = map quote1 *Main> quote ["foo", "baz", "bletch"] ["\"foo\"","\"baz\"","\"bletch\""] (BTW, is it standard terminology to say that quote lifts quote1 to lists?) In the above I have different names for the two functions. OO languages such as Java allow me to overload quote to accept either String or List (and return the same type). AFAICT, Haskell's parametric polymorphism allows me to define functions (e.g. length) which deal with "listness" concepts indifferent to the type contained by the list. Am I missing something, or should I admit to OO wrongheadedness and accept that my inability to write a single declaration unifying: quote :: String -> String quote :: [String] -> [String] is an emphatic clue that I should change my expectations? Thanks in advance for any enlightenment shed my way! -jn- -- Beauty of style and harmony and grace and good rhythm depend on simplicity. - Plato From jakubuv at gmail.com Mon May 4 12:23:59 2009 From: jakubuv at gmail.com (Jan Jakubuv) Date: Mon May 4 12:26:12 2009 Subject: [Haskell-beginners] Polymorphism question from an OO-speaking newbie In-Reply-To: <5e0b968a0905040834t56bf8c33v1e73d5f30f4d8c8f@mail.gmail.com> References: <5e0b968a0905040834t56bf8c33v1e73d5f30f4d8c8f@mail.gmail.com> Message-ID: <20090504162359.GA12421@lxultra2.macs.hw.ac.uk> On Mon, May 04, 2009 at 10:34:09AM -0500, Joel Neely wrote: > Short version: Is it possible/reasonable to define a single function > that accepts a single string or a list of strings (in which case it > maps the single-string "flavor" over the list)? > Hi Joel, you are looking for type classes (http://haskell.org/tutorial/classes.html). They allow you to use the same function name for different types. This is called ``ad-hoc'' polymorphism and you can define different implementation of the overloaded function for different types (which is your case). Haskell also support ``parametric'' polymorphism which can be used when you have the same implementation for different types (like the function `head` which returns the first member of a list independently on the type of elements). Try the following: {-# OPTIONS -fglasgow-exts #-} quoteS s = "\"" ++ s ++ "\"" quoteL l = map quoteS l class Quatable q where quote :: q -> q instance Quatable String where quote s = "\"" ++ s ++ "\"" instance Quatable [String] where quote = map quote Then you can use it as follows: *Main> "Quux said " ++ quote "foo" ++ " loudly" "Quux said \"foo\" loudly" *Main> quote ["foo", "baz", "bletch"] ["\"foo\"","\"baz\"","\"bletch\""] Don't forget the line `{-# OPTIONS -fglasgow-exts #-}` which enables some features of GHC you need for this example (but not necessarily for all examples using type classes). Alternatively you can use the line {-# LANGUAGE TypeSynonymInstances,FlexibleInstances #-} which lists necessary extensions explicitly. Sincerely, Jan. -- Heriot-Watt University is a Scottish charity registered under charity number SC000278. From jakubuv at gmail.com Mon May 4 12:29:26 2009 From: jakubuv at gmail.com (Jan Jakubuv) Date: Mon May 4 12:37:03 2009 Subject: [Haskell-beginners] Polymorphism question from an OO-speaking newbie In-Reply-To: <20090504162359.GA12421@lxultra2.macs.hw.ac.uk> References: <5e0b968a0905040834t56bf8c33v1e73d5f30f4d8c8f@mail.gmail.com> <20090504162359.GA12421@lxultra2.macs.hw.ac.uk> Message-ID: <20090504162926.GB12421@lxultra2.macs.hw.ac.uk> On Mon, May 04, 2009 at 05:23:59PM +0100, Jan Jakubuv wrote: > > Try the following: > > {-# OPTIONS -fglasgow-exts #-} > > quoteS s = "\"" ++ s ++ "\"" > quoteL l = map quoteS l I forgot to say that the functions `quoteS` and `quoteL` are just for illustration. You can miss them. Jan. -- Heriot-Watt University is a Scottish charity registered under charity number SC000278. From es at ertes.de Mon May 4 12:34:28 2009 From: es at ertes.de (Ertugrul Soeylemez) Date: Mon May 4 12:37:28 2009 Subject: [Haskell-beginners] Re: Polymorphism question from an OO-speaking newbie References: <5e0b968a0905040834t56bf8c33v1e73d5f30f4d8c8f@mail.gmail.com> Message-ID: <20090504183428.44dceaf0@tritium.xx> Hello Joel, the polymorphism concept in Haskell's type system is not the same as OOP's polymorphism. OOP polymorphism allows you to build abstract interfaces, whereas Haskell's polymorphism allows you to generalize both statements and functionality. There is no way to 'select' String and [String], as those two types have almost nothing in common. If you find types that have something in common, you can express it. "Having something in common" always means: there is a set of functions, which is defined for both. For example, the two types Integer and Float have in common that they are numeric types. Technically this means that (+), (-), (*) and a few other functions are defined for them. If that's all you need to know, you can generalize the function foo :: Integer -> Integer -> Integer foo a b = a*a + b*b to the following: foo :: Integral i => i -> i -> i foo a b = a*a + b*b Now 'foo' is defined for every type, which is an instance of the class Integral. Being an instance of that class precisely means that (+), (*) and some other functions are defined for the particular type. Since this is all you need to know for 'foo', there is no reason to restrict it to Integer. This can be applied to your example, but not verbatim. As said, the two types String and [String] have little in common. If you want to write a function with a polymorphic type, you always need to think in terms of things you know about the types. The types 'Maybe a' and '[a]' have a common property: [] and Maybe are both functors. Informally that means that they are types, which implement some notion of 'mapping a function over the value(s)'. Technically it means that the 'fmap' function is defined for both: fmap :: Functor f => (a -> b) -> f a -> f b Here is your original function: quote :: [String] -> [String] quote = map (\s -> "\"" ++ s ++ "\"") Since the 'map' function is actually just a special case of 'fmap', which is constrained to lists, knowing that you can generalize 'quote' to any functor: quote :: Functor f => f String -> f String quote = fmap (\s -> "\"" ++ s ++ "\"") To answer your original question: It is impossible to write a function, which handles both cases, because the two cases are inherently incompatible. In other words: Trying to write a Haskell function, which handles both cases, is pointless by concept. Greets, Ertugrul. Joel Neely wrote: > Short version: Is it possible/reasonable to define a single function > that accepts a single string or a list of strings (in which case it > maps the single-string "flavor" over the list)? > > Longer version: Thus far I know that Haskell allows me to define a > function on a single string, then define another function that maps > the first one across a list of strings, as in: > > *Main> let quote1 s = "\"" ++ s ++ "\"" > > *Main> "Quux said " ++ quote1 "foo" ++ " loudly" > "Quux said \"foo\" loudly" > > *Main> let quote = map quote1 > > *Main> quote ["foo", "baz", "bletch"] > ["\"foo\"","\"baz\"","\"bletch\""] > > (BTW, is it standard terminology to say that quote lifts quote1 to lists?) > > In the above I have different names for the two functions. OO > languages such as Java allow me to overload quote to accept either > String or List (and return the same type). AFAICT, Haskell's > parametric polymorphism allows me to define functions (e.g. length) > which deal with "listness" concepts indifferent to the type contained > by the list. > > Am I missing something, or should I admit to OO wrongheadedness and > accept that my inability to write a single declaration unifying: > > quote :: String -> String > quote :: [String] -> [String] > > is an emphatic clue that I should change my expectations? > > Thanks in advance for any enlightenment shed my way! > > -jn- > -- nightmare = unsafePerformIO (getWrongWife >>= sex) http://blog.ertes.de/ From byorgey at seas.upenn.edu Mon May 4 12:47:44 2009 From: byorgey at seas.upenn.edu (Brent Yorgey) Date: Mon May 4 12:48:50 2009 Subject: [Haskell-beginners] Polymorphism question from an OO-speaking newbie In-Reply-To: <5e0b968a0905040834t56bf8c33v1e73d5f30f4d8c8f@mail.gmail.com> References: <5e0b968a0905040834t56bf8c33v1e73d5f30f4d8c8f@mail.gmail.com> Message-ID: <20090504164744.GA14640@seas.upenn.edu> On Mon, May 04, 2009 at 10:34:09AM -0500, Joel Neely wrote: > Short version: Is it possible/reasonable to define a single function > that accepts a single string or a list of strings (in which case it > maps the single-string "flavor" over the list)? Short answer: no. For the long answer (no, but sort of yes) see below. =) > Longer version: Thus far I know that Haskell allows me to define a > function on a single string, then define another function that maps > the first one across a list of strings, as in: > > *Main> let quote1 s = "\"" ++ s ++ "\"" > > *Main> "Quux said " ++ quote1 "foo" ++ " loudly" > "Quux said \"foo\" loudly" > > *Main> let quote = map quote1 > > *Main> quote ["foo", "baz", "bletch"] > ["\"foo\"","\"baz\"","\"bletch\""] > > (BTW, is it standard terminology to say that quote lifts quote1 to lists?) There are several standard ways to say it, but that is indeed one of them. > In the above I have different names for the two functions. OO > languages such as Java allow me to overload quote to accept either > String or List (and return the same type). AFAICT, Haskell's > parametric polymorphism allows me to define functions (e.g. length) > which deal with "listness" concepts indifferent to the type contained > by the list. Right, and this is the key: *indifferent* to the type contained by the list. The implementation works in *exactly the same way* for any type you might wish to stick in. Java has this, too, with generics. List is defined independently of foo; you can stick any foo in you like and it works in the same way. However, the overloading you're talking about is known as 'ad-hoc' polymorphism. It means that you can give a *different* implementation for each type. You can't simulate this with parametric polymorphism. If you write a Haskell function foo :: a -> ... foo x = ... and you want foo to behave differently depending on the type of x, you can't do it. There's no way in the ... to decide what to do based on x's type; the implementation has to be the same no matter what type x is. However! You can get something similar to ad-hoc polymorphism with Haskell's type classes. Using type classes, you could do something like this: class Quotable q where -- any type which is an instance of Quotable must provide quote :: q -> q -- an implementation of quote instance Quotable String where quote = quote1 instance Quotable [String] where quote = map quote1 Now quote has type quote :: (Quotable q) => q -> q and can be used on either Strings or lists of Strings. Haskell will use type inference to figure out which version of quote to use depending on the context in which it is called. -Brent From es at ertes.de Mon May 4 12:57:03 2009 From: es at ertes.de (Ertugrul Soeylemez) Date: Mon May 4 13:00:18 2009 Subject: [Haskell-beginners] Re: Polymorphism question from an OO-speaking newbie References: <5e0b968a0905040834t56bf8c33v1e73d5f30f4d8c8f@mail.gmail.com> <20090504183428.44dceaf0@tritium.xx> Message-ID: <20090504185703.01295257@tritium.xx> Ertugrul Soeylemez wrote: > [...] If that's all you need to know, you can generalize the function > > foo :: Integer -> Integer -> Integer > foo a b = a*a + b*b > > to the following: > > foo :: Integral i => i -> i -> i > foo a b = a*a + b*b > > Now 'foo' is defined for every type, which is an instance of the class > Integral. Being an instance of that class precisely means that (+), > (*) and some other functions are defined for the particular type. > Since this is all you need to know for 'foo', there is no reason to > restrict it to Integer. Sorry, I actually meant the Num class, not the Integer class, but the example is still valid, because any Integral type is also a Num type. So here is a more general 'foo': foo :: Num a => a -> a -> a foo a b = a*a + b*b Greets, Ertugrul. -- nightmare = unsafePerformIO (getWrongWife >>= sex) http://blog.ertes.de/ From marcot at holoscopio.com Mon May 4 10:59:54 2009 From: marcot at holoscopio.com (Marco =?ISO-8859-1?Q?T=FAlio?= Gontijo e Silva) Date: Mon May 4 13:59:14 2009 Subject: [Haskell-beginners] Context reducion Stack overflow Message-ID: <1241449194.5122.181.camel@zezinho> Hello, I'm trying to compile the following code: > import Data.IORef > class F f where > m :: (a -> a) -> f a -> f a > class M m a where > mm :: (a -> a) -> m -> IO () > instance (F f, M m (f a)) => M m a where > mm f v = mm (m f) v And I'm getting: $ ghci teste.lhs -XMultiParamTypeClasses -XFlexibleInstances -XFlexibleContexts -XUndecidableInstances -XIncoherentInstances GHCi, version 6.10.1: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer ... linking ... done. Loading package base ... linking ... done. Failed, modules loaded: none. Prelude> ^DLeaving GHCi. GHCi, version 6.10.1: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer ... linking ... done. Loading package base ... linking ... done. [1 of 1] Compiling Main ( teste.lhs, interpreted ) teste.lhs:1:0: Context reduction stack overflow; size = 20 Use -fcontext-stack=N to increase stack size to N `$dM{atp} :: {M m (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f a))))))))))))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{atm} :: {M m (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f a)))))))))))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{atj} :: {M m (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f a))))))))))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{atg} :: {M m (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f a)))))))))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{atd} :: {M m (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f a))))))))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{ata} :: {M m (f (f (f (f (f (f (f (f (f (f (f (f (f (f (f a)))))))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{at7} :: {M m (f (f (f (f (f (f (f (f (f (f (f (f (f (f a))))))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{at4} :: {M m (f (f (f (f (f (f (f (f (f (f (f (f (f a)))))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{at1} :: {M m (f (f (f (f (f (f (f (f (f (f (f (f a))))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asY} :: {M m (f (f (f (f (f (f (f (f (f (f (f a)))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asV} :: {M m (f (f (f (f (f (f (f (f (f (f a))))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asS} :: {M m (f (f (f (f (f (f (f (f (f a)))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asP} :: {M m (f (f (f (f (f (f (f (f a))))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asM} :: {M m (f (f (f (f (f (f (f a)))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asJ} :: {M m (f (f (f (f (f (f a))))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asG} :: {M m (f (f (f (f (f a)))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asD} :: {M m (f (f (f (f a))))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asA} :: {M m (f (f (f a)))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asx} :: {M m (f (f a))}' arising from a use of `mm' at teste.lhs:10:13-22 `$dM{asj} :: {M m (f a)}' arising from a use of `mm' at teste.lhs:10:13-22 Failed, modules loaded: none. An example of instances of these classes: > instance F [] where > m = map > instance M (IORef a) a where > mm = flip modifyIORef I don't get what's the problem with my definition. Can someone clarify me on this? Greetings. -- marcot http://marcot.iaaeee.org/ From byorgey at seas.upenn.edu Mon May 4 15:54:28 2009 From: byorgey at seas.upenn.edu (Brent Yorgey) Date: Mon May 4 15:40:23 2009 Subject: [Haskell-beginners] Context reducion Stack overflow In-Reply-To: <1241449194.5122.181.camel@zezinho> References: <1241449194.5122.181.camel@zezinho> Message-ID: <20090504195427.GA29604@seas.upenn.edu> On Mon, May 04, 2009 at 11:59:54AM -0300, Marco T?lio Gontijo e Silva wrote: > > instance (F f, M m (f a)) => M m a where > > mm f v = mm (m f) v Perhaps you mean instance (F f, M m a) => M m (f a) where ... ? What you have written means that to have an instance M m a, you must have an instance M m (f a), for which you must have an instance M m (f (f a)), for which you must have an instance M m (f (f (f a)))... What I have written above expresses what I presume you meant: we can automatically derive an instance of (M m) for some structured type (f a) as long as we have a base instance M m a and can lift functions (a -> a) to (f a -> f a). -Brent From aditya.siram at gmail.com Tue May 5 09:11:23 2009 From: aditya.siram at gmail.com (aditya siram) Date: Tue May 5 08:56:49 2009 Subject: [Haskell-beginners] GHCI: Main function seems to mask other functions Message-ID: <594f78210905050611u26db87fdw5498c71749838cfe@mail.gmail.com> Hi all, I'm having a weird problem where when I add the 'main' function to my program and load it into GHCI, whenever I check the type or try to run my other functions I get a 'Not in scope: ' error. When I remove the 'main' function and reload the code, the problem goes away. I can't seem to reproduce this with a trivial example, but this has happened to me several times in my larger programs. thanks ... deech -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090505/f05a2b69/attachment.htm From marcot at holoscopio.com Tue May 5 09:28:10 2009 From: marcot at holoscopio.com (Marco =?ISO-8859-1?Q?T=FAlio?= Gontijo e Silva) Date: Tue May 5 09:14:14 2009 Subject: [Haskell-beginners] Context reducion Stack overflow In-Reply-To: <20090504195427.GA29604@seas.upenn.edu> References: <1241449194.5122.181.camel@zezinho> <20090504195427.GA29604@seas.upenn.edu> Message-ID: <1241530090.5122.264.camel@zezinho> Em Seg, 2009-05-04 ?s 15:54 -0400, Brent Yorgey escreveu: > On Mon, May 04, 2009 at 11:59:54AM -0300, Marco T?lio Gontijo e Silva wrote: > > > > instance (F f, M m (f a)) => M m a where > > > mm f v = mm (m f) v > > Perhaps you mean > > instance (F f, M m a) => M m (f a) where ... > > ? No, I really meant what I wrote. An example: If I the instances I wrote: > instance F [] where > m = map > instance M (IORef a) a where > mm = flip modifyIORef I want to define: > instance M (IORef [a]) a where > mm f v = mm (m f) v This could of course be written as: mm = mm . m I'd like to get this last instance automaticly from that definition, having: f = [] m = IORef [a] > What you have written means that to have an instance M m a, you must > have an instance M m (f a), for which you must have an instance M m (f > (f a)), for which you must have an instance M m (f (f (f a)))... After this explanation I think I located the problem. My problem seems to be with how IncoherentInstances work, by picking always the most generic instance. Thanks for that. Do you know if there a way to get around this, maybe using another GHC extension? Greetings. -- marcot http://marcot.iaaeee.org/ From daniel.is.fischer at web.de Tue May 5 10:42:06 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Tue May 5 10:28:00 2009 Subject: [Haskell-beginners] GHCI: Main function seems to mask other functions In-Reply-To: <594f78210905050611u26db87fdw5498c71749838cfe@mail.gmail.com> References: <594f78210905050611u26db87fdw5498c71749838cfe@mail.gmail.com> Message-ID: <200905051642.07053.daniel.is.fischer@web.de> Am Dienstag 05 Mai 2009 15:11:23 schrieb aditya siram: > Hi all, > I'm having a weird problem where when I add the 'main' function to my > program and load it into GHCI, whenever I check the type or try to run my > other functions I get a 'Not in scope: ' error. When I > remove the 'main' function and reload the code, the problem goes away. > > I can't seem to reproduce this with a trivial example, but this has > happened to me several times in my larger programs. > > thanks ... > deech Odd. Does it happen with the module header module Main where , with an explicit export list, or without any module header? Is it so that with the main function, you load the compiled function, then when you get the Not in scope, you remove the main function and reload the module without compiling? Then it's probably the correct behaviour, cf. the user's guide, section 3.4.3. What's really in scope at the prompt? From byorgey at seas.upenn.edu Tue May 5 11:33:58 2009 From: byorgey at seas.upenn.edu (Brent Yorgey) Date: Tue May 5 11:19:22 2009 Subject: [Haskell-beginners] Context reducion Stack overflow In-Reply-To: <1241530090.5122.264.camel@zezinho> References: <1241449194.5122.181.camel@zezinho> <20090504195427.GA29604@seas.upenn.edu> <1241530090.5122.264.camel@zezinho> Message-ID: <20090505153358.GA15119@seas.upenn.edu> On Tue, May 05, 2009 at 10:28:10AM -0300, Marco T?lio Gontijo e Silva wrote: > Em Seg, 2009-05-04 ?s 15:54 -0400, Brent Yorgey escreveu: > > On Mon, May 04, 2009 at 11:59:54AM -0300, Marco T?lio Gontijo e Silva wrote: > > > > > > instance (F f, M m (f a)) => M m a where > > > > mm f v = mm (m f) v > > > > Perhaps you mean > > > > instance (F f, M m a) => M m (f a) where ... > > > > ? > > No, I really meant what I wrote. An example: If I the instances I > wrote: > > > instance F [] where > > m = map > > > instance M (IORef a) a where > > mm = flip modifyIORef > > I want to define: > > > instance M (IORef [a]) a where > > mm f v = mm (m f) v > > This could of course be written as: > > mm = mm . m > > I'd like to get this last instance automaticly from that definition, > having: > > f = [] > m = IORef [a] I don't follow. In order to get instance M (IORef [a]) a from instance (F f, M m (f a)) => M m a would require instance M (IORef [a]) (f a) for some f, which you don't have. I might try rewriting M as class M f a where mm :: (a -> a) -> f a -> IO () and then your automatic lifting instance would be something like instance (F f, M f2 a) => M (f :.: f2) a where :.: denotes functor composition. -Brent From marcot at holoscopio.com Tue May 5 14:06:41 2009 From: marcot at holoscopio.com (Marco =?ISO-8859-1?Q?T=FAlio?= Gontijo e Silva) Date: Tue May 5 13:52:42 2009 Subject: [Haskell-beginners] Re: =?iso-8859-1?q?Programa=E7=E3o?= Funcional - TP1 In-Reply-To: References: Message-ID: <1241546801.5122.520.camel@zezinho> Ol? Henrique, seu problema Polepos.hs excedeu o tempo limite de execu??o no SPOJ. Algumas dicas: voc? pode escrever: cars = map fst startGrid positions = map snd startGrid (cars, positions) = unzip take (length positions) [0..] [0 .. pred $ length positions] do scoreBoard <- parseScoreboard n let startGrid = getStartGrid scoreBoard putStrLn $ showGrid startGrid main parseScoreboard n >>= putStrLn . showGrid . getStartGrid >> main showGrid Nothing = "-1" showGrid (Just xs) = foldl1 (++) $ intersperse " " $ map show xs showGrid = maybe "-1" (foldl1 (++) . intersperse " " . map show) Talvez fosse mais legal criar um tipo de dados e definir a classe Ord para eles do que usar tupla e sortBy. Essas dicas n?o est?o relacionadas com a demora no tempo de execu??o. Isso se deve, acredito eu, ao seu algoritmo. -- marcot http://marcot.iaaeee.org/ From marcot at holoscopio.com Tue May 5 14:07:16 2009 From: marcot at holoscopio.com (Marco =?ISO-8859-1?Q?T=FAlio?= Gontijo e Silva) Date: Tue May 5 13:53:17 2009 Subject: [Haskell-beginners] Context reducion Stack overflow In-Reply-To: <20090505153358.GA15119@seas.upenn.edu> References: <1241449194.5122.181.camel@zezinho> <20090504195427.GA29604@seas.upenn.edu> <1241530090.5122.264.camel@zezinho> <20090505153358.GA15119@seas.upenn.edu> Message-ID: <1241546836.5122.522.camel@zezinho> Em Ter, 2009-05-05 ?s 11:33 -0400, Brent Yorgey escreveu: > On Tue, May 05, 2009 at 10:28:10AM -0300, Marco T?lio Gontijo e Silva wrote: > > Em Seg, 2009-05-04 ?s 15:54 -0400, Brent Yorgey escreveu: > > > On Mon, May 04, 2009 at 11:59:54AM -0300, Marco T?lio Gontijo e Silva wrote: > > > > > > > > instance (F f, M m (f a)) => M m a where > > > > > mm f v = mm (m f) v > > > > > > Perhaps you mean > > > > > > instance (F f, M m a) => M m (f a) where ... > > > > > > ? > > > > No, I really meant what I wrote. An example: If I the instances I > > wrote: > > > > > instance F [] where > > > m = map > > > > > instance M (IORef a) a where > > > mm = flip modifyIORef > > > > I want to define: > > > > > instance M (IORef [a]) a where > > > mm f v = mm (m f) v > > > > This could of course be written as: > > > > mm = mm . m > > > > I'd like to get this last instance automaticly from that definition, > > having: > > > > f = [] > > m = IORef [a] > > I don't follow. In order to get > > instance M (IORef [a]) a > > from > > instance (F f, M m (f a)) => M m a > > would require > > instance M (IORef [a]) (f a) > > for some f, which you don't have. Yes, I have this instance defined for f = []: instance M (IORef [a]) [a] which is a instance of: instance M (IORef a) a > I might try rewriting M as > > class M f a where > mm :: (a -> a) -> f a -> IO () > > and then your automatic lifting instance would be something like > > instance (F f, M f2 a) => M (f :.: f2) a > > where :.: denotes functor composition. Ok, I think this is another possibility. But how could I define :.:? Greetings. -- marcot http://marcot.iaaeee.org/ From marcot at holoscopio.com Tue May 5 14:08:01 2009 From: marcot at holoscopio.com (Marco =?ISO-8859-1?Q?T=FAlio?= Gontijo e Silva) Date: Tue May 5 13:54:02 2009 Subject: [Haskell-beginners] Re: =?ISO-8859-1?Q?Programa=E7=E3o?= Funcional - TP1 In-Reply-To: <1241546801.5122.520.camel@zezinho> References: <1241546801.5122.520.camel@zezinho> Message-ID: <1241546881.5122.525.camel@zezinho> Sorry about this e-mail, I send it by mistake. Greetings. -- marcot http://marcot.iaaeee.org/ From jack at realmode.com Tue May 5 15:19:55 2009 From: jack at realmode.com (Jack Kennedy) Date: Tue May 5 15:05:20 2009 Subject: [Haskell-beginners] parallel program in haskell in 5 steps Message-ID: <1008bfc90905051219q7760c5ew9dd95b1a2b87db7c@mail.gmail.com> In step 4 of Haskell in 5 Steps [ http://haskell.org/haskellwiki/Haskell_in_5_steps], a parallel program is given. I changed it very slightly so it would run a long time (see below). It compiles and runs but my CPU meter is barely above 50%. I have a dual core processor. What in the world would keep this program from completely saturating the CPU? import Control.Parallel main = a `par` b `pseq` print (a + b) where a = ack 4 10 b = ack 4 10 fac 0 = 1 fac n = n * fac (n-1) ack 0 n = n+1 ack m 0 = ack (m-1) 1 ack m n = ack (m-1) (ack m (n-1)) fib 0 = 0 fib 1 = 1 fib n = fib (n-1) + fib (n-2) -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090505/2d831639/attachment.htm From michael at snoyman.com Tue May 5 15:51:59 2009 From: michael at snoyman.com (Michael Snoyman) Date: Tue May 5 15:37:23 2009 Subject: [Haskell-beginners] parallel program in haskell in 5 steps In-Reply-To: <1008bfc90905051219q7760c5ew9dd95b1a2b87db7c@mail.gmail.com> References: <1008bfc90905051219q7760c5ew9dd95b1a2b87db7c@mail.gmail.com> Message-ID: <29bf512f0905051251u28c94d97rd8c3ddd43c89203a@mail.gmail.com> How are you running the program? You have to explicitly tell the compiler/interpreter to use multiple system threads. Michael On Tue, May 5, 2009 at 10:19 PM, Jack Kennedy wrote: > In step 4 of Haskell in 5 Steps [ > http://haskell.org/haskellwiki/Haskell_in_5_steps], > a parallel program is given. I changed it very slightly so it would run a > long time (see below). > > It compiles and runs but my CPU meter is barely above 50%. I have a dual > core processor. > What in the world would keep this program from completely saturating the > CPU? > > import Control.Parallel > > main = a `par` b `pseq` print (a + b) > where > a = ack 4 10 > b = ack 4 10 > > fac 0 = 1 > fac n = n * fac (n-1) > > ack 0 n = n+1 > ack m 0 = ack (m-1) 1 > ack m n = ack (m-1) (ack m (n-1)) > > fib 0 = 0 > fib 1 = 1 > fib n = fib (n-1) + fib (n-2) > > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090505/f77d0f9e/attachment.htm From es at ertes.de Tue May 5 16:59:23 2009 From: es at ertes.de (Ertugrul Soeylemez) Date: Tue May 5 16:45:01 2009 Subject: [Haskell-beginners] Re: parallel program in haskell in 5 steps References: <1008bfc90905051219q7760c5ew9dd95b1a2b87db7c@mail.gmail.com> Message-ID: <20090505225923.2a34ae03@tritium.xx> Hello Jack, according to the source code your program should not run for a long time. If you're using GHC, use the '-threaded' flag when compiling and also pass the parameters '+RTS -N2' to your program to let it run with two parallel threads. However, my experience is that the pure parallel constructs are currently not really suitable for real world programs. With my dual core processor I get the expected ~190% CPU time using explicit parallelism using concurrency (i.e. forkIO and MVars), whereas with pure constructs like the ones from Control.Parallel and Control.Parallel.Strategies I get only about 140%. Greets, Ertugrul. Jack Kennedy wrote: > In step 4 of Haskell in 5 Steps [ > http://haskell.org/haskellwiki/Haskell_in_5_steps], > a parallel program is given. I changed it very slightly so it would run a > long time (see below). > > It compiles and runs but my CPU meter is barely above 50%. I have a dual > core processor. > What in the world would keep this program from completely saturating the > CPU? > > import Control.Parallel > > main = a `par` b `pseq` print (a + b) > where > a = ack 4 10 > b = ack 4 10 > > fac 0 = 1 > fac n = n * fac (n-1) > > ack 0 n = n+1 > ack m 0 = ack (m-1) 1 > ack m n = ack (m-1) (ack m (n-1)) > > fib 0 = 0 > fib 1 = 1 > fib n = fib (n-1) + fib (n-2) > -- nightmare = unsafePerformIO (getWrongWife >>= sex) http://blog.ertes.de/ From jack at realmode.com Tue May 5 17:05:56 2009 From: jack at realmode.com (Jack Kennedy) Date: Tue May 5 16:51:22 2009 Subject: [Haskell-beginners] parallel program in haskell in 5 steps In-Reply-To: <29bf512f0905051251u28c94d97rd8c3ddd43c89203a@mail.gmail.com> References: <1008bfc90905051219q7760c5ew9dd95b1a2b87db7c@mail.gmail.com> <29bf512f0905051251u28c94d97rd8c3ddd43c89203a@mail.gmail.com> Message-ID: <1008bfc90905051405v3ffc9da6tc828de863f6ee5f1@mail.gmail.com> I am compiling (Windows by the way) using the line from the tutorial: ghc -O2 --make par.hs -threaded and running with the line par +RTS -N2 CPU usage for the process flits around a little, but stays in the 45% - 55% range. On Tue, May 5, 2009 at 12:51 PM, Michael Snoyman wrote: > How are you running the program? You have to explicitly tell the > compiler/interpreter to use multiple system threads. > > Michael > > On Tue, May 5, 2009 at 10:19 PM, Jack Kennedy wrote: > >> In step 4 of Haskell in 5 Steps [ >> http://haskell.org/haskellwiki/Haskell_in_5_steps], >> a parallel program is given. I changed it very slightly so it would run a >> long time (see below). >> >> It compiles and runs but my CPU meter is barely above 50%. I have a dual >> core processor. >> What in the world would keep this program from completely saturating the >> CPU? >> >> import Control.Parallel >> >> main = a `par` b `pseq` print (a + b) >> where >> a = ack 4 10 >> b = ack 4 10 >> >> fac 0 = 1 >> fac n = n * fac (n-1) >> >> ack 0 n = n+1 >> ack m 0 = ack (m-1) 1 >> ack m n = ack (m-1) (ack m (n-1)) >> >> fib 0 = 0 >> fib 1 = 1 >> fib n = fib (n-1) + fib (n-2) >> >> >> _______________________________________________ >> Beginners mailing list >> Beginners@haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090505/ce07c344/attachment-0001.htm From joel.neely at gmail.com Wed May 6 09:23:10 2009 From: joel.neely at gmail.com (Joel Neely) Date: Wed May 6 09:08:33 2009 Subject: [Haskell-beginners] question >>= [style, efficiency] Message-ID: <5e0b968a0905060623k2e315345pda0f047326b4d35b@mail.gmail.com> I can think of a variety of ways to put a value at the end of a list of the same type. -- cutesy-clever putAtEnd :: [a] -> a -> [a] putAtEnd xs x = reverse (x : reverse xs) -- by hand atEnd :: [a] -> a -> [a] atEnd [] z = [z] atEnd (x:xs) z = x : atEnd xs z -- obvious brute force infixr 9 +: (+:) :: [a] -> a -> [a] (+:) xs x = xs ++ [x] I'd appreciate feedback/advice on whether: 1) I've missed any clearly-better alternatives; 2) any of those is obviously better or worse than the others; 3) I should have known where to look for the answer myself (instead of brute-force Googling). I've semi-inadvertently mixed two issues above: infix-vs-prefix and algorithm design. TIA for suggestions on either. -jn- -- Beauty of style and harmony and grace and good rhythm depend on simplicity. - Plato From applebiz89 at hotmail.com Wed May 6 09:28:18 2009 From: applebiz89 at hotmail.com (Ashley Banks) Date: Wed May 6 09:13:42 2009 Subject: [Haskell-beginners] (no subject) Message-ID: I havent done much IO at all in haskell, only within the function itself. However I want to get the input from the interface for the function and havent done this before. // Say I have this as my data type and list of films data Film = Film String String Int [String] -- List of films testDatabase :: [Film] testDatabase = [(Film "Casino Royale" "Martin Campbell" 2006 ["Garry", "Dave", "Zoe"]) ] // with functions such as: becomeFan :: String -> String -> [Film] -> [Film] becomeFan _ _ [] = [] becomeFan filmName fanName ((Film title director year fans):xs) | filmName == title = (Film title director year fanName:fans) : xs | otherwise = (Film title director year fans) : becomeFan filmName fanName xs filmsInGivenYear :: Int -> [Film] -> [String] filmsInGivenYear filmYear films = [ title | (Film title director year fans) <- films, year == filmYear] // I want to ask the user what function they want to use, I have this so far main :: IO() do putStr "Hi there! what is your name: " fanName = getLine do putStr "1 = Insert film, 2 = Become a Fan, 3 = The number of fans of a film, 4 = Film released in a year: " input = getLine read input :: Int (if input == 1 then main x = insertFilm [] else if input == 2 then main x = becomeFan [] else if input == 3 then main x = numberOfFans []) // I thought using the if statement would work, but now I cant think how to gather the needed input for the function they have chosen? thanks apple _________________________________________________________________ Beyond Hotmail ? see what else you can do with Windows Live. http://clk.atdmt.com/UKM/go/134665375/direct/01/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090506/1a624f35/attachment.htm From seideld at tcs.inf.tu-dresden.de Wed May 6 10:57:49 2009 From: seideld at tcs.inf.tu-dresden.de (Daniel Seidel) Date: Wed May 6 10:43:11 2009 Subject: [Haskell-beginners] (no subject) In-Reply-To: References: Message-ID: <4A01A56D.20504@tcs.inf.tu-dresden.de> Ashley Banks wrote: > I havent done much IO at all in haskell, only within the function itself. > However I want to get the input from the interface for the function and > havent done this before. > > // Say I have this as my data type and list of films > data Film = Film String String Int [String] > -- List of films > > testDatabase :: [Film] > testDatabase = [(Film "Casino Royale" "Martin Campbell" 2006 ["Garry", > "Dave", "Zoe"]) ] > > // with functions such as: > > becomeFan :: String -> String -> [Film] -> [Film] > becomeFan _ _ [] = [] > becomeFan filmName fanName ((Film title director year fans):xs) > | filmName == title = (Film title director year fanName:fans) : xs > | otherwise = (Film title director year fans) : becomeFan filmName fanName > xs > > filmsInGivenYear :: Int -> [Film] -> [String] > filmsInGivenYear filmYear films = [ title | (Film title director year fans) > <- films, year == filmYear] > > // I want to ask the user what function they want to use, I have this so far > > main :: IO() > do putStr "Hi there! what is your name: " > fanName = getLine > do putStr "1 = Insert film, 2 = Become a Fan, 3 = The number of fans of a > film, 4 = Film released in a year: " > input = getLine > read input :: Int > (if input == 1 then main x = insertFilm [] else if input == 2 then main x = > becomeFan [] else if input == 3 then main x = numberOfFans []) that part is kind of confusing. My Suggestion: separate functions for the "submenus". And a main function calling them: submenu "2" fanName = do putStr "What film you want to become a fan of? " filmName <- getLine putStr (show (becomeFan filmName fanName testDatabase)) return () submenu _ _ = do putStr "The function is not supported yet." return () main :: IO() main = do putStr "Hi there! what is your name: " fanName <- getLine putStr "1 = Insert film, 2 = Become a Fan, 3 = The number of fans of a film, 4 = Film released in a year, 5 = Exit: " input <- getLine (if input /= "5" then do submenu input fanName main else return ()) The example should work, but is not very useful, for just adding yourself to the fan list of a movie in testDatabase. You might need to pass around the database. It's not like manipulating a global object as you might do in imperative languages. > > // I thought using the if statement would work, but now I cant think how to > gather the needed input for the function they have chosen? > > thanks > > apple > > > ------------------------------------------------------------------------ > Get the New Internet Explore 8 Optimised for MSN. Download Now > > > > ------------------------------------------------------------------------ > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners From applebiz89 at hotmail.com Wed May 6 11:04:45 2009 From: applebiz89 at hotmail.com (Ashley Banks) Date: Wed May 6 10:50:09 2009 Subject: [Haskell-beginners] haskell IO help Message-ID: From: applebiz89@hotmail.com To: beginners@haskell.org Subject: Date: Wed, 6 May 2009 13:28:18 +0000 I havent done much IO at all in haskell, only within the function itself. However I want to get the input from the interface for the function and havent done this before. // Say I have this as my data type and list of films data Film = Film String String Int [String] -- List of films testDatabase :: [Film] testDatabase = [(Film "Casino Royale" "Martin Campbell" 2006 ["Garry", "Dave", "Zoe"]) ] // with functions such as: becomeFan :: String -> String -> [Film] -> [Film] becomeFan _ _ [] = [] becomeFan filmName fanName ((Film title director year fans):xs) | filmName == title = (Film title director year fanName:fans) : xs | otherwise = (Film title director year fans) : becomeFan filmName fanName xs filmsInGivenYear :: Int -> [Film] -> [String] filmsInGivenYear filmYear films = [ title | (Film title director year fans) <- films, year == filmYear] // I want to ask the user what function they want to use, I have this so far main :: IO() do putStr "Hi there! what is your name: " fanName = getLine do putStr "1 = Insert film, 2 = Become a Fan, 3 = The number of fans of a film, 4 = Film released in a year: " input = getLine read input :: Int (if input == 1 then main x = insertFilm [] else if input == 2 then main x = becomeFan [] else if input == 3 then main x = numberOfFans []) // I thought using the if statement would work, but now I cant think how to gather the needed input for the function they have chosen? thanks apple Get the New Internet Explore 8 Optimised for MSN. Download Now _________________________________________________________________ Share your photos with Windows Live Photos ? Free. http://clk.atdmt.com/UKM/go/134665338/direct/01/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090506/b8fe824e/attachment.htm From es at ertes.de Wed May 6 11:25:54 2009 From: es at ertes.de (Ertugrul Soeylemez) Date: Wed May 6 11:11:26 2009 Subject: [Haskell-beginners] Re: question >>= [style, efficiency] References: <5e0b968a0905060623k2e315345pda0f047326b4d35b@mail.gmail.com> Message-ID: <20090506172554.5dab4492@tritium.xx> Hello Joel, any algorithm to append something to a list must have at least O(n) complexity, where n is the length of the original list. That said, the third algorithm appears to be the best to me. It has exactly O(n) and is fully lazy, i.e. does not do anything at all until the list is consumed up to the appendage. Greets, Ertugrul. Joel Neely wrote: > I can think of a variety of ways to put a value at the end of a list > of the same type. > > -- cutesy-clever > putAtEnd :: [a] -> a -> [a] > putAtEnd xs x = reverse (x : reverse xs) > > -- by hand > atEnd :: [a] -> a -> [a] > atEnd [] z = [z] > atEnd (x:xs) z = x : atEnd xs z > > -- obvious brute force > infixr 9 +: > (+:) :: [a] -> a -> [a] > (+:) xs x = xs ++ [x] > > I'd appreciate feedback/advice on whether: > 1) I've missed any clearly-better alternatives; > 2) any of those is obviously better or worse than the others; > 3) I should have known where to look for the answer myself (instead of > brute-force Googling). > > I've semi-inadvertently mixed two issues above: infix-vs-prefix and > algorithm design. TIA for suggestions on either. > > -jn- > -- nightmare = unsafePerformIO (getWrongWife >>= sex) http://blog.ertes.de/ From haskell at colquitt.org Wed May 6 11:30:14 2009 From: haskell at colquitt.org (John Dorsey) Date: Wed May 6 11:15:36 2009 Subject: [Haskell-beginners] question >>= [style, efficiency] In-Reply-To: <5e0b968a0905060623k2e315345pda0f047326b4d35b@mail.gmail.com> References: <5e0b968a0905060623k2e315345pda0f047326b4d35b@mail.gmail.com> Message-ID: <20090506153014.GE3278@colquitt.org> Joel, > I can think of a variety of ways to put a value at the end of a list > of the same type. [...] > I'd appreciate feedback/advice on whether: > 1) I've missed any clearly-better alternatives; > 2) any of those is obviously better or worse than the others; > 3) I should have known where to look for the answer myself (instead of > brute-force Googling). Appending to a normal list is inherently messy and expensive. If it's doesn't have to be a normal list, though, things can get better. Look into "difference lists" as one example. Append for cheap while building, then get a list for linear-time cost when you're finished. The best data structure depends on what you need it to do. That said, your "obvious brute force" is the best of the ones you wrote, specifically because it's "obvious". It's clear what it does, it reuses existing functions well, and it's at least roughly as good as the others. John From byorgey at seas.upenn.edu Wed May 6 13:59:02 2009 From: byorgey at seas.upenn.edu (Brent Yorgey) Date: Wed May 6 13:44:21 2009 Subject: [Haskell-beginners] Context reducion Stack overflow In-Reply-To: <1241546836.5122.522.camel@zezinho> References: <1241449194.5122.181.camel@zezinho> <20090504195427.GA29604@seas.upenn.edu> <1241530090.5122.264.camel@zezinho> <20090505153358.GA15119@seas.upenn.edu> <1241546836.5122.522.camel@zezinho> Message-ID: <20090506175901.GA5176@seas.upenn.edu> On Tue, May 05, 2009 at 03:07:16PM -0300, Marco T?lio Gontijo e Silva wrote: > > and then your automatic lifting instance would be something like > > > > instance (F f, M f2 a) => M (f :.: f2) a > > > > where :.: denotes functor composition. > > Ok, I think this is another possibility. But how could I define :.:? newtype (g :.: f) a = O { unO :: g (f a) } Something like this is already defined in both the TypeCompose and category-extras libraries. -Brent From jack at realmode.com Wed May 6 21:33:18 2009 From: jack at realmode.com (Jack Kennedy) Date: Wed May 6 21:18:39 2009 Subject: [Haskell-beginners] parallel program in haskell in 5 steps In-Reply-To: <1008bfc90905051405v3ffc9da6tc828de863f6ee5f1@mail.gmail.com> References: <1008bfc90905051219q7760c5ew9dd95b1a2b87db7c@mail.gmail.com> <29bf512f0905051251u28c94d97rd8c3ddd43c89203a@mail.gmail.com> <1008bfc90905051405v3ffc9da6tc828de863f6ee5f1@mail.gmail.com> Message-ID: <1008bfc90905061833t53b7e8a4lddf4a8769c79071a@mail.gmail.com> Does this happen for everyone, or just me? On Tue, May 5, 2009 at 2:05 PM, Jack Kennedy wrote: > I am compiling (Windows by the way) using the line from the tutorial: > ghc -O2 --make par.hs -threaded > > and running with the line > > par +RTS -N2 > > CPU usage for the process flits around a little, but stays in the 45% - 55% > range. > > > On Tue, May 5, 2009 at 12:51 PM, Michael Snoyman wrote: > >> How are you running the program? You have to explicitly tell the >> compiler/interpreter to use multiple system threads. >> >> Michael >> >> On Tue, May 5, 2009 at 10:19 PM, Jack Kennedy wrote: >> >>> In step 4 of Haskell in 5 Steps [ >>> http://haskell.org/haskellwiki/Haskell_in_5_steps], >>> a parallel program is given. I changed it very slightly so it would run a >>> long time (see below). >>> >>> It compiles and runs but my CPU meter is barely above 50%. I have a dual >>> core processor. >>> What in the world would keep this program from completely saturating the >>> CPU? >>> >>> import Control.Parallel >>> >>> main = a `par` b `pseq` print (a + b) >>> where >>> a = ack 4 10 >>> b = ack 4 10 >>> >>> fac 0 = 1 >>> fac n = n * fac (n-1) >>> >>> ack 0 n = n+1 >>> ack m 0 = ack (m-1) 1 >>> ack m n = ack (m-1) (ack m (n-1)) >>> >>> fib 0 = 0 >>> fib 1 = 1 >>> fib n = fib (n-1) + fib (n-2) >>> >>> >>> _______________________________________________ >>> Beginners mailing list >>> Beginners@haskell.org >>> http://www.haskell.org/mailman/listinfo/beginners >>> >>> >> > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090506/b0d546ff/attachment-0001.htm From stevech1097 at yahoo.com.au Fri May 1 09:30:35 2009 From: stevech1097 at yahoo.com.au (Steve) Date: Wed May 6 22:15:03 2009 Subject: [Haskell-beginners] gcd In-Reply-To: <20090430173034.9E1253241DD@www.haskell.org> References: <20090430173034.9E1253241DD@www.haskell.org> Message-ID: <1241184635.4876.18.camel@host.localdomain> I had a look at the gcd definition in GHC 6.10.1 ghc-6.10.1/libraries/base/GHC/Real.lhs -- | @'gcd' x y@ is the greatest (positive) integer that divides both @x@ -- and @y@; for example @'gcd' (-3) 6@ = @3@, @'gcd' (-3) (-6)@ = @3@, -- @'gcd' 0 4@ = @4@. @'gcd' 0 0@ raises a runtime error. gcd :: (Integral a) => a -> a -> a gcd 0 0 = error "Prelude.gcd: gcd 0 0 is undefined" gcd x y = gcd' (abs x) (abs y) where gcd' a 0 = a gcd' a b = gcd' b (a `rem` b) Why is gcd 0 0 undefined? http://en.wikipedia.org/wiki/Greatest_common_divisor says: "It is useful to define gcd(0, 0) = 0 and lcm(0, 0) = 0 because then the natural numbers become a complete distributive lattice with gcd as meet and lcm as join operation. This extension of the definition is also compatible with the generalization for commutative rings given below." An added advantage, for haskell, of defining gcd 0 0 = 0 is that gcd would change from being a partial function to a total function. Regards, Steve From wingedtachikoma at gmail.com Wed May 6 23:25:43 2009 From: wingedtachikoma at gmail.com (Sean Bartell) Date: Wed May 6 23:11:02 2009 Subject: [Haskell-beginners] parallel program in haskell in 5 steps In-Reply-To: <1008bfc90905061833t53b7e8a4lddf4a8769c79071a@mail.gmail.com> References: <1008bfc90905051219q7760c5ew9dd95b1a2b87db7c@mail.gmail.com> <29bf512f0905051251u28c94d97rd8c3ddd43c89203a@mail.gmail.com> <1008bfc90905051405v3ffc9da6tc828de863f6ee5f1@mail.gmail.com> <1008bfc90905061833t53b7e8a4lddf4a8769c79071a@mail.gmail.com> Message-ID: > > Does this happen for everyone, or just me? I get the same result here. Changing a to fib 100 lets me get 65-75%. The compiler's probably just being smarter than you expect and combining both instances of ack 4 10. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090506/7b6f0769/attachment.htm From info at suud.de Thu May 7 12:36:08 2009 From: info at suud.de (Thomas Friedrich) Date: Thu May 7 12:21:52 2009 Subject: [Haskell-beginners] How to wait till a process is finished before invoking the next one? Message-ID: <4A030DF8.3040901@suud.de> Hi everyone, I have the following problem, and I hope that someone of yours might be able to help me. The Haskell program I am writing has the following setup: writeData :: [String] -> IO () writeData cs = ... runProgram:: [String] -> IO () runProgram cs = ... writeFeatures :: [String] -> IO () writeFeatures cs = ... runTestOnFeatures :: IO () runTestOnFeatures = ... main :: IO () main = do cs <- getArgs writeData cs runProgram cs writeFeatures cs runTestOnFeatures Each of the above function take a list of filenames, run certain command-line programs on them, which I invoke by runCommand, and each of them produce multiple output-files. Each function in main needs a couple of those output-files that are produced by the function directly above it. How do I get Haskell to wait, till all the data is written to the disk, before invoking the next command. The way the program is currently written, Haskell doesn't see that the input of one function depends on the output of another, and tries to run them all at the same time. Any ideas? Thanks everyone for your help. Cheers, Thomas From daniel.is.fischer at web.de Thu May 7 13:10:49 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Thu May 7 12:56:39 2009 Subject: [Haskell-beginners] How to wait till a process is finished before invoking the next one? In-Reply-To: <4A030DF8.3040901@suud.de> References: <4A030DF8.3040901@suud.de> Message-ID: <200905071910.50054.daniel.is.fischer@web.de> Am Donnerstag 07 Mai 2009 18:36:08 schrieb Thomas Friedrich: > Hi everyone, > > Each of the above function take a list of filenames, run certain > command-line programs on them, which I invoke by runCommand, and each of > them produce multiple output-files. Each function in main needs a > couple of those output-files that are produced by the function directly > above it. How do I get Haskell to wait, till all the data is written to > the disk, before invoking the next command. System.Process.waitForProcess should do it, conveniently runCommand returns a ProcessHandle. > The way the program is > currently written, Haskell doesn't see that the input of one function > depends on the output of another, and tries to run them all at the same > time. > > Any ideas? > > Thanks everyone for your help. > > Cheers, > Thomas From info at suud.de Thu May 7 14:53:11 2009 From: info at suud.de (Thomas Friedrich) Date: Thu May 7 14:38:53 2009 Subject: [Haskell-beginners] How to wait till a process is finished before invoking the next one? In-Reply-To: <200905071910.50054.daniel.is.fischer@web.de> References: <4A030DF8.3040901@suud.de> <200905071910.50054.daniel.is.fischer@web.de> Message-ID: <4A032E17.4030002@suud.de> Hi Daniel and everyone, Thanks for the reply! I thought of using waitForProcess, and in fact an earlier version of the program did. However, as the program got more complex, I don't really see how this is still possible. For example the runProgram function looks like this: runProgram :: [String] -> IO () runProgram [] = return () runProgram (c:cs) = do runCommand ("lalala " ++ c) runProgram cs It might be possible to write the function runProgram in a way, so that it returns an expression of type [IO ProcessHandle] and then try to work from there. But I have the feeling, that this will become messy very quickly, and there must be some more elegant way of doing this. The other thing is that actually not all functions are able to return ProcessHandles, e.g. writeFeatures :: [String] -> IO () writeFeatures cs = Exc.bracket (openFile training AppendMode) hClose (\h -> goo h) where goo h = go 1 cs where go :: Int -> [String] -> IO () go n [] = putStrLn "Features written." go n (c:cs) = do features <- makeFeatures n c -- makeFeatures :: Int -> String -> IO String hPutStr h features go (n+1) cs And the file that is produced here is needed in the next function. I hoped to do something with forkIO, as I would like to parallelize the whole program at the end. Especially the function runProgram would benefit hugely from this (I so don't have a clue how to do this yet;). I tried for example the following: main :: IO () main = do cs <- getArgs p1 <- forkIO $ writeData cs p2 <- forkIO $ runProgram cs p3 <- forkIO $ writeFeatures cs p4 <- forkIO $ runTestOnFeatures seq p1 (seq p2 (seq p3 (seq p4 (putStrLn "Done")))) But that of course doesn't work, because now I am not actually requesting anything. The program does in fact nothing, apart from printing out "Done". Any ideas? Cheers, Thomas Daniel Fischer wrote: > Am Donnerstag 07 Mai 2009 18:36:08 schrieb Thomas Friedrich: > >> Hi everyone, >> >> Each of the above function take a list of filenames, run certain >> command-line programs on them, which I invoke by runCommand, and each of >> them produce multiple output-files. Each function in main needs a >> couple of those output-files that are produced by the function directly >> above it. How do I get Haskell to wait, till all the data is written to >> the disk, before invoking the next command. >> > > System.Process.waitForProcess > > should do it, conveniently runCommand returns a ProcessHandle. > > >> The way the program is >> currently written, Haskell doesn't see that the input of one function >> depends on the output of another, and tries to run them all at the same >> time. >> >> Any ideas? >> >> Thanks everyone for your help. >> >> Cheers, >> Thomas >> > > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > From fitzpatrick.pj at googlemail.com Thu May 7 15:18:46 2009 From: fitzpatrick.pj at googlemail.com (PJ Fitzpatrick) Date: Thu May 7 15:04:10 2009 Subject: [Haskell-beginners] Real World Haskell Chapter 5 Message-ID: <32531bb40905071218k287b23b9xb1331d5af7fb0558@mail.gmail.com> Hi, I have compiled the SimpleJSON.hs file from chapter 5 and from another file attempted to import it. I am getting a could not find module error and the prelude tells me that the locations searched were SimpleJSON.hs and SimpleJSON.lhs. I am running Vista, GHC 6.10.2 Any ideas? tks, PJ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090507/6c391add/attachment.htm From daniel.is.fischer at web.de Thu May 7 16:04:51 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Thu May 7 15:50:39 2009 Subject: [Haskell-beginners] How to wait till a process is finished before invoking the next one? Message-ID: <200905072204.51750.daniel.is.fischer@web.de> Am Donnerstag 07 Mai 2009 20:53:11 schrieb Thomas Friedrich: > Hi Daniel and everyone, > > Thanks for the reply! > > I thought of using waitForProcess, and in fact an earlier version of the > program did. However, as the program got more complex, I don't really > see how this is still possible. > > For example the runProgram function looks like this: > > runProgram :: [String] -> IO () > runProgram [] = return () > runProgram (c:cs) = do > runCommand ("lalala " ++ c) > runProgram cs You could have runCommand ("lalala " ++ c) >>= waitForProcess in the penultimate line. That would ensure that the command has finished before its results are requested, but of course destroy all possibilities of parallelism :-( Another option would be runProgram [] = return [] runProgram (c:cs) = unsafeInterleaveIO $ do ph <- runCommand ("lalala " ++ c) phs <- runProgram cs return (ph:phs) and then, processHandles <- runProgram blah mapM_ waitForProcess (processHandles) next step that should work (I hope) and would allow the commands to be run in parallel while making sure all have finished before the next step is started. > > It might be possible to write the function runProgram in a way, so that > it returns an expression of type [IO ProcessHandle] and then try to work > from there. But I have the feeling, that this will become messy very > quickly, and there must be some more elegant way of doing this. The > other thing is that actually not all functions are able to return > ProcessHandles, e.g. > > writeFeatures :: [String] -> IO () > writeFeatures cs = Exc.bracket (openFile training AppendMode) hClose (\h > -> goo h) > where > goo h = go 1 cs > where > go :: Int -> [String] -> IO () > go n [] = putStrLn "Features written." > go n (c:cs) = do > features <- makeFeatures n c -- makeFeatures :: Int -> > String -> IO String > hPutStr h features > go (n+1) cs > > And the file that is produced here is needed in the next function. That can't appear in runProgram, though. I'm not sure it would work, but you coud give {-# LANGUAGE BangPatterns #-} ... !a <- writeFeatures blah more or go n (c:cs) = do features <- makeFeatures n c !a <- hPutStr h features go (n+1) cs a try. > > I hoped to do something with forkIO, as I would like to parallelize the > whole program at the end. Especially the function runProgram would > benefit hugely from this (I so don't have a clue how to do this yet;). I > tried for example the following: > > main :: IO () > main = do > cs <- getArgs > p1 <- forkIO $ writeData cs > p2 <- forkIO $ runProgram cs > p3 <- forkIO $ writeFeatures cs > p4 <- forkIO $ runTestOnFeatures > seq p1 (seq p2 (seq p3 (seq p4 (putStrLn "Done")))) > > But that of course doesn't work, because now I am not actually > requesting anything. The program does in fact nothing, apart from > printing out "Done". Huh. There doesn't seem to be a wait function for ThreadIds, so I guess you would have to communicate via MVars, QSemNs or some such means to signal that one task has been completed and the next can be started. > > Any ideas? > > Cheers, > Thomas From daniel.is.fischer at web.de Thu May 7 16:10:53 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Thu May 7 15:57:47 2009 Subject: [Haskell-beginners] Real World Haskell Chapter 5 In-Reply-To: <32531bb40905071218k287b23b9xb1331d5af7fb0558@mail.gmail.com> References: <32531bb40905071218k287b23b9xb1331d5af7fb0558@mail.gmail.com> Message-ID: <200905072210.53557.daniel.is.fischer@web.de> Am Donnerstag 07 Mai 2009 21:18:46 schrieb PJ Fitzpatrick: > Hi, > I have compiled the SimpleJSON.hs file from chapter 5 and from another file > attempted to import it. I am getting a could not find module error and the > prelude tells me that the locations searched were SimpleJSON.hs and > SimpleJSON.lhs. > I am running Vista, GHC 6.10.2 > Any ideas? > tks, > PJ If the other file is not in the same directory, you have to tell GHC where to look for it, ghc -ipath/to/SimpleJSONDir:path/to/OtherImport --make UseSimpleJSON or build a package with SimpleJSON in it (using Cabal) and register that, then ghc --make will know where to find it. If the file is in the same directory, it may be an encoding error (or some case-muck-up), but definitely something bad. From allbery at ece.cmu.edu Thu May 7 17:25:40 2009 From: allbery at ece.cmu.edu (Brandon S. Allbery KF8NH) Date: Thu May 7 17:11:12 2009 Subject: [Haskell-beginners] How to wait till a process is finished before invoking the next one? In-Reply-To: <4A032E17.4030002@suud.de> References: <4A030DF8.3040901@suud.de> <200905071910.50054.daniel.is.fischer@web.de> <4A032E17.4030002@suud.de> Message-ID: <3895BFDD-3C6C-46CB-ADBD-6712D7D971BC@ece.cmu.edu> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On May 7, 2009, at 14:53 , Thomas Friedrich wrote: > writeFeatures :: [String] -> IO () > writeFeatures cs = Exc.bracket (openFile training AppendMode) hClose > (\h -> goo h) > where > goo h = go 1 cs > where > go :: Int -> [String] -> IO () > go n [] = putStrLn "Features written." > go n (c:cs) = do > features <- makeFeatures n c -- makeFeatures :: Int -> > String -> IO String > hPutStr h features > go (n+1) cs > > And the file that is produced here is needed in the next function. You probably want to rethink how you're doing this. My own thought is that you have something like: > runThis :: FilePath -> [String] -> String -> IO (MVar String) > runThis cmd args inp = do > mvar <- newEmptyMVar > forkIO (readProcess cmd args inp >>= writeMVar mvar) > return mvar runOne launches in the background, you synchronize by doing takeMVar on the returned MVar (which will give you the output, if any). Even if there is no useful output you can still create data dependencies to insure things wait for what they need --- and that is exactly what you want to do: insure that there are data dependencies to constrain when programs are run. Otherwise, you'll have to settle for linear execution. - -- brandon s. allbery [solaris,freebsd,perl,pugs,haskell] allbery@kf8nh.com system administrator [openafs,heimdal,too many hats] allbery@ece.cmu.edu electrical and computer engineering, carnegie mellon university KF8NH -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.10 (Darwin) iEYEARECAAYFAkoDUeAACgkQIn7hlCsL25VH/gCghNldOSJnChoHrJwjeaboseU4 Z28An39bR2DAAlP6K9g00eb+NnHhKZSU =NHpz -----END PGP SIGNATURE----- From vze26m98 at optonline.net Thu May 7 19:06:37 2009 From: vze26m98 at optonline.net (Charles Turner) Date: Thu May 7 18:53:50 2009 Subject: [Haskell-beginners] Pros and Cons of Haskell libsndfile packages? Message-ID: <5F197860-D089-4387-9C75-E95259B54226@optonline.net> Hi all- I noticed that there were two different packages that provided a Haskell interface to the libsndfile library: HSoundFile: hsoundfile: Does anyone have any comments of the relative merits of either of these? Thanks! Charles From mle+hs at mega-nerd.com Thu May 7 20:09:25 2009 From: mle+hs at mega-nerd.com (Erik de Castro Lopo) Date: Thu May 7 19:54:46 2009 Subject: [Haskell-beginners] Pros and Cons of Haskell libsndfile packages? In-Reply-To: <5F197860-D089-4387-9C75-E95259B54226@optonline.net> References: <5F197860-D089-4387-9C75-E95259B54226@optonline.net> Message-ID: <20090508100925.37745149.mle+hs@mega-nerd.com> Charles Turner wrote: > Hi all- > > I noticed that there were two different packages that provided a > Haskell interface to the libsndfile library: > > hsndfile: HSoundFile: Does anyone have any comments of the relative merits of either of these? I'm probably biased, but I'm sure the former (hsndfile) opens more file formats and is more robust when parsing malformed files. Erik -- ---------------------------------------------------------------------- Erik de Castro Lopo http://www.mega-nerd.com/ From vze26m98 at optonline.net Fri May 8 13:40:55 2009 From: vze26m98 at optonline.net (Charles Turner) Date: Fri May 8 13:28:02 2009 Subject: [Haskell-beginners] Pros and Cons of Haskell libsndfile packages? In-Reply-To: <20090508100925.37745149.mle+hs@mega-nerd.com> References: <5F197860-D089-4387-9C75-E95259B54226@optonline.net> <20090508100925.37745149.mle+hs@mega-nerd.com> Message-ID: <38590B26-3068-4ECB-A4EF-1F1B847B8CB4@optonline.net> On May 7, 2009, at 8:09 PM, Erik de Castro Lopo wrote: > Charles Turner wrote: > That one is a indeed a haskell wrapper around libsndfile, while this > one: > >> HSoundFile: > Does anyone have any comments of the relative merits of either of >> these? > > I'm probably biased, but I'm sure the former (hsndfile) opens more > file > formats and is more robust when parsing malformed files. Hi Erik! In this case I think your bias is well-appreciated. Thanks so much for libsndfile, always one the first libraries I install on any new machine. Best, Charles From aditya.siram at gmail.com Fri May 8 14:22:23 2009 From: aditya.siram at gmail.com (aditya siram) Date: Fri May 8 14:07:44 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s Message-ID: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> Hi all, Is there a datatype in Haskell that can be used to represent only quantities >= 0? I got bitten by a bug because I forgot to reject an amount that was below zero after applying a decrementing operator. A simple unit test would have caught this, but I was wondering if there was some way of getting the type system to ensure this. thanks ... -deech -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090508/0a0a5703/attachment.htm From daniel.is.fischer at web.de Fri May 8 16:30:25 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Fri May 8 16:16:16 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s In-Reply-To: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> References: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> Message-ID: <200905082230.25873.daniel.is.fischer@web.de> Am Freitag 08 Mai 2009 20:22:23 schrieb aditya siram: > Hi all, > Is there a datatype in Haskell that can be used to represent only > quantities > > >= 0? I got bitten by a bug because I forgot to reject an amount that was > > below zero after applying a decrementing operator. A simple unit test would > have caught this, but I was wondering if there was some way of getting the > type system to ensure this. > > thanks ... > -deech I'm sure there are several implementations of a natural number datatype on Hackage (e.g. lazy Peano Numbers). And of course there are Word, Word32, Word64 (from Data.Word), but they wrap on underflow. Another option is to define an opaque type yourself: module Natural (Natural, mkNatural, maybeNatural) where newtype Natural = N Integer deriving (Eq, Ord, Show) mkNatural :: Integer -> Natural mkNatural n | n < 0 = error "Naturals must be non-negative" | otherwise = N n maybeNatural :: Integer -> Maybe Natural maybeNatural n | n < 0 = Nothing | otherwise = N n instance Num Natural where (N m) + (N n) = N (n+m) (N m) - (N n) = mkNatural (m-n) (N m) * (N n) = N (m*n) abs = id signum (N n) = if n == 0 then (N 0) else (N 1) negate n = error "Can't negate Naturals" fromInteger = mkNatural doesn't give you hard compile-time guarantees, but at least you get an error and not a hard-to-find bug. From daniel.carrera at theingots.org Fri May 8 16:54:29 2009 From: daniel.carrera at theingots.org (Daniel Carrera) Date: Fri May 8 16:39:46 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s In-Reply-To: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> References: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> Message-ID: <4A049C05.4080707@theingots.org> That's an interesting question. I would like to see the answer too. I can think of a completely impractical way to do it: data Quantity = 0 | 1 | 2 | 3 | ... | 65532 But of course, listing every integer seems like a dumb solution. But this gives me more confidence that there is a smart solution to your problem and we just need to find it. Daniel. aditya siram wrote: > Hi all, > Is there a datatype in Haskell that can be used to represent only > quantities >= 0? I got bitten by a bug because I forgot to reject an > amount that was below zero after applying a decrementing operator. A > simple unit test would have caught this, but I was wondering if there > was some way of getting the type system to ensure this. > > thanks ... > -deech > > > ------------------------------------------------------------------------ > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners From magnus at therning.org Fri May 8 18:06:37 2009 From: magnus at therning.org (Magnus Therning) Date: Fri May 8 17:52:06 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s In-Reply-To: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> References: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> Message-ID: <4A04ACED.5060405@therning.org> aditya siram wrote: > Hi all, > Is there a datatype in Haskell that can be used to represent only > quantities >= 0? I got bitten by a bug because I forgot to reject an > amount that was below zero after applying a decrementing operator. A > simple unit test would have caught this, but I was wondering if there > was some way of getting the type system to ensure this. Maybe Word32 (or one of it's siblings) would do? It's basically the same as 'unsigned int' in C, so it can under- and over-flow. /M -- Magnus Therning (OpenPGP: 0xAB4DFBA4) magnus?therning?org Jabber: magnus?therning?org http://therning.org/magnus identi.ca|twitter: magthe -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 197 bytes Desc: OpenPGP digital signature Url : http://www.haskell.org/pipermail/beginners/attachments/20090508/c11af4db/signature.bin From alexander.dunlap at gmail.com Fri May 8 18:36:55 2009 From: alexander.dunlap at gmail.com (Alexander Dunlap) Date: Fri May 8 18:22:30 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s In-Reply-To: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> References: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> Message-ID: <57526e770905081536x73e564a8i9edbfaecf9a38c7b@mail.gmail.com> On Fri, May 8, 2009 at 11:22 AM, aditya siram wrote: > Hi all, > Is there a datatype in Haskell that can be used to represent only quantities >>= 0?? I got bitten by a bug because I forgot to reject an amount that was > below zero after applying a decrementing operator. A simple unit test would > have caught this, but I was wondering if there was some way of getting the > type system to ensure this. > > thanks ... > -deech There are a number of ways of doing this, none of them entirely satisfactory. The essential problem is that subtraction becomes a partial function; i.e. if you try to subtract a and b, there's no way for the compiler to statically know if that's going to be nonnegative or not. One option would be to make a newtype of Integer, define (+), (*), etc. as with regular integers, define abs = id, negate = undefined. Then you can either make (-) a partial function (as in (NonNeg x) - (NonNeg y) = case x - y of z | z >= 0 -> z | otherwise -> undefined ) or leave subtraction undefined and make a maybeSubtract function that returns a Maybe value based on whether or not the result would be positive or negative. A third option would be to just take the absolute value of the result of a subtraction, although this should be used ONLY in situations where it is appropriate because usually you want to distinguish between x and -x. You could also use Peano numerals, but I don't think that those are very efficient. That's where you define data Nat = Zero | Succ Nat and so 3 :: Nat = Succ (Succ (Succ Zero)) etc. Hope that's of some help to you. Alex From patrick.leboutillier at gmail.com Fri May 8 21:53:04 2009 From: patrick.leboutillier at gmail.com (Patrick LeBoutillier) Date: Fri May 8 21:38:49 2009 Subject: [Haskell-beginners] SQL Lexer Message-ID: Hi all, In the process of writing an SQL parser I started by writing a lexer. The code can be found here: http://hpaste.org/fastcgi/hpaste.fcgi/view?id=4736#a4736 You can run it like this in ghci: Prelude SQL.Lexer> runLexer "select * from TABLE order by FIELD" [Token Reserved "select",Token Space " ",Token Operator "*",Token Space " ",Token Reserved "from",Token Space " ",Token Identifier "TABLE",Token Space " ",Token Reserved "order",Token Space " ",Token Reserved "by",Token Space " ",Token Identifier "FIELD"] Since this is pretty much my first Haskell project over 10 lines long, I'm looking for some feedback of any kind. Ultimately I would like to use this lexer to build a functional SQL parser using Parsec. Thanks, Patrick -- ===================== Patrick LeBoutillier Rosem?re, Qu?bec, Canada From scriptdevil.haskell at gmail.com Sat May 9 00:02:29 2009 From: scriptdevil.haskell at gmail.com (Ashok Gautham) Date: Fri May 8 23:47:47 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s In-Reply-To: <4A049C05.4080707@theingots.org> References: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> <4A049C05.4080707@theingots.org> Message-ID: On Sat, May 9, 2009 at 2:24 AM, Daniel Carrera wrote: > That's an interesting question. I would like to see the answer too. I can > think of a completely impractical way to do it: > > data Quantity = 0 | 1 | 2 | 3 | ... | 65532 > > > But of course, listing every integer seems like a dumb solution. But this > gives me more confidence that there is a smart solution to your problem and > we just need to find it. > > Daniel. What about operators? --- Ashok `ScriptDevil` Gautham From aditya.siram at gmail.com Sat May 9 03:07:10 2009 From: aditya.siram at gmail.com (aditya siram) Date: Sat May 9 02:52:27 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s In-Reply-To: <4A04ACED.5060405@therning.org> References: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> <4A04ACED.5060405@therning.org> Message-ID: <594f78210905090007v7381e280q430f609823e57a62@mail.gmail.com> Cool, that's really interesting! So is it accurate to say "predicated" datatypes [1] such as Int and String are provided by Haskell but you cannot create them or specialize on a subset of them? And there are really only two types of datatypes, tags [2] and the "predicated" types. Compound datatypes are just a combination of the two. -deech [1] By predicated I mean a datatype that would actually cause a compiler error ( not a runtime error ) if a method returned an out-of-bounds value. [2] By tags I mean the data constructors, but I find it easier to think of them as tags because they can appear alone, eg. data Coins = Penny | Nickel | Dime | Quarter where nothing is being constructed. On Fri, May 8, 2009 at 5:06 PM, Magnus Therning wrote: > aditya siram wrote: > >> Hi all, >> Is there a datatype in Haskell that can be used to represent only >> quantities >= 0? I got bitten by a bug because I forgot to reject an amount >> that was below zero after applying a decrementing operator. A simple unit >> test would have caught this, but I was wondering if there was some way of >> getting the type system to ensure this. >> > > Maybe Word32 (or one of it's siblings) would do? It's basically the same > as 'unsigned int' in C, so it can under- and over-flow. > > /M > > -- > Magnus Therning (OpenPGP: 0xAB4DFBA4) > magnus?therning?org Jabber: magnus?therning?org > http://therning.org/magnus identi.ca|twitter: magthe > > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090509/1c242a87/attachment.htm From daniel.carrera at theingots.org Sat May 9 03:37:53 2009 From: daniel.carrera at theingots.org (Daniel Carrera) Date: Sat May 9 03:23:11 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s In-Reply-To: <4A04ACED.5060405@therning.org> References: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> <4A04ACED.5060405@therning.org> Message-ID: <4A0532D1.9030403@theingots.org> Magnus Therning wrote: > aditya siram wrote: >> Hi all, >> Is there a datatype in Haskell that can be used to represent only >> quantities >= 0? I got bitten by a bug because I forgot to reject an >> amount that was below zero after applying a decrementing operator. A >> simple unit test would have caught this, but I was wondering if there >> was some way of getting the type system to ensure this. > > Maybe Word32 (or one of it's siblings) would do? It's basically the > same as 'unsigned int' in C, so it can under- and over-flow. What's under-flow? From keithshep at gmail.com Sat May 9 08:56:56 2009 From: keithshep at gmail.com (Keith Sheppard) Date: Sat May 9 08:42:13 2009 Subject: [Haskell-beginners] SQL Lexer In-Reply-To: References: Message-ID: <92e42b740905090556g4644f7ax71389bad2c0e6710@mail.gmail.com> This may be a better approach then my code. I didn't create a separate parser/lexer. -Keith On Fri, May 8, 2009 at 9:53 PM, Patrick LeBoutillier wrote: > Hi all, > > In the process of writing an SQL parser I started by writing a lexer. > The code can be found here: > http://hpaste.org/fastcgi/hpaste.fcgi/view?id=4736#a4736 > > You can run it like this in ghci: > > Prelude SQL.Lexer> runLexer "select * from TABLE order by FIELD" > [Token Reserved "select",Token Space " ",Token Operator "*",Token > Space " ",Token Reserved "from",Token Space " ",Token Identifier > "TABLE",Token Space " ",Token Reserved "order",Token Space " ",Token > Reserved "by",Token Space " ",Token Identifier "FIELD"] > > Since this is pretty much my first Haskell project over 10 lines long, > I'm looking for some feedback of any kind. > Ultimately I would like to use this lexer to build a functional SQL > parser using Parsec. > > > Thanks, > > Patrick > > -- > ===================== > Patrick LeBoutillier > Rosem?re, Qu?bec, Canada > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > From magnus at therning.org Sat May 9 12:30:02 2009 From: magnus at therning.org (Magnus Therning) Date: Sat May 9 12:16:28 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s In-Reply-To: <4A0532D1.9030403@theingots.org> References: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> <4A04ACED.5060405@therning.org> <4A0532D1.9030403@theingots.org> Message-ID: <4A05AF8A.9060000@therning.org> Daniel Carrera wrote: > Magnus Therning wrote: >> aditya siram wrote: >>> Hi all, >>> Is there a datatype in Haskell that can be used to represent only >>> quantities >= 0? I got bitten by a bug because I forgot to reject an >>> amount that was below zero after applying a decrementing operator. A >>> simple unit test would have caught this, but I was wondering if there >>> was some way of getting the type system to ensure this. >> >> Maybe Word32 (or one of it's siblings) would do? It's basically the >> same as 'unsigned int' in C, so it can under- and over-flow. > > What's under-flow? Negative overflow: > let one = 1 :: Word8 > one - 2 255 /M -- Magnus Therning (OpenPGP: 0xAB4DFBA4) magnus?therning?org Jabber: magnus?therning?org http://therning.org/magnus identi.ca|twitter: magthe -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 197 bytes Desc: OpenPGP digital signature Url : http://www.haskell.org/pipermail/beginners/attachments/20090509/2dd40e5a/signature.bin From kevin.haines at ntlworld.com Sat May 9 12:50:33 2009 From: kevin.haines at ntlworld.com (Kevin Haines) Date: Sat May 9 12:36:57 2009 Subject: [Haskell-beginners] Performance of function defined in a 'where' clause Message-ID: <4A05B459.3070903@ntlworld.com> Hi All, I'm trying to write a bit of code that maps each byte in a block of Word8's to 3xWord8 using an array; i.e. mapping from 8 bit to 24 bit colour (this is an OpenGL application, and I'm using textures). I should point out that this is experimental code, and I'm still learning Haskell (and *loving* it, by the way!), so it probably looks a little unpolished. First, some data: data Palette = Palette { palRed :: Word8, palGrn :: Word8, palBlu :: Word8 } palette = listArray (0,49) paletteList paletteList = [ Palette 0 0 0, Palette 0 0 0, Palette 0 0 0, ..... Then, my first implementation, which took 57% time under profiling, was: loadTile :: Int -> Int -> IO (Ptr Word8) loadTile lat lon = do terrainBytes <- readTile lat lon -- implementation #1 mapM_ (paletteMapper terrainBytes rgbBytes) [0..tileSize^2-1] free terrainBytes return rgbBytes where tileSize = 128 paletteMapper :: Ptr Word8 -> Ptr Word8 -> Int -> IO () paletteMapper tb rgb idx = do v <- peekElemOff tb idx pokeByteOff rgb (idx*3) (palRed (palette!v)) pokeByteOff rgb (idx*3+1) (palGrn (palette!v)) pokeByteOff rgb (idx*3+2) (palBlu (palette!v)) I tried moving paletterMapper out of the 'where' clause and into the top level, which then took only 26% of time - i.e. half the time: paletteMapper :: Ptr Word8 -> Ptr Word8 -> Int -> IO () paletteMapper tb rgb idx = do v <- peekElemOff tb idx pokeByteOff rgb (idx*3) (palRed (palette!v)) pokeByteOff rgb (idx*3+1) (palGrn (palette!v)) pokeByteOff rgb (idx*3+2) (palBlu (palette!v)) loadTile :: Int -> Int -> IO (Ptr Word8) loadTile lat lon = do terrainBytes <- readTile lat lon -- implementation #1 mapM_ (paletteMapper terrainBytes rgbBytes) [0..tileSize^2-1] free terrainBytes return rgbBytes where tileSize = 128 I don't understand why - the functions are the same, except for the scope they're in. Can anyone elaborate on what's happening? Incidentally, I now realise a faster way (14%) is: loadTile :: Int -> Int -> IO (Ptr Word8) loadTile lat lon = do terrainBytes <- readTile lat lon rgbBytes <- mallocBytes (3*(tileSize^2)) mapM_ (\x -> do v <- peekElemOff terrainBytes x pokeByteOff rgbBytes (x*3) (palRed (palette!v)) pokeByteOff rgbBytes (x*3+1) (palGrn (palette!v)) pokeByteOff rgbBytes (x*3+2) (palBlu (palette!v)) ) [0..tileSize^2-1] free terrainBytes return rgbBytes where tileSize = 128 (There may be faster/better ways still, I'm all ears :-) Cheers Kevin From byorgey at seas.upenn.edu Sat May 9 21:11:11 2009 From: byorgey at seas.upenn.edu (Brent Yorgey) Date: Sat May 9 20:56:37 2009 Subject: [Haskell-beginners] Performance of function defined in a 'where' clause In-Reply-To: <4A05B459.3070903@ntlworld.com> References: <4A05B459.3070903@ntlworld.com> Message-ID: <20090510011111.GA2728@seas.upenn.edu> On Sat, May 09, 2009 at 05:50:33PM +0100, Kevin Haines wrote: > > palette = listArray (0,49) paletteList > paletteList = [ > Palette 0 0 0, > Palette 0 0 0, > Palette 0 0 0, > ..... By the way, this should probably be something like palette = listArray (0,49) (replicate 50 $ Palette 0 0 0) Unless, of course, there are things other than Palette 0 0 0 in the remainder of the list. If you ever have to copy and paste *anything* in a Haskell program, warning bells should start going off in your head. =) Unfortunately, performance tuning is a black art of which I know precious little. Perhaps others will have some good suggestions. -Brent From andy.elvey at paradise.net.nz Sat May 9 22:37:07 2009 From: andy.elvey at paradise.net.nz (Andy Elvey) Date: Sat May 9 22:22:14 2009 Subject: [Haskell-beginners] SQL Lexer In-Reply-To: References: Message-ID: <4A063DD3.8080405@paradise.net.nz> Patrick LeBoutillier wrote: > Hi all, > > In the process of writing an SQL parser I started by writing a lexer. > The code can be found here: > http://hpaste.org/fastcgi/hpaste.fcgi/view?id=4736#a4736 > > You can run it like this in ghci: > > Prelude SQL.Lexer> runLexer "select * from TABLE order by FIELD" > [Token Reserved "select",Token Space " ",Token Operator "*",Token > Space " ",Token Reserved "from",Token Space " ",Token Identifier > "TABLE",Token Space " ",Token Reserved "order",Token Space " ",Token > Reserved "by",Token Space " ",Token Identifier "FIELD"] > > Since this is pretty much my first Haskell project over 10 lines long, > I'm looking for some feedback of any kind. > Ultimately I would like to use this lexer to build a functional SQL > parser using Parsec. > > > Thanks, > > Patrick > > Hi Patrick - I like it! I'm still a Haskell beginner, but even to me, your code seems very clear and easy to understand. I've been thinking of doing some simple parsers too, so I was wondering - may I use this code of yours as a base for them? Well done, and thanks for doing this! - Andy From bugfact at gmail.com Sun May 10 07:41:14 2009 From: bugfact at gmail.com (Peter Verswyvelen) Date: Sun May 10 07:26:28 2009 Subject: [Haskell-beginners] Performance of function defined in a 'where' clause In-Reply-To: <4A05B459.3070903@ntlworld.com> References: <4A05B459.3070903@ntlworld.com> Message-ID: Did you compile with -O or -O2? On Sat, May 9, 2009 at 6:50 PM, Kevin Haines wrote: > Hi All, > > I'm trying to write a bit of code that maps each byte in a block of Word8's > to 3xWord8 using an array; i.e. mapping from 8 bit to 24 bit colour (this is > an OpenGL application, and I'm using textures). > > I should point out that this is experimental code, and I'm still learning > Haskell (and *loving* it, by the way!), so it probably looks a little > unpolished. > > First, some data: > > data Palette = Palette { palRed :: Word8, palGrn :: Word8, palBlu :: Word8 > } > > palette = listArray (0,49) paletteList > paletteList = [ > Palette 0 0 0, > Palette 0 0 0, > Palette 0 0 0, > ..... > > > > Then, my first implementation, which took 57% time under profiling, was: > > loadTile :: Int -> Int -> IO (Ptr Word8) > loadTile lat lon = do > terrainBytes <- readTile lat lon > > -- implementation #1 > mapM_ (paletteMapper terrainBytes rgbBytes) [0..tileSize^2-1] > > free terrainBytes > return rgbBytes > > where tileSize = 128 > paletteMapper :: Ptr Word8 -> Ptr Word8 -> Int -> IO () > paletteMapper tb rgb idx = do > v <- peekElemOff tb idx > pokeByteOff rgb (idx*3) (palRed (palette!v)) > pokeByteOff rgb (idx*3+1) (palGrn (palette!v)) > pokeByteOff rgb (idx*3+2) (palBlu (palette!v)) > > > I tried moving paletterMapper out of the 'where' clause and into the top > level, which then took only 26% of time - i.e. half the time: > > > paletteMapper :: Ptr Word8 -> Ptr Word8 -> Int -> IO () > paletteMapper tb rgb idx = do > v <- peekElemOff tb idx > pokeByteOff rgb (idx*3) (palRed (palette!v)) > pokeByteOff rgb (idx*3+1) (palGrn (palette!v)) > pokeByteOff rgb (idx*3+2) (palBlu (palette!v)) > > loadTile :: Int -> Int -> IO (Ptr Word8) > loadTile lat lon = do > terrainBytes <- readTile lat lon > > -- implementation #1 > mapM_ (paletteMapper terrainBytes rgbBytes) [0..tileSize^2-1] > > free terrainBytes > return rgbBytes > > where tileSize = 128 > > > I don't understand why - the functions are the same, except for the scope > they're in. Can anyone elaborate on what's happening? > > > Incidentally, I now realise a faster way (14%) is: > > loadTile :: Int -> Int -> IO (Ptr Word8) > loadTile lat lon = do > > terrainBytes <- readTile lat lon > > rgbBytes <- mallocBytes (3*(tileSize^2)) > mapM_ (\x -> do > v <- peekElemOff terrainBytes x > pokeByteOff rgbBytes (x*3) (palRed (palette!v)) > pokeByteOff rgbBytes (x*3+1) (palGrn (palette!v)) > pokeByteOff rgbBytes (x*3+2) (palBlu (palette!v)) > ) [0..tileSize^2-1] > > > free terrainBytes > return rgbBytes > > where tileSize = 128 > > > (There may be faster/better ways still, I'm all ears :-) > > Cheers > > Kevin > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090510/47998c0d/attachment-0001.htm From kevin.haines at ntlworld.com Sun May 10 11:33:12 2009 From: kevin.haines at ntlworld.com (Kevin Haines) Date: Sun May 10 11:18:08 2009 Subject: [Haskell-beginners] Performance of function defined in a 'where' clause In-Reply-To: References: <4A05B459.3070903@ntlworld.com> Message-ID: <4A06F3B8.5020309@ntlworld.com> I used -O. Also, to answer Brent's posting, the Palette list is different further down but I did take the tip and replace the pasted lines with replicate, so thanks! Cheers Kevin Peter Verswyvelen wrote: > Did you compile with -O or -O2? > > On Sat, May 9, 2009 at 6:50 PM, Kevin Haines > wrote: > From patrick.leboutillier at gmail.com Sun May 10 12:55:49 2009 From: patrick.leboutillier at gmail.com (patrick.leboutillier@gmail.com) Date: Sun May 10 12:40:59 2009 Subject: [Haskell-beginners] SQL Lexer In-Reply-To: <4A063DD3.8080405@paradise.net.nz> Message-ID: <0015174c4458d8819d046991bba9@google.com> > Hi Patrick - I like it! I'm still a Haskell beginner, but even to me, > your code seems very clear and easy to understand. I've been thinking of > doing some simple parsers too, so I was wondering - may I use this code > of yours as a base for them? Sure, anyone here can use the code if they have any interest in it. Patrick > Well done, and thanks for doing this! > - Andy -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090510/62c0074c/attachment.html From jakubuv at gmail.com Mon May 11 06:33:20 2009 From: jakubuv at gmail.com (Jan Jakubuv) Date: Tue May 12 04:09:40 2009 Subject: [Haskell-beginners] SQL Lexer In-Reply-To: References: Message-ID: <20090511103320.GA11518@lxultra2.macs.hw.ac.uk> Hello Patrick, that's a nice one. I have two notes for you: 1) Why do you handle spaces as tokens? Can't you just skip them completely? It could make the parser simpler. 2) Your lexer does not provide any information about positions of tokens in the source code. You may want to do that in order to: a) report better error messages b) use parsec's `token` function [1] to make parsec work with your tokens. Note that parsec wants you to use its data type `SourcePos` to represent positions. See [2] for a simple example and further stuff. [1]: http://legacy.cs.uu.nl/daan/download/parsec/parsec.html#token [2]: http://legacy.cs.uu.nl/daan/download/parsec/parsec.html#SeperateScanners Sincerely, Jan. On Fri, May 08, 2009 at 09:53:04PM -0400, Patrick LeBoutillier wrote: > Hi all, > > In the process of writing an SQL parser I started by writing a lexer. > The code can be found here: > http://hpaste.org/fastcgi/hpaste.fcgi/view?id=4736#a4736 > > You can run it like this in ghci: > > Prelude SQL.Lexer> runLexer "select * from TABLE order by FIELD" > [Token Reserved "select",Token Space " ",Token Operator "*",Token > Space " ",Token Reserved "from",Token Space " ",Token Identifier > "TABLE",Token Space " ",Token Reserved "order",Token Space " ",Token > Reserved "by",Token Space " ",Token Identifier "FIELD"] > > Since this is pretty much my first Haskell project over 10 lines long, > I'm looking for some feedback of any kind. > Ultimately I would like to use this lexer to build a functional SQL > parser using Parsec. > > > Thanks, > > Patrick > > -- > ===================== > Patrick LeBoutillier > Rosem?re, Qu?bec, Canada > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners -- Heriot-Watt University is a Scottish charity registered under charity number SC000278. From patrick.leboutillier at gmail.com Mon May 11 09:28:53 2009 From: patrick.leboutillier at gmail.com (Patrick LeBoutillier) Date: Tue May 12 04:21:43 2009 Subject: [Haskell-beginners] SQL Lexer In-Reply-To: <20090511103320.GA11518@lxultra2.macs.hw.ac.uk> References: <20090511103320.GA11518@lxultra2.macs.hw.ac.uk> Message-ID: On Mon, May 11, 2009 at 6:33 AM, Jan Jakubuv wrote: > Hello Patrick, > > that's a nice one. I have two notes for you: > > 1) Why do you handle spaces as tokens? Can't you just skip them completely? > ? It could make the parser simpler. I added that as a debug tool (that way I can recreate the SQL exactly and use diff to make sure it tokenised properly). I agree that besides that it's not very useful. > > 2) Your lexer does not provide any information about positions of tokens in > ? the source code. You may want to do that in order to: > > ? a) report better error messages > ? b) use parsec's `token` function [1] to make parsec work with your tokens. > ? ? ?Note that parsec wants you to use its data type `SourcePos` to represent > ? ? ?positions. See [2] for a simple example and further stuff. > > [1]: http://legacy.cs.uu.nl/daan/download/parsec/parsec.html#token > [2]: http://legacy.cs.uu.nl/daan/download/parsec/parsec.html#SeperateScanners That's for phase 2... :). Actually I was planning on adding that in later. Thanks a lot for the links, I'll have a look at them. Thanks a lot for your feedback, Patrick > > Sincerely, > ? Jan. > > On Fri, May 08, 2009 at 09:53:04PM -0400, Patrick LeBoutillier wrote: >> Hi all, >> >> In the process of writing an SQL parser I started by writing a lexer. >> The code can be found here: >> http://hpaste.org/fastcgi/hpaste.fcgi/view?id=4736#a4736 >> >> You can run it like this in ghci: >> >> Prelude SQL.Lexer> runLexer "select * from TABLE order by FIELD" >> [Token Reserved "select",Token Space " ",Token Operator "*",Token >> Space " ",Token Reserved "from",Token Space " ",Token Identifier >> "TABLE",Token Space " ",Token Reserved "order",Token Space " ",Token >> Reserved "by",Token Space " ",Token Identifier "FIELD"] >> >> Since this is pretty much my first Haskell project over 10 lines long, >> I'm looking for some feedback of any kind. >> Ultimately I would like to use this lexer to build a functional SQL >> parser using Parsec. >> >> >> Thanks, >> >> Patrick >> >> -- >> ===================== >> Patrick LeBoutillier >> Rosem?re, Qu?bec, Canada >> _______________________________________________ >> Beginners mailing list >> Beginners@haskell.org >> http://www.haskell.org/mailman/listinfo/beginners > > > -- > Heriot-Watt University is a Scottish charity > registered under charity number SC000278. > > -- ===================== Patrick LeBoutillier Rosem?re, Qu?bec, Canada From mpm at alumni.caltech.edu Mon May 11 10:13:36 2009 From: mpm at alumni.caltech.edu (Michael Mossey) Date: Tue May 12 04:24:33 2009 Subject: [Haskell-beginners] defining 'init' in terms of 'foldr' Message-ID: <4A083290.3010509@alumni.caltech.edu> In S. Thompson's book, problem 9.13 asks us to define 'init' in terms of foldr. I was baffled at first because I didn't see a natural way to do this. It would look something like init xs = foldr f initialValue xs where f would cons on each character except the rightmost. f b = [] f b = a : b How does f "know" when it is passed the first character? initialValue has to signal this somehow. On #haskell, one person suggested doing it with some post-processing: init xs = snd $ foldr f (True,[]) xs where f _ (True,_) = (False,[]) f a (False,b) = (False,a:b) I had an idea. If the initial value is the entire list, then its length can function as the "signal" that we are dealing with the rightmost char. This requires no post-processing: init xs = foldr f xs xs where f a b | length b == length xs = [] | otherwise = a:b These seem contrived. I wonder if there is a more natural solution that Thompson had in mind. Any comments? -Mike From harold.stenger at gmail.com Mon May 11 17:03:44 2009 From: harold.stenger at gmail.com (Haroldo Stenger) Date: Tue May 12 04:45:58 2009 Subject: [Haskell-beginners] SQL Lexer In-Reply-To: References: Message-ID: <3678e0bb0905111403n50ae6f6eydd0c6e9d2002f012@mail.gmail.com> patrick, i was wondering if parts of the code could be more abstract. (i'm no genius in that area) but how about using some high order function for the numeric literal or the alpohanumeric literal? maybe a lexer is a hand-wise algorithm, but maybe it can be streched somehow :-) comments? best regards haroldo 2009/5/8 Patrick LeBoutillier > Hi all, > > In the process of writing an SQL parser I started by writing a lexer. > The code can be found here: > http://hpaste.org/fastcgi/hpaste.fcgi/view?id=4736#a4736 > > You can run it like this in ghci: > > Prelude SQL.Lexer> runLexer "select * from TABLE order by FIELD" > [Token Reserved "select",Token Space " ",Token Operator "*",Token > Space " ",Token Reserved "from",Token Space " ",Token Identifier > "TABLE",Token Space " ",Token Reserved "order",Token Space " ",Token > Reserved "by",Token Space " ",Token Identifier "FIELD"] > > Since this is pretty much my first Haskell project over 10 lines long, > I'm looking for some feedback of any kind. > Ultimately I would like to use this lexer to build a functional SQL > parser using Parsec. > > > Thanks, > > Patrick > > -- > ===================== > Patrick LeBoutillier > Rosem?re, Qu?bec, Canada > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090511/1d87cc00/attachment.html From byorgey at seas.upenn.edu Mon May 11 20:39:30 2009 From: byorgey at seas.upenn.edu (Brent Yorgey) Date: Tue May 12 04:55:24 2009 Subject: [Haskell-beginners] Performance of function defined in a 'where' clause In-Reply-To: <4A06F3B8.5020309@ntlworld.com> References: <4A05B459.3070903@ntlworld.com> <4A06F3B8.5020309@ntlworld.com> Message-ID: <20090512003930.GA23572@seas.upenn.edu> On Sun, May 10, 2009 at 04:33:12PM +0100, Kevin Haines wrote: > I used -O. Well, -O2 is almost always much better than -O. (And -O3 is worse, or at least used to be. Don't ask why.) -Brent From chaddai.fouche at gmail.com Mon May 11 23:55:24 2009 From: chaddai.fouche at gmail.com (=?UTF-8?B?Q2hhZGRhw68gRm91Y2jDqQ==?=) Date: Tue May 12 05:04:01 2009 Subject: [Haskell-beginners] A Quantity Type - Integer without the Negative #'s In-Reply-To: <594f78210905090007v7381e280q430f609823e57a62@mail.gmail.com> References: <594f78210905081122l31cc683aof7c0661c6ff56e2b@mail.gmail.com> <4A04ACED.5060405@therning.org> <594f78210905090007v7381e280q430f609823e57a62@mail.gmail.com> Message-ID: 2009/5/9 aditya siram : > Cool, that's really interesting! So is it accurate to say "predicated" > datatypes [1] such as Int and String are provided by Haskell but you cannot > create them or specialize on a subset of them? And there are really only two > types of datatypes, tags [2] and the "predicated" types. Compound datatypes > are just a combination of the two. > > -deech > > > [1] By predicated I mean a datatype that would actually cause a compiler > error ( not a runtime error ) if a method returned an out-of-bounds value. > [2] By? tags I mean the data constructors, but I find it easier to think of > them as tags because they can appear alone, eg. data Coins = Penny | Nickel > | Dime | Quarter > where nothing is being constructed. I don't see where the "predicated datatype" differs from the tags type : Int could as well be implemented as a "data Int = 0 | 1 | 2 |..." (sparing consideration about characters authorized in a tag), String isn't primitive at all, in fact it's just a list of characters.... You can easily create them. To specialize on a subset of them is very easy too, you can use newtype or data and create a smart constructor, as well as appropriate instances. The only problem is that it is pretty hard to get compile time guarantee on some of them but no other classic language provides more guarantee : the only languages that can do better are those that have dependant types (Agda, Coq, Epigram...) and as you'll know if you tried them they're frequently harder to use than Haskell. NB : You can use Template Haskell to get some compile-time errors for the constants of your code. -- Jeda? From apfelmus at quantentunnel.de Tue May 12 07:31:32 2009 From: apfelmus at quantentunnel.de (Heinrich Apfelmus) Date: Tue May 12 07:16:55 2009 Subject: [Haskell-beginners] Re: defining 'init' in terms of 'foldr' In-Reply-To: <4A083290.3010509@alumni.caltech.edu> References: <4A083290.3010509@alumni.caltech.edu> Message-ID: Michael Mossey wrote: > In S. Thompson's book, problem 9.13 asks us to define 'init' in terms of > foldr. I was baffled at first because I didn't see a natural way to do > this. It would look something like > > init xs = foldr f initialValue xs > > where f would cons on each character except the rightmost. > > f b = [] > f b = a : b > > How does f "know" when it is passed the first character? initialValue > has to signal this somehow. On #haskell, one person suggested doing it > with some post-processing: > > init xs = snd $ foldr f (True,[]) xs > where f _ (True,_) = (False,[]) > f a (False,b) = (False,a:b) > > I had an idea. If the initial value is the entire list, then its length > can function as the "signal" that we are dealing with the rightmost > char. This requires no post-processing: > > init xs = foldr f xs xs > where f a b | length b == length xs = [] > | otherwise = a:b > > These seem contrived. I wonder if there is a more natural solution that > Thompson had in mind. Any comments? It is best to see foldr f b as an operation that takes a list x0 : x1 : x2 : ... : [] and replaces every (:) with f and the [] with b : x0 `f` x1 `f` x2 `f` ... `f` b See also http://en.wikipedia.org/wiki/Fold_(higher-order_function) for Cale's nice pictures. It is then clear that we have to choose b to signal the end of the list. Furthermore, b should be the same as init [] . Unfortunately, this expression is a run-time error, but this is a fault of the type signature init :: [a] -> [a] which should really be init' :: [a] -> Maybe [a] to make it clear that some lists like the empty one simply don't have an initial segment. And this version has a natural implementation in terms of foldr : init' = foldr f Nothing where f _ Nothing = Just [] f x (Just xs) = Just (x:xs) Of course, we need some post-processing to obtain the original init from this, but I think that it's very natural. Regards, apfelmus -- http://apfelmus.nfshost.com From daniel.is.fischer at web.de Tue May 12 08:27:30 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Tue May 12 08:13:11 2009 Subject: [Haskell-beginners] defining 'init' in terms of 'foldr' In-Reply-To: <4A083290.3010509@alumni.caltech.edu> References: <4A083290.3010509@alumni.caltech.edu> Message-ID: <200905121427.30834.daniel.is.fischer@web.de> Am Montag 11 Mai 2009 16:13:36 schrieb Michael Mossey: > In S. Thompson's book, problem 9.13 asks us to define 'init' in terms of > foldr. Check the thread starting at http://www.haskell.org/pipermail/haskell-cafe/2005-April/009562.html That contains several interesting approaches, though I don't think any of those was lazy enough to deal with infinite lists. > I was baffled at first because I didn't see a natural way to do this. > It would look something like > > init xs = foldr f initialValue xs Since *FoldInit> init [] *** Exception: Prelude.init: empty list initialValue has to be (error "Prelude.init: empty list") if you don't do any post- processing. But then init [1] = f 1 (error "...") must be [], so f can't inspect its second argument (or it would return _|_ instead of []), but then you can't make init [1] = f 1 (error "...") return [] and also init [1,2] = f 1 (f 2 (error "...")) return [1]. So some post-processing is necessary. > > where f would cons on each character except the rightmost. > > f b = [] > f b = a : b > > How does f "know" when it is passed the first character? initialValue has > to signal this somehow. On #haskell, one person suggested doing it with > some post-processing: > > init xs = snd $ foldr f (True,[]) xs > where f _ (True,_) = (False,[]) > f a (False,b) = (False,a:b) That gives init [] = [], which is not correct. The starting value must be (True, error "..."). It doesn't work on infinite lists, and will produce a stack overflow on sufficiently long lists. The problem is that pattern matching is strict, so to determine which brach to take in f 1 (init [2 .. n]) we must know whether the first component of init [2 .. n] is True or False. So we must look at f 2 (init [3 .. n]), same problem there... Before *anything* of the overall result can be determined, the whole list has to be traversed. We can fix these issues by making it lazier: vinit = snd . foldr f (True, error "Prelude.init: empty list") where f x y = (False, if fst y then [] else x:snd y) Here, the first component of f's result is determined before looking at f's arguments, thus to determine which branch to take in the second component, all that has to be done is check if y comes from an application of f or the initial value. Works for infinite lists, errors on empty lists, specs fulfilled. > > I had an idea. If the initial value is the entire list, then its length can > function as the "signal" that we are dealing with the rightmost char. This > requires no post-processing: > > init xs = foldr f xs xs > where f a b | length b == length xs = [] > > | otherwise = a:b > That gives init [] = [], doesn't work on infinite lists and is badly inefficient. To fix at least two of these issues, you must do some post-processing, but that would lead again to something like above. > These seem contrived. I wonder if there is a more natural solution that > Thompson had in mind. Any comments? > > -Mike > From info at suud.de Tue May 12 12:27:06 2009 From: info at suud.de (Thomas Friedrich) Date: Tue May 12 12:12:49 2009 Subject: [Haskell-beginners] The main thread and waiting for other threads Message-ID: <4A09A35A.3020909@suud.de> Hi everyone, I have a problem with the following example in the Real World Haskell book, which aims to develop a module for controlling different threads. See, http://book.realworldhaskell.org/read/concurrent-and-multicore-programming.html in the chapter "The main thread and waiting for other threads". -- file: ch24/NiceFork.hs import Control.Concurrent import Control.Exception (Exception, try) import qualified Data.Map as M data ThreadStatus = Running | Finished -- terminated normally | Threw Exception -- killed by uncaught exception deriving (Eq, Show) -- | Create a new thread manager. newManager :: IO ThreadManager -- | Create a new managed thread. forkManaged :: ThreadManager -> IO () -> IO ThreadId -- | Immediately return the status of a managed thread. getStatus :: ThreadManager -> ThreadId -> IO (Maybe ThreadStatus) -- | Block until a specific managed thread terminates. waitFor :: ThreadManager -> ThreadId -> IO (Maybe ThreadStatus) -- | Block until all managed threads terminate. waitAll :: ThreadManager -> IO () When I run this through ghci I get the following failure: [1 of 1] Compiling NiceFork ( NiceFork.hs, interpreted ) NiceFork.hs:17:26: Class `Exception' used as a type In the type `Exception' In the data type declaration for `ThreadStatus' Failed, modules loaded: none. Any idea on how to solve this? Exception is a class not a type, so what to put there instead? Cheers, Thomas From daniel.is.fischer at web.de Tue May 12 17:57:15 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Tue May 12 17:43:40 2009 Subject: [Haskell-beginners] The main thread and waiting for other threads In-Reply-To: <4A09A35A.3020909@suud.de> References: <4A09A35A.3020909@suud.de> Message-ID: <200905122357.16244.daniel.is.fischer@web.de> Am Dienstag 12 Mai 2009 18:27:06 schrieb Thomas Friedrich: > Hi everyone, > > I have a problem with the following example in the Real World Haskell > book, which aims to develop a module for controlling different threads. > See, > > http://book.realworldhaskell.org/read/concurrent-and-multicore-programming. >html > > in the chapter "The main thread and waiting for other threads". > When I run this through ghci I get the following failure: > > [1 of 1] Compiling NiceFork ( NiceFork.hs, interpreted ) > > NiceFork.hs:17:26: > Class `Exception' used as a type > In the type `Exception' > In the data type declaration for `ThreadStatus' > Failed, modules loaded: none. > The book was written in the times of GHC 6.8.*, when Exception was a type. In GHC 6.10, it became a class because it was considered a bad idea to catch general exceptions, one should use adequate handlers for specific exceptions instead. Of course, that broke some code out there. > > Any idea on how to solve this? Exception is a class not a type, so what > to put there instead? For those who want to catch general exceptions, there is data SomeException = forall e . Exception e => SomeException e in Control.Exception, which should be roughly equivalent to the old Exception type. So replace Exception with SomeException in ThreadStatus, perhaps insert a few calls to toException in the appropriate places (the compiler will help you find them) and it should work. > > Cheers, > Thomas > Cheers, Daniel From RafaelGCPP.Linux at gmail.com Tue May 12 18:04:31 2009 From: RafaelGCPP.Linux at gmail.com (Rafael Gustavo da Cunha Pereira Pinto) Date: Tue May 12 17:49:38 2009 Subject: [Haskell-beginners] The main thread and waiting for other threads In-Reply-To: <4A09A35A.3020909@suud.de> References: <4A09A35A.3020909@suud.de> Message-ID: <351ff25e0905121504u4b5763ffrde45c7ee76d27524@mail.gmail.com> The problem is that Exception is a class. You should use SomeException, which is a type! On Tue, May 12, 2009 at 13:27, Thomas Friedrich wrote: > Hi everyone, > > I have a problem with the following example in the Real World Haskell book, > which aims to develop a module for controlling different threads. See, > > > http://book.realworldhaskell.org/read/concurrent-and-multicore-programming.html > > in the chapter "The main thread and waiting for other threads". > > -- file: ch24/NiceFork.hs > import Control.Concurrent > import Control.Exception (Exception, try) > import qualified Data.Map as M > > data ThreadStatus = Running > | Finished -- terminated normally > | Threw Exception -- killed by uncaught exception > deriving (Eq, Show) > > -- | Create a new thread manager. > newManager :: IO ThreadManager > > -- | Create a new managed thread. > forkManaged :: ThreadManager -> IO () -> IO ThreadId > > -- | Immediately return the status of a managed thread. > getStatus :: ThreadManager -> ThreadId -> IO (Maybe ThreadStatus) > > -- | Block until a specific managed thread terminates. > waitFor :: ThreadManager -> ThreadId -> IO (Maybe ThreadStatus) > > -- | Block until all managed threads terminate. > waitAll :: ThreadManager -> IO () > > When I run this through ghci I get the following failure: > > [1 of 1] Compiling NiceFork ( NiceFork.hs, interpreted ) > > NiceFork.hs:17:26: > Class `Exception' used as a type > In the type `Exception' > In the data type declaration for `ThreadStatus' > Failed, modules loaded: none. > > > Any idea on how to solve this? Exception is a class not a type, so what to > put there instead? > > Cheers, > Thomas > > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > -- Rafael Gustavo da Cunha Pereira Pinto Electronic Engineer, MSc. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090512/013e3853/attachment.html From quentin.moser at unifr.ch Wed May 13 02:47:41 2009 From: quentin.moser at unifr.ch (Quentin Moser) Date: Wed May 13 02:32:42 2009 Subject: [Haskell-beginners] The main thread and waiting for other threads In-Reply-To: <4A09A35A.3020909@suud.de> References: <4A09A35A.3020909@suud.de> Message-ID: <20090513084741.76120be7@unifr.ch> As Daniel Fischer already wrote, this is due to an API change in GHC 6.10. But the old exceptions interface is still available as Control.OldException; you can simply import that instead of Control.Exception. On Tue, 12 May 2009 12:27:06 -0400 Thomas Friedrich wrote: > Hi everyone, > > I have a problem with the following example in the Real World Haskell > book, which aims to develop a module for controlling different > threads. See, > > http://book.realworldhaskell.org/read/concurrent-and-multicore-programming.html > > in the chapter "The main thread and waiting for other threads". > > -- file: ch24/NiceFork.hs > import Control.Concurrent > import Control.Exception (Exception, try) > import qualified Data.Map as M > > data ThreadStatus = Running > | Finished -- terminated normally > | Threw Exception -- killed by uncaught exception > deriving (Eq, Show) > > -- | Create a new thread manager. > newManager :: IO ThreadManager > > -- | Create a new managed thread. > forkManaged :: ThreadManager -> IO () -> IO ThreadId > > -- | Immediately return the status of a managed thread. > getStatus :: ThreadManager -> ThreadId -> IO (Maybe ThreadStatus) > > -- | Block until a specific managed thread terminates. > waitFor :: ThreadManager -> ThreadId -> IO (Maybe ThreadStatus) > > -- | Block until all managed threads terminate. > waitAll :: ThreadManager -> IO () > > When I run this through ghci I get the following failure: > > [1 of 1] Compiling NiceFork ( NiceFork.hs, interpreted ) > > NiceFork.hs:17:26: > Class `Exception' used as a type > In the type `Exception' > In the data type declaration for `ThreadStatus' > Failed, modules loaded: none. > > > Any idea on how to solve this? Exception is a class not a type, so > what to put there instead? > > Cheers, > Thomas > > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners From voigt at tcs.inf.tu-dresden.de Wed May 13 04:45:26 2009 From: voigt at tcs.inf.tu-dresden.de (Janis Voigtlaender) Date: Wed May 13 04:30:34 2009 Subject: [Haskell-beginners] International Summer School on Advances in Programming Languages (precedes ICFP'09) Message-ID: <4A0A88A6.2090204@tcs.inf.tu-dresden.de> Potential participants of the below summer school should pre-register their interest now. The organizers need that information to go ahead with the planning. ===================================================================== International Summer School on Advances in Programming Languages 25th-28th August, 2009 Heriot-Watt University, Edinburgh, Scotland http://www.macs.hw.ac.uk/~greg/ISS-AiPL Overview ~~~~~~~~ This four-day residential International Summer School on Advances in Programming Languages has a major theme of Concurrency, Distribution, and Multicore. Intended primarily for postgraduate research students, the School offers lectures and practical sessions on an engaging blend of cutting edge theoretical and practical techniques from international experts. The Summer School is supported by the Scottish Informatics and Computer Science Alliance (http://www.sicsa.ac.uk/), a Scottish Funding Council Research Pool. Participants from SICSA member institutions may attend at no cost. Confirmed Topics/Speakers * Static and dynamic languages, Prof Philip Wadler, University of Edinburgh * Compiler technology for data-parallel languages, Dr Sven-Bodo Scholz, University of Hertfordshire * New applications of parametricity, Dr Janis Voigtlaender, Technical University of Dresden * Automatic vectorising compilation, Dr Paul Cockshott, University of Glasgow * Foundational aspects of size analysis, Prof Marko van Eekelen / Dr Olha Shakaravska, Radboud University Nijmegen * Context oriented programming, Dr Pascal Costanza, Vrije Universiteit Brussels * Multi-core programming, Dr Phil Trinder, Heriot-Watt University * Multi-core compilation, Dr Alastair Donaldson, Codeplay Software Ltd * Principles and Applications of Refinement Types, Dr Andrew D. Gordon, Microsoft Research, Cambridge * Resource aware programming in Hume, Prof Greg Michaelson, Heriot-Watt University / Prof Kevin Hammond, University of St Andrews * Haskell concurrency & parallelism, Dr Satnam Singh, Microsoft Research, Cambridge Location ~~~~~~~~ The Summer School is at Heriot-Watt University's Riccarton campus, set in pleasant parkland to the west of Edinburgh, with easy access to the airport, city and central Scotland (http://www.hw.ac.uk/welcome/directions.htm). The Summer School immediately precedes the 2009 International Conference on Functional Programming (http://www.cs.nott.ac.uk/~gmh/icfp09.html) and takes place during the Edinburgh International Festival (http://www.eif.co.uk/) , and the associated Edinburgh Festival Fringe (http://www.edfringe.com/) and Edinburgh International Book Festival (http://www.edbookfest.co.uk/) Steering Committee ~~~~~~~~~~~~~~~~~~ Prof Prof Greg Michaelson, Heriot-Watt University (Convenor), Prof Kevin Hammond, University of St Andrews Dr Patricia Johann, University of Strathclyde Prof Philip Wadler, University of Edinburgh Fee ~~~ Full rate: ?400; (free for SICSA students) Includes: four nights single room, en-suite accommodation with breakfast, lunch and dinner, plus coffee breaks and session materials. Day rate: ?200; (free for SICSA students) Includes: lunch, coffee breaks, session materials Registration of Interest ~~~~~~~~~~~~~~~~~~~~~~~~ If you are interested in attending the International Summer School, please complete the form available from (http://www.macs.hw.ac.uk/~greg/ISS-AiPL/ISS-AiPL%20register.doc) or below, and return it to: ********** International Summer School on Advances in Programming Languages 25th-28th August, 2009 Heriot-Watt University, Edinburgh, Scotland Registration of Interest Name: Address: Email: Phone: SICSA Uni: Yes / No Rate: Full / Day Accessibility requirements: Dietary requirements: Return to: ISS-AiPL-register@macs.hw.ac.uk ********** From mwinter at brocku.ca Wed May 13 14:30:51 2009 From: mwinter at brocku.ca (mwinter@brocku.ca) Date: Wed May 13 14:15:50 2009 Subject: [Haskell-beginners] Gtk2Hs + Sourceview on Windows Message-ID: <4A0AD99B.32613.12A3A11@mwinter.brocku.ca> Hi, I have ghc 6.10.1 and gtk2hs 0.10.0 installed on my windows vista computer. Both were installed using the installer on the webpages. I am able to use gtk, glade etc but not sourceview or cairo. In a forum I found that those have to be enabled using ./configure --enable-sourceview (similar for cairo). But my windows installation does not seem to have a script configure. What am I supposed to do? Thanks, Michael From aditya.siram at gmail.com Thu May 14 19:23:42 2009 From: aditya.siram at gmail.com (aditya siram) Date: Thu May 14 19:09:54 2009 Subject: [Haskell-beginners] Unable to Cabal Upgrade Message-ID: <594f78210905141623i702663fp9e85799d2f666477@mail.gmail.com> Hi all, I'm unable to cabal upgrade with the following error: Resolving dependencies... cabal: dependencies conflict: ghc-6.10.1 requires process ==1.0.1.1 however process-1.0.1.1 was excluded because ghc-6.10.1 requires process ==1.0.1.0 I am currently running ghc 6.10.1 and process 1.0.1.1. thanks ... deech -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090514/750fb3e3/attachment.html From lazycat.manatee at gmail.com Thu May 14 21:45:33 2009 From: lazycat.manatee at gmail.com (Andy Stewart) Date: Thu May 14 21:53:22 2009 Subject: [Haskell-beginners] Re: Unable to Cabal Upgrade References: <594f78210905141623i702663fp9e85799d2f666477@mail.gmail.com> Message-ID: <873ab7m94y.fsf@ubuntu.domain> Hi, aditya siram writes: > Hi all, > I'm unable to cabal upgrade with the following error: > Resolving dependencies... > cabal: dependencies conflict: ghc-6.10.1 requires process ==1.0.1.1 however > process-1.0.1.1 was excluded because ghc-6.10.1 requires process ==1.0.1.0 > > I am currently running ghc 6.10.1 and process 1.0.1.1. It's a bug of Cabal. I think developers will fix this problem in the feature. Now you can do below step for fix this problem: 1-> execute command "ghc-pkg list" Find same package from two list, and use command "ghc-pkg unregister PackageName --force" remove same pacakge. 2-> execute command "ghc-pkg unregister procss-1.0.1.1 --force" I think that's okay. And don't use "cabal upgrade" command until developers fix this problem. If you want install some package, please use command "cabal install" -- Andy From aditya.siram at gmail.com Fri May 15 00:06:03 2009 From: aditya.siram at gmail.com (aditya siram) Date: Thu May 14 23:51:31 2009 Subject: [Haskell-beginners] Re: Unable to Cabal Upgrade In-Reply-To: <873ab7m94y.fsf@ubuntu.domain> References: <594f78210905141623i702663fp9e85799d2f666477@mail.gmail.com> <873ab7m94y.fsf@ubuntu.domain> Message-ID: <594f78210905142106v78ea46cehf1c65ad1aa826edc@mail.gmail.com> I'm not sure why but installing GHC 6.10.3 fixed the issue. I was able to 'cabal upgrade' after that. -deech On Thu, May 14, 2009 at 8:45 PM, Andy Stewart wrote: > Hi, > > aditya siram writes: > > > Hi all, > > I'm unable to cabal upgrade with the following error: > > Resolving dependencies... > > cabal: dependencies conflict: ghc-6.10.1 requires process ==1.0.1.1 > however > > process-1.0.1.1 was excluded because ghc-6.10.1 requires process > ==1.0.1.0 > > > > I am currently running ghc 6.10.1 and process 1.0.1.1. > It's a bug of Cabal. > I think developers will fix this problem in the feature. > > Now you can do below step for fix this problem: > > 1-> execute command "ghc-pkg list" > Find same package from two list, and use command "ghc-pkg unregister > PackageName --force" remove same pacakge. > > 2-> execute command "ghc-pkg unregister procss-1.0.1.1 --force" > > I think that's okay. > > And don't use "cabal upgrade" command until developers fix this problem. > > If you want install some package, please use command "cabal install" > > -- Andy > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090514/f17c53bd/attachment.html From aditya.siram at gmail.com Mon May 18 22:56:00 2009 From: aditya.siram at gmail.com (aditya siram) Date: Mon May 18 22:40:43 2009 Subject: [Haskell-beginners] Searching Maybe lists Message-ID: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> Hi all, I would like to define a function that takes a list and a function that evaluates each member of the list to a Maybe value and output the first element in the list that evaluates to 'Just y', or 'Nothing' once the list has been completely processed. So something like: findMaybe :: [a] -> (a -> Maybe b) -> Maybe b The problem is that I don't want it to go through the entire list, but short-circuit when it hits a 'Just ...'. So far I have: orMaybe :: Maybe a -> Maybe a -> Maybe a orMaybe m1 m2 = case (m1,m2) of (_, Just a) -> Just a (Just a, _) -> Just a _ -> Nothing findMaybe :: [a] -> (a -> Maybe b) -> Maybe b findMaybe as f = foldr (\a sofar -> sofar `orMaybe` (f a)) Nothing as 'findMaybe', as far as I can tell, traverses the entire input list which is undesirable for long lists. How can I fix it? Curiously, the regular 'Data.List.find' function that applies a Boolean predicate to each member of the list also seems to first traverse the entire list using 'filter' and then grabs the head of the result. Thanks ... -deech -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090518/067465a8/attachment.html From aslatter at gmail.com Mon May 18 23:23:44 2009 From: aslatter at gmail.com (Antoine Latter) Date: Mon May 18 23:08:28 2009 Subject: [Haskell-beginners] Searching Maybe lists In-Reply-To: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> Message-ID: <694519c50905182023h30084ff1j504c3d7f370e2bbf@mail.gmail.com> On Mon, May 18, 2009 at 9:56 PM, aditya siram wrote: > Hi all, > I would like to define a function that takes a list and a function that > evaluates each member of the list to a Maybe value and output the first > element in the list that evaluates to 'Just y', or 'Nothing' once the list > has been completely processed. So something like: > > findMaybe :: [a] -> (a -> Maybe b) -> Maybe b > There are a couple of ways, the first one I could think of was: > findMaybe xs f = mconcat $ map f xs where mconcat is found in Data.Monoid Antoine From aslatter at gmail.com Mon May 18 23:26:46 2009 From: aslatter at gmail.com (Antoine Latter) Date: Mon May 18 23:11:29 2009 Subject: [Haskell-beginners] Searching Maybe lists In-Reply-To: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> Message-ID: <694519c50905182026ibe10d35o4fa4f42f48dc539c@mail.gmail.com> On Mon, May 18, 2009 at 9:56 PM, aditya siram wrote: > > Curiously, the regular 'Data.List.find' function that applies a Boolean > predicate to each member of the list also seems to first traverse the entire > list using 'filter' and then grabs the head of the result. > Ah! I just thought I'd point out that 'filter' does not necessarily traverse the entire list if all you do with the result is grab the head - it only traverses enough of the list to figure out what the head of the list should be (or even that it exists). Haskell is lazy :-) Antoine From alexander.dunlap at gmail.com Mon May 18 23:50:26 2009 From: alexander.dunlap at gmail.com (Alexander Dunlap) Date: Mon May 18 23:41:17 2009 Subject: [Haskell-beginners] Searching Maybe lists In-Reply-To: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> Message-ID: <57526e770905182050k345f193bl9ca377698946d94f@mail.gmail.com> On Mon, May 18, 2009 at 7:56 PM, aditya siram wrote: > Hi all, > I would like to define a function that takes a list and a function that > evaluates each member of the list to a Maybe value and output the first > element in the list that evaluates to 'Just y', or 'Nothing' once the list > has been completely processed. So something like: > > findMaybe :: [a] -> (a -> Maybe b) -> Maybe b > > The problem is that I don't want it to go through the entire list, but > short-circuit when it hits a 'Just ...'. So far I have: > > orMaybe :: Maybe a -> Maybe a -> Maybe a > orMaybe m1 m2 = case (m1,m2) of > ????????????????? (_, Just a) -> Just a > ????????????????? (Just a, _) -> Just a > ????????????????? _?????????? -> Nothing > > findMaybe :: [a] -> (a -> Maybe b) -> Maybe b > findMaybe as f = foldr (\a sofar -> sofar `orMaybe` (f a)) Nothing as > > 'findMaybe', as far as I can tell, traverses the entire input list which is > undesirable for long lists. How can I fix it? > > Curiously, the regular 'Data.List.find' function that applies a Boolean > predicate to each member of the list also seems to first traverse the entire > list using 'filter' and then grabs the head of the result. > > Thanks ... > -deech > find doesn't traverse the entire list. The filter function can traverse the whole list, but only if you observe all of its values afterwards. If you only look at the first value, like find does, then filter only goes until it produces one value. This is because of laziness. We can test that filter doesn't examine the entire list by trying it on an infinite list. If it traversed the entire list, it couldn't possibly work on an infinite list! $ ghci GHCi, version 6.10.3: http://www.haskell.org/ghc/ :? for help Loading package ghc-prim ... linking ... done. Loading package integer ... linking ... done. Loading package base ... linking ... done. Prelude> import Data.List Prelude Data.List> find (==3) [1..] Just 3 Prelude Data.List> But it does work! It only looks at values until it finds one that matches the predicate. Alex From DekuDekuplex at Yahoo.com Tue May 19 01:02:25 2009 From: DekuDekuplex at Yahoo.com (Benjamin L.Russell) Date: Tue May 19 00:47:20 2009 Subject: [Haskell-beginners] Re: Haskell vs Clean (speed) References: <49FDC109.4030302@theingots.org> Message-ID: On Sun, 03 May 2009 18:06:33 +0200, Daniel Carrera wrote: >I can't subscribe to Haskell-Cafe, so I'll post here. Thank you for your post. >Here I compare Haskell, Clean, OCaml, Lisp SBCL, C# Mono and >Fortran because they are all in the same ball mark: Which implementation of Fortran? According to "Fortran - Wikipedia, the free encyclopedia" (see http://en.wikipedia.org/wiki/Fortran), there are Absoft, Cray, GFortran, G95, Intel, Lahey/Fujitsu, Open Watcom, Pathscale, PGI, Silverfrost, Sun, XL Fortran, Visual Fortran, and other implementations. >* There are no results for OCaml for 32-bit. I'll be sure to ask Jon Harrop about that the next time he talks about OCaml and speed comparisons ;). Incidentally, why can't you also subscribe to Haskell-Cafe? -- Benjamin L. Russell -- Benjamin L. Russell / DekuDekuplex at Yahoo dot com http://dekudekuplex.wordpress.com/ Translator/Interpreter / Mobile: +011 81 80-3603-6725 "Furuike ya, kawazu tobikomu mizu no oto." -- Matsuo Basho^ From daniel.carrera at theingots.org Tue May 19 02:02:42 2009 From: daniel.carrera at theingots.org (Daniel Carrera) Date: Tue May 19 01:47:31 2009 Subject: [Haskell-beginners] Re: Haskell vs Clean (speed) In-Reply-To: References: <49FDC109.4030302@theingots.org> Message-ID: <4A124B82.8090206@theingots.org> Benjamin L.Russell wrote: > Incidentally, why can't you also subscribe to Haskell-Cafe? Eventually I found that somehow I couldn't get any emails at all from the Haskell server, but the problem magically fixed itself after several hours, so I did subscribe to haskell-cafe and there was a short thread about Haskell vs Clean. The message I got out of it is that it is very difficult to draw any conclusions from the Debian benchmark because there are more Haskell people willing to optimize the Haskell code. Daniel. From DekuDekuplex at Yahoo.com Tue May 19 03:21:49 2009 From: DekuDekuplex at Yahoo.com (Benjamin L.Russell) Date: Tue May 19 03:06:44 2009 Subject: [Haskell-beginners] Re: Haskell vs Clean (speed) References: <49FDC109.4030302@theingots.org> <4A124B82.8090206@theingots.org> Message-ID: <0lm415pevktrq9ocb9i824m4ud6hi1jpqv@4ax.com> On Tue, 19 May 2009 08:02:42 +0200, Daniel Carrera wrote: >Benjamin L.Russell wrote: >> Incidentally, why can't you also subscribe to Haskell-Cafe? > >Eventually I found that somehow I couldn't get any emails at all from >the Haskell server, but the problem magically fixed itself after several >hours, so I did subscribe to haskell-cafe and there was a short thread >about Haskell vs Clean. The message I got out of it is that it is very >difficult to draw any conclusions from the Debian benchmark because >there are more Haskell people willing to optimize the Haskell code. Yes, I just read that thread ("[Haskell-cafe] Haskell vs Clean (speed comparison)" at http://www.mail-archive.com/haskell-cafe@haskell.org/msg58988.html). Based on the discussions there, there seem to be two problems with this benchmark. In reference to Bulat Ziganshin's comments in that thread, On Sun, 3 May 2009 22:42:21 +0400, Bulat Ziganshin wrote: >Hello Daniel, > >Sunday, May 3, 2009, 10:24:52 PM, you wrote: > >> 32-bit sing core [1]: Lisp, Fortran > >:) this test measures speed of some programs, not "languages". >results are depends mainly on bundled libraries and RTS. by no means >it demonstrates speed of compiler-generated code of carefully-written >program what is typically considered as "language speed". the reasons >are: > >1) C++ people (and probably Fortran too) aren't so interested in >making fastest possible programs as Haskell community. it becomes >popular a few years ago, you can find that Haskell becomes several >faster at average since then, which doesn't reflect actual >improvements in GHC code generation (10-20%) What is the reason for this phenomenon, though? If participants don't optimize their programs, then what is the use of the benchmark? >2) Most programs there depend on speed of libraries. Moreover, there >is limitation that it should be *bundled* libraries, so results >greatly depends on what currently included in one compiler or another Is there a specific reason that the libraries need to be bundled? >3) it's prohibited to write your own fast code if bundled library is >too slow (for example, because it's too general) Again, this seems to give an advantage to programming languages with a preponderance of optimized bundled libraries. Based on the above arguments, it seems as if this benchmark does not really compare programming languages; rather, it seems to compare combinations of programming languages and their bundled libraries and communities. But then what happens if you have a great programming language with few optimized libraries and a small and relatively inactive community? -- Benjamin L. Russell -- Benjamin L. Russell / DekuDekuplex at Yahoo dot com http://dekudekuplex.wordpress.com/ Translator/Interpreter / Mobile: +011 81 80-3603-6725 "Furuike ya, kawazu tobikomu mizu no oto." -- Matsuo Basho^ From apfelmus at quantentunnel.de Tue May 19 04:05:00 2009 From: apfelmus at quantentunnel.de (Heinrich Apfelmus) Date: Tue May 19 03:49:52 2009 Subject: [Haskell-beginners] Re: Searching Maybe lists In-Reply-To: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> Message-ID: aditya siram wrote: > Hi all, > I would like to define a function that takes a list and a function that > evaluates each member of the list to a Maybe value and output the first > element in the list that evaluates to 'Just y', or 'Nothing' once the list > has been completely processed. So something like: > > findMaybe :: [a] -> (a -> Maybe b) -> Maybe b > > The problem is that I don't want it to go through the entire list, but > short-circuit when it hits a 'Just ...'. So far I have: > > orMaybe :: Maybe a -> Maybe a -> Maybe a > orMaybe m1 m2 = case (m1,m2) of > (_, Just a) -> Just a > (Just a, _) -> Just a > _ -> Nothing > > findMaybe :: [a] -> (a -> Maybe b) -> Maybe b > findMaybe as f = foldr (\a sofar -> sofar `orMaybe` (f a)) Nothing as > > > 'findMaybe', as far as I can tell, traverses the entire input list Thanks to lazy evaluation, this is not necessarily the case, see also http://en.wikibooks.org/wiki/Haskell/Performance_Introduction#Time You will have to write orMaybe as orMaybe Nothing y = y orMaybe x _ = x though. (By the way, your original code for orMaybe doesn't seem to do what you want.) This function has already been implemented for you, it's called mplus Regards, apfelmus -- http://apfelmus.nfshost.com From info at suud.de Tue May 19 09:39:16 2009 From: info at suud.de (Thomas Friedrich) Date: Tue May 19 09:24:33 2009 Subject: [Haskell-beginners] Searching Maybe lists In-Reply-To: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> Message-ID: <4A12B684.8080205@suud.de> Hi Aditya, Please try the following: findJust :: (Eq a) => [Maybe a] -> Maybe a findJust xs = case (dropWhile (==Nothing) xs) of [] -> Nothing cs -> head cs yourFunction :: (Eq b) => (a -> Maybe b) -> [a] -> Maybe b yourFunction f xs = findJust (map f xs) It only uses functions from the Prelude, and as Haskell evaluates lazy, it just does exactly what you wants. Happy Hacking, Thomas aditya siram wrote: > Hi all, > I would like to define a function that takes a list and a function > that evaluates each member of the list to a Maybe value and output the > first element in the list that evaluates to 'Just y', or 'Nothing' > once the list has been completely processed. So something like: > > findMaybe :: [a] -> (a -> Maybe b) -> Maybe b > > The problem is that I don't want it to go through the entire list, but > short-circuit when it hits a 'Just ...'. So far I have: > > orMaybe :: Maybe a -> Maybe a -> Maybe a > orMaybe m1 m2 = case (m1,m2) of > (_, Just a) -> Just a > (Just a, _) -> Just a > _ -> Nothing > > findMaybe :: [a] -> (a -> Maybe b) -> Maybe b > findMaybe as f = foldr (\a sofar -> sofar `orMaybe` (f a)) Nothing as > > 'findMaybe', as far as I can tell, traverses the entire input list > which is undesirable for long lists. How can I fix it? > > Curiously, the regular 'Data.List.find' function that applies a > Boolean predicate to each member of the list also seems to first > traverse the entire list using 'filter' and then grabs the head of the > result. > > Thanks ... > -deech > ------------------------------------------------------------------------ > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > From daniel.is.fischer at web.de Tue May 19 11:02:28 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Tue May 19 10:47:50 2009 Subject: [Haskell-beginners] Searching Maybe lists In-Reply-To: <4A12B684.8080205@suud.de> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> <4A12B684.8080205@suud.de> Message-ID: <200905191702.28642.daniel.is.fischer@web.de> Am Dienstag 19 Mai 2009 15:39:16 schrieb Thomas Friedrich: > Hi Aditya, > > Please try the following: > > findJust :: (Eq a) => [Maybe a] -> Maybe a > findJust xs = case (dropWhile (==Nothing) xs) of > [] -> Nothing > cs -> head cs > > yourFunction :: (Eq b) => (a -> Maybe b) -> [a] -> Maybe b > yourFunction f xs = findJust (map f xs) > > It only uses functions from the Prelude, and as Haskell evaluates lazy, > it just does exactly what you wants. No need for the Eq constraint, findJust xs = case dropWhile isNothing xs of [] -> Nothing (x:_) -> x isNothing could be imported from Data.Maybe or defined as isNothing Nothing = True isNothing _ = False if you don't want the import. Another method to define findJust is import Data.Maybe findJust = listToMaybe . catMaybes or import Control.Monad findJust = msum So for the original problem, we could use any of findMaybe :: [a] -> (a -> Maybe b) -> Maybe b findMaybe xs f = msum $ map f xs -- this indicates that the parameter order should be different findMaybe xs f = foldr mplus Nothing (map f xs) findMaybe xs f = listToMaybe . catMaybes $ map f xs findMaybe xs f = head (dropWhile isNothing (map f xs) ++ [Nothing]) findMaybe xs f = find isJust (map f xs) >>= id I find the first two best, but they bring Control.Monad into the game, if one wants to avoid that, I'd recommend defining 'mplus' for Maybe oneself, orMaybe m1@(Just _) _ = m1 orMaybe _ m2 = m2 findMaybe xs f = foldr orMaybe Nothing (map f xs). > > Happy Hacking, > Thomas > > aditya siram wrote: > > Hi all, > > I would like to define a function that takes a list and a function > > that evaluates each member of the list to a Maybe value and output the > > first element in the list that evaluates to 'Just y', or 'Nothing' > > once the list has been completely processed. So something like: > > > > findMaybe :: [a] -> (a -> Maybe b) -> Maybe b > > > > The problem is that I don't want it to go through the entire list, but > > short-circuit when it hits a 'Just ...'. So far I have: > > > > orMaybe :: Maybe a -> Maybe a -> Maybe a > > orMaybe m1 m2 = case (m1,m2) of > > (_, Just a) -> Just a > > (Just a, _) -> Just a > > _ -> Nothing > > > > findMaybe :: [a] -> (a -> Maybe b) -> Maybe b > > findMaybe as f = foldr (\a sofar -> sofar `orMaybe` (f a)) Nothing as > > > > 'findMaybe', as far as I can tell, traverses the entire input list > > which is undesirable for long lists. How can I fix it? > > > > Curiously, the regular 'Data.List.find' function that applies a > > Boolean predicate to each member of the list also seems to first > > traverse the entire list using 'filter' and then grabs the head of the > > result. > > > > Thanks ... > > -deech From aditya.siram at gmail.com Tue May 19 12:02:50 2009 From: aditya.siram at gmail.com (aditya siram) Date: Tue May 19 11:47:32 2009 Subject: [Haskell-beginners] Re: Searching Maybe lists In-Reply-To: References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> Message-ID: <594f78210905190902wbc4b1f0j77aaa0af99c9d52e@mail.gmail.com> > > > > You will have to write orMaybe as > > orMaybe Nothing y = y > orMaybe x _ = x > > though. (By the way, your original code for orMaybe doesn't seem to do > what you want.) This function has already been implemented for you, it's > called > > mplus My function 'orMaybe' takes two arguments m1 m2, if one of them is Nothing, it returns Nothing, if m1 is Just , it returns m1, if m2 is Just, it returns m2. This seems to be what I want. Why is this incorrect? However, your function 'orMaybe' is much more concise and elegant. thanks ... -deech -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090519/3d8e6f67/attachment.html From aditya.siram at gmail.com Tue May 19 12:05:09 2009 From: aditya.siram at gmail.com (aditya siram) Date: Tue May 19 11:49:51 2009 Subject: [Haskell-beginners] Searching Maybe lists In-Reply-To: <4A12B684.8080205@suud.de> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> <4A12B684.8080205@suud.de> Message-ID: <594f78210905190905r10cb87f6mc38d809c48e74a85@mail.gmail.com> Awesome! This seems to be an analog of the Data.List.find function. I originally didn't implement mine this way because I thought it went through the entire list, but I was obviously mistaken! Thanks for your help! -deech On Tue, May 19, 2009 at 8:39 AM, Thomas Friedrich wrote: > Hi Aditya, > > Please try the following: > > findJust :: (Eq a) => [Maybe a] -> Maybe a > findJust xs = case (dropWhile (==Nothing) xs) of > [] -> Nothing > cs -> head cs > > yourFunction :: (Eq b) => (a -> Maybe b) -> [a] -> Maybe b > yourFunction f xs = findJust (map f xs) > > It only uses functions from the Prelude, and as Haskell evaluates lazy, it > just does exactly what you wants. > > Happy Hacking, > Thomas > > > > aditya siram wrote: > >> Hi all, >> I would like to define a function that takes a list and a function that >> evaluates each member of the list to a Maybe value and output the first >> element in the list that evaluates to 'Just y', or 'Nothing' once the list >> has been completely processed. So something like: >> >> findMaybe :: [a] -> (a -> Maybe b) -> Maybe b >> >> The problem is that I don't want it to go through the entire list, but >> short-circuit when it hits a 'Just ...'. So far I have: >> >> orMaybe :: Maybe a -> Maybe a -> Maybe a >> orMaybe m1 m2 = case (m1,m2) of >> (_, Just a) -> Just a >> (Just a, _) -> Just a >> _ -> Nothing >> findMaybe :: [a] -> (a -> Maybe b) -> Maybe >> b >> findMaybe as f = foldr (\a sofar -> sofar `orMaybe` (f a)) Nothing as >> >> 'findMaybe', as far as I can tell, traverses the entire input list which >> is undesirable for long lists. How can I fix it? >> >> Curiously, the regular 'Data.List.find' function that applies a Boolean >> predicate to each member of the list also seems to first traverse the entire >> list using 'filter' and then grabs the head of the result. >> >> Thanks ... >> -deech >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Beginners mailing list >> Beginners@haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090519/7b5b1ad2/attachment-0001.html From apfelmus at quantentunnel.de Tue May 19 12:45:25 2009 From: apfelmus at quantentunnel.de (Heinrich Apfelmus) Date: Tue May 19 12:30:13 2009 Subject: [Haskell-beginners] Re: Searching Maybe lists In-Reply-To: <594f78210905190902wbc4b1f0j77aaa0af99c9d52e@mail.gmail.com> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> <594f78210905190902wbc4b1f0j77aaa0af99c9d52e@mail.gmail.com> Message-ID: aditya siram wrote: >> >> >> You will have to write orMaybe as >> >> orMaybe Nothing y = y >> orMaybe x _ = x >> >> though. (By the way, your original code for orMaybe doesn't seem to do >> what you want.) This function has already been implemented for you, it's >> called >> >> mplus > > > My function 'orMaybe' takes two arguments m1 m2, if one of them is Nothing, > it returns Nothing, if m1 is Just , it returns m1, if m2 is Just, it returns > m2. This seems to be what I want. Why is this incorrect? Ah, you're right, my bad. I was confused by the fact that you were using it from right to left, i.e. findMaybe f = foldr (\a sofar -> sofar `orMaybe` a) Nothing . map f = foldr (flip orMaybe) Nothing . map f instead of findMaybe f = foldr orMaybe Nothing . map f Regards, apfelmus -- http://apfelmus.nfshost.com From nathanmholden at gmail.com Tue May 19 14:43:04 2009 From: nathanmholden at gmail.com (Nathan Holden) Date: Tue May 19 14:27:46 2009 Subject: [Haskell-beginners] Intervals? Message-ID: <305228b20905191143wda9da5xe33c468c0be254b8@mail.gmail.com> I have been trying to optimize some research I was doing in my spare time. To spare complexity, it involves music-- I have a class, defining each note in an octave, and I want to write a function that will convert between two notes and an interval-- so, for instance: Foo As {- A sharp-- can't use # -} B = 1 because in the standard scale, A# and B are one semitone apart. My question, however, boils primarily down to this: Is there a way to make Foo :: Note -> Note -> Int Foo a b = a-b work? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090519/c6e23588/attachment.html From ezyang at MIT.EDU Tue May 19 14:53:36 2009 From: ezyang at MIT.EDU (Edward Z. Yang) Date: Tue May 19 14:38:20 2009 Subject: [Haskell-beginners] Intervals? In-Reply-To: <305228b20905191143wda9da5xe33c468c0be254b8@mail.gmail.com> References: <305228b20905191143wda9da5xe33c468c0be254b8@mail.gmail.com> Message-ID: <1242759184-sup-8220@javelin> Excerpts from nathanmholden's message of Tue May 19 14:43:04 -0400 2009: > My question, however, boils primarily down to this: Is there a way to make > > Foo :: Note -> Note -> Int > Foo a b = a-b A sensible implementation of the Num typeclass for Note should do the trick nicely. Cheers, Edward From nathanmholden at gmail.com Tue May 19 15:07:26 2009 From: nathanmholden at gmail.com (Nathan Holden) Date: Tue May 19 14:52:08 2009 Subject: [Haskell-beginners] Intervals? In-Reply-To: <4A130067.7050408@ebollocks.net> References: <305228b20905191143wda9da5xe33c468c0be254b8@mail.gmail.com> <4A130067.7050408@ebollocks.net> Message-ID: <305228b20905191207j205a3588h374f72cd2883680f@mail.gmail.com> Yeah, it would. But I always write code like that, it feels ungainly... writing super-long, case-by-case instance declarations (So that I can add in one little caveat at the end, or something), and stuff. Makes me feel like I must be dumb! On Tue, May 19, 2009 at 2:54 PM, Derek Gladding wrote: > Nathan Holden wrote: > >> [snip] >> My question, however, boils primarily down to this: Is there a way to make >> >> Foo :: Note -> Note -> Int >> Foo a b = a-b >> >> work? >> > > Would a function > > Bar :: Note -> Int > > make writing Foo easier? > > - Derek > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090519/c7112916/attachment.html From ezyang at MIT.EDU Tue May 19 15:16:36 2009 From: ezyang at MIT.EDU (Edward Z. Yang) Date: Tue May 19 15:01:19 2009 Subject: [Haskell-beginners] Intervals? In-Reply-To: <305228b20905191207j205a3588h374f72cd2883680f@mail.gmail.com> References: <305228b20905191143wda9da5xe33c468c0be254b8@mail.gmail.com> <4A130067.7050408@ebollocks.net> <305228b20905191207j205a3588h374f72cd2883680f@mail.gmail.com> Message-ID: <1242760523-sup-1152@javelin> Excerpts from nathanmholden's message of Tue May 19 15:07:26 -0400 2009: > Yeah, it would. But I always write code like that, it feels ungainly... > writing super-long, case-by-case instance declarations (So that I can add in > one little caveat at the end, or something), and stuff. Makes me feel like I > must be dumb! It seems to me, in this particular case, that you don't actually want an algebraic datatype (which seems to be what you're implementation is using). Cheers, Edward From aditya.siram at gmail.com Tue May 19 15:43:04 2009 From: aditya.siram at gmail.com (aditya siram) Date: Tue May 19 15:27:46 2009 Subject: [Haskell-beginners] Intervals? In-Reply-To: <305228b20905191143wda9da5xe33c468c0be254b8@mail.gmail.com> References: <305228b20905191143wda9da5xe33c468c0be254b8@mail.gmail.com> Message-ID: <594f78210905191243u7b6883cchc94b8c44183c11da@mail.gmail.com> How about something like this? import Data.List data Note = C | Cs | D | Eb | E | F | Fs | G | Ab | A | Bb | B deriving Enum --Counting forward from n1 to n2 interval :: Note -> Note -> Int n1 `interval` n2 = (12 - fromEnum n1) + fromEnum n2 --Counting back from n2 to n1 bInterval :: Note -> Note -> Int n1 `bInterval` n2 = fromEnum n1 - fromEnum n2 testInterval = A `interval` D testBInterval = A `bInterval` D I don't see how these operations could be made part of the Num because the output of the Num class has to be the same as the input. For instance, when adding two numbers, the output is a number, but for determining intervals, the inputs are Notes and the output is a number. -deech On Tue, May 19, 2009 at 1:43 PM, Nathan Holden wrote: > I have been trying to optimize some research I was doing in my spare time. > > To spare complexity, it involves music-- I have a class, defining each note > in an octave, and I want to write a function that will convert between two > notes and an interval-- so, for instance: > > Foo As {- A sharp-- can't use # -} B = 1 > > because in the standard scale, A# and B are one semitone apart. > > My question, however, boils primarily down to this: Is there a way to make > > Foo :: Note -> Note -> Int > Foo a b = a-b > > work? > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090519/0cb9f5fa/attachment.html From mpm at alumni.caltech.edu Tue May 19 18:14:12 2009 From: mpm at alumni.caltech.edu (Michael P Mossey) Date: Tue May 19 17:59:01 2009 Subject: [Haskell-beginners] "computation", "action" Message-ID: <4A132F34.3010201@alumni.caltech.edu> I've seen the following language used: - "computations" - "actions" - "effects" - "context" (For example, a functor is a context, and fmap applies a function to an object in that context without changing the context.) One words that seem to come up in Haskell are "algorithm" and "procedure." For example, a monad is sometimes called a "computation" or an "effect", but I've not seen it called an "algorithm." Can someone elaborate on these terms and why they are used? What is the difference between a computation and an algorithm? The word "effect" has a fairly obvious meaning (like an IO side-effect) but I suspect there's more to it. Why is the word "context" important, and when it is more appropriate than "container"? thanks, Mike From wagner.andrew at gmail.com Tue May 19 18:21:50 2009 From: wagner.andrew at gmail.com (Andrew Wagner) Date: Tue May 19 18:06:33 2009 Subject: [Haskell-beginners] "computation", "action" In-Reply-To: <4A132F34.3010201@alumni.caltech.edu> References: <4A132F34.3010201@alumni.caltech.edu> Message-ID: On Tue, May 19, 2009 at 6:14 PM, Michael P Mossey wrote: > I've seen the following language used: > - "computations" > - "actions" > - "effects" > - "context" (For example, a functor is a context, and fmap applies a > function to an object in that context without changing the context.) > These are all often used to describe expressions of type "m a" for some monad m. Different ones are more or less accurate depending on the particular monad you're talking about, and depending on your personal preferred way of looking at it. The thing with monads is that what they abstract is hard to put succinctly outside of code. > > One words that seem to come up in Haskell are "algorithm" and "procedure." > I've not seen these used in Haskell. They sound distinctly imperative to me, for some reason. But, until you define the words connotatively, all I can say is how they sound to me. > > For example, a monad is sometimes called a "computation" or an "effect", > but I've not seen it called an "algorithm." > > Can someone elaborate on these terms and why they are used? What is the > difference between a computation and an algorithm? The word "effect" has a > fairly obvious meaning (like an IO side-effect) but I suspect there's more > to it. > > Why is the word "context" important, and when it is more appropriate than > "container"? > I would put container in with the list at the beginning: it's another way that can be more or less appropriate for monadic values, depending on the monad and your perspective. > > thanks, > Mike > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090519/e0f9a553/attachment-0001.html From mpm at alumni.caltech.edu Tue May 19 18:34:03 2009 From: mpm at alumni.caltech.edu (Michael Mossey) Date: Tue May 19 18:19:03 2009 Subject: [Haskell-beginners] "computation", "action" In-Reply-To: References: <4A132F34.3010201@alumni.caltech.edu> Message-ID: <4A1333DB.80709@alumni.caltech.edu> Andrew Wagner wrote: > > > One words that seem to come up in Haskell are "algorithm" and > "procedure." > > > I've not seen these used in Haskell. They sound distinctly imperative to > me, for some reason. But, until you define the words connotatively, all > I can say is how they sound to me. > Yes, I meant to write they *don't* come up in Haskell. Regarding how words "sound", that's a good thing to talk about, in my opinion. It's related to the philosophy of epistemology. Words have connotations. They are used in certain contexts, with certain implications. Thanks, Mike From jason.dusek at gmail.com Tue May 19 19:27:00 2009 From: jason.dusek at gmail.com (Jason Dusek) Date: Tue May 19 19:11:42 2009 Subject: [Haskell-beginners] "computation", "action" In-Reply-To: <4A132F34.3010201@alumni.caltech.edu> References: <4A132F34.3010201@alumni.caltech.edu> Message-ID: <42784f260905191627k7c8c7d62ie79c5f6cee20e205@mail.gmail.com> 2009/05/19 Michael P Mossey : > For example, a monad is sometimes called a "computation" or an > "effect", but I've not seen it called an "algorithm." There is a difference between a monad and monadic value -- just as there is a difference between "List" as a structure and a particular list as an instantiation of that structure. > Can someone elaborate on these terms and why they are used? > What is the difference between a computation and an algorithm? This is a little like saying, what is the difference between an algorithm and a program? The term "algorithm" denotes not only a sequence of instructions to perform but also a solution (in a well-defined sense) to a particular problem. A computation is probably anything whatever that warms the computer. > The word "effect" has a fairly obvious meaning (like an IO > side-effect) but I suspect there's more to it. I believe the usual aphorism is "Monads sequence effects.". -- Jason Dusek From lemming at henning-thielemann.de Wed May 20 09:12:14 2009 From: lemming at henning-thielemann.de (Henning Thielemann) Date: Wed May 20 08:57:14 2009 Subject: [Haskell-beginners] HaL4: Haskell-Treffen in Halle/Saale, 12. Juni 2009 - Programm und Anmeldung Message-ID: Kindly excuse the German noise, please ... Liebe Haskell-Freunde! Das Programm fuer unser Haskell-Treffen steht jetzt so gut wie fest und man kann sich online anmelden. --------------------------------------------- HaL4 : Haskell - Tutorial + Workshop + Party am Freitag, dem 12. Juni 2009, in Halle/Saale --------------------------------------------- Das traditionsreiche HaL-Treffen bietet eine gute Mischung von Haskell-bezogenen Themen aus Forschung, Anwendung und Lehre mit vielen M?glichkeiten zu Diskussion und Unterhaltung bei der anschlie?enden Party. Der Workshop wird in diesem Jahr erg?nzt durch Tutorien f?r Haskell-Ein- und Umsteiger. Diesmal findet das Treffen in Halle/Saale im Institut f?r Informatik der Martin-Luther-Universit?t Halle-Wittenberg statt. Wir bieten: 10 Uhr: Entwurfsmuster vom Kopf auf die Fuesse gestellt: data, fold und laziness statt Kompositum, Visitor, Iterator. Johannes Waldmann, HTWK Leipzig. 11 Uhr: CAL/openquark: (openquark.org) Haskell + Java => CAL, eine praktische Einfuehrung (Eclipse als CAL-IDE, Excel lesen, typsicheres SQL = LINQ fuer Java?) Alf Richter, iba Consulting, Leipzig 12 Uhr: Wie man das Semikolon ueberlaedt: Code-Strukturierung und -Wiederverwendung durch Monaden 15 Uhr: Hayoo! Haskell API Search Timo Huebel, FH Wedel 16 Uhr: Haskell als reine Spezifikationssprache Baltasar Trancon y Widemann, Universit Bayreuth 17 Uhr: Funktional-logische Programmierung mit Curry Jan Christiansen, Universitaet Kiel 19 - 22 Uhr: Grillparty Wir freuen uns auf rege Teilnahme sowie spannende Vortr?ge mit hei?en Diskussionen und bitten um Anmeldung bis zum 31. Mai. Weitere Informationen auf ?http://www.iba-cg.de/hal4.html Beste Gruesse Henning Thielemann From aditya.siram at gmail.com Fri May 22 02:20:39 2009 From: aditya.siram at gmail.com (aditya siram) Date: Fri May 22 02:05:12 2009 Subject: [Haskell-beginners] Catching Network Connection Error Message-ID: <594f78210905212320i2443f11bs883bf9cbbc1f236e@mail.gmail.com> Hi all, I am trying to Network.HTTP to connect to a website and catch the error if the network is down. Here's what I have so far: data Errors = DocumentParseError | WebsiteUnreachableError | ISBNNotFoundError respHTML' :: ISBN -> IO (Either Errors (IO String)) respHTML' isbn = do rsp <- simpleHTTP (getRequest $ " http://isbndb.com/api/books.xml?results=details&access_key=" ++ key ++ "&index1=isbn&value1=" ++ isbn :: Request_String) return $ Right $ getResponseBody rsp `E.catch` (\(e :: E.SomeException ) -> return $ Left $ WebsiteUnreachableError) I get the following error when I load it into GHCI: Couldn't match expected type `[Char]' against inferred type `Either Errors b' Expected type: IO String Inferred type: IO (Either Errors b) In the expression: return $ Left $ WebsiteUnreachableError In the second argument of `E.catch', namely `(\ (e :: E.SomeException) -> return $ Left $ WebsiteUnreachableError)' Failed, modules loaded: none. Control.Exception.catch has the following signature : catch :: forall a. IO a -> (IOError -> IO a) -> IO a I would expect that the handler function needs to return an IO (Either Errors (IO String)). Why is it trying to return an IO String? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090522/48149092/attachment.html From miguel.pignatelli at uv.es Fri May 22 17:38:18 2009 From: miguel.pignatelli at uv.es (Miguel Pignatelli) Date: Fri May 22 17:22:57 2009 Subject: [Haskell-beginners] parMap on multicore computers In-Reply-To: <200905191702.28642.daniel.is.fischer@web.de> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> <4A12B684.8080205@suud.de> <200905191702.28642.daniel.is.fischer@web.de> Message-ID: <80BAF8FA-A95C-45EA-A761-16291A8D167E@uv.es> Hi all, I'm experimenting a bit with the parallelization capabilities of Haskell. What I am trying to do is to process in parallel all the lines of a text file, calculating the edit distance of each of these lines with a given string. This is my testing code: import System.IO import Control.Monad import Control.Parallel import Control.Parallel.Strategies edist :: String -> String -> Int -- edist calculates the edit distance of 2 strings -- see for example http://www.csse.monash.edu.au/~lloyd/tildeFP/Haskell/1998/Edit01/ getLines :: FilePath -> IO [Int] getLines = liftM ((parMap rnf (edist longString)) . lines) . readFile main :: IO () main = do list <- getLines "input.txt" mapM_ ( putStrLn . show ) list I am testing this code in a 2xQuadCore linux (Ubuntu 8.10) machine (8 cores in total). The code has been compiled with ghc --make -threaded mytest.hs I've been trying input files of different lengths, but the more cores I try to use, the worst performance I am getting. Here are some examples: # input.txt -> 10 lines (strings) of ~1200 letters each $ time ./mytest +RTS -N1 > /dev/null real 0m4.775s user 0m4.700s sys 0m0.080s $ time ./mytest +RTS -N4 > /dev/null real 0m6.272s user 0m8.220s sys 0m0.290s $ time ./mytest +RTS -N8 > /dev/null real 0m7.090s user 0m10.960s sys 0m0.400s # input.txt -> 100 lines (strings) of ~1200 letters each $ time ./mytest +RTS -N1 > /dev/null real 0m49.854s user 0m49.730s sys 0m0.120s $ time ./mytest +RTS -N4 > /dev/null real 1m11.303s user 1m36.210s sys 0m1.070s $ time ./mytest +RTS -N8 > /dev/null real 1m19.488s user 2m6.250s sys 0m1.270s What is going wrong in this code? Is this a problem of the "grain size" of the parallelization? Any help / advice would be very welcome, M; -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090522/5bd5abc8/attachment-0001.html From info at suud.de Fri May 22 18:02:14 2009 From: info at suud.de (Thomas Friedrich) Date: Fri May 22 17:48:00 2009 Subject: [Haskell-beginners] parMap on multicore computers In-Reply-To: <80BAF8FA-A95C-45EA-A761-16291A8D167E@uv.es> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> <4A12B684.8080205@suud.de> <200905191702.28642.daniel.is.fischer@web.de> <80BAF8FA-A95C-45EA-A761-16291A8D167E@uv.es> Message-ID: <4A1720E6.2090708@suud.de> Hi Miguel, I don't think that you can expect a program to process each line of a text file in parallel to very efficient. Opening a new thread is usually fairly cheap, however, there is some bookkeeping involved that shouldn't be underestimated. You only have 2 or 4 cores, so opening 100 threads or more, depending on the size of your file, will do you no good. You should rather split up your file in 4 chunks and and then process these *4* threads in parallel. That should make it more efficient! Parallel /= faster. At least not automatically. Happy Hacking, Thomas Miguel Pignatelli wrote: > Hi all, > > I'm experimenting a bit with the parallelization capabilities of Haskell. > What I am trying to do is to process in parallel all the lines of a > text file, calculating the edit distance of each of these lines with a > given string. > This is my testing code: > > > import System.IO > import Control.Monad > import Control.Parallel > import Control.Parallel.Strategies > > edist :: String -> String -> Int > -- edist calculates the edit distance of 2 strings > -- see for > example http://www.csse.monash.edu.au/~lloyd/tildeFP/Haskell/1998/Edit01/ > > > getLines :: FilePath -> IO [Int] > getLines = liftM ((parMap rnf (edist longString)) . lines) . readFile > > main :: IO () > main = do > list <- getLines "input.txt" > mapM_ ( putStrLn . show ) list > > I am testing this code in a 2xQuadCore linux (Ubuntu 8.10) machine (8 > cores in total). > The code has been compiled with > > ghc --make -threaded mytest.hs > > I've been trying input files of different lengths, but the more cores > I try to use, the worst performance I am getting. > Here are some examples: > > # input.txt -> 10 lines (strings) of ~1200 letters each > $ time ./mytest +RTS -N1 > /dev/null > > real 0m4.775s > user 0m4.700s > sys 0m0.080s > > $ time ./mytest +RTS -N4 > /dev/null > > real 0m6.272s > user 0m8.220s > sys 0m0.290s > > $ time ./mytest +RTS -N8 > /dev/null > > real 0m7.090s > user 0m10.960s > sys 0m0.400s > > # input.txt -> 100 lines (strings) of ~1200 letters each > $ time ./mytest +RTS -N1 > /dev/null > > real 0m49.854s > user 0m49.730s > sys 0m0.120s > > $ time ./mytest +RTS -N4 > /dev/null > > real 1m11.303s > user 1m36.210s > sys 0m1.070s > > $ time ./mytest +RTS -N8 > /dev/null > > real 1m19.488s > user 2m6.250s > sys 0m1.270s > > > What is going wrong in this code? Is this a problem of the "grain > size" of the parallelization? > Any help / advice would be very welcome, > > M; > > ------------------------------------------------------------------------ > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > From miguel.pignatelli at uv.es Sat May 23 06:33:42 2009 From: miguel.pignatelli at uv.es (Miguel Pignatelli) Date: Sat May 23 06:18:19 2009 Subject: [Haskell-beginners] parMap on multicore computers In-Reply-To: <4A1720E6.2090708@suud.de> References: <594f78210905181956n3005ddc3kaa3b69cfd390ba68@mail.gmail.com> <4A12B684.8080205@suud.de> <200905191702.28642.daniel.is.fischer@web.de> <80BAF8FA-A95C-45EA-A761-16291A8D167E@uv.es> <4A1720E6.2090708@suud.de> Message-ID: Hi Thomas, Thanks for your answer. I know that there is needed some tweak in the process of parallelization to get a performance gain. I agree with your reasoning regarding the 100 string input, but in the other example I gave, I use the 8-core machine to process a 10 strings (and each one takes half a second to be processed) why is this case failing too? M; El 23/05/2009, a las 0:02, Thomas Friedrich escribi?: > Hi Miguel, > > I don't think that you can expect a program to process each line of > a text file in parallel to very efficient. Opening a new thread is > usually fairly cheap, however, there is some bookkeeping involved > that shouldn't be underestimated. You only have 2 or 4 cores, so > opening 100 threads or more, depending on the size of your file, > will do you no good. You should rather split up your file in 4 > chunks and and then process these *4* threads in parallel. > > That should make it more efficient! Parallel /= faster. At least > not automatically. > > Happy Hacking, > Thomas > > > > Miguel Pignatelli wrote: >> Hi all, >> >> I'm experimenting a bit with the parallelization capabilities of >> Haskell. >> What I am trying to do is to process in parallel all the lines of a >> text file, calculating the edit distance of each of these lines >> with a given string. >> This is my testing code: >> >> >> import System.IO >> import Control.Monad >> import Control.Parallel >> import Control.Parallel.Strategies >> >> edist :: String -> String -> Int >> -- edist calculates the edit distance of 2 strings >> -- see for example http://www.csse.monash.edu.au/~lloyd/tildeFP/Haskell/1998/Edit01/ >> > Edit01/> >> >> getLines :: FilePath -> IO [Int] >> getLines = liftM ((parMap rnf (edist longString)) . lines) . readFile >> >> main :: IO () >> main = do >> list <- getLines "input.txt" >> mapM_ ( putStrLn . show ) list >> >> I am testing this code in a 2xQuadCore linux (Ubuntu 8.10) machine >> (8 cores in total). >> The code has been compiled with >> >> ghc --make -threaded mytest.hs >> >> I've been trying input files of different lengths, but the more >> cores I try to use, the worst performance I am getting. >> Here are some examples: >> >> # input.txt -> 10 lines (strings) of ~1200 letters each >> $ time ./mytest +RTS -N1 > /dev/null >> real 0m4.775s >> user 0m4.700s >> sys 0m0.080s >> >> $ time ./mytest +RTS -N4 > /dev/null >> >> real 0m6.272s >> user 0m8.220s >> sys 0m0.290s >> >> $ time ./mytest +RTS -N8 > /dev/null >> >> real 0m7.090s >> user 0m10.960s >> sys 0m0.400s >> >> # input.txt -> 100 lines (strings) of ~1200 letters each >> $ time ./mytest +RTS -N1 > /dev/null >> >> real 0m49.854s >> user 0m49.730s >> sys 0m0.120s >> >> $ time ./mytest +RTS -N4 > /dev/null >> >> real 1m11.303s >> user 1m36.210s >> sys 0m1.070s >> >> $ time ./mytest +RTS -N8 > /dev/null >> >> real 1m19.488s >> user 2m6.250s >> sys 0m1.270s >> >> >> What is going wrong in this code? Is this a problem of the "grain >> size" of the parallelization? >> Any help / advice would be very welcome, >> >> M; >> >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Beginners mailing list >> Beginners@haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> > > From byorgey at seas.upenn.edu Sat May 23 09:56:32 2009 From: byorgey at seas.upenn.edu (Brent Yorgey) Date: Sat May 23 09:41:02 2009 Subject: [Haskell-beginners] Catching Network Connection Error In-Reply-To: <594f78210905212320i2443f11bs883bf9cbbc1f236e@mail.gmail.com> References: <594f78210905212320i2443f11bs883bf9cbbc1f236e@mail.gmail.com> Message-ID: <20090523135632.GA25322@seas.upenn.edu> On Fri, May 22, 2009 at 01:20:39AM -0500, aditya siram wrote: > Hi all, > I am trying to Network.HTTP to connect to a website and catch the error if > the network is down. Here's what I have so far: > > data Errors = DocumentParseError > | WebsiteUnreachableError > | ISBNNotFoundError > > respHTML' :: ISBN -> IO (Either Errors (IO String)) > respHTML' isbn = > do > rsp <- simpleHTTP (getRequest $ " > http://isbndb.com/api/books.xml?results=details&access_key=" > ++ key > ++ "&index1=isbn&value1=" > ++ isbn :: Request_String) > return $ Right $ getResponseBody rsp > `E.catch` (\(e :: E.SomeException ) -> > return $ Left $ WebsiteUnreachableError) > > I get the following error when I load it into GHCI: > Couldn't match expected type `[Char]' > against inferred type `Either Errors b' > Expected type: IO String > Inferred type: IO (Either Errors b) > In the expression: return $ Left $ WebsiteUnreachableError > In the second argument of `E.catch', namely > `(\ (e :: E.SomeException) > -> return $ Left $ WebsiteUnreachableError)' > Failed, modules loaded: none. > > Control.Exception.catch has the following signature : > catch :: forall a. IO a -> (IOError -> IO a) -> IO a > > I would expect that the handler function needs to return an IO (Either > Errors (IO String)). Why is it trying to return an IO String? Because the call to 'catch' only applies to 'getResponseBody rsp', which is already inside the 'return $ Right $'. Perhaps you want to put parentheses around (return $ Right $ getResponseBody rsp) ? -Brent From info at suud.de Sat May 23 23:08:47 2009 From: info at suud.de (Thomas Friedrich) Date: Sat May 23 22:53:50 2009 Subject: [Haskell-beginners] QuickCheck Message-ID: <4A18BA3F.3070302@suud.de> Hallo everyone, I am currently playing around with the QuickCheck library. I came across the Test.QuickCheck.Batch module a couple of times. I don't seem to have this module installed? Where can I get it and how do I install it? I also would like to invoke the following in ghci. generate 10 (System.Random.mkStdGen 1) arbitrary :: [Int] However, the function `generate` doesn't get imported when importing QuickCheck, and I just cannot find out which module this one would be in. I came to the conclusion that `generate` is just not the best word for a google search :( Cheers, Thomas From lee.duhem at gmail.com Sat May 23 23:21:57 2009 From: lee.duhem at gmail.com (Lee Duhem) Date: Sat May 23 23:07:05 2009 Subject: [Haskell-beginners] QuickCheck In-Reply-To: <4A18BA3F.3070302@suud.de> References: <4A18BA3F.3070302@suud.de> Message-ID: On Sun, May 24, 2009 at 11:08 AM, Thomas Friedrich wrote: > Hallo everyone, > > I am currently playing around with the QuickCheck library. ?I came across > the > > Test.QuickCheck.Batch > > module a couple of times. ?I don't seem to have this module installed? > ?Where can I get it and how do I install it? If you use GHC, you already have it, try ghc-pkg list | grep -i quickcheck > > I also would like to invoke the following in ghci. > > generate 10 (System.Random.mkStdGen 1) arbitrary :: [Int] > > However, the function `generate` doesn't get imported when importing > QuickCheck, and I just cannot find out which module this one would be in. In GHCi, you need import Test.QuickCheck or Debug.QuickCheck before you use generate. > > I came to the conclusion that `generate` is just not the best word for a > google search :( Just check the document of GHC libraries: http://www.haskell.org/ghc/docs/latest/html/libraries/index.html http://www.haskell.org/ghc/docs/latest/html/libraries/QuickCheck/Test-QuickCheck.html lee From info at suud.de Sat May 23 23:25:23 2009 From: info at suud.de (Thomas Friedrich) Date: Sat May 23 23:10:27 2009 Subject: [Haskell-beginners] QuickCheck In-Reply-To: References: <4A18BA3F.3070302@suud.de> Message-ID: <4A18BE23.7050604@suud.de> Hi Lee, I do have GHC and no, I seem to not have it. $ ghc-pkg list | grep -i quickcheck Cabal-1.6.0.3, Chart-0.10.3, HUnit-1.2.0.3, QuickCheck-1.2.0.0, Diff-0.1.2, QuickCheck-2.1.0.1, binary-0.5.0.1, When I am in ghci, I have the following options: Prelude> :m +Test.QuickCheck Test.QuickCheck Test.QuickCheck.Function Test.QuickCheck.Monadic Test.QuickCheck.Property Test.QuickCheck.Test Test.QuickCheck.Arbitrary Test.QuickCheck.Gen Test.QuickCheck.Poly Test.QuickCheck.State Test.QuickCheck.Text Prelude> :m +Test.QuickCheck But no Test.QuickCheck.Batch :( Cheers, Thomas Lee Duhem wrote: > On Sun, May 24, 2009 at 11:08 AM, Thomas Friedrich wrote: > >> Hallo everyone, >> >> I am currently playing around with the QuickCheck library. I came across >> the >> >> Test.QuickCheck.Batch >> >> module a couple of times. I don't seem to have this module installed? >> Where can I get it and how do I install it? >> > > If you use GHC, you already have it, try > ghc-pkg list | grep -i quickcheck > > >> I also would like to invoke the following in ghci. >> >> generate 10 (System.Random.mkStdGen 1) arbitrary :: [Int] >> >> However, the function `generate` doesn't get imported when importing >> QuickCheck, and I just cannot find out which module this one would be in. >> > > In GHCi, you need import Test.QuickCheck or Debug.QuickCheck before > you use generate. > > >> I came to the conclusion that `generate` is just not the best word for a >> google search :( >> > > Just check the document of GHC libraries: > http://www.haskell.org/ghc/docs/latest/html/libraries/index.html > http://www.haskell.org/ghc/docs/latest/html/libraries/QuickCheck/Test-QuickCheck.html > > lee > From info at suud.de Sat May 23 23:47:06 2009 From: info at suud.de (Thomas Friedrich) Date: Sat May 23 23:32:08 2009 Subject: [Haskell-beginners] QuickCheck In-Reply-To: <4A18BE23.7050604@suud.de> References: <4A18BA3F.3070302@suud.de> <4A18BE23.7050604@suud.de> Message-ID: <4A18C33A.8050907@suud.de> Hi folks, I did the following and now it works. I had a closer look at my ghc-pkg list (thanks for the tip), which looked like the following: $ ghc-pkg list /usr/lib/ghc-6.10.3/./package.conf: Cabal-1.6.0.3, Chart-0.10.3, HUnit-1.2.0.3, QuickCheck-1.2.0.0, X11-1.4.5, array-0.2.0.0, base-3.0.3.1, base-4.1.0.0, bytestring-0.9.1.4, cairo-0.10.1, containers-0.2.0.1, data-accessor-0.2.0.2, data-accessor-template-0.2.1.1, directory-1.0.0.3, (dph-base-0.3), (dph-par-0.3), (dph-prim-interface-0.3), (dph-prim-par-0.3), (dph-prim-seq-0.3), (dph-seq-0.3), extensible-exceptions-0.1.1.0, filepath-1.1.0.2, gconf-0.10.1, (ghc-6.10.3), ghc-prim-0.1.0.0, glade-0.10.1, glib-0.10.1, gtk-0.10.1, gtkglext-0.10.1, gtksourceview2-0.10.1, haddock-2.4.2, haskell-src-1.0.1.3, haskell98-1.0.1.0, hpc-0.5.0.3, html-1.0.1.2, integer-0.1.0.1, mtl-1.1.0.2, network-2.2.1, old-locale-1.0.0.1, old-time-1.0.0.2, packedstring-0.1.0.1, parallel-1.1.0.1, parsec-2.1.0.1, pretty-1.0.1.0, process-1.0.1.1, random-1.0.0.1, regex-base-0.72.0.2, regex-compat-0.71.0.1, regex-posix-0.72.0.3, rts-1.0, soegtk-0.10.1, stm-2.1.1.2, svgcairo-0.10.1, syb-0.1.0.1, template-haskell-2.3.0.1, time-1.1.3, transformers-0.1.4.0, unix-2.3.2.0, utf8-string-0.3.4, utility-ht-0.0.5.1, xhtml-3000.2.0.1, xmonad-0.8.1, xmonad-contrib-0.8.1 /home/thomas/.ghc/i386-linux-6.10.3/package.conf: Diff-0.1.2, QuickCheck-2.1.0.1, binary-0.5.0.1, data-accessor-monads-fd-0.2, derive-0.1.4, dotgen-0.2, fingertree-0.0, ghc-paths-0.1.0.5, haskell-lexer-1.0, monads-fd-0.0.0.1, pointedlist-0.3.3, pureMD5-0.2.4, regex-base-0.93.1, regex-tdfa-1.0.0, rosezipper-0.1, split-0.1.1, terminfo-0.3.0.2, uniplate-1.2.0.3, unix-compat-0.1.2.1, vty-3.1.8.4, yi-0.6.0 I think because I installed yi, I must have also ended up with the QuickCheck-2.1.0.1 package. In this package there is NO `generate`, no `verboseCheck` and other functions available. (Why?? Does anyone know?) Also there is no QuickCheck.Batch. Simply hiding the package solved the problem, by ghc-pkg hide QuickCheck-2.1.0.1 Now, I can also use the Batch module. Does anyone know why the now version of QuickCheck is not compatible with the old one? Best, Thomas Thomas Friedrich wrote: > Hi Lee, > > I do have GHC and no, I seem to not have it. > > $ ghc-pkg list | grep -i quickcheck > Cabal-1.6.0.3, Chart-0.10.3, HUnit-1.2.0.3, QuickCheck-1.2.0.0, > Diff-0.1.2, QuickCheck-2.1.0.1, binary-0.5.0.1, > > > When I am in ghci, I have the following options: > > Prelude> :m +Test.QuickCheck > Test.QuickCheck Test.QuickCheck.Function > Test.QuickCheck.Monadic Test.QuickCheck.Property > Test.QuickCheck.Test > Test.QuickCheck.Arbitrary Test.QuickCheck.Gen > Test.QuickCheck.Poly Test.QuickCheck.State > Test.QuickCheck.Text > Prelude> :m +Test.QuickCheck > > But no Test.QuickCheck.Batch > > :( > > Cheers, > Thomas > > > Lee Duhem wrote: >> On Sun, May 24, 2009 at 11:08 AM, Thomas Friedrich wrote: >> >>> Hallo everyone, >>> >>> I am currently playing around with the QuickCheck library. I came >>> across >>> the >>> >>> Test.QuickCheck.Batch >>> >>> module a couple of times. I don't seem to have this module installed? >>> Where can I get it and how do I install it? >>> >> >> If you use GHC, you already have it, try >> ghc-pkg list | grep -i quickcheck >> >> >>> I also would like to invoke the following in ghci. >>> >>> generate 10 (System.Random.mkStdGen 1) arbitrary :: [Int] >>> >>> However, the function `generate` doesn't get imported when importing >>> QuickCheck, and I just cannot find out which module this one would >>> be in. >>> >> >> In GHCi, you need import Test.QuickCheck or Debug.QuickCheck before >> you use generate. >> >> >>> I came to the conclusion that `generate` is just not the best word >>> for a >>> google search :( >>> >> >> Just check the document of GHC libraries: >> http://www.haskell.org/ghc/docs/latest/html/libraries/index.html >> http://www.haskell.org/ghc/docs/latest/html/libraries/QuickCheck/Test-QuickCheck.html >> >> >> lee >> > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners From lee.duhem at gmail.com Sat May 23 23:49:17 2009 From: lee.duhem at gmail.com (Lee Duhem) Date: Sat May 23 23:34:33 2009 Subject: [Haskell-beginners] QuickCheck In-Reply-To: <4A18BE23.7050604@suud.de> References: <4A18BA3F.3070302@suud.de> <4A18BE23.7050604@suud.de> Message-ID: On Sun, May 24, 2009 at 11:25 AM, Thomas Friedrich wrote: > Hi Lee, > > I do have GHC and no, I seem to not have it. > > $ ghc-pkg list | grep -i quickcheck > ? Cabal-1.6.0.3, Chart-0.10.3, HUnit-1.2.0.3, QuickCheck-1.2.0.0, > ? Diff-0.1.2, QuickCheck-2.1.0.1, binary-0.5.0.1, Did you notice that you have two QuickCheck installed? > > > When I am in ghci, I have the following options: > > Prelude> :m +Test.QuickCheck > Test.QuickCheck ? ? ? ? ? ?Test.QuickCheck.Function > Test.QuickCheck.Monadic ? ?Test.QuickCheck.Property ? Test.QuickCheck.Test > Test.QuickCheck.Arbitrary ?Test.QuickCheck.Gen ? ? ? ?Test.QuickCheck.Poly > ? ? Test.QuickCheck.State ? ? ?Test.QuickCheck.Text > Prelude> :m +Test.QuickCheck > > But no Test.QuickCheck.Batch > I guess you use QuickCheck-2.1 instead of 1.2, and Test.QuickCheck.Batch in QuickCheck-1.2, not in 2.1 lee From lee.duhem at gmail.com Sat May 23 23:59:26 2009 From: lee.duhem at gmail.com (Lee Duhem) Date: Sat May 23 23:43:54 2009 Subject: [Haskell-beginners] QuickCheck In-Reply-To: <4A18C33A.8050907@suud.de> References: <4A18BA3F.3070302@suud.de> <4A18BE23.7050604@suud.de> <4A18C33A.8050907@suud.de> Message-ID: On Sun, May 24, 2009 at 11:47 AM, Thomas Friedrich wrote: > > Does anyone know why the now version of QuickCheck is not compatible with > the old one? You should ask the authors directly :-) lee From byorgey at seas.upenn.edu Sun May 24 10:57:57 2009 From: byorgey at seas.upenn.edu (Brent Yorgey) Date: Sun May 24 10:42:24 2009 Subject: [Haskell-beginners] QuickCheck In-Reply-To: <4A18C33A.8050907@suud.de> References: <4A18BA3F.3070302@suud.de> <4A18BE23.7050604@suud.de> <4A18C33A.8050907@suud.de> Message-ID: <20090524145757.GA30700@seas.upenn.edu> On Sat, May 23, 2009 at 11:47:06PM -0400, Thomas Friedrich wrote: > > Does anyone know why the now version of QuickCheck is not compatible with > the old one? Well, the new version is close to a complete rewrite. It's differently organized, has a number of new features, etc.; the fact that it's incompatible with 1.2 is a feature, not a bug. =) That said, I assume there are ways with version 2 to do the same things you were doing with 1.2, but they might have different names and so on. Unfortunately I'm not terribly familiar with the differences; you'll just have to dig around in the documentation. -Brent From manuel.a.castro at gmail.com Sun May 24 11:49:08 2009 From: manuel.a.castro at gmail.com (man) Date: Sun May 24 11:33:35 2009 Subject: [Haskell-beginners] cabal install readline: Missing HSReadlline.h on Mac OS X Leopard Message-ID: <47a7e7c60905240849h1cc823ddsd57c28bfef991b80@mail.gmail.com> I got this error installing readline (I really was installing lambdabot, it depends on readline): Configuring readline-1.0.1.0... checking for gcc... gcc checking for C compiler default output file name... a.out checking whether the C compiler works... yes checking whether we are cross compiling... no checking for suffix of executables... checking for suffix of object files... o checking whether we are using the GNU C compiler... yes checking whether gcc accepts -g... yes checking for gcc option to accept ISO C89... none needed checking for GNUreadline.framework... checking for readline... yes checking for rl_readline_version... yes checking for rl_begin_undo_group... yes checking for rl_erase_empty_line... yes checking for rl_free_undo_list... yes checking for rl_completion_word_break_hook in -lreadline... yes configure: creating ./config.status config.status: creating config.mk config.status: creating readline.buildinfo config.status: creating include/HsReadlineConfig.h cabal: Missing dependency on a foreign library: * Missing header file: HsReadline.h This problem can usually be solved by installing the system package that provides this library (you may need the "-dev" version). If the library is already installed but in a non-standard location then you can use the flags --extra-include-dirs= and --extra-lib-dirs= to specify where it is. cabal: Error: some packages failed to install: lambdabot-4.2.2.1 depends on readline-1.0.1.0 which failed to install. readline-1.0.1.0 failed during the configure step. The exception was: exit: ExitFailure 1 This is on Leopard, with GNUreadline installed. As far as I see, HSReadline is an internally generated header in readline-1.0.1.0, how possibly cabal may be missing it? I got nothing with google, only that readline seems to be problematic on OSX. PS: I just installed cabal-install in ~/.cabal, it installs other packages without fuss (gsl-random being an exception, but it may be the GSL part...). Any help welcome! -- happy hacking... man -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090524/f5835ab2/attachment.html From xykaumjunior at gmail.com Sun May 24 15:13:18 2009 From: xykaumjunior at gmail.com (=?ISO-8859-1?Q?Chic=E3o_Fernandes_Junior?=) Date: Sun May 24 14:57:45 2009 Subject: [Haskell-beginners] Empty list Exception Message-ID: Hello! I'm starting programming in haskell and I got some problems running in GHCi (version 6.8.2) a module I've wrote. The following ... --The first string is the name and the second is the user ID, just for test > type User = (String, String, Integer) > > users = [("Marcelo Castro", "1234", 2), > ("Joao Vicente Claudino", "1235", 2), > ("Wilson Brandao", "1236", 2), > ("Humberto Teixeira", "1237", 2), > ("Luiz Gonzaga", "1238", 2), > ("Severino Dias", "1239", 2), > ("Hermeto Pascoal", "1234", 2)] > > checkUser :: User -> [User] -> Bool > checkUser user userList > = if (head (filter (\userParse -> userParse == user) userList)) == > user > then True > else False > > > userRegistration :: User -> [User] -> [User] > userRegistration user list > | checkUser user list = list > | not(checkUser user list) = (user : list) > | otherwise = [] > -- ____________________________________________________________________________ Francisco Borges "Chic?o" Junior .......................................................................................... http://pedepinico.blogspot.com .......................................................................................... "Quem de boa vontade carrega o dif?cil, tamb?m carrega o menos dif?cil..." Lao Ts? - Tao Te Ching -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090524/10ee5b31/attachment.html From xykaumjunior at gmail.com Sun May 24 15:17:16 2009 From: xykaumjunior at gmail.com (=?ISO-8859-1?Q?Chic=E3o_Fernandes_Junior?=) Date: Sun May 24 15:01:44 2009 Subject: [Haskell-beginners] Re: Empty list Exception In-Reply-To: References: Message-ID: Sorry, I send it accidently. I did't finish my question. Anyway, when I run in GHCi: userRegistration ("Joao Claudino","1240",2) users > It gives me the exception: *** Exception: Prelude.head: empty list > Can anyone explain to me why the exception is thown and what should I do to fix it? Thanks -- ____________________________________________________________________________ Francisco Borges "Chic?o" Junior .......................................................................................... http://pedepinico.blogspot.com .......................................................................................... "Quem de boa vontade carrega o dif?cil, tamb?m carrega o menos dif?cil..." Lao Ts? - Tao Te Ching -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090524/c82826b2/attachment.html From wingedtachikoma at gmail.com Sun May 24 15:35:10 2009 From: wingedtachikoma at gmail.com (Sean Bartell) Date: Sun May 24 15:19:37 2009 Subject: [Haskell-beginners] Re: Empty list Exception In-Reply-To: References: Message-ID: When you use filter in this case, it's checking for users that are the same as ("Joao Claudino", "1240", 2). There aren't any in the list, so filter returns []. You can't use head on [], because there's no first element. Other notes about your code: - if x then True else False is the same as just x. - The otherwise case for userRegistration can never be reached. I would write it more like this: checkUser user userList = elem user userList userRegistration user list | checkUser user list = list | otherwise = user:list -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090524/c593ff74/attachment.html From xykaumjunior at gmail.com Sun May 24 22:14:42 2009 From: xykaumjunior at gmail.com (=?ISO-8859-1?Q?Chic=E3o_Fernandes_Junior?=) Date: Sun May 24 21:59:08 2009 Subject: [Haskell-beginners] Re: Empty list Exception In-Reply-To: <5e0b968a0905241235v1b467f8dn8ec70ddf71c77b92@mail.gmail.com> References: <5e0b968a0905241235v1b467f8dn8ec70ddf71c77b92@mail.gmail.com> Message-ID: Thanks everyone, for the answers. I didn't knew the *elem* function. That will make my life easier. For the sake of simplicity, I changed the code to a shorter *userRegistration* function, eliminating the *checkUser* function: userRegistration user list > | elem user list = list > | otherwise = user:list > It did worked fine. I'm glad that I can count with this list in my learning process. Thanks again. -- ____________________________________________________________________________ Francisco Borges "Chic?o" Junior .......................................................................................... http://pedepinico.blogspot.com .......................................................................................... "Quem de boa vontade carrega o dif?cil, tamb?m carrega o menos dif?cil..." Lao Ts? - Tao Te Ching -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090524/e07ea037/attachment-0001.html From alexey.skladnoy at gmail.com Mon May 25 07:41:16 2009 From: alexey.skladnoy at gmail.com (Khudyakov Alexey) Date: Mon May 25 07:22:48 2009 Subject: [Haskell-beginners] cabal install readline: Missing HSReadlline.h on Mac OS X Leopard In-Reply-To: <47a7e7c60905240849h1cc823ddsd57c28bfef991b80@mail.gmail.com> References: <47a7e7c60905240849h1cc823ddsd57c28bfef991b80@mail.gmail.com> Message-ID: <200905251541.16454.alexey.skladnoy@gmail.com> On Sunday 24 of May 2009 19:49:08 man wrote: > I got this error installing readline (I really was installing lambdabot, it > depends on readline): > > config.status: creating config.mk > config.status: creating readline.buildinfo > config.status: creating include/HsReadlineConfig.h > __cabal: Missing dependency on a foreign library:__ > * Missing header file: HsReadline.h > This problem can usually be solved by installing the system package that > provides this library (you may need the "-dev" version). If the library is > already installed but in a non-standard location then you can use the flags > --extra-include-dirs= and --extra-lib-dirs= to specify where it is. > cabal: Error: some packages failed to install: > I suppose that cabal could not find headers for readline. Check that they are installed and cabal is able to find them. From voigt at tcs.inf.tu-dresden.de Tue May 26 04:12:58 2009 From: voigt at tcs.inf.tu-dresden.de (Janis Voigtlaender) Date: Tue May 26 03:57:22 2009 Subject: [Haskell-beginners] ANNOUNCE: Haskell Communities and Activities Report (16th ed., May 2009) Message-ID: <4A1BA48A.5060901@tcs.inf.tu-dresden.de> On behalf of the many, many contributors, I am pleased to announce that the Haskell Communities and Activities Report (16th edition, May 2009) http://www.haskell.org/communities/ is now available from the Haskell Communities home page in PDF and HTML formats. Many thanks go to all the people that contributed to this report, both directly, by sending in descriptions, and indirectly, by doing all the interesting things that are reported. I hope you will find it as interesting a read as I did. If you have not encountered the Haskell Communities and Activities Reports before, you may like to know that the first of these reports was published in November 2001. Their goal is to improve the communication between the increasingly diverse groups, projects, and individuals working on, with, or inspired by Haskell. The idea behind these reports is simple: Every six months, a call goes out to all of you enjoying Haskell to contribute brief summaries of your own area of work. Many of you respond (eagerly, unprompted, and sometimes in time for the actual deadline ;-) to the call. The editor collects all the contributions into a single report and feeds that back to the community. When I try for the next update, six months from now, you might want to report on your own work, project, research area or group as well. So, please put the following into your diaries now: ---------------------------------------- End of October 2009: target deadline for contributions to the November 2009 edition of the HC&A Report ---------------------------------------- Unfortunately, many Haskellers working on interesting projects are so busy with their work that they seem to have lost the time to follow the Haskell related mailing lists and newsgroups, and have trouble even finding time to report on their work. If you are a member, user or friend of a project so burdened, please find someone willing to make time to report and ask them to "register" with the editor for a simple e-mail reminder in October (you could point me to them as well, and I can then politely ask if they want to contribute, but it might work better if you do the initial asking). Of course, they will still have to find the ten to fifteen minutes to draw up their report, but maybe we can increase our coverage of all that is going on in the community. Feel free to circulate this announcement further in order to reach people who might otherwise not see it. Enjoy! Janis Voigtlaender -- Dr. Janis Voigtlaender http://wwwtcs.inf.tu-dresden.de/~voigt/ mailto:voigt@tcs.inf.tu-dresden.de From lee.duhem at gmail.com Tue May 26 05:36:19 2009 From: lee.duhem at gmail.com (Lee Duhem) Date: Tue May 26 05:20:40 2009 Subject: [Haskell-beginners] Re: [Haskell] ANNOUNCE: Haskell Communities and Activities Report (16th ed., May 2009) In-Reply-To: <4A1BA48A.5060901@tcs.inf.tu-dresden.de> References: <4A1BA48A.5060901@tcs.inf.tu-dresden.de> Message-ID: On Tue, May 26, 2009 at 4:12 PM, Janis Voigtlaender wrote: > On behalf of the many, many contributors, I am pleased to announce > that the > > ? ? ? ? ? ?Haskell Communities and Activities Report > ? ? ? ? ? ? ? ? ?(16th edition, May 2009) > > ? ? ? ? ? ? http://www.haskell.org/communities/ > > is now available from the Haskell Communities home page in PDF and > HTML formats. > Congratulation and thank you for your work. BTW, I notice Haskell-Cafe isn't in the To list, I hope this isn't in purpose :-) lee From simonpj at microsoft.com Tue May 26 05:22:21 2009 From: simonpj at microsoft.com (Simon Peyton-Jones) Date: Tue May 26 22:21:59 2009 Subject: [Haskell-beginners] RE: [Haskell] ANNOUNCE: Haskell Communities and Activities Report (16th ed., May 2009) In-Reply-To: <4A1BA48A.5060901@tcs.inf.tu-dresden.de> References: <4A1BA48A.5060901@tcs.inf.tu-dresden.de> Message-ID: <638ABD0A29C8884A91BC5FB5C349B1C337FC667310@EA-EXMSG-C334.europe.corp.microsoft.com> | On behalf of the many, many contributors, I am pleased to announce | that the | Haskell Communities and Activities Report | (16th edition, May 2009) | is now available from the Haskell Communities home page in PDF and | HTML formats. Congratulations Janis! Thank you from all of us for editing the HCAR. Simon From info at suud.de Wed May 27 19:55:05 2009 From: info at suud.de (Thomas Friedrich) Date: Wed May 27 19:39:58 2009 Subject: [Haskell-beginners] System.Random Message-ID: <4A1DD2D9.8090600@suud.de> Hi everyone, Using random numbers in Haskell is not entirely trivial (at least, still not for me) and again I am bagging my head against the Gen-Monad. I'd like to write a kind of bootstrap function sample :: Int -> [a] -> [a] sample n xs = ... that samples uniformly n elements from the list xs. I am not sure how to go about this. Would you try something like sample1 :: StdGen -> Int -> [a] -> [a] and later use this in an IO Monad, something along the lines of: do {g <- mkStdGen; return $ sample g n xs}, or would you write it like sample2 :: Int -> [a] -> Gen [a] and then use a function from QuickCheck like `generate` to get your samples? You see, I don't even know how to start thinking about the problem. If anyone got an idea, I'd be pleased if you could help me. Cheers, Thomas From alexander.dunlap at gmail.com Wed May 27 23:08:52 2009 From: alexander.dunlap at gmail.com (Alexander Dunlap) Date: Wed May 27 23:01:07 2009 Subject: [Haskell-beginners] System.Random In-Reply-To: <4A1DD2D9.8090600@suud.de> References: <4A1DD2D9.8090600@suud.de> Message-ID: <57526e770905272008g3e6b50b7q2fb38b6ea9288bb2@mail.gmail.com> On Wed, May 27, 2009 at 4:55 PM, Thomas Friedrich wrote: > Hi everyone, > > Using random numbers in Haskell is not entirely trivial (at least, still not > for me) and again I am bagging my head against the Gen-Monad. ?I'd like to > write a kind of bootstrap function > > sample :: Int -> [a] -> [a] > sample n xs = ... > > that samples uniformly n elements from the list xs. ?I am not sure how to go > about this. ?Would you try something like > > sample1 :: StdGen -> Int -> [a] -> [a] > > and later use this in an IO Monad, something along the lines of: ?do {g <- > mkStdGen; return $ sample g n xs}, or would you write it like > > sample2 :: Int -> [a] -> Gen [a] > > and then use a function from QuickCheck like `generate` to get your samples? > > You see, I don't even know how to start thinking about the problem. > > If anyone got an idea, I'd be pleased if you could help me. > > Cheers, > Thomas > > In general, I don't think you'd use QuickCheck to generate random numbers. QuickCheck is, for the most part, solely for testing. I would write sample1 :: [a] -> StdGen -> a sample :: [a] -> StdGen -> [a] -- defined in terms of sample1 and use the take function to trim the list to the number of items you want. (Lazy evaluation guarantees that the rest won't get evaluated.) Use mkStdGen to create a StdGen to pass to it. Alex From regis.saint-paul at create-net.org Thu May 28 04:06:25 2009 From: regis.saint-paul at create-net.org (Regis Saint-Paul) Date: Thu May 28 03:51:05 2009 Subject: [Haskell-beginners] System.Random In-Reply-To: <4A1DD2D9.8090600@suud.de> References: <4A1DD2D9.8090600@suud.de> Message-ID: <0ECE801A1FC4469E91ECB541D2758EBF@createnet.org> Hi, Maybe you'd like to check this page: http://apfelmus.nfshost.com/random-permutations.html which gives also interesting pointers to other references. Your sampling from a list can be seen as taking the first n elements of a permutation of the list. Of course, computing the full permutation of a long list to just take the first two or three elements would not be very wise. So there are other solutions depending on your setting (sampling from a finite list with or without replacement, sampling from an infinite list (a stream) as data comes in, etc... I just give some rough idea: - for fixed size lists, a random number can be used to skip some elements of the list, if you start from the beginning of the list after reaching its end, then you sample with replacement. - for infinite list, a list of the n sampled element is maintained. This list is first filled with the first n elements of the source list and then, for each new incoming element, you decide randomly if you want to keep it or not. If you decide to keep it, you decide, also randomly, which element of the sample list it replaces. You can find all the details in [1]. Do you need that for performing tests with QuickCheck or was using QuickCheck just an idea on how to go about the problem? Best, Regis [1] Vitter, Jeffrey S. 1985. "Random sampling with a reservoir." ACM Trans. Math. Softw. 11(1):37-57. > -----Original Message----- > From: beginners-bounces@haskell.org [mailto:beginners-bounces@haskell.org] > On Behalf Of Thomas Friedrich > Sent: Thursday, 28 May 2009 1:55 AM > To: beginners > Subject: [Haskell-beginners] System.Random > > Hi everyone, > > Using random numbers in Haskell is not entirely trivial (at least, still > not for me) and again I am bagging my head against the Gen-Monad. I'd > like to write a kind of bootstrap function > > sample :: Int -> [a] -> [a] > sample n xs = ... > > that samples uniformly n elements from the list xs. I am not sure how > to go about this. Would you try something like > > sample1 :: StdGen -> Int -> [a] -> [a] > > and later use this in an IO Monad, something along the lines of: do {g > <- mkStdGen; return $ sample g n xs}, or would you write it like > > sample2 :: Int -> [a] -> Gen [a] > > and then use a function from QuickCheck like `generate` to get your > samples? > > You see, I don't even know how to start thinking about the problem. > > If anyone got an idea, I'd be pleased if you could help me. > > Cheers, > Thomas > > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners From info at suud.de Thu May 28 10:22:21 2009 From: info at suud.de (Thomas Friedrich) Date: Thu May 28 10:07:13 2009 Subject: [Haskell-beginners] System.Random In-Reply-To: <0ECE801A1FC4469E91ECB541D2758EBF@createnet.org> References: <4A1DD2D9.8090600@suud.de> <0ECE801A1FC4469E91ECB541D2758EBF@createnet.org> Message-ID: <4A1E9E1D.7000504@suud.de> Hi, thanks for the web-link. The idea of putting the whole thing into a random shuffling framework is in deed a really neat idea. And it perfectly suits my purpose. I don't need it for QuickCheck, it was just an idea. Cheers, Thomas Regis Saint-Paul wrote: > Hi, > > Maybe you'd like to check this page: > http://apfelmus.nfshost.com/random-permutations.html > which gives also interesting pointers to other references. > > Your sampling from a list can be seen as taking the first n elements of a > permutation of the list. Of course, computing the full permutation of a long > list to just take the first two or three elements would not be very wise. So > there are other solutions depending on your setting (sampling from a finite > list with or without replacement, sampling from an infinite list (a stream) > as data comes in, etc... > I just give some rough idea: > - for fixed size lists, a random number can be used to skip some elements of > the list, if you start from the beginning of the list after reaching its > end, then you sample with replacement. > - for infinite list, a list of the n sampled element is maintained. This > list is first filled with the first n elements of the source list and then, > for each new incoming element, you decide randomly if you want to keep it or > not. If you decide to keep it, you decide, also randomly, which element of > the sample list it replaces. You can find all the details in [1]. > > Do you need that for performing tests with QuickCheck or was using > QuickCheck just an idea on how to go about the problem? > > Best, > Regis > > [1] Vitter, Jeffrey S. 1985. "Random sampling with a reservoir." ACM Trans. > Math. Softw. 11(1):37-57. > > > > > > >> -----Original Message----- >> From: beginners-bounces@haskell.org [mailto:beginners-bounces@haskell.org] >> On Behalf Of Thomas Friedrich >> Sent: Thursday, 28 May 2009 1:55 AM >> To: beginners >> Subject: [Haskell-beginners] System.Random >> >> Hi everyone, >> >> Using random numbers in Haskell is not entirely trivial (at least, still >> not for me) and again I am bagging my head against the Gen-Monad. I'd >> like to write a kind of bootstrap function >> >> sample :: Int -> [a] -> [a] >> sample n xs = ... >> >> that samples uniformly n elements from the list xs. I am not sure how >> to go about this. Would you try something like >> >> sample1 :: StdGen -> Int -> [a] -> [a] >> >> and later use this in an IO Monad, something along the lines of: do {g >> <- mkStdGen; return $ sample g n xs}, or would you write it like >> >> sample2 :: Int -> [a] -> Gen [a] >> >> and then use a function from QuickCheck like `generate` to get your >> samples? >> >> You see, I don't even know how to start thinking about the problem. >> >> If anyone got an idea, I'd be pleased if you could help me. >> >> Cheers, >> Thomas >> >> >> _______________________________________________ >> Beginners mailing list >> Beginners@haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> > > From gilbertw1 at gmail.com Thu May 28 11:50:36 2009 From: gilbertw1 at gmail.com (William Gilbert) Date: Thu May 28 11:34:50 2009 Subject: [Haskell-beginners] Haskell Generic Function Question Message-ID: I am trying to write a function that will covert either an integer or an int into a list containing its digits. ex. toIntegralList 123 -> [1,2,3] I have written the following definition that tries to use read to generically cast a string value to an Integral type that is the same as the Integral passed in: toIntegralList :: (Integral a) => a -> [a] toIntegralList x = map (\c -> read [c] :: a) (show x) I understand it would be very simple to just create two functions, one that converts an Int and one that converts an Integer, however I was wondering if there were any way to accomplish what I am trying to do here. Thanks In Advance, Bryan From mail at paulvisschers.net Thu May 28 12:49:15 2009 From: mail at paulvisschers.net (Paul Visschers) Date: Thu May 28 12:33:30 2009 Subject: [Haskell-beginners] Haskell Generic Function Question In-Reply-To: References: Message-ID: <4A1EC08B.3060506@paulvisschers.net> I assume the problem is that the function doesn't compile. This should work: > toIntegralList :: (Read a, Show a) => a -> [a] > toIntegralList x = map (\c -> read [c]) (show x) This adds the required Read and Show instances, which are necessary because of the read and show functions, respectively. Also note that I have omitted your extra type annotation, which also causes an compile error. The problem with this functions is that you can use it on a lot of stuff that isn't a number, and you'll get a runtime read error, to remedy this, just reinsert the Integral type class requirement: > toIntegralList :: (Integral a, Read a, Show a) => a -> [a] > toIntegralList x = map (\c -> read [c]) (show x) Hope this helps, Paul William Gilbert wrote: > I am trying to write a function that will covert either an integer or > an int into a list containing its digits. > > ex. toIntegralList 123 -> [1,2,3] > > I have written the following definition that tries to use read to > generically cast a string value to an Integral type that is the same > as the Integral passed in: > > toIntegralList :: (Integral a) => a -> [a] > toIntegralList x = map (\c -> read [c] :: a) (show x) > > I understand it would be very simple to just create two functions, one > that converts an Int and one that converts an Integer, however I was > wondering if there were any way to accomplish what I am trying to do > here. > > Thanks In Advance, > Bryan > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners From es at ertes.de Thu May 28 12:55:06 2009 From: es at ertes.de (Ertugrul Soeylemez) Date: Thu May 28 12:39:34 2009 Subject: [Haskell-beginners] Re: Haskell Generic Function Question References: Message-ID: <20090528185506.1e62c685@tritium.xx> William Gilbert wrote: > I am trying to write a function that will covert either an integer or > an int into a list containing its digits. > > ex. toIntegralList 123 -> [1,2,3] > > I have written the following definition that tries to use read to > generically cast a string value to an Integral type that is the same > as the Integral passed in: > > toIntegralList :: (Integral a) => a -> [a] > toIntegralList x = map (\c -> read [c] :: a) (show x) > > I understand it would be very simple to just create two functions, one > that converts an Int and one that converts an Integer, however I was > wondering if there were any way to accomplish what I am trying to do > here. Of course you can use read and show for that, but personally I find it more appropriate to write the algorithm yourself. It will be faster and give you a much more useful digit ordering, namely starting with the least significant digit: toDigits :: Integral i => i -> i -> [i] toDigits base = takeWhile (>0) . map (`rem` base) . iterate (`div` base) toDecimalDigits :: Integral i => i -> [i] toDecimalDigits = toDigits 10 fromDigits :: Num a => a -> [a] -> a fromDigits base = foldr (\d c -> base*c + d) 0 fromDecimalDigits :: Num a => [a] -> a fromDecimalDigits = fromDigits 10 Greets, Ertugrul. -- nightmare = unsafePerformIO (getWrongWife >>= sex) http://blog.ertes.de/ From info at suud.de Thu May 28 13:05:12 2009 From: info at suud.de (Thomas Friedrich) Date: Thu May 28 12:50:11 2009 Subject: [Haskell-beginners] Haskell Generic Function Question In-Reply-To: References: Message-ID: <4A1EC448.9040805@suud.de> Hi William, I would also strongly success not to make the detour using read and show. Keep things simple. Here is my suggestion: toDigit x = case f x of (0,b) -> [b] (a,b) -> toDigit a ++ [b] f = \x -> (x `quot` 10, x `mod` 10) Best, Thomas William Gilbert wrote: > I am trying to write a function that will covert either an integer or > an int into a list containing its digits. > > ex. toIntegralList 123 -> [1,2,3] > > I have written the following definition that tries to use read to > generically cast a string value to an Integral type that is the same > as the Integral passed in: > > toIntegralList :: (Integral a) => a -> [a] > toIntegralList x = map (\c -> read [c] :: a) (show x) > > I understand it would be very simple to just create two functions, one > that converts an Int and one that converts an Integer, however I was > wondering if there were any way to accomplish what I am trying to do > here. > > Thanks In Advance, > Bryan > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > From jeff at nokrev.com Thu May 28 13:18:22 2009 From: jeff at nokrev.com (Jeff Wheeler) Date: Thu May 28 13:02:42 2009 Subject: [Haskell-beginners] Haskell Generic Function Question In-Reply-To: <4A1EC448.9040805@suud.de> References: <4A1EC448.9040805@suud.de> Message-ID: <1243531102.10719.2.camel@ulysses> On Thu, 2009-05-28 at 13:05 -0400, Thomas Friedrich wrote: > toDigit x = case f x of > (0,b) -> [b] > (a,b) -> toDigit a ++ [b] > > f = \x -> (x `quot` 10, x `mod` 10) Your function f is almost the same as divMod in Prelude. Also, using a lambda function seems odd; this is simpler: > f x = (x `quot` 10, x `mod` 10) Anyways, because that's essentially just divMod, toDigit can be simplified thusly: > toDigit x = case x `divMod` 10 of > (0, b) -> [b] > (a, b) -> toDigit a ++ [b] Jeff Wheeler From nathanmholden at gmail.com Thu May 28 17:53:43 2009 From: nathanmholden at gmail.com (Nathan Holden) Date: Thu May 28 17:43:38 2009 Subject: [Haskell-beginners] Case Expressions Message-ID: <305228b20905281453m227462f2iab0b4337ff8a5151@mail.gmail.com> I've been playing with my musical interval problem from before, and I got a little further, but I came up with a problem. I have two types-- Interval and BasicInterval. Interval has 12 constructors, Unison, MinorSecond, MajorSecond, MinorThird, et cetera. BasicInterval has 8 constructors, BasicUnison, Second, Third, Fourth, and so on. I want to be able to convert between them somewhat; I have function interToBasic, which at the moment looks like: interToBasic :: Interval -> BasicInterval interToBasic a = if (b == 1 || b == 2) then Second else if (b == 3 || b == 4) then Third .. where b = fromEnum a What I wanted to do, and figure is probably doable, but I can't seem to find it, is to say something like case (fromEnum a) of 1 || 2 -> Second 3 || 4 -> Third .. Is this doable? If so, what's the syntax? -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090528/02601f12/attachment.html From byorgey at seas.upenn.edu Thu May 28 18:27:31 2009 From: byorgey at seas.upenn.edu (Brent Yorgey) Date: Thu May 28 18:10:46 2009 Subject: [Haskell-beginners] Case Expressions In-Reply-To: <305228b20905281453m227462f2iab0b4337ff8a5151@mail.gmail.com> References: <305228b20905281453m227462f2iab0b4337ff8a5151@mail.gmail.com> Message-ID: <20090528222731.GA7510@seas.upenn.edu> On Thu, May 28, 2009 at 05:53:43PM -0400, Nathan Holden wrote: > > interToBasic :: Interval -> BasicInterval > interToBasic a = if (b == 1 || b == 2) then Second > else if (b == 3 || b == 4) then Third > .. > where b = fromEnum a > > What I wanted to do, and figure is probably doable, but I can't seem to find > it, is to say something like > > case (fromEnum a) of > 1 || 2 -> Second > 3 || 4 -> Third > .. > > Is this doable? If so, what's the syntax? You can do this with pattern guards, like so: interToBasic a | b == 1 || b == 2 = Second | b == 3 || b == 4 = Third ... where b = fromEnum a You could also use b `elem` [1,2] as an alternative to b == 1 || b == 2 (especially nice when there are more than two options). -Brent From jj36 at hw.ac.uk Thu May 28 13:01:58 2009 From: jj36 at hw.ac.uk (Jan Jakubuv) Date: Thu May 28 22:24:13 2009 Subject: [Haskell-beginners] Haskell Generic Function Question In-Reply-To: References: Message-ID: <20090528170158.GA24741@lxultra2.macs.hw.ac.uk> Hi Bryan, I think that it isn't a very good idea to use `read/show` to do some numeric computations. You can use standard functions `div` and `mod` which work with any Integrals. digits :: (Integral a) => a -> [a] digits 0 = [] digits n = digits (n `div` 10) ++ [n `mod` 10] This code behaves differently on 0 then your one (also on negative numbers). You can fix it easily, and moreover, you may want to use `divMod` and some accumulator to improve efficiency: digits2 :: (Integral a) => a -> [a] digits2 0 = [0] digits2 n = digits2' n [] where digits2' 0 acc = acc digits2' n acc = let (r,d) = divMod n 10 in digits2' r (d:acc) I hope I understood well what you were asking about ;-) Btw, to make your code working I needed to write it as: toIntegralList :: (Read a, Integral a) => a -> [a] toIntegralList (x :: a) = map (\c -> read [c] :: a) (show x) Sincerely, Jan. On Thu, May 28, 2009 at 11:50:36AM -0400, William Gilbert wrote: > I am trying to write a function that will covert either an integer or > an int into a list containing its digits. > > ex. toIntegralList 123 -> [1,2,3] > > I have written the following definition that tries to use read to > generically cast a string value to an Integral type that is the same > as the Integral passed in: > > toIntegralList :: (Integral a) => a -> [a] > toIntegralList x = map (\c -> read [c] :: a) (show x) > > I understand it would be very simple to just create two functions, one > that converts an Int and one that converts an Integer, however I was > wondering if there were any way to accomplish what I am trying to do > here. > > Thanks In Advance, > Bryan > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners -- Heriot-Watt University is a Scottish charity registered under charity number SC000278. From heringtonlacey at mindspring.com Fri May 29 01:40:07 2009 From: heringtonlacey at mindspring.com (Dean Herington) Date: Fri May 29 01:24:42 2009 Subject: [Haskell-beginners] Case Expressions In-Reply-To: <20090528222731.GA7510@seas.upenn.edu> References: <305228b20905281453m227462f2iab0b4337ff8a5151@mail.gmail.com> <20090528222731.GA7510@seas.upenn.edu> Message-ID: And I'm not sure what going via integers (via `fromEnum`) buys you. Why not: interToBasic :: Interval -> BasicInterval interToBasic a | a `elem` [MinorSecond, MajorSecond] = Second | a `elem` [MinorThird, MajorThird] = Third ... Dean At 6:27 PM -0400 5/28/09, Brent Yorgey wrote: >On Thu, May 28, 2009 at 05:53:43PM -0400, Nathan Holden wrote: >> > > interToBasic :: Interval -> BasicInterval > > interToBasic a = if (b == 1 || b == 2) then Second >> else if (b == 3 || b == 4) then Third >> .. >> where b = fromEnum a >> >> What I wanted to do, and figure is probably doable, but I can't seem to find >> it, is to say something like >> >> case (fromEnum a) of >> 1 || 2 -> Second >> 3 || 4 -> Third >> .. >> >> Is this doable? If so, what's the syntax? > >You can do this with pattern guards, like so: > >interToBasic a | b == 1 || b == 2 = Second > | b == 3 || b == 4 = Third > ... > where b = fromEnum a > >You could also use b `elem` [1,2] as an alternative to b == 1 || b == >2 (especially nice when there are more than two options). > >-Brent From info at suud.de Fri May 29 10:27:26 2009 From: info at suud.de (Thomas Friedrich) Date: Fri May 29 10:12:24 2009 Subject: [Haskell-beginners] Case Expressions In-Reply-To: <305228b20905281453m227462f2iab0b4337ff8a5151@mail.gmail.com> References: <305228b20905281453m227462f2iab0b4337ff8a5151@mail.gmail.com> Message-ID: <4A1FF0CE.2090304@suud.de> Hi Nathan, do you mean something like the following? data Interval = Unison | MinorSecond | MajorSecond | MinorThird | MajorThird deriving (Show) data BasicInterval = BasicUnison | Second | Third | Fourth deriving (Show) inter2basic :: Interval -> BasicInterval inter2basic x = case x of Unison -> BasicUnison MinorSecond -> Second MajorSecond -> Second MinorThird -> Third MajorThird -> Third Happy Hacking, Thomas Nathan Holden wrote: > I've been playing with my musical interval problem from before, and I > got a little further, but I came up with a problem. > > I have two types-- Interval and BasicInterval. > > Interval has 12 constructors, Unison, MinorSecond, MajorSecond, > MinorThird, et cetera. > BasicInterval has 8 constructors, BasicUnison, Second, Third, Fourth, > and so on. > > I want to be able to convert between them somewhat; > > I have function interToBasic, which at the moment looks like: > > interToBasic :: Interval -> BasicInterval > interToBasic a = if (b == 1 || b == 2) then Second > else if (b == 3 || b == 4) then Third > .. > where b = fromEnum a > > What I wanted to do, and figure is probably doable, but I can't seem > to find it, is to say something like > > case (fromEnum a) of > 1 || 2 -> Second > 3 || 4 -> Third > .. > > Is this doable? If so, what's the syntax? > ------------------------------------------------------------------------ > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > From nathanmholden at gmail.com Fri May 29 10:38:23 2009 From: nathanmholden at gmail.com (Nathan Holden) Date: Fri May 29 10:22:35 2009 Subject: [Haskell-beginners] Case Expressions In-Reply-To: <4A1FF0CE.2090304@suud.de> References: <305228b20905281453m227462f2iab0b4337ff8a5151@mail.gmail.com> <4A1FF0CE.2090304@suud.de> Message-ID: <305228b20905290738g35317299m7f31b0d84af12cd7@mail.gmail.com> Yes. I did. In fact, that's what it looks like. But in general I also derive Enum, and use Ints instead, because that's how I conveniently switch between notes and intervals. On Fri, May 29, 2009 at 10:27 AM, Thomas Friedrich wrote: > Hi Nathan, > > do you mean something like the following? > > data Interval = Unison > | MinorSecond > | MajorSecond > | MinorThird > | MajorThird > deriving (Show) > > data BasicInterval = BasicUnison > | Second > | Third > | Fourth > deriving (Show) > > inter2basic :: Interval -> BasicInterval > inter2basic x = case x of > Unison -> BasicUnison > MinorSecond -> Second > MajorSecond -> Second > MinorThird -> Third > MajorThird -> Third > > > Happy Hacking, > Thomas > > > > Nathan Holden wrote: > >> I've been playing with my musical interval problem from before, and I got >> a little further, but I came up with a problem. >> >> I have two types-- Interval and BasicInterval. >> >> Interval has 12 constructors, Unison, MinorSecond, MajorSecond, >> MinorThird, et cetera. >> BasicInterval has 8 constructors, BasicUnison, Second, Third, Fourth, and >> so on. >> >> I want to be able to convert between them somewhat; >> >> I have function interToBasic, which at the moment looks like: >> >> interToBasic :: Interval -> BasicInterval >> interToBasic a = if (b == 1 || b == 2) then Second >> else if (b == 3 || b == 4) then Third >> .. >> where b = fromEnum a >> >> What I wanted to do, and figure is probably doable, but I can't seem to >> find it, is to say something like >> >> case (fromEnum a) of >> 1 || 2 -> Second >> 3 || 4 -> Third >> .. >> >> Is this doable? If so, what's the syntax? >> ------------------------------------------------------------------------ >> >> _______________________________________________ >> Beginners mailing list >> Beginners@haskell.org >> http://www.haskell.org/mailman/listinfo/beginners >> >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090529/0d25bb03/attachment.html From es at ertes.de Fri May 29 19:32:48 2009 From: es at ertes.de (Ertugrul Soeylemez) Date: Fri May 29 19:17:13 2009 Subject: [Haskell-beginners] Re: Haskell Generic Function Question References: <4A1EC448.9040805@suud.de> <1243531102.10719.2.camel@ulysses> Message-ID: <20090530013248.77299a54@tritium.xx> Jeff Wheeler wrote: > > f = \x -> (x `quot` 10, x `mod` 10) > > Your function f is almost the same as divMod in Prelude. Also, using a > lambda function seems odd; this is simpler: > > > f x = (x `quot` 10, x `mod` 10) Hmm, simplicity. f = (`quot` 10) &&& (`mod` 10) SCNR. Greets, Ertugrul. -- nightmare = unsafePerformIO (getWrongWife >>= sex) http://blog.ertes.de/ From mpm at alumni.caltech.edu Sat May 30 03:09:23 2009 From: mpm at alumni.caltech.edu (Michael Mossey) Date: Sat May 30 02:53:55 2009 Subject: [Haskell-beginners] Re: [Haskell-cafe] can someone point me to more help about Database.HDBC.ODBC? In-Reply-To: <4A20C057.9020708@post.ustu.ru> References: <4A20A190.4090809@alumni.caltech.edu> <4A20C057.9020708@post.ustu.ru> Message-ID: <4A20DBA3.6050805@alumni.caltech.edu> This is on Windows, by the way. Do you know the equivalent file names and locations? What is DSN? Thanks, Mike ??????? ??????? wrote: > Usually you should pass only DSN name in the string, if you used InnoDB > database in MySQL. > > For example, /etc/odbc.ini file: > ; > ; odbc.ini configuration for MyODBC and MyODBC 3.51 Drivers > ; > [.....] > > [Default] > Driver = MySQL > Description = MySQL ODBC 3.51 Driver DSN > Server = localhost > Port = > User = root > Password = > Database = test > Option = 3 > Socket = > > [......] > > Next step, you can test a connection with command: > $ isql Default > +---------------------------------------+ > | Connected! | > | | > | sql-statement | > | help [tablename] | > | quit | > | | > +---------------------------------------+ > SQL> > > And finally, the haskell code: > > conn <- connectODBC "DSN=Default;" > > From ivan at llaisdy.com Sat May 30 16:37:05 2009 From: ivan at llaisdy.com (Ivan Uemlianin) Date: Sat May 30 16:21:14 2009 Subject: [Haskell-beginners] Maybe a and Maybe t Message-ID: <4A2198F1.4020306@llaisdy.com> Dear All I have just started learning Haskell, using the O'Reilly book Real World Haskell [1]. I'm finding it a very exciting language, and I'm working my way painstakingly through the book (I'm up to the exercises at the end of Chapter 3). I have a query based on some code from the book. This question has been asked on the book web site [2], but no answer. I have also Googled about and scanned the documentation. So, here is some code (quoted from [3]): two implementations of a function to return the second element of a list. I have commented out the type signatures. -- file: rwh_examples2/ch03/MySecond.hs -- safeSecond :: [a] -> Maybe a safeSecond [] = Nothing safeSecond xs = if null (tail xs) then Nothing else Just (head (tail xs)) -- tidySecond :: [a] -> Maybe a tidySecond (_:x:_) = Just x tidySecond _ = Nothing My query concerns the inferred types of these functions --- the types inferred when the type signatures are commented out. Here's a ghci session: Prelude> :load seconds.hs [1 of 1] Compiling Main ( seconds.hs, interpreted ) Ok, modules loaded: Main. *Main> *Main> :type safeSecond safeSecond :: [a] -> Maybe a *Main> *Main> :type tidySecond tidySecond :: [t] -> Maybe t *Main> Why is safeSecond type [a] -> Maybe a, and tidySecond type [t] -> Maybe t? I mean why does one use "a" and the other "t"? What does it mean? If I change the order of the two functions in the source file, safeSecond still uses "a" and tidySecond still uses "t". Can anyone help, or point me to the right place in the documentation? In the meantime, if I find out from elsewhere I'll report back here. Thanks and best wishes Ivan [1] http://book.realworldhaskell.org/ [2] http://book.realworldhaskell.org/read/defining-types-streamlining-functions.html#comment8512 [3] http://book.realworldhaskell.org/read/defining-types-streamlining-functions.html#deftypes.morecontrolled -- ============================================================ Ivan A. Uemlianin Speech Technology Research and Development ivan@llaisdy.com www.llaisdy.com llaisdy.wordpress.com www.linkedin.com/in/ivanuemlianin "Froh, froh! Wie seine Sonnen, seine Sonnen fliegen" (Schiller, Beethoven) ============================================================ From wagner.andrew at gmail.com Sat May 30 16:50:39 2009 From: wagner.andrew at gmail.com (Andrew Wagner) Date: Sat May 30 16:34:51 2009 Subject: [Haskell-beginners] Maybe a and Maybe t In-Reply-To: <4A2198F1.4020306@llaisdy.com> References: <4A2198F1.4020306@llaisdy.com> Message-ID: Good question, but I'm afraid the answer will be a bit disappointing. The real answer is, it doesn't matter - the following types are all the same: a -> a b -> b t -> t supercalifragilisticexpialodocious -> supercalifragilisticexpialodocious As for why it actually happens in this case, it's no doubt related to the particular algorithm ghci uses to do the type inference. On Sat, May 30, 2009 at 4:37 PM, Ivan Uemlianin wrote: > Dear All > > I have just started learning Haskell, using the O'Reilly book Real World > Haskell [1]. I'm finding it a very exciting language, and I'm working my > way painstakingly through the book (I'm up to the exercises at the end of > Chapter 3). > > I have a query based on some code from the book. This question has been > asked on the book web site [2], but no answer. I have also Googled about > and scanned the documentation. > > So, here is some code (quoted from [3]): two implementations of a function > to return the second element of a list. I have commented out the type > signatures. > > -- file: rwh_examples2/ch03/MySecond.hs > > -- safeSecond :: [a] -> Maybe a > > safeSecond [] = Nothing > safeSecond xs = if null (tail xs) > then Nothing > else Just (head (tail xs)) > > -- tidySecond :: [a] -> Maybe a > > tidySecond (_:x:_) = Just x > tidySecond _ = Nothing > > My query concerns the inferred types of these functions --- the types > inferred when the type signatures are commented out. Here's a ghci session: > > Prelude> :load seconds.hs > [1 of 1] Compiling Main ( seconds.hs, interpreted ) > Ok, modules loaded: Main. > *Main> > *Main> :type safeSecond > safeSecond :: [a] -> Maybe a > *Main> > *Main> :type tidySecond > tidySecond :: [t] -> Maybe t > *Main> > > Why is safeSecond type [a] -> Maybe a, and tidySecond type [t] -> Maybe t? > I mean why does one use "a" and the other "t"? What does it mean? > If I change the order of the two functions in the source file, safeSecond > still uses "a" and tidySecond still uses "t". > > Can anyone help, or point me to the right place in the documentation? In > the meantime, if I find out from elsewhere I'll report back here. > > Thanks and best wishes > > Ivan > > [1] http://book.realworldhaskell.org/ > [2] > http://book.realworldhaskell.org/read/defining-types-streamlining-functions.html#comment8512 > [3] > http://book.realworldhaskell.org/read/defining-types-streamlining-functions.html#deftypes.morecontrolled > > > -- > ============================================================ > Ivan A. Uemlianin > Speech Technology Research and Development > > ivan@llaisdy.com > www.llaisdy.com > llaisdy.wordpress.com > www.linkedin.com/in/ivanuemlianin > > "Froh, froh! Wie seine Sonnen, seine Sonnen fliegen" > (Schiller, Beethoven) > ============================================================ > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.haskell.org/pipermail/beginners/attachments/20090530/11c69ba9/attachment.html From lee.duhem at gmail.com Sat May 30 20:29:01 2009 From: lee.duhem at gmail.com (Lee Duhem) Date: Sat May 30 20:13:09 2009 Subject: [Haskell-beginners] Maybe a and Maybe t In-Reply-To: References: <4A2198F1.4020306@llaisdy.com> Message-ID: On Sun, May 31, 2009 at 4:50 AM, Andrew Wagner wrote: > Good question, but I'm afraid the answer will be a bit disappointing. The > real answer is, it doesn't matter - the following types are all the same: > a -> a > b -> b > t -> t > supercalifragilisticexpialodocious -> supercalifragilisticexpialodocious And by the way, they are called 'type variables', as other kind of variables, when without name conflict, their names don't matter. lee From ivan at llaisdy.com Sun May 31 09:37:27 2009 From: ivan at llaisdy.com (Ivan Uemlianin) Date: Sun May 31 09:21:33 2009 Subject: [Haskell-beginners] Maybe a and Maybe t In-Reply-To: References: <4A2198F1.4020306@llaisdy.com> Message-ID: <4A228817.2060206@llaisdy.com> Dear Andrew and Lee Thanks for your comments. Andrew Wagner wrote: > As for why it actually happens in this case, it's no doubt related to > the particular algorithm ghci uses to do the type inference. Interesting. I tried it in hugs and it doesn't happen: Main> :type safeSecond safeSecond :: [a] -> Maybe a Main> :type tidySecond tidySecond :: [a] -> Maybe a Main> So, it's a property of the ghci interpreter rather than of the language itself. Just out of curiousity, I'd be interested to know what's going on in ghci to produce this effect. Are there different types of type variable? Learning Haskell is reminding me of things I studied when I was an undergrad (linguistics)--- pattern matching in Prolog; the type system reminds me of categorial grammar; we looked at lamda calculus as part of formal semantics. So I'd like to look into this more deeply, if anyone can give me a pointer into the ghci docs. Thanks again Ivan -- ============================================================ Ivan A. Uemlianin Speech Technology Research and Development ivan@llaisdy.com www.llaisdy.com llaisdy.wordpress.com www.linkedin.com/in/ivanuemlianin "Froh, froh! Wie seine Sonnen, seine Sonnen fliegen" (Schiller, Beethoven) ============================================================ From daniel.is.fischer at web.de Sun May 31 11:42:44 2009 From: daniel.is.fischer at web.de (Daniel Fischer) Date: Sun May 31 11:27:36 2009 Subject: [Haskell-beginners] Maybe a and Maybe t In-Reply-To: <4A228817.2060206@llaisdy.com> References: <4A2198F1.4020306@llaisdy.com> <4A228817.2060206@llaisdy.com> Message-ID: <200905311742.45381.daniel.is.fischer@web.de> Am Sonntag 31 Mai 2009 15:37:27 schrieb Ivan Uemlianin: > Dear Andrew and Lee > > Thanks for your comments. > > Andrew Wagner wrote: > > As for why it actually happens in this case, it's no doubt related to > > the particular algorithm ghci uses to do the type inference. In particular the part where GHC assigns names to type variables. > > Interesting. I tried it in hugs and it doesn't happen: > > Main> :type safeSecond > safeSecond :: [a] -> Maybe a > Main> :type tidySecond > tidySecond :: [a] -> Maybe a > Main> > > So, it's a property of the ghci interpreter rather than of the language > itself. Yes. > > Just out of curiousity, I'd be interested to know what's going on in > ghci to produce this effect. Are there different types of type variable? There are type variables of different *kind*, e.g. in class Functor f where fmap :: (a -> b) -> f a -> f b a and b are type variables of kind * (the kind of ordinary types) and f is a type variable of kind (* -> *) (the kind of type constructors which take an ordinary type and produce another ordinary type, examples are [] and Maybe). But that has nothing to do with the phenomenon, in the inferred type signatures of safeSecond and tidySecond, the 'a' resp. 't' are both type variables of kind *. The only difference is that in one case the name supply delivered 'a' and in the other it delivered 't'. From b.lehnert at gmx.de Sun May 31 17:19:02 2009 From: b.lehnert at gmx.de (Bernhard Lehnert) Date: Sun May 31 17:03:09 2009 Subject: [Haskell-beginners] A try on a bank machine algorithm... Message-ID: <1243804742.10638.2.camel@sol> Hi, in a Python Forum someone came up with his homework. Just beginning to learn Haskell I thougt I give it a try. I came up with a working function that I present here to ask you, if I am doing this right or how I could do better. The job was: A cash machine has to compute the division of bank notes for any given amount of money. According to German wikipedia a common algorithm is to start giving one 5$ bill, one 10$ and so on as long as possible (I call this uphill). After that the maximum possible amount of large bank notes is given... Thus every customer gets some small bills and still the number of bills is limited. My answer is a function called 'paymixture' which first calls a function 'uphill' and afterwards a function 'downhill', each of which realizes one part of the above mentioned algorithm. The downhill function might also be usefull apart from the rest which is why indentation ist different... Thanks a lot, Bernhard ------------------------------------------------------------- paymixture :: (Ord t, Num t)=> t->[t]->[t] -- usage: paymixture (sum to pay) (list of possible bills) -- list of bills has to be sorted, like in -- 'paymixture 175 [5,10,20,50,100]' -- last element of the returned list is what could not be paid. paymixture n list = init(up) ++ down where up = uphill n list down = downhill (last(up)) (reverse list) uphill :: (Ord t, Num t)=> t->[t]->[t] uphill n [] = [n] uphill 0 _ = [0] uphill n (x:xs) = if n>=x then x : uphill (n-x) xs else uphill n xs downhill :: (Ord t, Num t)=> t ->[t]->[t] -- usage: downhill (sum to pay) (list of possible bills) -- list if bills has to be reverse sorted, like e. g. -- 'downhill 175 [200, 100, 50, 20, 10,5]' -- last elemt of return value is the rest that could -- not be paid. downhill n supply | (n References: <4A2198F1.4020306@llaisdy.com> <4A228817.2060206@llaisdy.com> <200905311742.45381.daniel.is.fischer@web.de> Message-ID: <20090601001034.GA18806@seas.upenn.edu> On Sun, May 31, 2009 at 05:42:44PM +0200, Daniel Fischer wrote: > > But that has nothing to do with the phenomenon, in the inferred type signatures of > safeSecond and tidySecond, the 'a' resp. 't' are both type variables of kind *. > The only difference is that in one case the name supply delivered 'a' and in the other it > delivered 't'. I think one source of difference is that ghci tries to maintain type variable names from declared type signatures. So perhaps one of the library functions used to define 'safeSecond' has an explicitly declared type signature that mentions 'a', and a library function used to defined 'tidySecond' has one that mentions 't'. -Brent From alexander.dunlap at gmail.com Sun May 31 23:39:53 2009 From: alexander.dunlap at gmail.com (Alexander Dunlap) Date: Sun May 31 23:24:16 2009 Subject: [Haskell-beginners] A try on a bank machine algorithm... In-Reply-To: <1243804742.10638.2.camel@sol> References: <1243804742.10638.2.camel@sol> Message-ID: <57526e770905312039r720e2c17y671293ec0f06e4c0@mail.gmail.com> On Sun, May 31, 2009 at 2:19 PM, Bernhard Lehnert wrote: > Hi, > > in a Python Forum someone came up with his homework. Just beginning to > learn Haskell I thougt I give it a try. I came up with a working > function that I present here to ask you, if I am doing this right or how > I could do better. > > The job was: A cash machine has to compute the division of bank notes > for any given amount of money. > According to German wikipedia a common algorithm is to start giving one > 5$ bill, one 10$ and so on as long as possible (I call this uphill). > After that the maximum possible amount of large bank notes is given... > > Thus every customer gets some small bills and still the number of bills > is limited. My answer is a function called 'paymixture' which first > calls a function 'uphill' and afterwards a function 'downhill', each of > which realizes one part of the above mentioned algorithm. The downhill > function might also be usefull apart from the rest which is why > indentation ist different... > > Thanks a lot, > Bernhard > ------------------------------------------------------------- > paymixture :: (Ord t, Num t)=> t->[t]->[t] > -- usage: paymixture (sum to pay) (list of possible bills) > -- list of bills has to be sorted, like in > -- 'paymixture 175 [5,10,20,50,100]' > -- last element of the returned list is what could not be paid. > > paymixture n list = init(up) ++ down > ? ? ? where > ? ? ? up = uphill n list > ? ? ? down = downhill (last(up)) (reverse list) > > ? ? ? uphill :: (Ord t, Num t)=> t->[t]->[t] > ? ? ? uphill n [] = [n] > ? ? ? uphill 0 _ ?= [0] > ? ? ? uphill n (x:xs) = if n>=x then x : uphill (n-x) xs > ? ? ? ? ? ? ? ? ? ? ? ? ? ? else uphill n xs > > downhill :: (Ord t, Num t)=> t ->[t]->[t] > -- usage: downhill (sum to pay) (list of possible bills) > -- list if bills has to be reverse sorted, like e. g. > -- 'downhill 175 [200, 100, 50, 20, 10,5]' > -- last elemt of return value is the rest that could > -- not be paid. > > downhill n supply | (n ? ? ? ? ? ? ? ? ?| otherwise = max : downhill (n-max) supply > ? ? ? ? ? ? ? ? ? ? where max = head([x|x<-supply, x<=n]) > > _______________________________________________ > Beginners mailing list > Beginners@haskell.org > http://www.haskell.org/mailman/listinfo/beginners > Some comments on things that would be done differently by someone with more experience: - You have more parentheses than you need. If you have head xs, it doesn't need to be head(xs). Parentheses are only needed when you want it to associate in the other direction (i.e. you need them for head (tail xs) but not take 3 xs because 3 and xs are both arguments of take but (tail xs) is one argument of head. It's actually even more complicated than that when you get into currying (although it makes a lot more sense then) but that's the simple way of thinking about it. - The "last" function is terribly inefficient - it has to traverse the entire list to get to the end. It would be better for uphill to return a pair of (all-but-last-element,last-element) because then the list wouldn't have to be traversed again to get the last element. It looks like you're doing well; good luck learning Haskell! Alex