Yet Another Haskell Tutorial

193 Pages·2016·0.77 MB·English

by  Daume Hal.

Yet Another Haskell Tutorial
Revision 2
Hal Daume III
http://www.isi.edu/~hdaume/htut/
Copyright (c) Hal Daume III, 2002−2006.
The preprint version of this tutorial is intended to be available to the entire Haskell community,
so we grant permission to copy and distribute it for any purpose, provided that it is reproduced
in its entirety, including this notice. Modified versions may not be distributed without prior
consent of the authfreeor, and must still maintain a copy of this notice. The author retains the
right to change or modify this copyright at any time, as well as to make the book no longer
free of charge.
ii
About This Report
The goal of the Yet Another Haskell Tutorial is to providea completeintoductionto
theHaskellprogramminglanguage.ItassumesnoknowledgeoftheHaskelllanguage
or familiarity with functionalprogrammingin general. However, general familiarity
withprogrammingconcepts(suchasalgorithms)willbehelpful. Thisisnotintended
to be an introductionto programmingin general; rather, to programmingin Haskell.
Suf(cid:2)cient familiarity with your operating system and a text editor is also necessary
(thisreportonlydiscussesinstallationoncon(cid:2)gurationonWindowsand*Nixsystem;
otheroperatingsystemsmaybesupported(cid:150)consultthedocumentationofyourchosen
compilerformoreinformationoninstallingonotherplatforms).
What is Haskell?
Haskell is called a lazy, purefunctionalprogramminglanguage. It is called lazy be-
causeexpressionswhicharenotneededtodeterminetheanswertoaproblemarenot
evaluated. The opposize of lazy is strict, which is the evaluation strategry of most
commonprogramminglanguages(C,C++,Java,evenML).Astrictlanguageisonein
whicheveryexpressionisevaluated,whethertheresultofitscomputationisimportant
ornot. (Thisisprobablynotentirelytrueasoptimizingcompilersforstrictlanguages
oftendowhat’scalled(cid:147)deadcodeelmination(cid:148)(cid:150)thisremovesunusedexpressionsfrom
the program.) It is called pure because it does not allow side effects (A side effect
is somethingthataffectsthe(cid:147)state(cid:148)oftheworld. Forinstance, afunctionthatprints
somethingtothescreenissaidtobeside-effecting,asisafunctionwhichaffectsthe
valueofaglobalvariable.) (cid:150)ofcourse,aprogramminglanguagewithoutsideeffects
wouldbehorriblyuseless;Haskellusesasystemofmonadstoisolateallimpurecom-
putationsfromtherestoftheprogramandperformtheminthesafeway(seeChapter9
fora discussionof monadsproperor Chapter 5 forhow to do input/outputin a pure
language).
Haskell is called a functional language because the evaluation of a program is
equivalentto evaluatinga functionin the pure mathematicalsense. This also differs
from standard languages (like C and Java) which evaluate a sequence of statements,
oneaftertheother(thisistermedanimperativelangauge).
iii
iv
The History of Haskell
ThehistoryofHaskellisbestdescribedusingthewordsoftheauthors. Thefollowing
textisquotedfromthepublishedversionoftheHaskell98Report:
InSeptemberof1987ameetingwasheldattheconferenceonFunctional
ProgrammingLanguagesandComputerArchitecture(FPCA’87)inPort-
land, Oregon, to discuss an unfortunate situation in the functional pro-
grammingcommunity:therehadcomeintobeingmorethanadozennon-
strict,purelyfunctionalprogramminglanguages,allsimilarinexpressive
powerandsemanticunderpinnings. Therewasastrongconsensusatthis
meetingthatmorewidespreaduseofthisclassoffunctionallanguageswas
beinghamperedbythelackofacommonlanguage. Itwasdecidedthata
committeeshouldbeformedto designsucha language,providingfaster
communicationofnewideas,astablefoundationforrealapplicationsde-
velopment, and a vehicle through which others would be encouraged to
usefunctionallanguages.Thisdocumentdescribestheresultofthatcom-
mittee’sefforts:apurelyfunctionalprogramminglanguagecalledHaskell,
namedafterthelogicianHaskellB.Currywhoseworkprovidesthelogical
basisformuchofours.
Thecommittee’sprimarygoalwastodesignalanguagethatsatis(cid:2)edthese
constraints:
1. Itshouldbesuitableforteaching,research,andapplications,includ-
ingbuildinglargesystems.
2. It should be completely described via the publication of a formal
syntaxandsemantics.
3. Itshouldbefreelyavailable. Anyoneshouldbepermittedtoimple-
mentthelanguageanddistributeittowhomevertheyplease.
4. Itshouldbebasedonideasthatenjoyawideconsensus.
5. It should reduce unnecessary diversity in functional programming
languages.
The committee intended that Haskell would serve as a basis for future
researchinlanguagedesign,andhopedthatextensionsorvariantsofthe
languagewouldappear,incorporatingexperimentalfeatures.
Haskellhasindeedevolvedcontinuouslysinceitsoriginalpublication.By
themiddleof1997,therehadbeenfouriterationsofthelanguagedesign
(thelatestatthatpointbeingHaskell1.4).Atthe1997HaskellWorkshop
inAmsterdam,itwasdecidedthatastablevariantofHaskellwasneeded;
this stable language is the subject of this Report, and is called (cid:147)Haskell
98(cid:148).
Haskell 98 was conceived as a relatively minor tidy-up of Haskell 1.4,
makingsomesimpli(cid:2)cations,andremovingsomepitfallsfortheunwary.
v
Itisintendedtobea(cid:147)stable(cid:148)languageinsensetheimplementorsarecom-
mitted to supporting Haskell 98 exactly as speci(cid:2)ed, for the foreseeable
future.
The original Haskell Report covered only the language, together with a
standardlibrarycalledthePrelude.BythetimeHaskell98wasstabilised,
ithadbecomeclearthatmanyprogramsneedaccesstoalargersetofli-
brary functions (notably concerning input/output and simple interaction
withtheoperatingsystem). Iftheseprogramweretobeportable,asetof
librarieswouldhaveto bestandardisedtoo. Aseparateeffortwas there-
forebegunbyadistinct(butoverlapping)committeeto(cid:2)xtheHaskell98
Libraries.
Why Use Haskell?
Clearlyyou’reinterestedinHaskellsinceyou’rereadingthistutorial. Therearemany
motivations for using Haskell. My personal reason for using Haskell is that I have
foundthat I write more bug-freecode in less time using Haskell than any other lan-
guage.Ialso(cid:2)nditveryreadableandextensible.
Perhapsmostimportantly,however,IhaveconsistentlyfoundtheHaskellcommu-
nity to be incredibly helpful. The language is constantly evolving (that’s not to say
it’s instable; ratherthat thereare numerousextensionsthat havebeenaddedto some
compilerswhichI(cid:2)ndveryuseful)andusersuggestionsareoftenheededwhennew
extensionsaretobeimplemented.
Why Not Use Haskell?
Mytwobiggestcomplaints,andthecomplaintsofmostHaskellersIknow,are:(1)the
generatedcodetendstobeslowerthanequivalentprogramswritteninalanguagelike
C;and(2)ittendstobedif(cid:2)culttodebug.
Thesecondproblemtendsnotbetoaverybigissue: mostofthecodeI’vewritten
isnotbuggy,asmostofthecommonsourcesofbugsinotherlanguagessimplydon’t
existinHaskell. The(cid:2)rstissue certainlyhascomeupafewtimes inmyexperience;
however,CPU time is almost always cheaperthan programmertime and if I haveto
wait a little longer for my results after having saved a few days programming and
debugging.
Of course, this isn’t the case of all applications. Some people may (cid:2)nd that the
speedhittakenforusingHaskellisunbearable. However,Haskellhasastandardized
foreign-functioninterfacewhichallowyoutolinkincodewritteninotherlanguages,
for when you need to get the most speed out of your code. If you don’t (cid:2)nd this
suf(cid:2)cient, I would suggest taking a look at the language O’Caml, which often out-
performsevenC++,yetalsohasmanyofthebene(cid:2)tsofHaskell.
vi
Target Audience
TherehavebeenmanybooksandtutorialswrittenaboutHaskell; fora(nearly)com-
pletelist,visitthehttp://haskell.org/bookshelf(HaskellBookshelf)atthe
Haskellhomepage.Abriefsurveyofthetutorialsavailableyields:
A Gentle Introduction to Haskell is an introduction to Haskell, given that the
(cid:15)
readerisfamiliarwithfunctionalprogrammingenlarge.
HaskellCompanionisashortreferenceofcommonconceptsandde(cid:2)nitions.
(cid:15)
OnlineHaskellCourseisashortcourse(inGerman)forbeginningwithHaskell.
(cid:15)
Two Dozen Short Lessons in Haskell is the draft of an excellent textbook that
(cid:15)
emphasizesuserinvolvement.
Haskell Tutorial is based on a course given at the 3rd International Summer
(cid:15)
SchoolonAdvancedFunctionalProgramming.
HaskellforMirandaProgrammersassumesknowledgeofthelanguageMiranda.
(cid:15)
PLEAC-HaskellisatutorialinthestyleofthePerlCookbook.
(cid:15)
Thoughall of these tutorials is excellent, they are on their own incomplete: The
(cid:147)GentleIntroduction(cid:148)isfartooadvancedforbeginningHaskellersandtheotherstend
toendtooearly,ornotcovereverything.Haskellisfullofpitfallsfornewprogrammers
and experienced non-functional programmers alike, as can be witnessed by reading
throughthearchivesoftheHaskellmailinglist.
Itbecameclearthatthereisastrongneedforatutorialwhichisintroductoryinthe
sensethatitdoesnotassumeknowledgeoffunctionalprogramming,butwhichisad-
vancedinthesensethatitdoesassumesomebackgroundinprogramming.Moreover,
noneoftheknowntutorialsintroduceinput/outputanditeractivitysoonenough(Paul
Hudak’sbookisanexceptioninthatitdoesintroduceIObypage35,thoughthefocus
andaimofthatbookandthistutorialareverydifferent).Thistutorialisnotforbegin-
ningprogrammers;someexperienceandknowledgeofprogrammingandcomputersis
assumed(thoughtheappendixdoescontainsomebackgroundinformation).
TheHaskelllanguageunderwentastandardizationprocessandtheresultiscalled
Haskell 98. The majority of this book will cover the Haskell 98 standard. Any de-
viations from the standard will be noted (for instance, many compilers offer certain
extensionstothestandardwhichareuseful;someofthesemaybediscussed).
Thegoalsofthistutorialare:
tobepracticalaboveallelse
(cid:15)
toprovideacomprehensive,freeintroductiontotheHaskelllanguage
(cid:15)
topointoutcommonpitfallsandtheirsolutions
(cid:15)
toprovideagoodsenseofhowHaskellcanbeusedintherealworld
(cid:15)
vii
Additional Online Sources of Information
AShortIntroductiontoHaskell :
http://haskell.org/aboutHaskell.html
HaskellWiki :
http://haskell.org/hawiki/
Haskell-Tutorial :
ftp://ftp.geoinfo.tuwien.ac.at/navratil/HaskellTutorial.
pdf
TouroftheHaskellPrelude :
http://www.cs.uu.nl/(cid:152)afie/haskell/tourofprelude.html
CoursesinHaskell :
http://haskell.org/classes/
Acknowledgements
It wouldbe inappropriatenot to givecredit also to the originaldesigners ofHaskell.
Thoseare: Arvind,LennartAugustsson,DaveBarton, Brian Boutel, WarrenBurton,
Jon Fairbairn, Joseph Fasel, Andy Gordon, Maria Guzman, Kevin Hammond, Ralf
Hinze,PaulHudak,JohnHughes,ThomasJohnsson,MarkJones,DickKieburtz,John
Launchbury,ErikMeijer, RishiyurNikhil,JohnPeterson,SimonPeytonJones,Mike
Reeve,AlastairReid,ColinRunciman,PhilipWadler,DavidWise,JonathanYoung.
Finally, I would like to speci(cid:2)cally thank Simon Peyton Jones, Simon Marlow,
JohnHughes, Alastair Reid, KoenClassen, ManuelChakravarty,SigbjornFinneand
SvenPanne,allofwhomhavemademylifelearningHaskellallthemoreenjoyableby
alwaysbeingsupportive. Thereweredoubtlessotherswhohelpedandarenotlisted,
butthesearethosewhocometomind.
Alsothankstothemanypeoplewhohavereported(cid:147)bugs(cid:148)inthe(cid:2)rstedition.
-HalDaume·III
viii
Contents
1 Introduction 3
2 GettingStarted 5
2.1 Hugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.1.1 Wheretogetit . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.2 Installationprocedures . . . . . . . . . . . . . . . . . . . . . 6
2.1.3 Howtorunit . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1.4 Programoptions . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.5 Howtogethelp . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 GlasgowHaskellCompiler . . . . . . . . . . . . . . . . . . . . . . . 7
2.2.1 Wheretogetit . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2.2 Installationprocedures . . . . . . . . . . . . . . . . . . . . . 8
2.2.3 Howtorunthecompiler . . . . . . . . . . . . . . . . . . . . 8
2.2.4 Howtoruntheinterpreter . . . . . . . . . . . . . . . . . . . 8
2.2.5 Programoptions . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2.6 Howtogethelp . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3 NHC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3.1 Wheretogetit . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3.2 Installationprocedures . . . . . . . . . . . . . . . . . . . . . 9
2.3.3 Howtorunit . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3.4 Programoptions . . . . . . . . . . . . . . . . . . . . . . . . 9
2.3.5 Howtogethelp . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.4 Editors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3 LanguageBasics 11
3.1 Arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2 Pairs,TriplesandMore . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3 Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.3.1 Strings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.3.2 SimpleListFunctions . . . . . . . . . . . . . . . . . . . . . 18
3.4 SourceCodeFiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.5 Functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.5.1 LetBindings . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.5.2 In(cid:2)x . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
ix
x CONTENTS
3.6 Comments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.7 Recursion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.8 Interactivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4 TypeBasics 37
4.1 SimpleTypes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
4.2 PolymorphicTypes . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
4.3 TypeClasses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.3.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.3.2 EqualityTesting . . . . . . . . . . . . . . . . . . . . . . . . 41
4.3.3 TheNumClass . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.3.4 TheShowClass. . . . . . . . . . . . . . . . . . . . . . . . . 41
4.4 FunctionTypes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4.4.1 LambdaCalculus . . . . . . . . . . . . . . . . . . . . . . . . 42
4.4.2 Higher-OrderTypes . . . . . . . . . . . . . . . . . . . . . . 42
4.4.3 ThatPeskyIOType. . . . . . . . . . . . . . . . . . . . . . . 44
4.4.4 ExplicitTypeDeclarations . . . . . . . . . . . . . . . . . . . 45
4.4.5 FunctionalArguments . . . . . . . . . . . . . . . . . . . . . 46
4.5 DataTypes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.5.1 Pairs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4.5.2 MultipleConstructors . . . . . . . . . . . . . . . . . . . . . 49
4.5.3 RecursiveDatatypes . . . . . . . . . . . . . . . . . . . . . . 51
4.5.4 BinaryTrees . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.5.5 EnumeratedSets . . . . . . . . . . . . . . . . . . . . . . . . 52
4.5.6 TheUnittype . . . . . . . . . . . . . . . . . . . . . . . . . . 53
4.6 ContinuationPassingStyle . . . . . . . . . . . . . . . . . . . . . . . 53
5 BasicInput/Output 57
5.1 TheRealWorldSolution . . . . . . . . . . . . . . . . . . . . . . . . 57
5.2 Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
5.3 TheIOLibrary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.4 AFileReadingProgram . . . . . . . . . . . . . . . . . . . . . . . . 64
6 Modules 67
6.1 Exports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
6.2 Imports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
6.3 HierarchicalImports . . . . . . . . . . . . . . . . . . . . . . . . . . 70
6.4 LiterateVersusNon-Literate . . . . . . . . . . . . . . . . . . . . . . 71
6.4.1 Bird-scripts . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
6.4.2 LaTeX-scripts. . . . . . . . . . . . . . . . . . . . . . . . . . 72
7 AdvancedFeatures 73
7.1 SectionsandIn(cid:2)xOperators . . . . . . . . . . . . . . . . . . . . . . 73
7.2 LocalDeclarations . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
7.3 PartialApplication . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
7.4 PatternMatching . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80