book

Learn Haskell by Example

by Philipp Hagenlocher

December 2024

Beginner

456 pages

13h 37m

English

Manning Publications

Read now

Unlock full access

Who should read this bookHow this book is organized: A road mapAbout the codeliveBook discussion forumOther online resources
1.1 What is Haskell?1.1.1 Abstraction and theory1.1.2 A safe place1.2 The pure functional way1.2.1 A declarative recipe1.2.2 From cake to program1.2.3 It’s all for simplicity1.3 Usage of abstraction1.3.1 The good parts1.3.2 The bad parts1.4 The things we learnSummary

2.1 A primer on Haskell2.1.1 Caesar’s cipher2.1.2 A new project2.1.3 The first module2.2 Typical types and fantastic functions2.2.1 Types on the atomic level2.2.2 Lists and tuples2.2.3 Function types2.2.4 Adding types to math2.3 A little help with the alphabet2.3.1 Synonymous types for readability2.3.2 The kinds of letters2.3.3 Logical combinations2.4 Rotating the wheel2.4.1 Finding an element’s index2.4.2 Finding the element at an index2.4.3 Guarding control flow2.5 Transforming a string2.5.1 A higher-order mapping2.5.2 Parameterizing types2.5.3 A finished cipherSummary
3.1 Talking to the outside3.1.1 Simple actions for input and output3.1.2 Simulating a loop3.1.3 Breaking out of a recursive action3.2 Pure functions inside of actions3.2.1 Reading and modifying user input3.2.2 Data flow between pure and impure code3.3 Reading from the environment3.3.1 Parsing command line arguments3.3.2 Encoding errors with Maybe3.4 Example: Reading and printing a command line argument3.4.1 The let keyword3.4.2 Running the program with stackSummary
4.1 Reading files and transforming their content4.1.1 Writing a pure library4.1.2 Hiding auxiliary arguments4.2 Parametrized behavior in higher-order functions4.2.1 Partial function application4.3 Algebraic data structures as an encoding of possibilities4.3.1 Sum types or tagged unions4.3.2 Don’t repeat yourself4.3.3 The zip function4.3.4 Working with missing values4.3.5 Printing a list of values with mapM4.4 From library to executable4.4.1 Encoding and parsing command line options4.4.2 An overview of the projectSummary
5.1 Building a graph5.1.1 Polymorphic types5.1.2 Introducing a new module5.1.3 The Eq type class and type constraints5.1.4 The flip function5.2 Encapsulating implementations5.2.1 Adding and removing entries5.2.2 Using export lists to hide constructors5.2.3 The Show type class5.3 Using and reusing code5.3.1 Qualified imports5.3.2 Building maps for permutations5.3.3 Creating a permutation map from a dictionary5.4 Parameterizing transformations5.4.1 List comprehensionsSummary
6.1 Constructing a breadth-first search6.1.1 Overview of the algorithm6.1.2 Keeping track of search state6.1.3 Finding the solution by backtracking6.2 Type variable scoping6.2.1 Universal quantification6.2.2 Language extensions6.2.3 Using lexically scoped type variables6.3 Improving performance with hashmaps6.3.1 Analyzing performance with profiling6.3.2 Adding project dependencies6.3.3 Lazy evaluationSummary
7.1 Modeling CSV data7.1.1 Record syntax7.1.2 Encoding errors with Either7.2 Smart constructors7.2.1 Ensuring a property at the time of construction7.2.2 Providing an unsafe alternative7.2.3 The dollar sign operator7.3 Using type classes7.3.1 Semigroup and Monoid7.3.2 The IsString type class7.4 Creating a new type class7.4.1 A type class for slicing data structures7.4.2 Re-exporting modulesSummary
8.1 Parsing data8.1.1 Parsing numeric values8.2 Folding data structures8.2.1 The concept of folding8.2.2 A structure for parsing8.2.3 The Functor type class8.2.4 Using folding for parsing8.3 Printing a CSV8.3.1 Operations on CSVs8.4 A simple command-line parser8.4.1 Supporting flags and complicated argumentsSummary
9.1 How to test9.1.1 Property testing9.1.2 Generating random values9.1.3 Random and Uniform9.1.4 Using a global random value generator9.1.5 A basic property test9.1.6 Defining postconditions for random values9.2 Randomized testing9.2.1 The benefit of referential transparency9.3 The QuickCheck testing framework9.3.1 Using Property in QuickCheck9.4 Generating random values for testing9.4.1 The random generator Genrandom value function9.4.2 Example: AssocMap9.4.3 Shrinking test cases9.5 Practical usage of property testing9.5.1 Verbosity and coverage reports9.5.2 Modifying a test’s parameters9.5.3 Constructing test suites9.5.4 The effectiveness of testingSummary
10.1 Modeling sound with numbers10.1.1 The zoo of numeric type classes10.1.2 Creating periodic functions10.2 Using infinite lists10.2.1 Attack, decay, sustain, and release10.2.2 Building and working with infinite lists10.3 Controlling synthesis10.3.1 Partial field selectors10.3.2 A function as a type10.4 Note models10.4.1 A type class for pitches10.4.2 The Ratio type10.4.3 Different kinds of exponentiationSummary
11.1 Polymorphic data structures with multiple types11.1.1 Existential quantification11.1.2 Using existentially quantified types11.2 Interpreting structures11.2.1 Mixing signals11.2.2 Groups of polyphony11.3 Implementing a domain-specific language11.3.1 Simplifying syntax11.3.2 Custom operators for list-like data structures11.3.3 Fixity declarationsSummary
12.1 Writing a parser12.1.1 Portable images from the past12.1.2 How to parse a file12.1.3 Composing effects12.1.4 Choosing alternatively12.1.5 Introducing monads12.1.6 A discussion on monads12.1.7 How to fail12.2 Parsing on a bigger scale12.2.1 An introduction to Attoparsec12.2.2 Parsing images12.2.3 Choosing between formats12.2.4 Putting parsers togetherSummary
13.1 Providing type information to the caller13.1.1 Problems with return-type polymorphism13.1.2 Generalized algebraic data types13.1.3 The Vector type13.1.4 Dynamic types with existential quantification13.2 Validation of parsed data13.3 A generic algorithm for image conversion13.3.1 Image algorithm for conversion matrices13.3.2 Exporting images as PNG13.4 Using parallelism to transform data13.4.1 Measuring time13.4.2 How parallelism works13.4.3 All about sparks image process HECs (Haskell execution contexts)Summary
14.1 Opening and reading files14.1.1 System.IO and Handle14.1.2 Buffering and Seeking14.2 Reading bytes from a file14.2.1 Resource acquisition and bracket14.3 Working with the filesystem and exceptions14.3.1 System.Directory and System.FilePath14.3.2 Listing files and directories14.3.3 The basics of exceptions14.4 Throwing and catching exceptions14.4.1 Handling an errorSummary
15.1 Monad transformers15.1.1 Reading an environment with ReaderT15.1.2 StateT and WriterT15.1.3 Stacking multiple transformers with RWST15.2 Implementing an application15.3 Providing a CLI15.3.1 Parsing arguments with optparse-applicative15.3.2 Enumerating custom types15.3.3 The Enum class15.3.4 A CLI for AppSummary
16.1 Encoding values as JSON16.1.1 Aeson and JSON parsing16.1.2 Serializing to JSON16.2 Deriving type classes with Generic16.3 Using a SQLite database16.3.1 The basics of sqlite-simple16.3.2 ToRow and FromRow16.3.3 Defining actions for database accessSummary
17.1 Defining a typesafe API17.1.1 Typed APIs with Servant17.1.2 Phantom types17.1.3 Implementing the API17.2 Running the application17.2.1 The WAI application17.2.2 Application middleware17.2.3 Producing HTML from data17.3 Deriving a client17.3.1 Using servant-client17.3.2 Defining commands for the CLI17.3.3 Implementing the clientSummary
A.1 Choosing tools and editorA.2 Installing Haskell toolsA.2.1 Running Stack and GHCiA.2.2 Installing other toolsA.3 All in one Docker file
B.1 Lazy data structuresB.2 How it worksB.3 Leaking space and timeB.4 undefined and newtype

Content preview from Learn Haskell by Example

13 Parallel image processing

This chapter covers

Working with generalized algebraic data types
Safely avoiding return-type polymorphism
Writing a generic algorithm for transforming image data
Improving performance using parallelism

In the last chapter, we built a parser for PNM files, a simple image format. While we have covered the possibility of reading these files, we have not yet created a way of transforming them and writing them back to the filesystem. That will change with this chapter.

We want to cover using Haskell’s type system to create a data type that can dynamically store data of various types without needing to use parametric polymorphism while still being able to retrieve type information on this data later on. Then, ...