https://www.haskell.org/definition/haskell98-report.pdf

REPL

ghci is repl for haskell

To know all flags in ghci: :show language

• use :t variable to get type
• use :k variable to get kind
• use :i binding to get info on binding
• use :l filename to load .hs file into repl
• To return to the Prelude> prompt, use the command :m, which is short for :module. This will unload the file from GHCi, so the code in that file will no longer be in scope in your REPL
• use :set -Wall to get compile time warnings in ghci

Enabling extensions in GHCi

The extension name after -X e.g. :set -XRankNTypes

Note - In GHCi, ":t" only accepts terms/values, and ":k" only accepts types/type-exprs

faster docs is accessed by SPC m h H

Write code blocks in ghci using :{ and :}. e.g.

prelude>:{
  do
    putStrLn "hello world"
:}

• The it value in ghci: it is the last evaluated expression in ghci.

We need to use let in order to declare variables in REPL, but it is not required in file.

Stack project

Needs to have a Main.hs module with a main declaration, which will be entry point for the project/app

Note: you can load modules in gchi optionally without having a main block.

Building a stack project with warnings etc. stack build --ghc-options="-Wall"

Effects in haskell/fp

1. printing/rendering
2. Network requests
3. Hardware events

Prefix and infix functions

All functions are prefix functions by default and calling syntax is fn arg1 arg2. To use a prefix function as an infix form, you put it in back quotes e.g . fn arg1 arg2 is same as arg1 `fn` arg2

To use an infix function, you put it at first position with paranthenses e.g. (+) 2 6 is same as 2 + 6

Parentheses

Only for grouping, does not imply strict execution. e.g. f (1 + 2), but f returns a constant value, then (1 + 2) is never evaluated Similarly f [1..] would return easily, if f is eager and makes decision which does not need all elements

Sectioning

Sectioning is a concise way to partially apply arguments to infix operators.

(++ "ing") "sleep"
(++ "ing") "walk"

Haskell Stack

Recommended tool for beginners to manage tools/dependencies/projects.

Names and namespaces

Six kinds of names:

1. variables - lowercase
2. Constructors - uppoercase
3. Type constructors - uppercase
4. Type variables - lowercase
5. Type classes - uppercase
6. Modules - uppercase

Typeclassess

Equivalent to interface or protocol in other languages, Usually a collection of methods supported grouped under a name.

Top level expressions

The interactive environment ghci would lead you to believe that you can punch some expressions into an .hs file and run the thing (in a similar fashion to languages like swift and ruby). This is not the case.

Haskell needs an entrypoint called main. Instead module header, import declaration or top-level declaration expected.

In haskell you can put a top-level binding/declaration e.g. k = 99, but you cannot put a top-level expression like putStrLn "wow.

Type variable in constraint error

In plain Haskell 2010, constraints can only be applied to type variables, like a or b or whatever, but not to anything more complex.

Legal: Show a => a -> String

Illegal: Show (Maybe a) => a -> String

Basics

• blocks are indentation based.

• Space stands for function application.

• To force execution order, use () round parentheses, default execution order is lazy (undetermined depending on runtime)

• $ vs. (): The $ is for avoiding parantheses. Anything that comes after $ will be evaluated first. e.g. all below are quivalent

putStrLn (show (1 + 1))
putStrLn $ show (1 + 1)
putStrLn $ show $ 1 + 1

• function is defined with =. i.e. A function definition:

1. starts with the name of the function,
2. followed by its formal parameters separated by white spaces,
3. an equal sign,
4. and an expression that is the body of the function.

e.g. a b c = d e

• Function application -- in most cases just the "whitespace operator" --has the highest precedence

• There is no order of execution (lazy)

• Order of evaluation is present in do.

Checking if given term/type is part of typeclass

:t (id :: Num a => a -> a) "test" -- string is not num, will error out.

Evaluation orders in lambda calculus

1. Normal order: Evaluate leftmost, outermost lambdas first, there is a possibility to apply functions before evaluating arguments. similar to call by name but call by name does not evaluate inside the body of an unapplied function.

2. Applicative order: Evaluate leftmost, innermost lambdas first, i.e. strategy in which the arguments of a function are evaluated from left to right in a post-order traversal of reducible expressions (redexes). Also known as call by value.

3. Lazy evaluation: Call by need is a memoized variant of call by name where, if the function argument is evaluated, that value is stored for subsequent uses.

Reading error messages

Understanding bottom-up typechecking and unification can help in polymorphic cases.s

http://ics.p.lodz.pl/~stolarek/_media/pl:research:stolarek_understanding_basic_haskell_error_messages.pdf

https://stackoverflow.com/questions/10375532/haskell-understanding-no-instance-for-error-messages-in-ghci

The (b ~ a) notation indicates equality of the types a and b.

The term rigid type variable indicates that our types a and b have been directly specified by the type annotation [5] and the compiler is not free to unify them.

Telling compiler that unification can happen:

{-# LANGUAGE TypeFamilies #-}
isEq :: (Eq a, Eq b, a ~ b) => a -> b -> Bool
isEq x y = x == y

Functions dont derive Show

The Haskell compiler doesn't maintain the expressions as they are, but translates them to machine code or some other low-level representation. You would usually see error with an arrow, it means you were trying to Show a function e.g.No instance for (Show ([Char] -> [Char])), which also means you passed fewer arguments to a function.

Theoritically, Functional programming is about functions. A mathematical (precisely, set-theoretic) function is entirely defined by its graph, that is by pairs of objects (argument, value).

only sensible way to show functions is to show their graph. can be done by http://hackage.haskell.org/package/universe-reverse-instances.

User defined data types

There are three main ways:

1. Algebraic data types, using data keyword
2. Type synonym declarations using type keyword
3. Datatype renamings using newtype keyword
4. Generalized Algebraic data types GADTs with data + where keywords

normal constructor application has higher precedence than infix constructor application (thus a : Foo a parses a : (Foo a))

Algebraic data types uisng data

topdecl	→	data [context =>] simpletype [= constrs] [deriving]

The parts in brackets are optional An example is

  data Maybe a = Just a | Nothing

An example with context/constraint is

  data Eq a => Set a = NilSet | ConsSet a (Set a)

The definition also involves constructors

simpletype	→	tycon tyvar1 … tyvark	    (k ≥ 0) // Maybe is a type constructor and a is a typevar in above defition
 
constrs	→	constr1 | … | constrn	    (n ≥ 1)     // data constructors always appear after equals "=" e.g. Just, Nothing

constr	→	con [!] atype1 … [!] atypek	    (arity con  =  k, k ≥ 0)
        |	(btype | ! atype) conop (btype | ! atype)	    (infix conop)
        |	con { fielddecl1 , … , fielddecln }	    (n ≥ 0)

fielddecl	→	vars :: (type | ! atype)
 
deriving	→	deriving (dclass | (dclass1, … , dclassn))	    (n ≥ 0)

An algebraic datatype declaration has the form: data cx => T u1 … uk = K1 t11 … t1k1 | ⋅⋅⋅ | Kn tn1 … tnkn where cx is a context. This declaration introduces a new type constructor T with zero or more constituent data constructors K1, …, Kn.

Here tij could be type variables or type constants.

The type variables u1 through uk must be distinct and may appear in cx and the tij; it is a static error for any other type variable to appear in cx or on the right-hand-side.

Another example:

data Shape = Circle Float Float Float | Rectangle Float Float Float Float

data Person = Person { firstName :: String  
                     , lastName :: String  
                     , age :: Int  
                     , height :: Float  
                     , phoneNumber :: String  
                     , flavor :: String  
                     } deriving (Show)

Field Labels: For large datatypes it is useful to assign field labels to the components of a data object. This allows a specific field to be referenced independently of its location within the constructor.

The arguments to the positional constructor occur in the same order as the labeled fields. For example, the declaration

  data C = F { f1,f2 :: Int, f3 :: Bool }

defines a type and constructor identical to the one produced by

  data C = F Int Int Bool

You can now access values with labels as functions e..g

k = F 2 4 True
(f1 k) // prints 2
(f2 k) // prints 4
(f3 k) // prints True
:t k // prints C

-- define data type using records
data Car = Car {company :: String, model :: String, year :: Int} deriving (Show)

-- create values with Constructors with records with label"="value
ghci> Car {company="Ford", model="Mustang", year=1967}

-- destructuring with record/labels
tellCar :: Car -> String  
tellCar (Car {company = c, model = m, year = y}) = "This " ++ c ++ " " ++ m ++ " was made in " ++ show y

Field names share the top level namespace with ordinary variables and class methods and must not conflict with other top level names in scope.

The pattern F {} matches any value built with constructor F, whether or not F was declared with record syntax.

Type synonym declarations via type

type tycon tyvar1 .. tyvark = existingtype

It has the form type T u1 … uk = t which introduces a new type constructor, T . The type (T t1 …tk) is equivalent to the type t[t1∕u1, …, tk∕uk]. The type variables u1 through uk must be distinct and are scoped only over t; it is a static error for any other type variable to appear in t. The kind of the new type constructor T is of the form κ1 →… → κk → κ where the kinds κi of the arguments ui and κ of the right hand side t are determined by kind inference.

Recursive type synonyms are not allowed i.e.

  type Rec a   =  [Circ a]        -- invalid  
  type Circ a  =  [Rec a]         -- invalid

Datatype renamings via newtype

When using newtype, you're restricted to just one constructor with one field.(See N and t below)

Why newtype? (better alternative to type alias) Direct correspondence with field type(isomorphic). Newtype guarantees that there is no overhead for using it, and that it is operationally equivalent to the thing it wraps. e.g.

-- direct correspondence with Name type is string type
-- no other constructors or fields possible.
newtype Name = Name String

type MyCharArray = [Char]

myname :: MyCharArray
myname = "afdaf"

name :: [Char]
name = "simplename"

take 3 name -- works over string

-- also works over type alias
take 3 myname

-- with newtype
newtype MyNewTypeCharArray = MyNewTypeCharArray [Char]

mynewytpename :: MyNewTypeCharArray
mynewytpename = MyNewTypeCharArray "adfaf"

-- will not typecheck with newtype
take 3 mynewytpename

A declaration of the form newtype cx => T u1 … uk = N t introduces a new type whose representation is the same as an existing type. The type (T u1… uk) renames the datatype t.

YOu can think of N as NewConstructor. The constructor N in an expression coerces a value from type t to type (T u1 … uk). Using N in a pattern coerces a value from type (T u1 … uk) to type t.

But if you use newtype, Haskell knows that you're just using it to wrap an existing type into a new type (hence the name), because you want it to be the same internally but have a different type. With that in mind, Haskell can get rid of the wrapping and unwrapping once it resolves which value is of what type.

Unlike algebraic datatypes, the newtype constructor N is unlifted, so that N ⊥ is the same as ⊥.

The following examples clarify the differences between data (algebraic datatypes), type (type synonyms), and newtype (renaming types.) Given the declarations

  data D1 = D1 Int  
  data D2 = D2 !Int  
  type S = Int  
  newtype N = N Int  
  d1 (D1 i) = 42  
  d2 (D2 i) = 42  
  s i = 42  
  n (N i) = 42

the expressions (d1 ⊥), (d2 ⊥) and (d2 (D2 ⊥)) are all equivalent to ⊥, whereas (n ⊥), (n (N ⊥)), (d1 (D1 ⊥)) and (s ⊥) are all equivalent to 42. In particular, (N ⊥) is equivalent to ⊥ while (D1 ⊥) is not equivalent to ⊥.

Both newtype and the single-constructor data introduce a single value constructor, but the value constructor introduced by newtype is strict and the value constructor introduced by data is lazy. So if you have

data D = D Int
newtype N = N Int

Then N undefined is equivalent to undefined and causes an error when evaluated. But D undefined is not equivalent to undefined, and it can be evaluated as long as you don't try to peek inside.

A newtype declaration may use field-naming syntax, though of course there may only be one field. Thus:

  newtype Age = Age { unAge :: Int }

brings into scope both a constructor and a de-constructor:

  Age   :: Int -> Age  
  unAge :: Age -> Int

Generalized algebraic data types GADTs

data Maybe a where
   Nothing  :: Maybe a
   Just :: a -> Maybe a

This syntax is made available by the language option {-#LANGUAGE GADTs #-}. It should be familiar to you in that it closely resembles the syntax of type class declarations. It's also easy to remember if you already like to think of constructors as just being functions. Each constructor is just defined by a type signature.

Data constructors vs Type Constructors

In a data declaration, a type constructor is the thing on the left hand side of the equals sign. The data constructor(s) are the things on the right hand side of the equals sign.

You use type constructors where a type is expected, and you use data constructors where a value is expected.

-- Colour is a Type constructor, Red/Green/Blue are data constructors.
data Colour = Red | Green | Blue

If you'd want to construct a binary tree to store Strings, you could imagine doing something like

-- SBTree is a type constructor, and Leaf/Branch are data constructor functions that produce values of type SBTree
data SBTree = Leaf String
            | Branch String SBTree SBTree

What we see here is a type SBTree that contains two data constructors. In other words, there are two functions (namely Leaf and Branch) that will construct values of the SBTree type. If you're not familiar with how binary trees work, just hang in there. You don't actually need to know how binary trees work, only that this one stores Strings in some way.

Kinds

To ensure that they are valid, type expressions are classified into different kinds, which take one of two possible forms:

The symbol ∗ represents the kind of all nullary type constructors.

Char, Int, Integer, Float, Double and Bool are type constants with kind ∗.

If κ1 and κ2 are kinds, then κ1 → κ2 is the kind of types that take a type of kind κ1 and return a type of kind κ2.

Maybe and IO are unary type constructors, and treated as types with kind ∗ → ∗

1. The trivial type is written as () and has kind ∗. It denotes the nullary tuple type, and has exactly one value, also written (). (think already a concrete type)
2. The function type is written as (->) and has kind ∗ → ∗ → ∗. (think that it takes two concrete types and returns a new concrete function type)
3. The list type is written as [] and has kind ∗ → ∗. (think it takes a concrete type and returns the concrete array type)
4. The tuple types are written as (,), (,,), and so on. Their kinds are ∗→∗→∗, ∗→∗→∗→∗, and so on.
5. Type application: Type application. If t1 is a type of kind κ1 → κ2 and t2 is a type of kind κ1, then t1 t2 is a type expression of kind κ2.
6. A parenthesized type, having form (t), is identical to the type t.

Kind inference checks the validity of type expressions in a similar way that type inference checks the validity of value expressions. However, unlike types, kinds are entirely implicit and are not a visible part of the language

Operators

Functions in Haskell default to prefix syntax, meaning that the function being applied is at the beginning of the expression rather than the middle.

An operator is a function that can be applied using infix syntax, or partially applied using a section (Section 3.5).

One can use a prefix function in infix style by wrapping it in backticks. And one can use an infix operator in prefix style by wrapping it in parentheses.

Know more about operators using :i, e.g. :i (*) The infixl in the output signifies that it is an infix operator and left associative

An operator is either an operator symbol, such as + or $$, or is an ordinary identifier enclosed in grave accents (backquotes), such as op. For example, instead of writing the prefix application op x y, one can write the infix application xop y. If no fixity declaration is given for op then it defaults to highest precedence and left associativity

Dually, an operator symbol can be converted to an ordinary identifier by enclosing it in parentheses. For example, (+) x y is equivalent to x + y, and foldr (⋆) 1 xs is equivalent to foldr (\x y -> x⋆y) 1 xs.

Tuples

Combines multiple values into a single value. Same syntax is the type constructor as well as data constructor: (,)

(,) is a data constructor for tuples.

(,) 8 "hi" generates a tuple (8, "hi")

One cannot partially applied variable binding using (,), E.g. (,) 1 will throw error.

data (,) a b = (,) a b -- as there are two different typevars, tuple can be made of two different concrete-types 
fst :: (a, b) -> a
snd :: (a, b) -> b

-- multi tuple constructors are weird!!
(, , ,) 1 2 3 4
-- output: (1, 2, 3, 4)

Utilities present in Data.Tuple

Lists

Define as:

data [] a = [] | a : [a] -- involves cons constructor (:)

From the type definition, restriction on the typevar is list allows only elements of single parametric type a

Note: We can see [] in both terms and types, i.e. it acts as both a data constructor and type constructor respectively.

Referencing length on a list is generally bad idea being O(n) as it is traversing whole list to give answer.

Values and types

An expression evaluates to a value and has a static type. Values and types are not mixed in Haskell.

However, the type system allows user-defined datatypes of various sorts, and permits not only parametric polymorphism (using a traditional Hindley-Milner type structure) but also ad hoc polymorphism, or overloading (using type classes).

Errors in Haskell are semantically equivalent to ⊥ (“bottom”). Technically, they are indistinguishable from nontermination, so the language includes no mechanism for detecting or acting upon errors.

Namespaces

There are six kinds of names in Haskell: those for variables and constructors denote values; those for type variables, type constructors, and type classes refer to entities related to the type system; and module names refer to modules. There are two constraints on naming:

1. Names for variables and type variables are identifiers beginning with lowercase letters or underscore; the other four kinds of names are identifiers beginning with uppercase letters.

2. An identifier must not be used as the name of a type constructor and a class in the same scope. These are the only constraints; for example, Int may simultaneously be the name of a module, class, and constructor within a single scope.

Function application and definition

Function application is written e1 e2. Application associates to the left, so the parentheses may be omitted in (f x) y. Because e1 could be a data constructor, partial applications of data constructors are allowed.

Lambda abstractions are written \ p1 … pn -> e, where the pi are patterns. An expression such as \x:xs->x is syntactically incorrect; it may legally be written as \(x:xs)->x.

Runtime exceptions

1. non total functions - usually warned during compilation by non-exhaustive pattern matching.

Reading types

The constraints are specified before => and the type signature is shown after =>.

One can see the type of any binding using :t on terminal.

show :: Show a => a -> String
-- show takes a binding (which should be a Showable) and returns a string represetnation of that binding.

Verifying type

-- :t 1 returns Num
-- but 1 :: Num throws error.

1 :: Int -- does not throw any error
1 :: Bool -- throws error

common types

1. concat operator for lists

-- :t (++)
(++) :: [a] -> [a] -> [a]

2. String is a type alias for [Char]

-- :t 'A'   -- Char
-- :t "A"   -- [Char]

3. readables are read using read which takes in a string and returns

-- :t read
-- read :: Read a => String -> a

4. fromInteger

instance Num x where
    fromInteger :: Integer -> x

Whenever x doesn not confirm to Num, we get erros like No instance for (Num [Char]) etc. when doing 3 + "4"

5. cons symbol (:)

-- :t (:)
-- (:) :: a -> [a] -> [a]

6: putStrLn

-- putStrLn :: St
String -> IO ()

7. getLine

-- getLine :: IO String
-- value from an IO is extracted using <-

8. sequence in Control.Monad Converting traversable monads to monadic traversable.

-- sequence :: (Traversable t, Monad m) =>
-- t (m a) -> m (t a)
-- e.g. [IO Char] -> IO [Char]

9 read Read interprets an object/data from a string.

-- read :: Read a => String -> a

it's always a good idea to instead use the readMaybe function from Text.Read

10 show Show converts an object/data to string

-- show :: Show a => a -> String

11 error throw error with given string message.

-- error :: [Char] -> a
error "hello"

12. chr converts Int to Char e.g.

chr :: Int -> Char
chr 65
-- 'A'

13. ord

14. iterate given a starting value and a function to be applied repeatedly, return the infitie list of the same.

iterate :: (a -> a) -> a -> [a]
take 10 $ iterate (+1) 0
-- [0,1,2,3,4,5,6,7,8,9]

14. unfoldr: Part of Data.List package. unfoldr: the dual to foldr, unfoldr builds a list from a seed value. The first argument is a function that takes the element and returns Nothing if it is done or returns Just (a,b) in which case, a is prepended to the list and b is used as the next element in recursive call. The second argument is the seed value to start the generation with

-- Nothing can be returned by generating fn to halt generating
unfoldr :: (b -> Maybe (a, b) -> b -> [a])

iterate f == unfoldr (\x -> Just (x, f x)) == [x, f x, f f x, f f f x, .... ]
unfoldr (\b -> if b == 0 then Nothing else Just (b, b-1)) 10
-- [10,9,8,7,6,5,4,3,2,1]

Numerics

When we query the types of numeric values, we see typeclass information instead of a concrete type, because the compiler doesn’t know which specific numeric type a value is until the type is either declared or the compiler is forced to infer a specific type based on the function.

Data Types: Int, Float, Double, Rational are data types where as

Four sorts of nubmer: Integral, RealFloat, Ratio, Complex. This is not finely demarcated though. It is for our understanding. Integral and RealFloat are type classes.

Typeclasses: Real, Num, Integral, Fractional are type classes

Operators under Num: (+), (-), (*), Members are: Integer, Int, Float, Double. Note how fractional division (/) is not a part of Num, since it may not satisfy a -> a -> a required by Num. See Fractional.

Operators under Integral: quot, div, mod, rem, toInteger, Members of this typcleassSs: Int, Integer.

Operators under RealFloat: floatRadix, floatDigits, floatRange, decodeFloat, encodeFloat, exponent, significand, scaleFloat, isNaN, isInfinite, isDenormalized, isNegativeZero, isIEEE, atan2. Members of RealFloat are Float, Double. RealFloat extends upon RealFrac and Floating

Operators under Real: toRational, Members: Integer, Int, Float, Double

Operators under Fractional -> (/), recip, fromRational. Members are: Float, Double. Since Int is not a member of Fractional type class, / operator does not work on Int. One has to do something before it.

Operators under RealFrac : properFraction, truncate, round, ceiling, floor. Instances/members of RealFrac are Float and Double. Also RealFrac extends upon Real and Fractional.

Operators under Floating: pi, exp, log, sqrt, **, sin,cos,tan,etc. Instances/members of floating are Float and Double.

Ratio: It is made of Integral number pair data (Integral a) => Ratio a = a % a, Also type Rational = Ratio Integer. Import can be done via Data.Ratio.

Complex: Made of RealFloat number pair

Defaulting: "I need a type (a) with (Num a => a) and (Fractional a => a). (Fractional a) requires (Num a) already, so that part is redundant and I can proceed with just the (Fractional a)". it picks a reasonable instance of Fractional via defaulting, which gives it Double.

https://stackoverflow.com/questions/42820603/why-can-a-num-act-like-a-fractional

Integer vs Int?

Integer is for BigIntegers i.e. it can hold arbitrary large integers in the sense of natural numbers.

Int is bounded by Int64 i.e. range -9223372036854775808..9223372036854775807

Other integers with smaller ranges are Int8, Int16, Int32 and Int64 which are availble under importing GHC.Int. and are also members of Bounded type class.

Bindings

Using let and where and pattern matching by defining functions, We introduced bindings. Bindings are essentially names for expressions. And cannot be rebound in same scope.

Reverse thin arrow operator <-

In case of List comprehensions and case guards, <- means a pattern match.

In case of mondadic values, <- means run an action or thing of it as unwrapping a monad value, equivalent to >>= in a normal expression.

let expression

let expression kinds of act as a lambda, that introduces scope.

let ... in ... is an expression, that is, it can be written wherever expressions are allowed.

Let unlike in other languages, is lazy in haskell,

Below code does not cause error until x or y is evaluated.

let (x,y) = undefined in e

Scope of bindings declared in let, are only valid till the body of let expression. e.g

j = 4
main = do
  let k = 2
  let u = 4 in
    putStrLn $ show $ k + j + u
  putStrLn $ show k
  putStrLn $ show j
  -- putStrLn $ show u -- u not in scope

Multiline let can be achieved by using careful indentation:

up arg1 =
  let p = 22
      t = 11
      in putStrLn $ show $ p + t

-- up 0
-- 33

let without in: introduces binding in the current scope. Also known as let-statement. e.g.

do statements
    let variable = exp -- variable binding available inside do block
    statements

Let vs Where

Let and Where are fundamentally different.

Let introduces an expression hence we say let expression can be used wherever an expression is expected.

Where is a declaration that bounds to a surrounding construct.

Where clause

Where is a useful clause that bounds to surrounding construct and can have multiple declarations (each on a new line) within itself (they can also use each other) e.g.

fn a b c = x + y + z
  where
    x = a + b + c
    y = if (2 ^ 2) == 4
          then 10
          else 99
    z = a * b * c * m
    m = 1

when to use a do block?

When doing a bunch of computation, and then returning a value via return. We need to combine all of it into a single expression via do block. Especially useful in let declarations in expression. or resultExpression in

case expression of pattern -> resultExpression
                   pattern -> resultExpression
                   pattern -> resultExpression

List comprehensions

A list comprehension has the form [ e | q1, ..., qn ], n>=1, where the qi qualifiers are either

1. generators of the form p <- e, where p is a pattern (see Section 3.17) of type t and e is an expression of type [t]
2. guards, which are arbitrary expressions of type Bool
3. local bindings that provide new definitions for use in the generated expression e or subsequent guards and generators.

Such a list comprehension returns the list of elements produced by evaluating e in the successive environments created by the nested, depth-first evaluation of the generators in the qualifier list. Binding of variables occurs according to the normal pattern matching rules (see Section 3.17), and if a match fails then that element of the list is simply skipped over. Thus:

[ x |  xs   <- [ [(1,2),(3,4)], [(5,4),(3,2)] ], 
      (3,x) <- xs ]

gives [4,2].

where bindings vs let bindings

IO Actions

An IO Action has the type IO t.

Actions can be created, assigned, and passed anywhere. However, they may only be performed (executed) from within another I/O action

main itself is an I/O action with type IO ()

Performing (executing) an action of type IO t may perform I/O and will ultimately deliver a result of type t

File IO

import System.IO is necessary.

Handle is the core class that represents file handle. openFile :: FilePathString -> IOMode -> IO Handle i.e. returns handle given file path and IO Mode. hGetLine :: Handle -> IO String. hGetContents :: Handle -> IO String. hIsEOF :: Handle -> IO Bool. Always close your file handles with hClose

        fHandle <- openFile "testdata.txt" ReadMode
        ln <- hGetLine fHandle
        putStrLn ln
        hClose fHandle

        -- get all content in file using hGetContents
        fH2 <- openFile "good_data.txt" ReadMode
        allContent <- hGetContents fH2
        putStrLn allContent
        hClose fH2

Write to handle is done via hPutStrLn hPutStrLn :: Handle -> String -> IO ()

In Haskell, return is the opposite of <-. That is, return takes a pure value and wraps it inside IO.

return is used to wrap data in a monad. When speaking about I/O, return is used to take pure data and bring it into the IO monad

There are three well known pre defined handles in System.IO:

stdin :: Handle
stdout :: Handle
stderr :: Handle
getLine = hGetLine stdin
putStrLn = hPutStrLn stdout
print = hPrint stdout

Program to combine contents of files in current directory

import System.IO
import System.Directory
import Data.Text

getFileContents :: String -> IO String
getFileContents fileName = do
  case fileName of
    -- ignoring directories
    "." -> return ("")
    ".." -> return ("..")
    ".ghc" -> return (".ghc")
    -- for regular files
    otherwise -> do
      fH <- openFile fileName ReadMode
      contents <- hGetContents fH
      let copycontent = contents
      return (copycontent)

main = do
        putStrLn "Greetings! what is your name?"
        inpStr <- getLine
        putStrLn $ "Welcome to haskell, " ++ inpStr ++ "!"
        
        dirs <- getDirectoryContents "."
        mapM_ putStrLn dirs
        let allContentIO = (Prelude.map getFileContents dirs) in
          do
            let allContent = (sequence allContentIO) in
              do
                allContentString <- allContent
                mapM_ putStrLn allContentString

sequence is usefule when dealing with monad arrays

sequence  :: Monad m => [m a] -> m [a]

Map/dictionary data structure

Introduced via Data.Map.Map The Map k v type represents a finite map (sometimes called a dictionary) from keys of type k to values of type v.

lm = Data.Map.fromList [('a', '1'), ('b','2')]
Data.Map.lookup 'z' lm
-- Nothing
Data.Map.lookup 'a' lm
-- Just '1'