learn you some erlang - chap3 to chap5

learn you some erlang - chap3 to chap5

• 1.
  LEARN U ERLANG WEEK-2PRESENTED BY NOLLEH
• 2.
  3. SYNTAX INFUNCTIONS PATTERN MATCHING ▸ greet without patten matching function greet(Gender,Name) if Gender == male then print("Hello, Mr. %s!", Name) else if Gender == female then print("Hello, Mrs. %s!", Name) else print("Hello, %s!", Name) end
• 3.
  3. SYNTAX INFUNCTIONS PATTERN MATCHING ▸ greet with patten matching greet(male, Name) -> io:format("Hello, Mr. ~s!", [Name]); greet(female, Name) -> io:format("Hello, Mrs. ~s!", [Name]); greet(_, Name) -> io:format("Hello, ~s!", [Name]).
• 4.
  3. SYNTAX INFUNCTIONS PATTERN MATCHING ▸ disusing / using function(X) -> Expression; function(Y) -> Expression; function(_) -> Expression. function(Args) if X then Expression else if Y then Expression else Expression
• 5.
  3. SYNTAX INFUNCTIONS PATTERN MATCHING ▸ example head([H|_]) -> H. second([_,X|_]) -> X. same(X,X) -> true; same(_,_) -> false.
• 6.
  3. SYNTAX INFUNCTIONS PATTERN MATCHING ▸ how’s it works? ▸ error occurs unless the new value is the same as the old one
• 7.
  3. SYNTAX INFUNCTIONS PATTERN MATCHING ▸ example2 ▸ functions head’s ‘=‘ operator valid_time({Date = {Y,M,D}, Time = {H,Min,S}}) -> io:format("The Date tuple (~p) says today is: ~p/~p/~p,~n”, [Date,Y,M,D]), io:format("The time tuple (~p) indicates: ~p:~p:~p.~n”, [Time,H,Min,S]); valid_time(_) -> io:format("Stop feeding me wrong data!~n").
• 8.
  3. SYNTAX INFUNCTIONS PATTERN MATCHING ▸ example2 ▸ prob: It also recv just tuple! and too precise sometimes. 4> c(functions). {ok, functions} 5> functions:valid_time({{2011,09,06},{09,04,43}}). The Date tuple ({2011,9,6}) says today is: 2011/9/6, The time tuple ({9,4,43}) indicates: 9:4:43. ok 6> functions:valid_time({{2011,09,06},{09,04}}). Stop feeding me wrong data!
• 9.
  3. SYNTAX INFUNCTIONS GUARDS,GUARDS! ▸ needs expressive way on sometimes… ▸ range of value ▸ not limited as certain types of data
• 10.
  3. SYNTAX INFUNCTIONS GUARDS,GUARDS! ▸ impractical vs practical old_enough(0) -> false; old_enough(1) -> false; old_enough(2) -> false; ... old_enough(14) -> false; old_enough(15) -> false; old_enough(_) -> true. ok old_enough(X) when X >= 16 -> true; old_enough(_) -> false.
• 11.
  3. SYNTAX INFUNCTIONS GUARDS,GUARDS! ▸ simillar with andalso (little diff aspect on exceptions) ▸ orelse right_age(X) when X >= 16, X =< 104 -> true; right_age(_) -> false. wrong_age(X) when X < 16; X > 104 -> true; wrong_age(_) -> false.
• 12.
  3. SYNTAX INFUNCTIONS GUARDS,GUARDS! ▸ in guard, You will be ▸ able to use functions like ▸ A*B/C >= 0 ▸ is_integer/1, is_atom/1 … ▸ unable to use user defined function ▸ because of side-effect
• 13.
  3. SYNTAX INFUNCTIONS WHAT THE IF!? ▸ similiar with guard but outside of function clauses head ▸ different from other language
• 14.
  3. SYNTAX INFUNCTIONS WHAT THE IF!? -module(what_the_if). -export([heh_fine/0]). heh_fine() -> if 1 =:= 1 -> works end, if 1 =:= 2; 1 =:= 1 -> works end, if 1 =:= 2, 1 =:= 1 -> fails end.
• 15.
  3. SYNTAX INFUNCTIONS WHAT THE IF!? 1> c(what_the_if). ./what_the_if.erl:12: Warning: no clause will ever match ./what_the_if.erl:12: Warning: the guard for this clause evaluates to 'false' {ok,what_the_if} 2> what_the_if:heh_fine(). ** exception error: no true branch found when evaluating an if expression in function what_the_if:heh_fine/0
• 16.
  3. SYNTAX INFUNCTIONS WHAT THE IF!? ▸ true branch oh_god(N) -> if N =:= 2 -> might_succeed; true -> always_does %% this is Erlang's if's 'else!' end. 4> what_the_if:oh_god(2). might_succeed 5> what_the_if:oh_god(3). always_does
• 17.
  3. SYNTAX INFUNCTIONS WHAT IS IF!? ▸ why not else ? ▸ both branch should be avoided ▸ if is usually easier ▸ guard has only limited set
• 18.
  3. SYNTAX INFUNCTIONS IN CASE … OF ▸ example insert(X,[]) -> [X]; insert(X,Set) -> case lists:member(X,Set) of true -> Set; false -> [X|Set] end.
• 19.
  3. SYNTAX INFUNCTIONS IN CASE … OF ▸ pattern matching + guard beach(Temperature) -> case Temperature of {celsius, N} when N >= 20, N =< 45 -> 'favorable'; {kelvin, N} when N >= 293, N =< 318 -> 'scientifically favorable'; {fahrenheit, N} when N >= 68, N =< 113 -> 'favorable in the US’; _ -> 'avoid beach' end.
• 20.
  3. SYNTAX INFUNCTIONS IN CASE … OF ▸ instead, we can replace with bunch of functions. beachf({celsius, N}) when N >= 20, N =< 45 -> 'favorable'; ... beachf(_) -> 'avoid beach'.
• 21.
  3. SYNTAX INFUNCTIONS WHICH TO USE? ▸ function call vs case of ▸ same way at a lower level ▸ only one difference when arg is more than one case {A,B} of Pattern Guards -> ... end.
• 22.
  3. SYNTAX INFUNCTIONS WHICH TO USE? ▸ function call vs case of ▸ arguably cleaner insert(X,[]) -> [X]; insert(X,Set) -> case lists:member(X,Set) of true -> Set; false -> [X|Set] end.
• 23.
  3. SYNTAX INFUNCTIONS WHICH TO USE? ▸ if vs if through guard? ▸ use where doesn’t need whole pattern matching ▸ personal preference
• 24.
  4. TYPES (ORLACK OF THERE OF) DYNAMITE-STRONG TYPING ▸ as you’ve seen, no need to type Type! ▸ elang is dynamically typed ▸ compiler won't always yell at you ▸ statically typed language is safer..? ▸ elang is reported as nine nine (99.999 % available) ▸ strongly typed 1> 6 + "1".
• 25.
  4. TYPES (ORLACK OF THERE OF) DYNAMITE-STRONG TYPING ▸ type conversion <type>_to_<type> 1> erlang:list_to_integer("54"). 54 2> erlang:integer_to_list(54). "54" 3> erlang:list_to_integer("54.32"). ** exception error: bad argument in function list_to_integer/1 called as list_to_integer("54.32")
• 26.
  4. TYPES (ORLACK OF THERE OF) TYPE CONVERSION ▸ type conversion <type>_to_<type> 4> erlang:list_to_float("54.32"). 54.32 5> erlang:atom_to_list(true). "true" 6> erlang:list_to_bitstring("hi there"). <<"hi there">> 7> erlang:bitstring_to_list(<<"hi there">>). "hi there"
• 27.
  4. TYPES (ORLACK OF THERE OF) TO GUARD A DATA TYPE ▸ type check ▸ why not typeof ▸ force user to make program that surely knowing the effect ▸ can used in guard expression is_atom/1 is_binary/1 is_bitstring/1 is_boolean/1 is_builtin/3 is_float/1 is_function/1 is_function/2 is_integer/1 is_list/1 is_number/1 is_pid/1 is_port/1 is_record/2 is_record/3 is_reference/1 is_tuple/1
• 28.
  4. TYPES (ORLACK OF THERE OF) FOR TYPE JUNKIES ▸ briefly describe tools used to do static type analysis ▸ first try in 1997 ▸ success type ▸ will not exact type every expression ▸ type it infers are right, type errors it finds are really error
• 29.
  4. TYPES (ORLACK OF THERE OF) FOR TYPE JUNKIES ▸ success type and(_,_) -> bool() and(false, _) -> false; and(_, false) -> false; and(true,true) -> true.
• 30.
  4. TYPES (ORLACK OF THERE OF) FOR TYPE JUNKIES ▸ if you interested $ typer --help $ dialyzer --help
• 31.
  5. RECURSION HELLO RECURSION! ▸functional programming do not offer loop
• 32.
  5. RECURSION HELLO RECURSION! ▸factorial -module(recursive). -export([fac/1]). fac(N) when N == 0 -> 1; fac(N) when N > 0 -> N*fac(N-1). fac(0) -> 1; fac(N) when N > 0 -> N*fac(N-1).
• 33.
  5. RECURSION HELLO RECURSION! ▸recursion ▸ function that calls itself ▸ need to have stopping condition (base case)
• 34.
  5. RECURSION LENGTH ▸ weneed ▸ simplest - empty list a base case; a function that calls itself; a list to try our function on. fac(0) -> 1; fac(N) when N > 0 -> N*fac(N-1).
• 35.
  5, RECURSION LENGTH ▸ listis recursively [1 | [2| ... [n | []]]]. len([]) -> 0; len([_|T]) -> 1 + len(T).
• 36.
  5, RECURSION LENGTH ▸ howit works? len([1,2,3,4]) = len([1 | [2,3,4]) = 1 + len([2 | [3,4]]) = 1 + 1 + len([3 | [4]]) = 1 + 1 + 1 + len([4 | []]) = 1 + 1 + 1 + 1 + len([]) = 1 + 1 + 1 + 1 + 0 = 1 + 1 + 1 + 1 = 1 + 1 + 2 = 1 + 3 = 4
• 37.
  5. RECURSION LENGTH OFA TAIL RECURSION ▸ it is problematic ▸ keep millions of numbers in memory for such a simple calculation. ▸ let’s tail recursion
• 38.
  5. RECURSION LENGTH OFA TAIL RECURSION ▸ linear -> iterative one ▸ need to be alone ▸ additional is stacked ▸ accumulator ▸ need to extra variable to hold the intermediate result
• 39.
  5. RECURSION LENGTH OFA TAIL RECURSION ▸ using accumulator tail_fac(N) -> tail_fac(N,1). tail_fac(0,Acc) -> Acc; tail_fac(N,Acc) when N > 0 -> tail_fac(N-1,N*Acc). tail_fac(4) = tail_fac(4,1) tail_fac(4,1) = tail_fac(4-1, 4*1) tail_fac(3,4) = tail_fac(3-1, 3*4) tail_fac(2,12) = tail_fac(2-1, 2*12) tail_fac(1,24) = tail_fac(1-1, 1*24) tail_fac(0,24) = 24
• 40.
  5. RECURSION LENGTH OFA TAIL RECURSION ▸ length tail recursion tail_len(L) -> tail_len(L,0). tail_len([], Acc) -> Acc; tail_len([_|T], Acc) -> tail_len(T,Acc+1). len([]) -> 0; len([_|T]) -> 1 + len(T).
• 41.
  5. RECURSION MORE RECURSIVEFUNCTIONS ▸ After all, recursion being the only looping construct that exists in Erlang ▸ except list comprehension ▸ duplicate duplicate(0,_) -> []; duplicate(N,Term) when N > 0 -> [Term|duplicate(N-1,Term)].
• 42.
  5. RECURSION MORE RECURSIVEFUNCTIONS ▸ duplicate tail_duplicate(N,Term) -> tail_duplicate(N,Term,[]). tail_duplicate(0,_,List) -> List; tail_duplicate(N,Term,List) when N > 0 -> tail_duplicate(N-1, Term, [Term|List]). function(N, Term) -> while N > 0 -> List = [Term|List], N = N-1 end, List.
• 43.
  5. RECURSION MORE RECURSIVEFUNCTIONS ▸ true nightmare which is not tail recursion reverse([]) -> []; reverse([H|T]) -> reverse(T)++[H]. reverse([1,2,3,4]) = [4]++[3]++[2]++[1] ↑ ↵ = [4,3]++[2]++[1] ↑ ↑ ↵ = [4,3,2]++[1] ↑ ↑ ↑ ↵ = [4,3,2,1]
• 44.
  5. RECURSION MORE RECURSIVEFUNCTIONS ▸ let’s rescue tail_reverse(L) -> tail_reverse(L,[]). tail_reverse([],Acc) -> Acc; tail_reverse([H|T],Acc) -> tail_reverse(T, [H|Acc]). tail_reverse([1,2,3,4]) = tail_reverse([2,3,4], [1]) = tail_reverse([3,4], [2,1]) = tail_reverse([4], [3,2,1]) = tail_reverse([], [4,3,2,1]) = [4,3,2,1]
• 45.
  5. RECURSION MORE RECURSIVEFUNCTIONS ▸ sublist/2 ▸ a little different sublist(_,0) -> []; sublist([],_) -> []; sublist([H|T],N) when N > 0 -> [H|sublist(T,N-1)]. tail_sublist(L, N) -> tail_sublist(L, N, []). tail_sublist(_, 0, SubList) -> SubList; tail_sublist([], _, SubList) -> SubList; tail_sublist([H|T], N, SubList) when N > 0 -> tail_sublist(T, N-1, [H|SubList]).
• 46.
  5. RECURSION MORE RECURSIVEFUNCTIONS ▸ problems.. ▸ solve sublist([1,2,3,4,5,6],3) tail_sublist(L, N) -> reverse(tail_sublist(L, N, [])).
• 47.
  5. RECURSION MORE RECURSIVEFUNCTIONS ▸ zip/2 ▸ implements 1> recursive:zip([a,b,c],[1,2,3]). [{a,1},{b,2},{c,3}] zip([],[]) -> []; zip([X|Xs],[Y|Ys]) -> [{X,Y}|zip(Xs,Ys)]. lenient_zip([],_) -> []; lenient_zip(_,[]) -> []; lenient_zip([X|Xs],[Y|Ys]) -> [{X,Y}|lenient_zip(Xs,Ys)].
• 48.
  5. RECURSION MORE RECURSIVEFUNCTIONS ▸ TCO (tail call optimization) ▸ vm does eliminate current stack frame ▸ LCO (last call optimization) ▸ more general
• 49.
  5. RECURSION QUICK, SORT! ▸naive version ▸ pivot ▸ smallers; equals | lagers ▸ until empty list to sort
• 50.
  5. RECURSION QUICK, SORT! ▸naive version
• 51.
  5. RECURSION QUICK, SORT! ▸as two parts ▸ partioning ▸ apply the partitioning to each parts, and glue quicksort([]) -> []; quicksort([Pivot|Rest]) -> {Smaller, Larger} = partition(Pivot,Rest,[],[]), quicksort(Smaller) ++ [Pivot] ++ quicksort(Larger).
• 52.
  5. RECURSION QUICK, SORT! ▸partitioning partition(_,[], Smaller, Larger) -> {Smaller, Larger}; partition(Pivot, [H|T], Smaller, Larger) -> if H =< Pivot -> partition(Pivot, T, [H|Smaller], Larger); H > Pivot -> partition(Pivot, T, Smaller, [H|Larger]) end. lc_quicksort([]) -> []; lc_quicksort([Pivot|Rest]) -> lc_quicksort([Smaller || Smaller <- Rest, Smaller =< Pivot]) ++ [Pivot] ++ lc_quicksort([Larger || Larger <- Rest, Larger > Pivot]).
• 53.
  5. RECURSION MORE THANLISTS ▸ tree ▸ key / two other node (smaller, larger) ▸ also able to contain empty node
• 54.
  5. RECURSION MORE THANLISTS ▸ lets choice tuple! ▸ empty {node, {Key, Value, Smaller, Larger}} {node, nil} -module(tree). -export([empty/0, insert/3, lookup/2]). empty() -> {node, 'nil'}.
• 55.
  5. RECURSION MORE THANLISTS ▸ base case is empty node ▸ where to put content ▸ compare, larger / smaller
• 56.
  5. RECURSION MORE THANLISTS ▸ compare, larger / smaller insert(Key, Val, {node, 'nil'}) -> {node, {Key, Val, {node, 'nil'}, {node, 'nil'}}}; insert(NewKey, NewVal, {node, {Key, Val, Smaller, Larger}}) when NewKey < Key -> {node, {Key, Val, insert(NewKey, NewVal, Smaller), Larger}}; insert(NewKey, NewVal, {node, {Key, Val, Smaller, Larger}}) when NewKey > Key -> {node, {Key, Val, Smaller, insert(NewKey, NewVal, Larger)}}; insert(Key, Val, {node, {Key, _, Smaller, Larger}}) -> {node, {Key, Val, Smaller, Larger}}.
• 57.
  5. RECURSION MORE THANLISTS ▸ returns completely new tree ▸ sometimes shared by vm ▸ look up lookup(_, {node, 'nil'}) -> undefined; lookup(Key, {node, {Key, Val, _, _}}) -> {ok, Val}; lookup(Key, {node, {NodeKey, _, Smaller, _}}) when Key < NodeKey -> lookup(Key, Smaller); lookup(Key, {node, {_, _, _, Larger}}) -> lookup(Key, Larger).
• 58.
  ▸ using example 5.RECURSION MORE THAN LISTS 1> T1 = tree:insert("Jim Woodland", "[email protected]", tree:empty()). 2> T2 = tree:insert("Mark Anderson", "[email protected]", T1). 3> Addresses = tree:insert("Anita Bath", “[email protected]", tree:insert("Kevin Robert", "[email protected]", tree:insert(“Wilson Longbrow", "[email protected]", T2))).
• 59.
  5. RECURSION MORE THANLISTS {node,{"Jim Woodland","[email protected]", {node,{"Anita Bath","[email protected]", {node,nil}, {node,nil}}}, {node,{"Mark Anderson","[email protected]", {node,{"Kevin Robert","[email protected]", {node,nil}, {node,nil}}}, {node,{"Wilson Longbrow","[email protected]", {node,nil}, {node,nil}}}}}}}
• 60.
  5. RECURSION MORE THANLISTS ▸ using example 4> tree:lookup("Anita Bath", Addresses). {ok, "[email protected]"} 5> tree:lookup("Jacques Requin", Addresses). undefined
• 61.
  5. RECURSION THINKING RECURSIVELY ▸our approach is more declarative ▸ consise algorithm easy to understand ▸ divide - and - conquer! ▸ you will learn how to abstract, next time.