Scala objects not changing their internal state - list

I am seeing a problem with some Scala 2.7.7 code I'm working on, that should not happen if it the equivalent was written in Java. Loosely, the code goes creates a bunch of card players and assigns them to tables.
class Player(val playerNumber : Int)
class Table (val tableNumber : Int) {
var players : List[Player] = List()
def registerPlayer(player : Player) {
println("Registering player " + player.playerNumber + " on table " + tableNumber)
players = player :: players
}
}
object PlayerRegistrar {
def assignPlayersToTables(playSamplesToExecute : Int, playersPerTable:Int) = {
val numTables = playSamplesToExecute / playersPerTable
val tables = (1 to numTables).map(new Table(_))
assert(tables.size == numTables)
(0 until playSamplesToExecute).foreach {playSample =>
val tableNumber : Int = playSample % numTables
tables(tableNumber).registerPlayer(new Player(playSample))
}
tables
}
}
The PlayerRegistrar assigns a number of players between tables. First, it works out how many tables it will need to break up the players between and creates a List of them.
Then in the second part of the code, it works out which table a player should be assigned to, pulls that table from the list and registers a new player on that table.
The list of players on a table is a var, and is overwritten each time registerPlayer() is called. I have checked that this works correctly through a simple TestNG test:
#Test def testRegisterPlayer_multiplePlayers() {
val table = new Table(1)
(1 to 10).foreach { playerNumber =>
val player = new Player(playerNumber)
table.registerPlayer(player)
assert(table.players.contains(player))
assert(table.players.length == playerNumber)
}
}
I then test the table assignment:
#Test def testAssignPlayerToTables_1table() = {
val tables = PlayerRegistrar.assignPlayersToTables(10, 10)
assertEquals(tables.length, 1)
assertEquals(tables(0).players.length, 10)
}
The test fails with "expected:<10> but was:<0>". I've been scratching my head, but can't work out why registerPlayer() isn't mutating the table in the list. Any help would be appreciated.

The reason is that in the assignPlayersToTables method, you are creating a new Table object. You can confirm this by adding some debugging into the loop:
val tableNumber : Int = playSample % numTables
println(tables(tableNumber))
tables(tableNumber).registerPlayer(new Player(playSample))
Yielding something like:
Main$$anon$1$Table#5c73a7ab
Registering player 0 on table 1
Main$$anon$1$Table#21f8c6df
Registering player 1 on table 1
Main$$anon$1$Table#53c86be5
Registering player 2 on table 1
Note how the memory address of the table is different for each call.
The reason for this behaviour is that a Range is non-strict in Scala (until Scala 2.8, anyway). This means that the call to the range is not evaluated until it's needed. So you think you're getting back a list of Table objects, but actually you're getting back a range which is evaluated (instantiating a new Table object) each time you call it. Again, you can confirm this by adding some debugging:
val tables = (1 to numTables).map(new Table(_))
println(tables)
Which gives you:
RangeM(Main$$anon$1$Table#5492bbba)
To do what you want, add a toList to the end:
val tables = (1 to numTables).map(new Table(_)).toList

val tables = (1 to numTables).map(new Table(_))
This line seems to be causing all the trouble - mapping over 1 to n gives you a RandomAccessSeq.Projection, and to be honest, I don't know how exactly they work, but a bit less clever initialising technique does the job.
var tables: Array[Table] = new Array(numTables)
for (i <- 0 to numTables) tables(i) = new Table(i)
Using the first initialisation method I wasn't able to change the objects (just like you), but using a simple array everything seems to be working.

Related

DynamoDB - Get all items which overlap a search time interval

My application manages bookings of a user. These bookings are composed by a start_date and end_date, and their current partition in dynamodb is the following:
PK SK DATA
USER#1#BOOKINGS BOOKING#1 {s: '20190601', e: '20190801'}
[GOAL] I would query all reservations which overlap a search time interval as the following:
I tried to find a solution for this issue but I found only a way to query all items inside a search time interval, which solves only this problem:
I decided to make an implementation of it to try to make some change to solve my problem but I didn't found a solution, following you can find my implementation of "query inside interval" (this is not a dynamodb implementation, but I will replace isBetween function with BETWEEN operand):
import { zip } from 'lodash';
const bookings = [
{ s: '20190601', e: '20190801', i: '' },
{ s: '20180702', e: '20190102', i: '' }
];
const search_start = '20190602'.split('');
const search_end = '20190630'.split('');
// s:20190601 e:20190801 -> i:2200119900680011
for (const b of bookings) {
b['i'] = zip(b.s.split(''), b.e.split(''))
.reduce((p, c) => p + c.join(''), '');
}
// (start_search: 20190502, end_search: 20190905) => 22001199005
const start_clause: string[] = [];
for (let i = 0; i < search_start.length; i += 1) {
if (search_start[i] === search_end[i]) {
start_clause.push(search_start[i] + search_end[i]);
} else {
start_clause.push(search_start[i]);
break;
}
}
const s_index = start_clause.join('');
// (end_search: 20190905, start_search: 20190502) => 22001199009
const end_clause: string[] = [];
for (let i = 0; i < search_end.length; i += 1) {
if (search_end[i] === search_start[i]) {
end_clause.push(search_end[i] + search_start[i]);
} else {
end_clause.push(search_end[i]);
break;
}
}
const e_index = (parseInt(end_clause.join('')) + 1).toString();
const isBetween = (s: string, e: string, v: string) => {
const sorted = [s,e,v].sort();
console.info(`sorted: ${sorted}`)
return sorted[1] === v;
}
const filtered_bookings = bookings
.filter(b => isBetween(s_index, e_index, b.i));
console.info(`filtered_bookings: ${JSON.stringify(filtered_bookings)}`)
There’s not going to be a beautiful and simple yet generic answer.
Probably the best approach is to pre-define your time period size (days, hours, minutes, seconds, whatever) and use the value of that as the PK so for each day (or hour or whatever) you have in that item collection a list of the items touching that day with the sort key of the start time (so you can do the inequality there) and you can use a filter on the end time attribute.
If your chosen time period is days and you need to query across a week then you’ll issue seven queries. So pick a time unit that’s around the same size as your selected time periods.
Remember you need to put all items touching that day (or whatever) into the day collection. If an item spans a week it needs to be inserted 7 times.
Disclaimer: This is a very use-case-specific and non-general approach I took when trying to solve the same problem; it picks up on #hunterhacker 's approach.
Observations from my use case:
The data I'm dealing with is financial/stock data, which spans back roughly 50 years in the past up to 150 years into the future.
I have many thousands of items per year, and I would like to avoid pulling in all 200 years of information
The vast majority of the items I want to query spans a time that fits within a year (ie. most items don't go from 30-Dec-2001 to 02-Jan-2002, but rather from 05-Mar-2005 to 10-Mar-2005)
Based on the above, I decided to add an LSI and save the relevant year for every item whose start-to-end time is within a single year. The items that straddle a year (or more) I set that LSI with 0.
The querying looks like:
if query_straddles_year:
# This doesn't happen often in my use case
result = query_all_and_filter_after()
else:
# Most cases end up here (looking for a single day, for instance)
year_constrained_result = query_using_lsi_for_that_year()
result_on_straddling_bins = query_using_lsi_marked_with_0() # <-- this is to get any of the indexes that do straddle a year
filter_and_combine(year_constrained_result, result_on_straddling_bins)

Retrieve values from an array - get "cannot call value of non-function type String"

I'm trying to retrieve a value from an array, based on an index parsed from a string of digits. I'm stuck on this error, and the other answers to similar questions in this forum appear to be for more advanced developers (this is my first iOS app).
The app will eventually look up weather reports ("MAFOR" groupings of 5 digits each) from a web site, parse each group and lookup values from arrays for wind direction, speed, forecast period etc using each character.
The playground code is below, appreciate any help on where I am going wrong (look for ***)
//: Playground - noun: a place where people can play
import UIKit
var str = "Hello, playground"
// create array for Forecast Period
let forecastPeriodArray = ["Existing conditions at beginning","3 hours","6 hours","9 hours","12 hours","18 hours","24 hours","48 hours","72 hours","Occasionally"]
// create array for Wind Direction
let windDirectionArray = ["Calm","Northeast","East","Southeast","South","Southwest","West","Northwest","North","Variable"]
// create array for Wind Velocity
let windVelocityArray = ["0-10 knots","11-16 knots","17-21 knots","22-27 knots","28-33 knots","34-40 knots","41-47 knots","48-55 knots","56-63 knots","64-71 knots"]
// create array for Forecast Weather
let forecastWeatherArray = ["Moderate or good visibility (> 3 nm.","Risk of ice accumulation (temp 0C to -5C","Strong risk of ice accumulkation (air temp < -5C)","Mist (visibility 1/2 to 3 nm.)","Fog (visibility less than 1/2 nm.)","Drizzle","Rain","Snow, or rain and snow","Squally weather with or without showers","Thunderstorms"]
// retrieve full MAFOR line of several information groups (this will be pulled from a web site)
var myMaforLineString = "11747 19741 13757 19751 11730 19731 11730 13900 11630 13637"
// split into array components wherever " " is encountered
var myMaforArray = myMaforLineString.components(separatedBy: " ")
let count = myMaforArray.count
print("There are \(count) items in the array")
// Go through each group and parse out the needed digits
for maforGroup in myMaforArray {
print("MAFOR group \(maforGroup)")
// get Forecast Period
var idx = maforGroup.index(maforGroup.startIndex, offsetBy: 1)
var periodInt = maforGroup[idx]
print("periodInt is \(periodInt)")
// *** here is where I am stuck... trying to use the periodInt index value to retrieve the description from the ForecastPeriodArray
var periodDescription = forecastPeriodArray(periodInt)
print("Forecast period = (forecastPeriodArray(periodInt)")
// get Wind Direction
idx = maforGroup.index(maforGroup.startIndex, offsetBy: 2)
var directionInt = maforGroup[idx]
print("directionInt is \(directionInt)")
// get Wind Velocity
idx = maforGroup.index(maforGroup.startIndex, offsetBy: 3)
var velocityInt = maforGroup[idx]
print("velocityInt is \(velocityInt)")
// get Weather Forecast
idx = maforGroup.index(maforGroup.startIndex, offsetBy: 4)
var weatherInt = maforGroup[idx]
print("weatherInt is \(weatherInt)")
}
#shallowThought was close.
You are trying to access an array by its index, therefore use the array[index] notation. But your index has to be of the correct type. forecastPeriodArray[periodInt] therefore does not work since periodInt is not an Int as the name would suggest. Currently it is of type Character which does not make much sense.
What you are probably trying to achieve is convert the character to an integer and use that to access the array:
var periodInt = Int(String(maforGroup[idx]))!
You might want to add error handling for the case when the character does not actually represent an integer.

How to maintain an immutable list when you impact object linked to each other into this list

I'm trying to code the fast Non Dominated Sorting algorithm (NDS) of Deb used in NSGA2 in immutable way using Scala.
But the problem seems more difficult than i think, so i simplify here the problem to make a MWE.
Imagine a population of Seq[A], and each A element is decoratedA with a list which contains pointers to other elements of the population Seq[A].
A function evalA(a:decoratedA) take the list of linkedA it contains, and decrement value of each.
Next i take a subset list decoratedAPopulation of population A, and call evalA on each. I have a problem, because between each iteration on element on this subset list decoratedAPopulation, i need to update my population of A with the new decoratedA and the new updated linkedA it contain ...
More problematic, each element of population need an update of 'linkedA' to replace the linked element if it change ...
Hum as you can see, it seem complicated to maintain all linked list synchronized in this way. I propose another solution bottom, which probably need recursion to return after each EvalA a new Population with element replaced.
How can i do that correctly in an immutable way ?
It's easy to code in a mutable way, but i don't find a good way to do this in an immutable way, do you have a path or an idea to do that ?
object test extends App{
case class A(value:Int) {def decrement()= new A(value - 1)}
case class decoratedA(oneAdecorated:A, listOfLinkedA:Seq[A])
// We start algorithm loop with A element with value = 0
val population = Seq(new A(0), new A(0), new A(8), new A(1))
val decoratedApopulation = Seq(new decoratedA(population(1),Seq(population(2),population(3))),
new decoratedA(population(2),Seq(population(1),population(3))))
def evalA(a:decoratedA) = {
val newListOfLinked = a.listOfLinkedA.map{e => e.decrement()
new decoratedA(a.oneAdecorated,newListOfLinked)}
}
def run()= {
//decoratedApopulation.map{
// ?
//}
}
}
Update 1:
About the input / output of the initial algorithm.
The first part of Deb algorithm (Step 1 to Step 3) analyse a list of Individual, and compute for each A : (a) domination count, the number of A which dominate me (the value attribute of A) (b) a list of A i dominate (listOfLinkedA).
So it return a Population of decoratedA totally initialized, and for the entry of Step 4 (my problem) i take the first non dominated front, cf. the subset of elements of decoratedA with A value = 0.
My problem start here, with a list of decoratedA with A value = 0; and i search the next front into this list by computing each listOfLinkedA of each of this A
At each iteration between step 4 to step 6, i need to compute a new B subset list of decoratedA with A value = 0. For each , i decrementing first the domination count attribute of each element into listOfLinkedA, then i filter to get the element equal to 0. A the end of step 6, B is saved to a list List[Seq[DecoratedA]], then i restart to step 4 with B, and compute a new C, etc.
Something like that in my code, i call explore() for each element of B, with Q equal at the end to new subset of decoratedA with value (fitness here) = 0 :
case class PopulationElement(popElement:Seq[Double]){
implicit def poptodouble():Seq[Double] = {
popElement
}
}
class SolutionElement(values: PopulationElement, fitness:Double, dominates: Seq[SolutionElement]) {
def decrement()= if (fitness == 0) this else new SolutionElement(values,fitness - 1, dominates)
def explore(Q:Seq[SolutionElement]):(SolutionElement, Seq[SolutionElement])={
// return all dominates elements with fitness - 1
val newSolutionSet = dominates.map{_.decrement()}
val filteredSolution:Seq[SolutionElement] = newSolutionSet.filter{s => s.fitness == 0.0}.diff{Q}
filteredSolution
}
}
A the end of algorithm, i have a final list of seq of decoratedA List[Seq[DecoratedA]] which contain all my fronts computed.
Update 2
A sample of value extracted from this example.
I take only the pareto front (red) and the {f,h,l} next front with dominated count = 1.
case class p(x: Int, y: Int)
val a = A(p(3.5, 1.0),0)
val b = A(p(3.0, 1.5),0)
val c = A(p(2.0, 2.0),0)
val d = A(p(1.0, 3.0),0)
val e = A(p(0.5, 4.0),0)
val f = A(p(0.5, 4.5),1)
val h = A(p(1.5, 3.5),1)
val l = A(p(4.5, 1.0),1)
case class A(XY:p, value:Int) {def decrement()= new A(XY, value - 1)}
case class ARoot(node:A, children:Seq[A])
val population = Seq(
ARoot(a,Seq(f,h,l),
ARoot(b,Seq(f,h,l)),
ARoot(c,Seq(f,h,l)),
ARoot(d,Seq(f,h,l)),
ARoot(e,Seq(f,h,l)),
ARoot(f,Nil),
ARoot(h,Nil),
ARoot(l,Nil))
Algorithm return List(List(a,b,c,d,e), List(f,h,l))
Update 3
After 2 hour, and some pattern matching problems (Ahum...) i'm comming back with complete example which compute automaticaly the dominated counter, and the children of each ARoot.
But i have the same problem, my children list computation is not totally correct, because each element A is possibly a shared member of another ARoot children list, so i need to think about your answer to modify it :/ At this time i only compute children list of Seq[p], and i need list of seq[A]
case class p(x: Double, y: Double){
def toSeq():Seq[Double] = Seq(x,y)
}
case class A(XY:p, dominatedCounter:Int) {def decrement()= new A(XY, dominatedCounter - 1)}
case class ARoot(node:A, children:Seq[A])
case class ARootRaw(node:A, children:Seq[p])
object test_stackoverflow extends App {
val a = new p(3.5, 1.0)
val b = new p(3.0, 1.5)
val c = new p(2.0, 2.0)
val d = new p(1.0, 3.0)
val e = new p(0.5, 4.0)
val f = new p(0.5, 4.5)
val g = new p(1.5, 4.5)
val h = new p(1.5, 3.5)
val i = new p(2.0, 3.5)
val j = new p(2.5, 3.0)
val k = new p(3.5, 2.0)
val l = new p(4.5, 1.0)
val m = new p(4.5, 2.5)
val n = new p(4.0, 4.0)
val o = new p(3.0, 4.0)
val p = new p(5.0, 4.5)
def isStriclyDominated(p1: p, p2: p): Boolean = {
(p1.toSeq zip p2.toSeq).exists { case (g1, g2) => g1 < g2 }
}
def sortedByRank(population: Seq[p]) = {
def paretoRanking(values: Set[p]) = {
//comment from #dk14: I suppose order of values isn't matter here, otherwise use SortedSet
values.map { v1 =>
val t = (values - v1).filter(isStriclyDominated(v1, _)).toSeq
val a = new A(v1, values.size - t.size - 1)
val root = new ARootRaw(a, t)
println("Root value ", root)
root
}
}
val listOfARootRaw = paretoRanking(population.toSet)
//From #dk14: Here is convertion from Seq[p] to Seq[A]
val dominations: Map[p, Int] = listOfARootRaw.map(a => a.node.XY -> a.node.dominatedCounter) //From #dk14: It's a map with dominatedCounter for each point
val listOfARoot = listOfARootRaw.map(raw => ARoot(raw.node, raw.children.map(p => A(p, dominations.getOrElse(p, 0)))))
listOfARoot.groupBy(_.node.dominatedCounter)
}
//Get the first front, a subset of ARoot, and start the step 4
println(sortedByRank(Seq(a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p)).head)
}
Talking about your problem with distinguishing fronts (after update 2):
val (left,right) = population.partition(_.node.value == 0)
List(left, right.map(_.copy(node = node.copy(value = node.value - 1))))
No need for mutating anything here. copy will copy everything but fields you specified with new values. Talking about the code, the new copy will be linked to the same list of children, but new value = value - 1.
P.S. I have a feeling you may actually want to do something like this:
case class A(id: String, level: Int)
val a = A("a", 1)
val b = A("b", 2)
val c = A("c", 2)
val d = A("d", 3)
clusterize(List(a,b,c,d)) === List(List(a), List(b,c), List(d))
It's simple to implement:
def clusterize(list: List[A]) =
list.groupBy(_.level).toList.sortBy(_._1).map(_._2)
Test:
scala> clusterize(List(A("a", 1), A("b", 2), A("c", 2), A("d", 3)))
res2: List[List[A]] = List(List(A(a,1)), List(A(b,2), A(c,2)), List(A(d,3)))
P.S.2. Please consider better naming conventions, like here.
Talking about "mutating" elements in some complex structure:
The idea of "immutable mutating" some shared (between parts of a structure) value is to separate your "mutation" from the structure. Or simply saying, divide and conquerror:
calculate changes in advance
apply them
The code:
case class A(v: Int)
case class AA(a: A, seq: Seq[A]) //decoratedA
def update(input: Seq[AA]) = {
//shows how to decrement each value wherever it is:
val stats = input.map(_.a).groupBy(identity).mapValues(_.size) //domination count for each A
def upd(a: A) = A(a.v - stats.getOrElse(a, 0)) //apply decrement
input.map(aa => aa.copy(aa = aa.seq.map(upd))) //traverse and "update" original structure
}
So, I've introduced new Map[A, Int] structure, that shows how to modify the original one. This approach is based on highly simplified version of Applicative Functor concept. In general case, it should be Map[A, A => A] or even Map[K, A => B] or even Map[K, Zipper[A] => B] as applicative functor (input <*> map). *Zipper (see 1, 2) actually could give you information about current element's context.
Notes:
I assumed that As with same value are same; that's default behaviour for case classess, otherwise you need to provide some additional id's (or redefine hashCode/equals).
If you need more levels - like AA(AA(AA(...)))) - just make stats and upd recursive, if dеcrement's weight depends on nesting level - just add nesting level as parameter to your recursive function.
If decrement depends on parent node (like decrement only A(3)'s, which belongs to A(3)) - add parent node(s) as part of stats's key and analise it during upd.
If there is some dependency between stats calculation (how much to decrement) of let's say input(1) from input(0) - you should use foldLeft with partial stats as accumulator: val stats = input.foldLeft(Map[A, Int]())((partialStats, elem) => partialStats ++ analize(partialStats, elem))
Btw, it takes O(N) here (linear memory and cpu usage)
Example:
scala> val population = Seq(A(3), A(6), A(8), A(3))
population: Seq[A] = List(A(3), A(6), A(8), A(3))
scala> val input = Seq(AA(population(1),Seq(population(2),population(3))), AA(population(2),Seq(population(1),population(3))))
input: Seq[AA] = List(AA(A(6),List(A(8), A(3))), AA(A(8),List(A(6), A(3))))
scala> update(input)
res34: Seq[AA] = List(AA(A(5),List(A(7), A(3))), AA(A(7),List(A(5), A(3))))

Insertion order of a list based on order of another list

I have a sorting problem in Scala that I could certainly solve with brute-force, but I'm hopeful there is a more clever/elegant solution available. Suppose I have a list of strings in no particular order:
val keys = List("john", "jill", "ganesh", "wei", "bruce", "123", "Pantera")
Then at random, I receive the values for these keys at random (full-disclosure, I'm experiencing this problem in an akka actor, so events are not in order):
def receive:Receive = {
case Value(key, otherStuff) => // key is an element in keys ...
And I want to store these results in a List where the Value objects appear in the same order as their key fields in the keys list. For instance, I may have this list after receiving the first two Value messages:
List(Value("ganesh", stuff1), Value("bruce", stuff2))
ganesh appears before bruce merely because he appears earlier in the keys list. Once the third message is received, I should insert it into this list in the correct location per the ordering established by keys. For instance, on receiving wei I should insert him into the middle:
List(Value("ganesh", stuff1), Value("wei", stuff3), Value("bruce", stuff2))
At any point during this process, my list may be incomplete but in the expected order. Since the keys are redundant with my Value data, I throw them away once the list of values is complete.
Show me what you've got!
I assume you want no worse than O(n log n) performance. So:
val order = keys.zipWithIndex.toMap
var part = collection.immutable.TreeSet.empty[Value](
math.Ordering.by(v => order(v.key))
)
Then you just add your items.
scala> part = part + Value("ganesh", 0.1)
part: scala.collection.immutable.TreeSet[Value] =
TreeSet(Value(ganesh,0.1))
scala> part = part + Value("bruce", 0.2)
part: scala.collection.immutable.TreeSet[Value] =
TreeSet(Value(ganesh,0.1), Value(bruce,0.2))
scala> part = part + Value("wei", 0.3)
part: scala.collection.immutable.TreeSet[Value] =
TreeSet(Value(ganesh,0.1), Value(wei,0.3), Value(bruce,0.2))
When you're done, you can .toList it. While you're building it, you probably don't want to, since updating a list in random order so that it is in a desired sorted order is an obligatory O(n^2) cost.
Edit: with your example of seven items, my solution takes about 1/3 the time of Jean-Philippe's. For 25 items, it's 1/10th the time. 1/30th for 200 (which is the difference between 6 ms and 0.2 ms on my machine).
If you can use a ListMap instead of a list of tuples to store values while they're gathered, this could work. ListMap preserves insertion order.
class MyActor(keys: List[String]) extends Actor {
def initial(values: ListMap[String, Option[Value]]): Receive = {
case v # Value(key, otherStuff) =>
if(values.forall(_._2.isDefined))
context.become(valuesReceived(values.updated(key, Some(v)).collect { case (_, Some(v)) => v))
else
context.become(initial(keys, values.updated(key, Some(v))))
}
def valuesReceived(values: Seq[Value]): Receive = { } // whatever you need
def receive = initial(keys.map { k => (k -> None) })
}
(warning: not compiled)

Erlang record item list

For example i have erlang record:
-record(state, {clients
}).
Can i make from clients field list?
That I could keep in client filed as in normal list? And how can i add some values in this list?
Thank you.
Maybe you mean something like:
-module(reclist).
-export([empty_state/0, some_state/0,
add_client/1, del_client/1,
get_clients/1]).
-record(state,
{
clients = [] ::[pos_integer()],
dbname ::char()
}).
empty_state() ->
#state{}.
some_state() ->
#state{
clients = [1,2,3],
dbname = "QA"}.
del_client(Client) ->
S = some_state(),
C = S#state.clients,
S#state{clients = lists:delete(Client, C)}.
add_client(Client) ->
S = some_state(),
C = S#state.clients,
S#state{clients = [Client|C]}.
get_clients(#state{clients = C, dbname = _D}) ->
C.
Test:
1> reclist:empty_state().
{state,[],undefined}
2> reclist:some_state().
{state,[1,2,3],"QA"}
3> reclist:add_client(4).
{state,[4,1,2,3],"QA"}
4> reclist:del_client(2).
{state,[1,3],"QA"}
::[pos_integer()] means that the type of the field is a list of positive integer values, starting from 1; it's the hint for the analysis tool dialyzer, when it performs type checking.
Erlang also allows you use pattern matching on records:
5> reclist:get_clients(reclist:some_state()).
[1,2,3]
Further reading:
Records
Types and Function Specifications
dialyzer(1)
#JUST MY correct OPINION's answer made me remember that I love how Haskell goes about getting the values of the fields in the data type.
Here's a definition of a data type, stolen from Learn You a Haskell for Great Good!, which leverages record syntax:
data Car = Car {company :: String
,model :: String
,year :: Int
} deriving (Show)
It creates functions company, model and year, that lookup fields in the data type. We first make a new car:
ghci> Car "Toyota" "Supra" 2005
Car {company = "Toyota", model = "Supra", year = 2005}
Or, using record syntax (the order of fields doesn't matter):
ghci> Car {model = "Supra", year = 2005, company = "Toyota"}
Car {company = "Toyota", model = "Supra", year = 2005}
ghci> let supra = Car {model = "Supra", year = 2005, company = "Toyota"}
ghci> year supra
2005
We can even use pattern matching:
ghci> let (Car {company = c, model = m, year = y}) = supra
ghci> "This " ++ c ++ " " ++ m ++ " was made in " ++ show y
"This Toyota Supra was made in 2005"
I remember there were attempts to implement something similar to Haskell's record syntax in Erlang, but not sure if they were successful.
Some posts, concerning these attempts:
In Response to "What Sucks About Erlang"
Geeking out with Lisp Flavoured Erlang. However I would ignore parameterized modules here.
It seems that LFE uses macros, which are similar to what provides Scheme (Racket, for instance), when you want to create a new value of some structure:
> (define-struct car (company model year))
> (define supra (make-car "Toyota" "Supra" 2005))
> (car-model supra)
"Supra"
I hope we'll have something close to Haskell record syntax in the future, that would be really practically useful and handy.
Yasir's answer is the correct one, but I'm going to show you WHY it works the way it works so you can understand records a bit better.
Records in Erlang are a hack (and a pretty ugly one). Using the record definition from Yasir's answer...
-record(state,
{
clients = [] ::[pos_integer()],
dbname ::char()
}).
...when you instantiate this with #state{} (as Yasir did in empty_state/0 function), what you really get back is this:
{state, [], undefined}
That is to say your "record" is just a tuple tagged with the name of the record (state in this case) followed by the record's contents. Inside BEAM itself there is no record. It's just another tuple with Erlang data types contained within it. This is the key to understanding how things work (and the limitations of records to boot).
Now when Yasir did this...
add_client(Client) ->
S = some_state(),
C = S#state.clients,
S#state{clients = [Client|C]}.
...the S#state.clients bit translates into code internally that looks like element(2,S). You're using, in other words, standard tuple manipulation functions. S#state.clients is just a symbolic way of saying the same thing, but in a way that lets you know what element 2 actually is. It's syntactic saccharine that's an improvement over keeping track of individual fields in your tuples in an error-prone way.
Now for that last S#state{clients = [Client|C]} bit, I'm not absolutely positive as to what code is generated behind the scenes, but it is likely just straightforward stuff that does the equivalent of {state, [Client|C], element(3,S)}. It:
tags a new tuple with the name of the record (provided as #state),
copies the elements from S (dictated by the S# portion),
except for the clients piece overridden by {clients = [Client|C]}.
All of this magic is done via a preprocessing hack behind the scenes.
Understanding how records work behind the scenes is beneficial both for understanding code written using records as well as for understanding how to use them yourself (not to mention understanding why things that seem to "make sense" don't work with records -- because they don't actually exist down in the abstract machine...yet).
If you are only adding or removing single items from the clients list in the state you could cut down on typing with a macro.
-record(state, {clients = [] }).
-define(AddClientToState(Client,State),
State#state{clients = lists:append([Client], State#state.clients) } ).
-define(RemoveClientFromState(Client,State),
State#state{clients = lists:delete(Client, State#state.clients) } ).
Here is a test escript that demonstrates:
#!/usr/bin/env escript
-record(state, {clients = [] }).
-define(AddClientToState(Client,State),
State#state{clients = lists:append([Client], State#state.clients)} ).
-define(RemoveClientFromState(Client,State),
State#state{clients = lists:delete(Client, State#state.clients)} ).
main(_) ->
%Start with a state with a empty list of clients.
State0 = #state{},
io:format("Empty State: ~p~n",[State0]),
%Add foo to the list
State1 = ?AddClientToState(foo,State0),
io:format("State after adding foo: ~p~n",[State1]),
%Add bar to the list.
State2 = ?AddClientToState(bar,State1),
io:format("State after adding bar: ~p~n",[State2]),
%Add baz to the list.
State3 = ?AddClientToState(baz,State2),
io:format("State after adding baz: ~p~n",[State3]),
%Remove bar from the list.
State4 = ?RemoveClientFromState(bar,State3),
io:format("State after removing bar: ~p~n",[State4]).
Result:
Empty State: {state,[]}
State after adding foo: {state,[foo]}
State after adding bar: {state,[bar,foo]}
State after adding baz: {state,[baz,bar,foo]}
State after removing bar: {state,[baz,foo]}