Chart of IEnumerable LINQ equivalents in Scala?

Question

Possible Duplicate:
LINQ analogues in Scala

I am looking for chart which shows equivalents in Scala of LINQ methods for IEnumerable:

• First is head
• Select is map
• SingleOrDefault is ... (I don't know)
• ... and so on

Does anyone know anything of such "translate" table?

Answer 1

I am only listing out the equivalents of functions from Enumerable<A>. This is incomplete as of now. I will try to update this later with more.

xs.Aggregate(accumFunc)              -> xs.reduceLeft(accumFunc)
xs.Aggregate(seed, accumFunc)        -> xs.foldLeft(seed)(accumFunc)
xs.Aggregate(seed, accumFunc, trans) -> trans(xs.foldLeft(seed)(accumFunc))
xs.All(pred)                         -> xs.forall(pred)
xs.Any()                             -> xs.nonEmpty
xs.Any(pred)                         -> xs.exists(pred)
xs.AsEnumerable()                    -> xs.asTraversable // roughly
xs.Average()                         -> xs.sum / xs.length
xs.Average(trans)                    -> trans(xs.sum / xs.length)
xs.Cast<A>()                         -> xs.map(_.asInstanceOf[A])
xs.Concat(ys)                        -> xs ++ ys
xs.Contains(x)                       -> xs.contains(x) //////
xs.Contains(x, eq)                   -> xs.exists(eq(x, _))
xs.Count()                           -> xs.size
xs.Count(pred)                       -> xs.count(pred)
xs.DefaultIfEmpty()                  -> if(xs.isEmpty) List(0) else xs // Use `mzero` (from Scalaz) instead of 0 for more genericity
xs.DefaultIfEmpty(v)                 -> if(xs.isEmpty) List(v) else xs
xs.Distinct()                        -> xs.distinct
xs.ElementAt(i)                      -> xs(i)
xs.ElementAtOrDefault(i)             -> xs.lift(i).orZero // `orZero` is from Scalaz
xs.Except(ys)                        -> xs.diff(ys)
xs.First()                           -> xs.head
xs.First(pred)                       -> xs.find(pred) // returns an `Option`
xs.FirstOrDefault()                  -> xs.headOption.orZero
xs.FirstOrDefault(pred)              -> xs.find(pred).orZero
xs.GroupBy(f)                        -> xs.groupBy(f)
xs.GroupBy(f, g)                     -> xs.groupBy(f).mapValues(_.map(g))
xs.Intersect(ys)                     -> xs.intersect(ys)
xs.Last()                            -> xs.last
xs.Last(pred)                        -> xs.reverseIterator.find(pred) // returns an `Option`
xs.LastOrDefault()                   -> xs.lastOption.orZero
xs.LastOrDefault(pred)               -> xs.reverseIterator.find(pred).orZero
xs.Max()                             -> xs.max
xs.Max(f)                            -> xs.maxBy(f)
xs.Min()                             -> xs.min
xs.Min(f)                            -> xs.minBy(f)
xs.OfType<A>()                       -> xs.collect { case x: A => x }
xs.OrderBy(f)                        -> xs.sortBy(f)
xs.OrderBy(f, comp)                  -> xs.sortBy(f)(comp) // `comp` is an `Ordering`.
xs.OrderByDescending(f)              -> xs.sortBy(f)(implicitly[Ordering[A]].reverse)
xs.OrderByDescending(f, comp)        -> xs.sortBy(f)(comp.reverse)
Enumerable.Range(start, count)       -> start until start + count
Enumerable.Repeat(x, times)          -> Iterator.continually(x).take(times)
xs.Reverse()                         -> xs.reverse
xs.Select(trans)                     -> xs.map(trans) // For indexed overload, first `zipWithIndex` and then `map`.
xs.SelectMany(trans)                 -> xs.flatMap(trans)
xs.SequenceEqual(ys)                 -> xs.sameElements(ys)
xs.Skip(n)                           -> xs.drop(n)
xs.SkipWhile(pred)                   -> xs.dropWhile(pred)
xs.Sum()                             -> xs.sum
xs.Sum(f)                            -> xs.map(f).sum // or `xs.foldMap(f)`. Requires Scalaz.
xs.Take(n)                           -> xs.take(n)
xs.TakeWhile(pred)                   -> xs.takeWhile(pred)
xs.OrderBy(f).ThenBy(g)              -> xs.sortBy(x => (f(x), g(x))) // Or: xs.sortBy(f &&& g). `&&&` is from Scalaz.
xs.ToArray()                         -> xs.toArray // Use `xs.toIndexedSeq` for immutable indexed sequence.
xs.ToDictionary(f)                   -> xs.map(f.first).toMap // `first` is from Scalaz. When f = identity, you can just write `xs.toMap`.
xs.ToList()                          -> xs.toList // This returns an immutable list. Use `xs.toBuffer` if you want a mutable list.
xs.Union(ys)                         -> xs.union(ys)
xs.Where(pred)                       -> xs.filter(pred)
xs.Zip(ys, f)                        -> (xs, ys).zipped.map(f) // When f = identity, use `xs.zip(ys)`

---

There is no direct equivalent of some functions, but it's fairly easy to roll your own. Here are some such functions.

Single:

def single[A](xs: Traversable[A]): A = {
  if(xs.isEmpty) sys error "Empty sequence!"
  else if(xs.size > 1) sys error "More than one elements!"
  else xs.head
}

SingleOrDefault:

def singleOrDefault[A : Zero](xs: Traversable[A]): A = {
  if(xs.isEmpty) mzero
  else if(xs.size > 1) sys error "More than one elements!"
  else xs.head
}

Join:

def join[A, B, K, R](outer: Traversable[A], inner: Traversable[B])
    (outKey: A => K, inKey: B => K, f: (A, B) => R): Traversable[R] = {
  for(o <- outer; i <- inner; if outKey(o) == inKey(i)) yield f(o, i)
}

GroupJoin:

def groupJoin[A, B, K, R](outer: Traversable[A], inner: Traversable[B])
    (outKey: A => K, inKey: B => K, f: (A, Traversable[B]) => R): Traversable[R] = {
  for(o <- outer) yield {
    val zs = for(i <- inner; if outKey(o) == inKey(i)) yield i
    f(o, zs)
  }
}

---

Notes:

1. In idiomatic Scala, total functions are generally preferred over partial functions. So, idiomatic implementation of single and singleOrDefault would produce a value of type Either[Exception, A] instead of A. For example, here is refined implementation of single that returns Either[Exception, A].

   def single[A](xs: Traversable[A]): Either[Exception, A] = {
     if(xs.isEmpty) Left(new RuntimeException("Empty sequence!"))
     else if(xs.size > 1) Left(new RuntimeException("More than one elements!"))
     else Right(xs.head)
   }
   

2. Scalaz's Zero/mzero are not quite same as C#'s default value mechanism. For details, you can refer to this post I wrote on this topic some time back.

3. You can use enrich-my-library pattern to achieve the same effect as C#'s extension methods. Refer to this and this for details.