Query the result of for expression inside the loop

Question

for (i <- Range(1,7); j <- Range(1,7))
  yield (i,j) // want to yield only if there is no such a pair

Is it possible to have access to the list, formed by yield, inside the loop? For example, if I don't want to add duplicates.

P.S. The main question is not how to do it in this particular case. But how to avoid duplicates in more complex case, where I want to check, what was already yielded.

score 3 · Accepted Answer · answered Jan 14 '12 at 19:31

You cannot access members of an immutable collection being built from within the for comprehension.

First of all, I'm going to modify your example so it actually produces duplicates:

for (i <- Range(1,3); j <- Range(1,3)) yield (i min j, i max j)
//scala.collection.immutable.IndexedSeq[(Int, Int)] = Vector((1,1), (1,2), (1,2), (2,2))

For comprehensions are just syntactic sugar, so here's the equivalent using map and flatMap that produces the exact same result

Range(1,3).flatMap{i => Range(1,3).map{ j => (i min j, i max j)}}

As you can see, you're not actually building a single collection, but rather a collection of collections, and then merging them together as they're created. The inner map is taking each j in the range and mapping it to a pair, then the outer flatMap is mapping each i to a sequence of pairs and merging them together. If I change the flatMap to just map, the result is this:

Range(1,3).map{i => Range(1,3).map{ j => (i min j, i max j)}}
//Vector(Vector((1,1), (1,2)), Vector((1,2), (2,2)))

So only after the whole operation is finished can you access the result as a single collection. The result is a Vector which extends IndexedSeq[(Int, Int)], so you can use any of that trait's methods on the result, one of them is distinct:

(for (i <- Range(1,3); j <- Range(1,3)) yield (i min j, i max j)).distinct
//Vector((1,1), (1,2), (2,2))

score 2 · Answer 2 · answered Jan 14 '12 at 18:05

2

Like this

for (i <- Range(1,7); j <- Range(1,7); if i != j ) yield (i,j)

or this

for (i <- Range(1,7); j <- Range(1,7); if i < j ) yield (i,j)

?

answered Jan 14 '12 at 18:05

0__

66,707
21
171
266

The main question is not how to do it in this particular case. But how to avoid duplicates in more complex case, where I want to check, what was already yielded. – Konstantin Milyutin Jan 14 '12 at 18:22

score 1 · Answer 3 · answered Jan 14 '12 at 18:05

1

(for {i <- (1 to 7); j <- (1 to 7)} yield (i,j)).distinct

This would return list of tuples without duplicates.

And no, there is no way to have access to the list inside for loop. That would break the functional programming, after all.

answered Jan 14 '12 at 18:05

Rogach

26,050
21
93
172

thanks. can you explain what functions I can apply on yield. If you apply distinct, then yield represents a list? – Konstantin Milyutin Jan 14 '12 at 18:21
@damluar I think you misunderstand what "yield" means. See http://stackoverflow.com/questions/1052476/can-someone-explain-scalas-yield – Dan Burton Jan 14 '12 at 23:36
No, I understand it right. I just didnt't notice brackets surrounding for. – Konstantin Milyutin Jan 15 '12 at 00:16

score 1 · Answer 4 · answered Jan 14 '12 at 21:10

I'm not sure, whether you want to prevent pairs of identic entries, or identic pairs, ignoring which comes first.

In the second case, you might use an ordering, and just add cases where a < b or a <= b:

for (i <- Range(1,6); j <- Range(i+1, 7)) yield (i,j)

score -1 · Answer 5 · answered Jan 14 '12 at 18:09

-1

for (i <- Range(1,7); j <- Range( i ,7))
  yield (i,j)

answered Jan 14 '12 at 18:09

Infinity

3,431
3
25
46

Query the result of for expression inside the loop

5 Answers5