anti-join does not work as expected in presence of missing values

[real-iceman]

If a value of the join column is available in both datasets of an anti-join, no rows for this join column value should appear in the result set. However, they are included if another column (of the left-join) contains a missing-value.

Example from the documentation:

ds1:

:a	:b	:c
1	101	a
2	102	b
1	103	s
2	104
3	105	t
4	106	r
	107	a
	108	c
4	109	t

ds2:

:a	:b	:c	:d	:e
	110	d	X	3
1	109	a	X	4
2	108	t	X	5
5	107	a	X	6
4	106	t	X	7
3	105	a	X
2	104	b	X	8
1	103	l	X	1
	102	e	X	1

(tc/anti-join ds1 ds2 :b)

anti-join [5 3]:

:b	:a	:c
108		c
107		a
105	3	t
102	2	b
101	1	a

I would expect

:b	:a	:c
101	1	a

because values 102, ..., 108 are all found in column :b of dataset ds2.

Or am I misunderstanding the nature of an anti-join?

[genmeblog]

Oh, thanks, I will verify this today or tomorrow.

[genmeblog]

Yes, I can confirm, it's wrong (also semi-join). I don't know why I haven't verified it from the very beginning :/

Anyway, I'll prepare a patch soon.

[genmeblog]

Fixed in 6.094.1, verify against dplyr (+ created tests)