Feature Request: expose isEqualCollection to cypher

[dan-kez]

See: CollectionUtils#isEqualCollection

This would be very useful to check if two collections are the same. As far as I'm aware this isn't exposed to cypher.

[jexp]

Did you try in cypher [1,2,3] = [1,2,3] ?
Or where doesn't that hold true?

[InverseFalcon]

Unless I'm mistaken, isEqualCollection() ignores ordering, it returns true if all elements in both collections are the same, and the cardinality (# of occurrences) of each element are the same.

In Cypher, [1,2,3] = [3,2,1] returns false. An isEqualCollection() call on these would return true.

I wanted to check if apoc.coll.containsAll(coll, values) already does this (seems it would ignore order but probably miss cardinality checking), but I can't seem to get it to work...I don't know what the return key is to use for YIELD (am I overlooking something simple, or is this a gap in the documentation of how to get a boolean result from APOC procedures?)

[jexp]

So length(c1) = length(c2) AND apoc.coll.containsAll(c1,c2) should do it?

[InverseFalcon]

Oddly enough, containsAll() doesn't seem to be available with apoc-3.1.0.2-all, I get an error if attempting to call it:

There is no procedure with the name apoc.coll.containsAll registered for this database instance. Please ensure you've spelled the procedure name correctly and that the procedure is properly deployed.

However, I don't think this would work, at least where elements of the collection repeat. This would work for [1,2,3] and [3,2,1], but would not work for [1,1,2,3] and [1,2,2,3], containsAll() in either direction would return true, and the length would be the same, but the cardinalities of the elements would be different. The backing collection used for this couldn't be a set. We would need some kind of counted bag collection or similar.

[jexp]

It's now a function, so you can call it in-line. But then isCollectionEqual is rather a disjunct() on sets which also exists as function.

…

On Tue, Dec 13, 2016 at 1:32 AM, InverseFalcon ***@***.***> wrote: Oddly enough, containsAll() doesn't seem to be available with apoc-3.1.0.2-all, I get an error if attempting to call it: There is no procedure with the name apoc.coll.containsAll registered for this database instance. Please ensure you've spelled the procedure name correctly and that the procedure is properly deployed. However, I don't think this would work, at least where elements of the collection repeat. This would work for [1,2,3] and [3,2,1], but would not work for [1,1,2,3] and [1,2,2,3], containsAll() in either direction would return true, and the length would be the same, but the cardinalities of the elements would be different. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#151 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAEHY-C0759wZ9g60JMLGe-fCzWpVXk2ks5rHeergaJpZM4KB0hN> .