Feature #18135: Introduce Enumerable#detect_only

• Status: Open
• Assignee: -
• Target version: -

[meisel (Michael Eisel)]

It can be important to get the only element for which a block returns true, and to assert that this is the only element as such. For example, this can be a very helpful sanity check when one is working with data that's outside of their control and is not perfectly understood. They may have a guess as to how to get a specific element matching some criteria, but if they used Enumerable#detect might be hiding the fact that they have written an incorrect block and that there's in fact more than one element that matches it. It could also be a parameter on Typically, I'd do it like this:

matches = array.select { |elem| some_method(elem) }
raise if matches.size != 0
match = matches.first

Here, it would be shortened to:

match = array.detect_only { |elem| some_method(elem) }

It could also be a parameter on Enumerable#detect instead of a separate method.

[jeremyevans0 (Jeremy Evans)]

• Tracker changed from Bug to Feature
• Backport deleted (2.6: UNKNOWN, 2.7: UNKNOWN, 3.0: UNKNOWN)

[meisel (Michael Eisel)]

Also, another alternative could be Enumerable#take_only, where it takes the only element from the enumerable or else fails. So instead of:

array.detect_only { |elem| some_method(elem) }

We'd have:

array.select { |elem| some_method(elem) }.take_only

[duerst (Martin Dürst)]

meisel (Michael Eisel) wrote in #note-2:

Also, another alternative could be Enumerable#take_only, where it takes the only element from the enumerable or else fails. So instead of:

array.detect_only { |elem| some_method(elem) }

We'd have:

array.select { |elem| some_method(elem) }.take_only

It's a good idea to separate things out, but this could be highly inefficient (checking and selecting lots of elements only to find out later that there's more than one).

[mame (Yusuke Endoh)]

• Related to Feature #13683: Add strict Enumerable#single added

[mame (Yusuke Endoh)]

Do you mean raise if matches.size != 1 instead of ... != 0?

The idea of take_only is already proposed in #13683.

[meisel (Michael Eisel)]

Yes, I meant ... != 1, thanks. And I think we can close this issue, and I'll add my thoughts to the issue that this dupes.