Documentation is not correct about behaviour of recurse builtin

[dakusui]

Describe the bug
jq-1.6 documentation says,

recurse(f) is identical to recurse(f; . != null) and can be used without concerns about recursion depth.

however, from jq-1.6, the filter for null was abolished, which is good and consistent, I believe.

To Reproduce

Just see the documentations and search for definition for the one parameter recurse function.

• https://stedolan.github.io/jq/manual/v1.6/#Builtinoperatorsandfunctions
• https://stedolan.github.io/jq/manual/#Builtinoperatorsandfunctions

Expected behavior

The description should be like following

recurse(f) is identical to recurse(f; true) and can be used without concerns about recursion depth.

since recurse(f) is defined def recurse(f): def r: ., (f | r); r; in builtin.jq

Environment (please complete the following information):

• ALL
• jq version : 1.6 or leter

Additional context
None

[dakusui]

Hmm,
Judging from the behavior of jq-1.6, I think I was wrong.
But strangely, the definition of the recurse/1 function has changed not to handle null in a special way...
Maybe there's a logic that I don't know...

[dakusui]

I think I misunderstood the behavour of jq-1.6.
Please mark this ticket "invalid".

[nicowilliams]

I don't quite understand what the documentation says about this either...

[dakusui]

To me, the statement looks trying to justify the jq's behavior where nulls are ignored by recurse built-in and therefore also ignored by paths/1 builtin, which internally uses it...

From the commit you made 3 years ago at this line,

https://github.com/stedolan/jq/blame/master/src/builtin.jq#L18

I thought they (and you) intended to abolish such an inconsistent behavior and treat nulls as others are, but on my local, it's showing a following behavior and puzzling me.

$ jq --version
jq-1.6
$ echo '{"k":null}' | jq -r -c 'paths(scalars)'
$ echo '{"k":null}' | jq -r -c 'paths'
["k"]
$

[dakusui]

Re-opening this issue since something in documentation should be improved apparently.

[TWiStErRob]

Bump, this is really confusing on an otherwise already advanced topic.

Ref https://unix.stackexchange.com/a/654730/23398