Named and unnamed operators

[jkeenan]

As part of my research on Fatalization of currently deprecated goto
construct I have been
re-reading our current documentation of goto in blead and have been trying
to locate in our test suite examples of each claim made in that documentation.
What we find -- or fail to find -- there will have an impact on the
documentation updates provided as part of a pull request like the current
#23782. I've posted the preliminary results
of that documentation review
here.

This is the fifth in a series of GH issues addressing the items which I
believe are lacking in that documentation.

perldoc -f goto states:

Also, unlike most named operators, [`goto LABEL`] has
the same precedence as assignment.

"[N]amed operators" implies the existence of unnamed operators. Do we have
any? Where are they discussed?

How do we test precedence?

[mauke]

The opposite of named operators (like length or sort) are symbolic operators (like ++ or >). See also the discussion in https://perldoc.perl.org/perlop#Terms-and-List-Operators-(Leftward).

[jkeenan]

The opposite of named operators (like length or sort) are symbolic operators ...

In that case, I think there should be some place in our docs where we describe unnamed or symbolic operators.

$ ack -i '(?<=un)?named\soperator' .
pod/perlop.pod
521:but, because C<"*"> is higher precedence than named operators:

pod/perlfunc.pod
105:functions, like some keywords and named operators)
2120:Unlike most named operators, this has the same precedence as assignment.
3726:Also, unlike most named operators, this has the same precedence as
4140:Unlike most named operators, this has the same precedence as assignment.
4745:Unlike most named operators, this has the same precedence as assignment.
6951:Unlike most named operators, this has the same precedence as assignment.
7465:Unlike most named operators, this is also exempt from the

toke.c
9035:    /* Check for plugged-in named operator */
[perl] 2304 $ ack -i '(symbolic|unnamed)\soperator' .
toke.c
9138:    /* Check for plugged-in symbolic operator */

I didn't find the string unnamed operator anywhere, and I found the string symbolic operator only in a comment deep inside a guts file.