perlguts: Fix apidoc lines

[khwilliamson]

The functions have different signatures, so can't be considered one group, due to the constraints of autodoc. This change causes perlintern.pod to now have the correct pod for the affected items.

[jkeenan]

The functions have different signatures, so can't be considered one group, due to the constraints of autodoc. This change causes perlintern.pod to now have the correct pod for the affected items.

In perl-5.40.0 and in blead, when I call perldoc perlintern and scroll down to the subsection on Concurrency, I see:

Concurrency
    "CVf_SLABBED"
    "CvROOT"
    "CvSTART"
        Described in perlguts.

    "CX_CUR"
        Described in perlguts.

              CX_CUR()
...

When I build your branch and do likewise, I see:

Concurrency
    "CVf_SLABBED"
        Described in perlguts.

    "CvROOT"
        Described in perlguts.

            OP *  CvROOT(CV * sv)

    "CvSTART"
        Described in perlguts.

            OP *  CvSTART(CV * sv)

    "CX_CUR"
        Described in perlguts.

              CX_CUR()

    "CXINC"
        Described in perlguts.

...

Is that what you are aiming for in this p.r.?

[khwilliamson]

Yes, though a commit in the pipeline will cut down the textual expansion. The prototypes for all non-trivial calls have been traditionally displayed in perlapi and perlintern even when the details of what happens with those calls are given in another pod, as here, perlguts. I made that choice back when, perhaps without realizing it. But when I did, I left it that way, as this was a way to make sure they did get displayed at least somewhere. Someone coming here first will see them before going off to perlguts. Presumably this is the place to start.

What's happening in this case is that CVf_SLABBED has a trivial prototype -- just its name, so it is omitted. The other two have more complicated ones, so theirs should have been displayed, but weren't, because of the bug this commit fixes. Now, their prototypes do get displayed.