Lars Ingebrigtsen <larsi@gnus.org> writes:

> "Drew Adams" <drew.adams@oracle.com> writes:
>
>> Actually, the real bug is that `dired-get-subdir-min' has no doc string, so the
>> doc string of `cdr' is used in its stead.  The whole point of defining and using
>> a separate function for this, instead of just coding with `cdr', is to enable
>> readers of the code - and a fortiori the doc - to think in terms of the data
>> structure of directories and not just abstract cons cells.  There should be doc
>> at the proper level of abstraction here.
[...]
> ;; can't use macro,  must be redefinable for other alist format in dired-nstd.
> (defalias 'dired-get-subdir-min 'cdr)
>
> Which is kinda er.
>
> I don't know what nstd is, and the function doesn't seem to be redefined
> anywhere, so is this just legacy gunk?  Should `dired-get-subdir-min'
> just be replaced by `cdr' in all the calling sites?

From searching the web, I think dired-nstd is "dired nested", which
indeed seems to be some old, now non-existent stuff from Emacs 19 or
older.  See this for example:
https://www.cs.cmu.edu/cgi-bin/info2www?(dired.info)Dired%20Known%20Problems

I therefore think it's safe to obsolete this function, like in the below
patch.  Comments?

Best regards,
Stefan Kangas