RE: feature request: indicator of minibuffer-recursion depth

["Drew Adams" <drew.adams@oracle.com>]

    > However, personally I would prefer not to burden the prompt (which can
    > already be quite complex) and instead put the indicator in
    > the mode line somehow.

    At startup my Emacs has 1 frame, with 1 (mini) window, and 0 mode-lines.
    Also, even after I've opened a few other frames, there's still only ever
    1 miniwindow and it has no mode-line next to it (and all the frames with
    their mode lines may be iconified or displayed on some other display).

Well, then you have the same problem already with recursive edits, no? And
the same problem with other stuff that Emacs puts into the mode line.
Obviously, if someone chooses not to show mode lines, then mode line stuff
is, well, not visible ;-).

A minibuffer recursion-depth indicator (which you seem to favor, in general)
would be a new feature. If you've gotten by so far without any such
indication, and you often do without similar stuff (edit recursion-depth
indicator) that is already in the mode line, then perhaps you'll not miss
it?

My guess is that most people do show mode lines most - nay all - of the
time. I could be wrong.

    I.e. putting the info on the mode line is not a good idea.

for those people who don't display mode lines.

Where would you put the info? So far, the mode line and the minibuffer have
been suggested.

Here's another idea: Use the minibuffer window fringe as a recursion-depth
indicator somehow. (That's probably not a great idea; I'm not sure what uses
the fringe might already serve in the minibuffer, or what uses it might
better serve there in the future.)

To me, it makes sense to put the two recursion-depth indicators in the same
place (mode line).  You would see things like [[[...]]] for recursive edits
and things like {{...}} for recursive minibuffers. If, for some reason, you
entered a recursive minibuffer from a recursive edit, the indicator would be
[{...}]. The opposite order of events would be indicated by {[...]}.