Re: bug/feature - emacs doesn't tell you about keys with multiple prefixes

[Richard Stallman]

The new code has a bug: it finds just the last occurrence of the same
keymap and tests whether the prefix for that occurrence is an initial
segment.  What if the last occurrence isn't but another occurrence is?
In that case, it won't detect the cycle.  I suspect that when M-a,
M-b, and M-a M-b are the same keymap, this code won't compare the M-a
with the M-a M-b.

It can't use rassq now; it has to check each previous occurrence of the same
keymap.