Re: syntax-ppss bug?

[Richard Stallman]

    > I tracked down the problem below to an infloop in
    > python-beginning-of-statement which occurs because syntax-ppss returns -1
    > as the depth.  I think that is the wrong value.  Would you
    > please investigate?

    Actually, parse-partial-sexp returns the same -1.  And if you close more
    parens, you can get to -2, -3, ... it seems it's done on purpose.

The job of parse-partial-sexp is to describe the changes from point A
to point B.  When it scans from a place deep within parens into a
place less deep, it should return a negative value.  That value is
accurate as an answer--to this question.

However, syntax-ppss is supposed to describe the status with respect
to top level.  If the depth value is negative, it means that something
is wrong.  Perhaps this is only possible when the code is nonsensical.

Can you document this, in the docs of syntax-ppss in the Lisp manual?