add math.h header that we missed from one of the previous Pd Vanilla port commits
fix two format specifiers that were the wrong type
commit 06cbbf40ef0f89cfb17462f66eaf86da172a3033

commit 4ec6b6437261342b979f5f818d33f5141cb91386

(bug report https://sourceforge.net/p/pure-data/bugs/158/)
commit e454201dc1dfd773813dc9274abb1786f965af06

and stopped while spooling to disk)
https://sourceforge.net/p/pure-data/bugs/1077/
commit 060a5d945c6fb70b4d82040665b53a4abc6beb5c

tabread4~ needs at least 4 samples to interpolate;
this patch adds guards in order to not access out-of-bound indices in the table,
in case the table is smaller than 4 samples.
commit e667894370ec22bc1d1d844443d90f1a76adc03e

if the entire soundfile fits into the fifo, then [readsf~] will start
to loop the file, since the 'drained' test in d_soundfile.c:1937 does
not trigger.
the fix is to properly set the fifohead (in the read thread) to the new
position, even if the read() returned '0' (indicating EOF)

commit 057cb3ee6cc58597a1b74a00a7880c2cfd0a466a

tabread4~ needs at least 4 samples to interpolate;
this patch adds guards in order to not access out-of-bound indices in the table,
in case the table is smaller than 4 samples.

commit e667894370ec22bc1d1d844443d90f1a76adc03e

commit 420bd1fffa0336650309d4c158c6dd208d993310

port from Pd Vanilla: aiff (and maybe wav) file reading bugfix: odd-sized chunks are padded up to even number of bytes
commit a567e4398d14901840f24ac82ecfc79306e22903

commit dc79deff7b8e89edc4e97557e7b76c0afce0fea3

Pd Vanilla commit 98b4959f03f0fd9acfd7c117d8061589504c89e6

This reverts commit 16973649, reversing
changes made to 15094e4c.

Conflicts:
	pd/src/m_pd.h

fix error where the "$@" variable was overwriting the bounds variable for the loop used to check for its existence

fix a very nasty utf8 code-point counter/offsetter that can easily dereferencing garbage thanks to Pd's use of non-null-terminated strings

fix edge case where menu was appearing on top of the dropdown button
more commenting of oddities

guard against Chromium's counter-intuitive mouseenter/mouseover/mousemove events that get generated when a scroll event moves the element under the mouse
Also, use "mousemove" instead of "mouseover" event as it properly highlights
the element under the mouse even for single pixel mouse movements. "mouseover"
would fail to highlight if the keyboard navigation had moved the highlight
to a different element within view.
These changes fix the problem of generating a chain of mouse events that
would mess up keyboard navigation when the mouse pointer happened to be over
the <ol>. These changes make the interaction between mouse and keyboard much
more like what's expected in most GUI toolkits. (The only difference now is
that the menu can't overflow the browser window, but there's nothing we can
do about that.)

height from the maximum menu height

strings

Don't multiply requested memory size by sizeof(char*)