Re: Floating point improvements

[MOON@MIT-AI (Sent by MOON0@MIT-AI)]

UCADR 434 fixes bugs in floating-point divide and bignum remainder.
These bug fixes fix a bug in floating-point read.  I made a few cosmetic
changes to read which I don't think fixed any bugs.

I have extensively revised the floating-point printer.  It no longer has
an array, which prevented it from working recursively and reentrantly,
it has bugs fixed, and it prints flonums with full precision instead of
small-flonum precision.  The necessary routines are there to allow
~E and ~F in FORMAT to be implemented easily.  Note: this version of PRINT
will not work in microcode older than 434, due to above-mentioned bugs.