Decimal floating point number to binary: Difference between revisions

Content added Content deleted
m (→‎{{header|Phix}}: simplified output)
m (→‎{{header|Phix}}: added limiter)
Line 1,081: Line 1,081:
if d>0 then
if d>0 then
res &= '.'
res &= '.'
while d>0 do
-- while d>0 do -- (see update below)
while d>0
and (base=2 or length(res)<15) do
d *= base
d *= base
integer digit = floor(d)
integer digit = floor(d)
Line 1,114: Line 1,116:
procedure test(atom f, integer base=2)
procedure test(atom f, integer base=2)
string s = dec_to(f,base)
string s = dec_to(f,base)
atom g = to_dec(s,base)
atom g = to_dec(s,base),
h = f-g
⚫
printf(1,"%.8g => 0(%d):%s => %.8g\n", {f,base,s,g})
string e = iff(h=0?"":sprintf(" (error: %g)",h))
⚫
printf(1,"%.8g => 0(%d):%s => %.8g%s\n", {f,base,s,g,e})
end procedure
end procedure
test(23.34375)
test(23.34375)
Line 1,141: Line 1,145:
0 => 0(2):0 => 0
0 => 0(2):0 => 0
65535 => 0(16):FFFF => 65535
65535 => 0(16):FFFF => 65535
23.7 => 0(35):N.OHHHHHHHHFIVE => 23.7 (error: -3.55271e-15)
23.7
23.7
"23.699999999999"
"23.699999999999999289457264239899814128875732421875"
</pre>
</pre>
Aside: I was quite surprised to get 100% accuracy on these tests, but actually it is more of
Aside: I was quite surprised to get 100% accuracy on these tests, but actually it is more of
a lucky coincidence in the way it is written, as the last test shows. The truth of the matter
a lucky coincidence in the way it is written, as the last test shows. The truth of the matter
is simply that you ''can'' extract a float to a binary text representation exactly, in a way that
is simply that you ''can'' extract a float to a binary text representation exactly, in a way that
you just cannot do for most other (ie non-power-2) bases.
you just cannot do for most other (ie non-power-2) bases.<br>
Update: Added a limiter for non-base-2 fractions, as per 1/3 -> 0.333 forever in decimal. Base 2 is guaranteed to
terminate anyway, but for other bases we need some limit - the 15 I opted for is completely arbitrary.


=={{header|Python}}==
=={{header|Python}}==
Retrieved from "https://rosettacode.org/wiki/Decimal_floating_point_number_to_binary"