Raymond Toy wrote:
> Viktor T. Toth wrote:
>
>> c. $RTEST_ROMBERG, problem: (error break), and
>> share/contrib/stringproc/rtestprintf.mac, problems: (7 34 35) under CMUCL
>> 19c
>>
>
> I get many more failures with CMUCL. But I think most of them are due
> to different expectations. For example, some tests expect results like
> 1.2E+3, but CMUCL uses a lower case e. Some are questionable because of
> the number of digits printed.
>
FWIW, current CMUCL fails on tests 29, 38, 39, 40, 47, and 48.
Prob 29 fails because CMUCL prints "/hot .../hamburger .../ice" instead
of "/hot .../hamburger". I think CMUCL is correct here. I think it
should print out 3 items, not 2.
Prob 38 fails because CMUCL uses "e" instead of "E" as the exponent
marker, and because it prints pi with one less digit than the expected
value.
Prob 39 fails because pi has one less digit than expected.
Prob 40 fails because CMUCL prints out "42. " (4 spaces) instead of
"42.". I'm pretty sure the CLHS says ~g must print the answer as
"42. ", so the expected result is wrong.
Prob 47 fails because 1.985 isn't rounded to 1.98 as expected. I can't
tell from the CLHS if the round to even is supposed to happen with the
printed value or the actual value. 1.985 doesn't have an exact
single-float representation, and the nearest value is slightly above 1.985.
I can't quite figure out from the CLHS the expected value for Prob 48.
Ray