Padded exponent - #644

Closed

BebeSparkelSparkel wants to merge 34 commits into

haskell:masterfrom

BebeSparkelSparkel:padded-exponent

Commits on Jan 9, 2024

improved RealFloat benchmarks
BebeSparkelSparkel
committed
332bcf6
better names for RealFloat tests
BebeSparkelSparkel
committed
36241cc

Commits on Jan 12, 2024

averaged realfloat specal strings instead of checking each special value since order changes the bench mark
BebeSparkelSparkel
committed
b3a2276
improved test and add bench for small doubles
BebeSparkelSparkel
committed
9a86e45

Commits on Jan 15, 2024

improved tests for FStandard
BebeSparkelSparkel
committed
a0998bf
added better labels to real float tests
BebeSparkelSparkel
committed
ce7d20e
differencated special values from basic values and Float from Double
BebeSparkelSparkel
committed
b98fc5d

Commits on Jan 16, 2024

put float test in the correct group
BebeSparkelSparkel
committed
edd64cd
combined FloatFormat and FormatMode
BebeSparkelSparkel
committed
b20d7a6
customized FGeneric exponent range
BebeSparkelSparkel
committed
b22b6b3
FScientific now has a selectable case E
BebeSparkelSparkel
committed
37e8d22
generaized FloatingDecimal and intermediate
BebeSparkelSparkel
committed
1b16c67
generailized decimalLength
BebeSparkelSparkel
committed
d3cdedc
generalized mantissa to Word64
BebeSparkelSparkel
committed
e11b303
generalized f2s and d2s
BebeSparkelSparkel
committed
aa15ac2
added formatFloating which combines the logic of formatFlat and formatDouble
BebeSparkelSparkel
committed
053f87a
added SpecialStrings for scientific non-normal float values
BebeSparkelSparkel
committed
40f90a4
added SpecialStrings to standard floating point non-normal values
BebeSparkelSparkel
committed
b1e3e30
RealFloat optimizations
BebeSparkelSparkel
committed
945916b
generalized breakdown
BebeSparkelSparkel
committed
08c5050
added some INLINABLE to RealFloat.Internal
BebeSparkelSparkel
committed
d87507b
toCharsNonNumbersAndZero now accepts the sign, mantissa, and exponent parameters and returns a maybe
BebeSparkelSparkel
committed
0ea0a35
toCharsNonNumberAndZero now takes the float and only uses bit operations to determine special string
BebeSparkelSparkel
committed
e181e2a
removed f2s d2s
BebeSparkelSparkel
committed
cbeeef8
removed f2s f2s' f2Intermediate
BebeSparkelSparkel
committed
bf287a5
removed specialStr and replaced with improved version of toCharsNonNumbersAndZero
BebeSparkelSparkel
committed
f399638
clean up
BebeSparkelSparkel
committed
d9ebd68
cleaned up function formatFloating
BebeSparkelSparkel
committed
bd2b685
fixed precison printing of zero and neg zero for FStandard
BebeSparkelSparkel
committed
a9cbf58
labels for RealFloat format parameters
BebeSparkelSparkel
committed
bbd4f76
fix possible overflow error when converting String to Builder
BebeSparkelSparkel
committed
a1e556e
specialized maxEncodeLength to Float and Double
BebeSparkelSparkel
committed
182d76f
moved FloatFormat to Internal so that it can be exported and users can manipulate it beyond the regular formants
BebeSparkelSparkel
committed
8479796

Commits on Jan 17, 2024

added zero padded exponent for scientific floating builder
BebeSparkelSparkel
committed
a2c7324