cosmos-sdk/math/CHANGELOG.md

Changelog

[Unreleased]

• #11783 feat(math): Upstream GDA based decimal type

math/v1.4.0 - 2024-01-20

Features

• #20034 Significantly speedup LegacyDec.QuoTruncate and LegacyDec.QuoRoundUp.

Bug fixes

• Fix ASA-2024-010: Math Bit length differences between Int and Dec

math/v1.3.0 - 2024-02-22

Features

• #18552 Add safe arithmetic operations for math.Int that return an error in case of an overflow or any mishap.
• #18421 Add mutative api for LegacyDec.BigInt().
• #18874 Speedup math.Int.Mul by removing a duplicate overflow check
• #19386 Speedup math.Int overflow checks
• #19466 Speedup math.NewLegacyDec by one heap allocation
• #19467 Slightly speedup math.LegacyDec Ceil and MarshalTo methods
• #19479 Speedup math.LegacyDec functions that involve math.Int by removing a heap allocation. (Ceil, TruncateInt, NewLegacyDecFromInt)

Bug Fixes

• #18519 Prevent Overflow in Dec.Ceil().

math/v1.2.0 - 2023-11-07

Features

• #18247 Add mutative api for Uint.BigInt().
• #17803 Add mutative api for Int.BigInt().
• #18030 Add mutative api for NewIntFromBigInt.

Bug Fixes

• #18228 Fix panic when calling BigInt() on an uninitialized Uint.
• #18214 Ensure that modifying the argument to NewUIntFromBigInt doesn't mutate the returned value.
• #18211 RelativePow now returns 1 when 0^0, before it was returning the scale factor.
• #17725 Fix state break in ApproxRoot. This has been present since math/v1.0.1. It changed the rounding behavior at precision end in an intermediary division from banker's to truncation. The truncation occurs from binary right shift in the case of square roots. The change is now reverted back to banker's rounding universally for any root.

math/v1.1.2 - 2023-08-21

Bug Fixes

• #17489 Revert #16263.

math/v1.1.1 - 2023-08-21

Bug Fixes

• #17480 Fix panic when calling .Size() on a nil math.Int value.

math/v1.1.0 - 2023-08-19

Features

• #17427 Implement LegacyDec.MulRoundUp that rounds up at precision end.

Improvements

• #17109 Add .ToLegacyDec() method on math.Int type for converting to math.LegacyDec.
• #16263 Improved math/Int.Size by computing the decimal digits count instead of firstly invoking .Marshal() then checking the length

Bug Fixes

• #17352 Ensure that modifying the argument to NewIntFromBigInt doesn't mutate the returned value.
• #16266 Fix legacy dec power mut zero exponent precision.

math/v1.0.1 - 2023-05-15

Improvements

• #15768 Removed the second call to the init method for the global variable grand.
• #16141 Speedup LegacyDec.ApproxRoot and LegacyDec.ApproxSqrt.

Bug Fixes

• #15714 FormatInt returns an error on empty string.

math/v1.0.0 - 2023-03-23

Bug Fixes

• #15506 Dec marshal shouldn't have side effects

math/v1.0.0-rc.0 - 2023-03-13

Features

• #15043 add rand funcs to math

Bug Fixes

• #14922 check for negative precision

Testing

• #15215 fix FormatDec test

math/v1.0.0-beta.6 - 2023-02-06

Features

• #14760 add collections key encoders and value encoders for common types.
• #14166 math: add generics versions of Max, Min to cater to all numeric types
• #13381 add uint IsNil method

Improvements

• #14010 math: optimize and test FormatInt + simplify LegacyNewDecFromStr
• #12794 math: precompute & use square of precisionReuse instead of 2 repeated computations

Bug Fixes

• #14691 do not flatten events attributes by event types
• #14252 math: add LegacyNewDecFromStr fuzzers + remove unnecessary error wrapping

Testing

• #14576 Added test cases for precisionMultiplier

math/v1.0.0-beta.3 - 2022-07-20

Bug Fixes

• #11996 math: fix Uint.Unmarshal's lack of negative value checking