You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This commit was created on GitHub.com and signed with GitHub’s verified signature.
0.10.0 (2024-03-31)
Bug Fix Fix broken CLRG
broke when 'REGS' was replaced by 'RPN83REG'
Bug Fix Parse floating numbers equivalent to 0.0 more correctly.
The canonical internal representation of 0.0 in TI-OS has an exponent
value of $80 (i.e. 0), with all mantissa digits set to 0.
The previous code set the mantissa digits correctly, but incorrectly
set the exponent to $7F or some other value depending on the
position of the decimal point relative to the 0 digits.
It made almost no difference because various floating point routines
seem to canonicalize the exponent to the correct $80 before
continuing.
However, in an upcoming feature, the validation CkPosInt() is called
before canonicalization can take place, the CkPosInt() returns an
incorrect result.
The fix correctly detects all variations of a 0.0 (e.g. an empty
string "", "0.0", "000.0", "-000.000E1", "00.00E0") and correctly
returns the canonical representation of 0.0 which works with CkPosInt().
Bug Fix Validate data type for STAT functions
allow only Real numbers for Sigma+ and Sigma-
Bug Fix Validate data type for TVM functions
allow only Real numbers for TVM functions
Bug Fix Validate data type for % function
allow only Real numbers for X and Y
Bug Fix Validate data type when storing to ANS upon app exit
allow only Real or Complex numbers to stored to ANS
Bug Fix Support real arguments for CPLX menu functions (REAL, IMAG, CONJ, CABS, CANG)