Bug fixes:

• Fix test_hexadecimal_with_libc_bulk(), see #1049 (Doug Torrance)
• Keep available deprecated aliases for mpc/mpf_log() (Sergey B Kirpichev)
• Use version_file option of setuptools-scm to keep version info, see #1048
  (Sergey B Kirpichev)
• Add workaround for test on s390x, see #1061 (Sergey B Kirpichev)

Full Changelog: 1.4.0...1.4.1

Features:

• Support underscores as digit separators per PEP 515, see #661 (Sergey B Kirpichev)
• Add rationals converter for mpf's, see #666 (Sergey B Kirpichev)
• Rewrite bernpoly/eulerpoly to avoid dependency on bernoulli(1) convention, see #700 (Sergey B Kirpichev)
• Support base kwarg for from_str(), see #703 (Jonathan Warner, Sergey B Kirpichev)
• Support randmatrix() for mp.iv and mp contexts, see #527 (Maximilian Gaukler)
• Added rank() function for matrices, see #610 (Jan-Philipp Hoffmann)
• Add plus flag to select the B_1 sign convention for bernoulli/bernfrac, see #724 (Jeremy Tan Jie Rui, Sergey B Kirpichev)
• Add mpf.as_integer_ratio() method, support construction of mpf from Decimal objects, see #731 (Sergey B Kirpichev)
• Expose lower/upper_gamma functions, see #740 (Sergey B Kirpichev)
• Support mpc initialization from string, see #743 (Sergey B Kirpichev)
• Support asinh/acosh/atanh in the fp context, see #750 (Sergey B Kirpichev)
• Support binary/octal/hexadecimal string output, see #711 (Jonathan Warner, Sergey B Kirpichev)
• Support pickling for matrices and mpi, see #761 (Sergey B Kirpichev)
• Support matrix.__array__() dunder method, see #767 (Sergey B Kirpichev)
• Support more number syntaxes, see #778 (Sergey B Kirpichev)
• Run mpmath as a module for interactive work, see #773, #923, #931, #936, #939 and #954 (Sergey B Kirpichev)
• Add signed option to to_man_exp(), see #783 (Sergey B Kirpichev)
• Add fp.hypot, see #798 (Sergey B Kirpichev)
• Support inf/nan's in ctx.almosteq(), #802 (Sergey B Kirpichev)
• Implement mpf.__format__(), see #819, #831, #850, #859, #857, #862, #881, #944 and #966 (Javier Garcia, Sergey B Kirpichev)
• Support conversion from scalar ndarray's, see #821 (Sergey B Kirpichev)
• Support rounding modes in mpf.__format__, see #823, #831, #834 and #969 (Javier Garcia, Sergey B Kirpichev)
• Support '%' presentation type for mpf, see #847 (Sergey B Kirpichev)
• Support gmpy2-like rounding modes in to_str(), see #830 (Javier Garcia)
• Implement 'a'/'A' formating types for mpf.__format__, see #841 and #870 (Sergey B Kirpichev)
• Add mpc.__format__(), see #855 (Sergey B Kirpichev)
• Now mpf.__round__() returns mpf, see #826 and #966 (Sergey B Kirpichev)
• Support 'b' (binary) format type for mpf/mpc, see #867 (Sergey B Kirpichev)
• Implement mpf.__floordiv__() and mpf.__divmod__(), see #873 (Sergey B Kirpichev)
• Add parameters for MPContext constructor, see #876 and #963 (Sergey B Kirpichev)
• Add MPFR-compatible aliases for rounding modes, see #892 (Sergey B Kirpichev)
• Support negative indexes in matrix, see #897 (Riccardo Orsi)
• Better introspection support for decorated functions, see #900 (Sergey B Kirpichev)
• Add moving sofa demo, see #924 (Sergey B Kirpichev)
• Support spherical Bessel functions (jn/yn), #935 (Sergey B Kirpichev)
• Add pretty_dps context property to control number of printed digits, see #933 (Sergey B Kirpichev)
• Support thousands separators for formatting of fractional part, see #925 and #936 (Sergey B Kirpichev)
• Use PyREPL, as fallback (no IPython), see #941 (Sergey B Kirpichev)
• Add exp2() and log2(), see #948 (Sergey B Kirpichev)
• Support rounding property for the mp context, see #963 (Sergey B Kirpichev)
• Add Fox H-function with rational A/B parameters (foxh()), see #982 (Hongren Zheng)
• Provide experimental support for free-threading builds, see #993 (Sergey B Kirpichev)

Compatibility:

• Drop Python 2 support, see #629 (Fangchen Li)
• Drop support for Python versions < 3.9, see #675 and #911 (Sergey B Kirpichev)
• Drop private mpq class, use Rational's, provided by backend, see #691 and #769 (Sergey B Kirpichev)
• Drop to_pickable()/from_pickable() helpers, see #667 and #769 (Sergey B Kirpichev)
• Direct access to _mpf_ tuples now deprecated, please use from/to_man_exp() functions and special constants (finf, fninf and fnan), see #783 (Sergey B Kirpichev)
• Removed sage backend, see #732 (Sergey B Kirpichev)
• Drop MPMATH_STRICT environment variable, see #759 (Sergey B Kirpichev)
• Deprecate current (descending) order of coefficients in polyval(), etc, see #779, #844 and #845 (Sergey B Kirpichev, Warren Weckesser)
• Deprecate mpmath.math2, see #769 (Sergey B Kirpichev)
• Importing from the mpmath.libmp submodules is deprecated, use instead from mpmath.libmp import foo, see issue #704 (comment) for available functions (Sergey B Kirpichev)
• Deprecate bitcount function, see #721 and #955 (Sergey B Kirpichev)
• Deprecate mpf/mpc_log, see #989 (Sergey B Kirpichev)
• Deprecate fp.is_special(), see #1042 (Sergey B Kirpichev)

Bug fixes:

• sum_accurately(), betainc() and power() fixes, see #664 (Sergey B Kirpichev)
• Warn users about Python's true division, see #670 (Sergey B Kirpichev)
• Propagate nan's in ei_asympt(), see #672 (Sergey B Kirpichev)
• Fix matrix.__eq__, fix string parsing with underscores, see #679 (Sergey B Kirpichev)
• Raise IndexError if matrix index out of bounds, see #689 (Sergey B Kirpichev)
• Fix nan handling in fp.mag() and hyper(), see #688 (Sergey B Kirpichev)
• Optimize sparse matrix dot product, see #450 (Tyler Chen)
• Correct pow() for mpf's to be consistent with mpfr/float's, see #690 (Sergey B Kirpichev)
• Improve hypsum non-convergence behaviour, see #703 (Benjamin Fischer)
• Fixed TypeError in LU_comp, see #610 (Jan-Philipp Hoffmann)
• Fix disagreement fp.mod vs mp.mod, see #710 (Sergey B Kirpichev)
• Skip eigenvectors if left=right=False for one-dimentional matrix, see #713 (Sergey B Kirpichev)
• Fix mpc() constructor to be compatible with complex(), fix disagreement fp.pow vs mp.pow, see #731 (Sergey B Kirpichev)
• Add derivative keyword argument for besselk(), see #735 (Sergey B Kirpichev)
• Fix quadosc(), fast case in _hyp2f1(), correct fp.gamma() and fp.isnprint(), see #740 (Sergey B Kirpichev)
• Pevent erroneous setting of dps/prec on mpmath module, see #678 (Colin B. Macdonald)
• Update splot() for recent matplotlib, see #747 (Sergey B Kirpichev)
• acos_asin(): don't try to normalize special numbers, fix repr(mp.eps), see #750 (Sergey B Kirpichev)
• Fix choleksy_solve for complex matrix, see #755 (Qiming Sun)
• Fix hang in polylog_general(), fix demos, use cholesky_solve() for overdetermined complex linear systems, fix several issues with empty matrices, see #759
• sqrt(z): special case for infinite z.imag, see #777 (Sergey B Kirpichev)
• Correct atan2(±inf, ±inf), see #775 (Sergey B Kirpichev)
• Handle infinite arguments in tan/tanh, see #785 (Sergey B Kirpichev)
• Drop mpf.__complex__(), workaround 1/z division for acot/asec/acsc/acoth, see #797 (Sergey B Kirpichev)
• Handle more special cases in besselk and hyp1f1, see #792, #800 and #801 (Sergey B Kirpichev)
• Correct asin/acos for infinite arguments, see #795 (Sergey B Kirpichev)
• Reduce memory usage for QuadratureRule cache, see #812 (Sergey B Kirpichev)
• Drop sign from nstr(mpf('inf')) output, see #828 (Sergey B Kirpichev)
• Improve accuracy of log1p(), see #803 and #854 (Tim Peters, Sergey B Kirpichev)
• Normalize mpf in mp.npconvert(), fix special cases in bernpoly(), see #839 (Sergey B Kirpichev)
• Fix mpf_div() for prec=0, see #849 (Sergey B Kirpichev)
• Raise ValueError for complex infininity condition in zeta(s, a), see #864 (Sergey B Kirpichev)
• Enable trap_complex for MDNewton, see #870 (Sergey B Kirpichev)
• Use correct mixed-mode functions in fsub/fdiv, add special cases for infinities in mpc_*div(), see #873 (Sergey B Kirpichev)
• Revert "fix ellippi to return an inf instead of raising an exception", see #875 (Sergey B Kirpichev)
• Reject invalid strings in from_str(), see #886 (Sergey B Kirpichev)
• Use parity formula for besseli, see #889 (Sergey B Kirpichev)
• Special case in ctx.hypsum for infinite z, see #902 (Sergey B Kirpichev)
• Raise an exception if iv's comparison can't be decided, see #903 (Sergey B Kirpichev)
• Add special case for ±inf in polylog_continuation(), see #904, #1034 and #1037 (Sergey B Kirpichev, Colin B. Macdonald)
• Increase working precision in polylog_general() for negative s, see #898 (Sergey B Kirpichev)
• Correct case for integer n in besselj/besseli, see #909 (Sergey B Kirpichev)
• Use sum_accurately() in hankel1/2(), see #926 (Sergey B Kirpichev)
• Ensure mpf_bernoulli() returns normalized answer, see #939 (Sergey B Kirpichev)
• Use mpf_log1p in acos_asin() helper (implementing Hull et al algorithm), see #948 and #1036 (Sergey B Kirpichev)
• Fix kwargs passing in the nstr() for mpc, see #964 (David Walker)
• Fix exception type for int(inf), see #966 (Sergey B Kirpichev)
• Ensure exp, sin, tan, etc have a correct __name__ attribute, see #997 (Warren Weckesser)
• Matrix raise ValueError in case of negative dimensions, see #1004 (Ayush Baranwal)
• Support lists in sinm() and cosm(), see #1003 (Ayush Baranwal)
• Properly handle nan's elliprj(), see #1038 (Sergey B Kirpichev)
• Raise ValueError for logm(0), see #1017 (Ayush Baranwal)
• Fix erf(z) with re(z) of large magnitude, see #1039 (Sergey B Kirpichev)
• Return nan's for polylog(s, nan) or polylog(s, nan+nanj), see #1041 (Sergey B Kirpichev)
• Fix fp.isnormal() for subnormals, see #1042 (Sergey B Kirpichev)

Maintenance:

• Use codecov/coverage-action, see #674 (Sergey B Kirpichev)
• Run CI tests with pytest-xdist, see #685 (Sergey B Kirpichev)
• Add pyproject.toml, depend on flake518, see #684 (Sergey B Kirpichev)
• Port torture.py/extratest_zeta.py/extratest_gamma.py to the pytest framework, see #687 (Sergey B Kirpichev)
• Create CITATION.bib, see #681 (Devharsh Trivedi)
• Avoid using star imports in tests and documentation, see #698 (Sergey B Kirpichev)
• Fix some py2's remnants and remove old gmpy workarounds, see #699 (Sergey B Kirpichev)
• Use math.isqrt in isqrt_python calculations, see #695 (Daiki Tak...

Security issues:

• Fixed ReDOS vulnerability in mpmathify() (CVE-2021-29063) (Vinzent Steinberg)

Features:

• Added quadsubdiv() for numerical integration with adaptive path splitting
  (Fredrik Johansson)
• Added the Cohen algorithm for inverse Laplace transforms
  (Guillermo Navas-Palencia)
• Some speedup of matrix multiplication (Fredrik Johansson)
• Optimizations to Carlson elliptic integrals (Paul Masson)
• Added signal functions (squarew(), trianglew(), sawtoothw(), unit_triangle()
  sigmoidw()) (Nike Dattani, Deyan Mihaylov, Tina Yu)

Bug fixes:

• Correct mpf initialization from tuple for finf and fninf (Sergey B Kirpichev)
• Support QR decomposition for matrices of width 0 and 1 (Clemens Hofreither)
• Fixed some cases where elliprj() gave inaccurate results (Fredrik Johansson)
• Fixed cases where digamma() hangs for complex input (Fredrik Johansson)
• Fixed cases of polylog() with integer-valued parameter with complex type
  (Fredrik Johansson)
• Fixed fp.nsum() with Euler-Maclaurin algorithm (Fredrik Johansson)

Maintenance:

• Dropped support for Python 3.4 (Sergey B Kirpichev)
• Documentation cleanup (Sergey B Kirpichev)
• Removed obsolete files (Sergey B Kirpichev)
• Added options to runtests.py to skip tests and exit on failure
  (Jonathan Warner)