+ jk4^RI5^RI5^RItRtRtRtR#))*Nc^\n\^4R,R8XgQh\^@4R,^8XgQh\^@4R,R8XgQh\^4R,R8XgQh\^ 4R,PR4'gQh\^ 4R,PR4'gQh\^ 4R,PR4'gQh\^ 4R,PR 4'gQh\^ 4R ,PR 4'gQh\^ 4R,PR 4'gQhR #)g@ig?g?gP?gR[:XL @gXs@gIHb=?g3i?g333333?g[-?gRB3Ϲ ?Nggg333333ӿ)mpdpsmpfaeX/home/luca/screen-checking/.venv/lib/python3.14/site-packages/mpmath/tests/test_power.pytest_fractional_powr s! BF r7c>T !! ! r7c>Q   r7d?e ## # r7d?l ** * GsN  1 2 22 2 GsN  0 1 11 1 GtO   3 4 44 4 GtO   5 6 66 6 GsN  1 2 22 2 GtO   3 4 44 4r c \P!R4RFp\^24Fp\P!^V^, ,^V,4p\P!^^d4pW#,p\ \ \ V4\ V4V\44V8gQh\ \ \ V4\ V4V\44V8dKQh K R#)zF Test that inexact integer powers are rounded in the right direction. iN) 5) randomseedrangerandintto_intmpf_powfrom_int round_downround_up)preciababs r test_pow_integer_directionr s  KKrAq46{AtG4Aq#&AB'(1+x{D*MNQSS SS'(1+x{D(KLrQ QQ r c RpREFwrV\n\^4\^4V)^,,,p\^4\^4V)^,,, pV)pV)pV\n\pV!V^V4^8gQhV!V^V4^8XgQhV!V^V4R8gQhV!V^V4R8XgQhV!V^V4^8gQhV!V^V4^8XgQhV!V^V4^8gQhV!V^V4^8XgQhV!VRV4^8XgQhV!VRV4^8gQhV!VRV4R8XgQhV!VRV4R8gQhV!VRV4^8XgQhV!VRV4^8gQhV!VRV4^8XgQhV!VRV4^8gQh\pV!V^V4^8XgQhV!V^V4^8gQhV!V^V4R8XgQhV!V^V4R8gQhV!V^V4^8XgQhV!V^V4^8gQhV!V^V4^8XgQhV!V^V4^8gQhV!VRV4^8gQhV!VRV4^8XgQhV!VRV4R8gQhV!VRV4R8XgQhV!VRV4^8gQhV!VRV4^8XgQhV!VRV4^8gQhV!VRV4^8XgQh\ pV!V^V4^8gQhV!V^V4^8XgQhV!V^V4R8XgQhV!V^V4R8gQhV!V^V4^8gQhV!V^V4^8XgQhV!V^V4^8gQhV!V^V4^8XgQhV!VRV4^8XgQhV!VRV4^8gQhV!VRV4R8gQhV!VRV4R8XgQhV!VRV4^8XgQhV!VRV4^8gQhV!VRV4^8XgQhV!VRV4^8gQh\ pV!V^V4^8XgQhV!V^V4^8gQhV!V^V4R8gQhV!V^V4R8XgQhV!V^V4^8XgQhV!V^V4^8gQhV!V^V4^8XgQhV!V^V4^8gQhV!VRV4^8gQhV!VRV4^8XgQhV!VRV4R8XgQhV!VRV4R8gQhV!VRV4^8gQhV!VRV4^8XgQhV!VRV4^8gQhV!VRV4^8XdEKQh ^\nR#)aQ Stress test directed rounding for powers with integer exponents. Basically, we look at the following cases: >>> 1.0001 ** -5 # doctest: +SKIP 0.99950014996500702 >>> 0.9999 ** -5 # doctest: +SKIP 1.000500150035007 >>> (-1.0001) ** -5 # doctest: +SKIP -0.99950014996500702 >>> (-0.9999) ** -5 # doctest: +SKIP -1.000500150035007 >>> 1.0001 ** -6 # doctest: +SKIP 0.99940020994401269 >>> 0.9999 ** -6 # doctest: +SKIP 1.0006002100560125 >>> (-1.0001) ** -6 # doctest: +SKIP 0.99940020994401269 >>> (-0.9999) ** -6 # doctest: +SKIP 1.0006002100560125 etc. We run the tests with values a very small epsilon away from 1: small enough that the result is indistinguishable from 1 when rounded to nearest at the output precision. We check that the result is not erroneously rounded to 1 in cases where the rounding should be done strictly away from 1. c`\\VPV\PV44#)N)make_mpf mpf_pow_int_mpf_rr)xnrs&&&r powr'test_pow_epsilon_rounding..powrCs  AGGQ;<r:s  5 R yr