/* mpfr_get_f -- convert a MPFR number to a GNU MPF number Copyright 2005 Free Software Foundation, Inc. This file is part of the MPFR Library. The MPFR Library is free software; you can redistribute it and/or modify it under the terms of the GNU Lesser General Public License as published by the Free Software Foundation; either version 2.1 of the License, or (at your option) any later version. The MPFR Library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more details. You should have received a copy of the GNU Lesser General Public License along with the MPFR Library; see the file COPYING.LIB. If not, write to the Free Software Foundation, Inc., 51 Franklin Place, Fifth Floor, Boston, MA 02110-1301, USA. */ #include "mpfr-impl.h" /* return value is 0 iff no error occurred in the conversion (1 for NaN, +Inf, -Inf that have no equivalent in mpf) */ int mpfr_get_f (mpf_ptr x, mpfr_srcptr y, mp_rnd_t rnd_mode) { unsigned long sx, sy, precx, precy, sh; mp_exp_t ey; if (MPFR_UNLIKELY(MPFR_IS_SINGULAR(y))) { if (MPFR_IS_ZERO(y)) { mpf_set_ui (x, 0); return 0; } else /* NaN or Inf */ return 1; } sx = PREC(x); /* number of limbs of the mantissa of x */ precy = MPFR_PREC(y); precx = sx * BITS_PER_MP_LIMB; sy = 1 + (MPFR_PREC(y) - 1) / BITS_PER_MP_LIMB; /* since mpf numbers are represented in base 2^BITS_PER_MP_LIMB, we loose -EXP(y) % BITS_PER_MP_LIMB bits in the most significant limb */ ey = MPFR_GET_EXP(y) % BITS_PER_MP_LIMB; if (ey <= 0) sh = (unsigned long) (-ey); else /* 0 < ey < BITS_PER_MP_LIMB */ sh = BITS_PER_MP_LIMB - (unsigned long) ey; if (precy + sh <= precx) /* we can copy directly */ { /* necessarily sy <= sx */ if (sh) mpn_rshift (PTR(x) + sx - sy, MPFR_MANT(y), sy, sh); else MPN_COPY (PTR(x) + sx - sy, MPFR_MANT(y), sy); if (sx > sy) MPN_ZERO (PTR(x), sx - sy); EXP(x) = (MPFR_GET_EXP(y) + sh) / BITS_PER_MP_LIMB; } else /* we have to round to precx - sh bits */ { mpfr_t z; unsigned long sz; mpfr_init2 (z, precx - sh); sz = 1 + (MPFR_PREC(z) - 1) / BITS_PER_MP_LIMB; mpfr_set (z, y, rnd_mode); /* warning, sh may change due to rounding, but then z is a power of two, thus we can safely ignore its last bit which is 0 */ ey = MPFR_GET_EXP(z) % BITS_PER_MP_LIMB; sh = (ey <= 0) ? (unsigned long) (-ey) : BITS_PER_MP_LIMB - (unsigned long) ey; if (sh) mpn_rshift (PTR(x) + sx - sz, MPFR_MANT(z), sz, sh); else MPN_COPY (PTR(x) + sx - sz, MPFR_MANT(z), sz); if (sx > sz) MPN_ZERO (PTR(x), sx - sz); EXP(x) = (MPFR_GET_EXP(z) + sh) / BITS_PER_MP_LIMB; mpfr_clear (z); } /* set size and sign */ SIZ(x) = (MPFR_FROM_SIGN_TO_INT(MPFR_SIGN(y)) < 0) ? -sx : sx; return 0; }