Import bigdecimal-1.4.0.pre-20181130a

* https://github.com/ruby/bigdecimal/compare/74d25ef..v1.4.0.pre.20181130a

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@66124 b2dd03c8-39d4-4d8f-98ff-823fe69b080e

11 files changed, 391 insertions, 201 deletions

diff --git a/ext/bigdecimal/bigdecimal.c b/ext/bigdecimal/bigdecimal.c
index d1aa56c121..d2701cfb8e 100644
--- a/ext/bigdecimal/bigdecimal.c
+++ b/ext/bigdecimal/bigdecimal.c
@@ -136,24 +136,6 @@ rb_rational_den(VALUE rat)
 #define DoSomeOne(x,y,f) rb_num_coerce_bin(x,y,f)
 
 /*
- * Returns the BigDecimal version number.
- */
-static VALUE
-BigDecimal_version(VALUE self)
-{
-  /*
-   * 1.0.0: Ruby 1.8.0
-   * 1.0.1: Ruby 1.8.1
-   * 1.1.0: Ruby 1.9.3
-   */
-#ifndef RUBY_BIGDECIMAL_VERSION
-# error RUBY_BIGDECIMAL_VERSION is not defined
-#endif
-  rb_warning("BigDecimal.ver is deprecated; use BigDecimal::VERSION instead.");
-  return rb_str_new2(RUBY_BIGDECIMAL_VERSION);
-}
-
-/*
  *   VP routines used in BigDecimal part
  */
 static unsigned short VpGetException(void);
@@ -664,9 +646,10 @@ VP_EXPORT Real *
 VpNewRbClass(size_t mx, const char *str, VALUE klass)
 {
     VALUE obj = TypedData_Wrap_Struct(klass, &BigDecimal_data_type, 0);
-    Real *pv = VpAlloc(mx,str);
+    Real *pv = VpAlloc(mx, str, 1);
     RTYPEDDATA_DATA(obj) = pv;
     pv->obj = obj;
+    RB_OBJ_FREEZE(obj);
     return pv;
 }
 
@@ -2165,15 +2148,10 @@ BigDecimal_exponent(VALUE self)
     return INT2NUM(e);
 }
 
-/* Returns debugging information about the value as a string of comma-separated
- * values in angle brackets with a leading #:
+/* Returns a string representation of self.
  *
  *   BigDecimal("1234.5678").inspect
  *     #=> "0.12345678e4"
- *
- * The first part is the address, the second is the value as a string, and
- * the final part ss(mm) is the current number of significant digits and the
- * maximum number of significant digits, respectively.
  */
 static VALUE
 BigDecimal_inspect(VALUE self)
@@ -2335,7 +2313,7 @@ BigDecimal_power(int argc, VALUE*argv, VALUE self)
     n = NIL_P(prec) ? (ssize_t)(x->Prec*VpBaseFig()) : NUM2SSIZET(prec);
 
     if (VpIsNaN(x)) {
-	y = VpCreateRbObject(n, "0#");
+	y = VpCreateRbObject(n, "0");
 	RB_GC_GUARD(y->obj);
 	VpSetNaN(y);
 	return ToValue(y);
@@ -2459,7 +2437,7 @@ BigDecimal_power(int argc, VALUE*argv, VALUE self)
 	    }
 	}
 	else {
-	    y = VpCreateRbObject(n, "0#");
+	    y = VpCreateRbObject(n, "0");
 	    if (BIGDECIMAL_NEGATIVE_P(x)) {
 		if (is_integer(vexp)) {
 		    if (is_even(vexp)) {
@@ -2492,7 +2470,7 @@ BigDecimal_power(int argc, VALUE*argv, VALUE self)
 	}
 	else if (RTEST(rb_funcall(abs_value, '<', 1, INT2FIX(1)))) {
 	    if (is_negative(vexp)) {
-		y = VpCreateRbObject(n, "0#");
+		y = VpCreateRbObject(n, "0");
 		if (is_even(vexp)) {
 		    VpSetInf(y, VpGetSign(x));
 		}
@@ -2510,7 +2488,7 @@ BigDecimal_power(int argc, VALUE*argv, VALUE self)
 	}
 	else {
 	    if (is_positive(vexp)) {
-		y = VpCreateRbObject(n, "0#");
+		y = VpCreateRbObject(n, "0");
 		if (is_even(vexp)) {
 		    VpSetInf(y, VpGetSign(x));
 		}
@@ -2560,52 +2538,8 @@ BigDecimal_power_op(VALUE self, VALUE exp)
     return BigDecimal_power(1, &exp, self);
 }
 
-static VALUE
-BigDecimal_s_allocate(VALUE klass)
-{
-    return VpNewRbClass(0, NULL, klass)->obj;
-}
-
 static Real *BigDecimal_new(int argc, VALUE *argv);
 
-/* call-seq:
- *   new(initial, digits)
- *
- * Create a new BigDecimal object.
- *
- * initial:: The initial value, as an Integer, a Float, a Rational,
- *           a BigDecimal, or a String.
- *
- *           If it is a String, spaces are ignored and unrecognized characters
- *           terminate the value.
- *
- * digits:: The number of significant digits, as an Integer. If omitted or 0,
- *          the number of significant digits is determined from the initial
- *          value.
- *
- * The actual number of significant digits used in computation is usually
- * larger than the specified number.
- *
- * ==== Exceptions
- *
- * TypeError:: If the +initial+ type is neither Integer, Float,
- *             Rational, nor BigDecimal, this exception is raised.
- *
- * TypeError:: If the +digits+ is not an Integer, this exception is raised.
- *
- * ArgumentError:: If +initial+ is a Float, and the +digits+ is larger than
- *                 Float::DIG + 1, this exception is raised.
- *
- * ArgumentError:: If the +initial+ is a Float or Rational, and the +digits+
- *                 value is omitted, this exception is raised.
- */
-static VALUE
-BigDecimal_s_new(int argc, VALUE *argv, VALUE self)
-{
-  rb_warning("BigDecimal.new is deprecated; use Kernel.BigDecimal method instead.");
-  return rb_call_super(argc, argv);
-}
-
 static VALUE
 BigDecimal_initialize(int argc, VALUE *argv, VALUE self)
 {
@@ -2700,10 +2634,40 @@ BigDecimal_new(int argc, VALUE *argv)
 	break;
     }
     StringValueCStr(iniValue);
-    return VpAlloc(mf, RSTRING_PTR(iniValue));
+    return VpAlloc(mf, RSTRING_PTR(iniValue), 1);
 }
 
-/* See also BigDecimal.new */
+/* call-seq:
+ *   BigDecimal(initial, digits)
+ *
+ * Create a new BigDecimal object.
+ *
+ * initial:: The initial value, as an Integer, a Float, a Rational,
+ *           a BigDecimal, or a String.
+ *
+ *           If it is a String, spaces are ignored and unrecognized characters
+ *           terminate the value.
+ *
+ * digits:: The number of significant digits, as an Integer. If omitted or 0,
+ *          the number of significant digits is determined from the initial
+ *          value.
+ *
+ * The actual number of significant digits used in computation is usually
+ * larger than the specified number.
+ *
+ * ==== Exceptions
+ *
+ * TypeError:: If the +initial+ type is neither Integer, Float,
+ *             Rational, nor BigDecimal, this exception is raised.
+ *
+ * TypeError:: If the +digits+ is not an Integer, this exception is raised.
+ *
+ * ArgumentError:: If +initial+ is a Float, and the +digits+ is larger than
+ *                 Float::DIG + 1, this exception is raised.
+ *
+ * ArgumentError:: If the +initial+ is a Float or Rational, and the +digits+
+ *                 value is omitted, this exception is raised.
+ */
 static VALUE
 BigDecimal_global_new(int argc, VALUE *argv, VALUE self)
 {
@@ -2715,6 +2679,7 @@ BigDecimal_global_new(int argc, VALUE *argv, VALUE self)
     GUARD_OBJ(pv, BigDecimal_new(argc, argv));
     if (ToValue(pv)) pv = VpCopy(NULL, pv);
     RTYPEDDATA_DATA(obj) = pv;
+    RB_OBJ_FREEZE(obj);
     return pv->obj = obj;
 }
 
@@ -3138,6 +3103,21 @@ get_vp_value:
     return y;
 }
 
+VALUE
+rmpd_util_str_to_d(VALUE str)
+{
+  ENTER(1);
+  char const *c_str;
+  Real *pv;
+  VALUE obj;
+
+  c_str = StringValueCStr(str);
+  GUARD_OBJ(pv, VpAlloc(0, c_str, 0));
+  obj = TypedData_Wrap_Struct(rb_cBigDecimal, &BigDecimal_data_type, pv);
+  RB_OBJ_FREEZE(obj);
+  return obj;
+}
+
 /* Document-class: BigDecimal
  * BigDecimal provides arbitrary-precision floating point decimal arithmetic.
  *
@@ -3277,18 +3257,17 @@ Init_bigdecimal(void)
 
     /* Class and method registration */
     rb_cBigDecimal = rb_define_class("BigDecimal", rb_cNumeric);
-    rb_define_alloc_func(rb_cBigDecimal, BigDecimal_s_allocate);
 
     /* Global function */
     rb_define_global_function("BigDecimal", BigDecimal_global_new, -1);
 
     /* Class methods */
-    rb_define_singleton_method(rb_cBigDecimal, "new", BigDecimal_s_new, -1);
+    rb_undef_method(CLASS_OF(rb_cBigDecimal), "allocate");
+    rb_undef_method(CLASS_OF(rb_cBigDecimal), "new");
     rb_define_singleton_method(rb_cBigDecimal, "mode", BigDecimal_mode, -1);
     rb_define_singleton_method(rb_cBigDecimal, "limit", BigDecimal_limit, -1);
     rb_define_singleton_method(rb_cBigDecimal, "double_fig", BigDecimal_double_fig, 0);
     rb_define_singleton_method(rb_cBigDecimal, "_load", BigDecimal_load, 1);
-    rb_define_singleton_method(rb_cBigDecimal, "ver", BigDecimal_version, 0);
 
     rb_define_singleton_method(rb_cBigDecimal, "save_exception_mode", BigDecimal_save_exception_mode, 0);
     rb_define_singleton_method(rb_cBigDecimal, "save_rounding_mode", BigDecimal_save_rounding_mode, 0);
@@ -3745,25 +3724,19 @@ One(void)
 VP_EXPORT double
 VpGetDoubleNaN(void) /* Returns the value of NaN */
 {
-    static double fNaN = 0.0;
-    if (fNaN == 0.0) fNaN = Zero()/Zero();
-    return fNaN;
+    return nan("");
 }
 
 VP_EXPORT double
 VpGetDoublePosInf(void) /* Returns the value of +Infinity */
 {
-    static double fInf = 0.0;
-    if (fInf == 0.0) fInf = One()/Zero();
-    return fInf;
+    return HUGE_VAL;
 }
 
 VP_EXPORT double
 VpGetDoubleNegInf(void) /* Returns the value of -Infinity */
 {
-    static double fInf = 0.0;
-    if (fInf == 0.0) fInf = -(One()/Zero());
-    return fInf;
+    return -HUGE_VAL;
 }
 
 VP_EXPORT double
@@ -3964,8 +3937,8 @@ VpInit(BDIGIT BaseVal)
     VpGetDoubleNegZero();
 
     /* Allocates Vp constants. */
-    VpConstOne = VpAlloc(1UL, "1");
-    VpPt5 = VpAlloc(1UL, ".5");
+    VpConstOne = VpAlloc(1UL, "1", 1);
+    VpPt5 = VpAlloc(1UL, ".5", 1);
 
 #ifdef BIGDECIMAL_DEBUG
     gnAlloc = 0;
@@ -4029,6 +4002,52 @@ overflow:
     return VpException(VP_EXCEPTION_OVERFLOW, "Exponent overflow", 0);
 }
 
+Real *
+rmpd_parse_special_string(const char *str)
+{
+    static const struct {
+        const char *str;
+        size_t len;
+        int sign;
+    } table[] = {
+        { SZ_INF,  sizeof(SZ_INF)  - 1, VP_SIGN_POSITIVE_INFINITE },
+        { SZ_PINF, sizeof(SZ_PINF) - 1, VP_SIGN_POSITIVE_INFINITE },
+        { SZ_NINF, sizeof(SZ_NINF) - 1, VP_SIGN_NEGATIVE_INFINITE },
+        { SZ_NaN,  sizeof(SZ_NaN)  - 1, VP_SIGN_NaN               }
+    };
+    static const size_t table_length = sizeof(table) / sizeof(table[0]);
+    size_t i;
+
+    for (i = 0; i < table_length; ++i) {
+        const char *p;
+        if (strncmp(str, table[i].str, table[i].len) != 0) {
+            continue;
+        }
+
+        p = str + table[i].len;
+        while (*p && ISSPACE(*p)) ++p;
+        if (*p == '\0') {
+            Real *vp = VpAllocReal(1);
+            vp->MaxPrec = 1;
+            switch (table[i].sign) {
+              default:
+                UNREACHABLE; break;
+              case VP_SIGN_POSITIVE_INFINITE:
+                VpSetPosInf(vp);
+                return vp;
+              case VP_SIGN_NEGATIVE_INFINITE:
+                VpSetNegInf(vp);
+                return vp;
+              case VP_SIGN_NaN:
+                VpSetNaN(vp);
+                return vp;
+            }
+        }
+    }
+
+    return NULL;
+}
+
 /*
  * Allocates variable.
  * [Input]
@@ -4043,10 +4062,10 @@ overflow:
  *   NULL be returned if memory allocation is failed,or any error.
  */
 VP_EXPORT Real *
-VpAlloc(size_t mx, const char *szVal)
+VpAlloc(size_t mx, const char *szVal, int strict_p)
 {
     const char *orig_szVal = szVal;
-    size_t i, ni, ipn, ipf, nf, ipe, ne, dot_seen, exp_seen, nalloc;
+    size_t i, j, ni, ipf, nf, ipe, ne, dot_seen, exp_seen, nalloc;
     char v, *psz;
     int  sign=1;
     Real *vp = NULL;
@@ -4057,7 +4076,10 @@ VpAlloc(size_t mx, const char *szVal)
     if (mx == 0) ++mx;
 
     if (szVal) {
+        /* Skipping leading spaces */
         while (ISSPACE(*szVal)) szVal++;
+
+        /* Processing the leading one `#` */
         if (*szVal != '#') {
             if (mf) {
                 mf = (mf + BASE_FIG - 1) / BASE_FIG + 2; /* Needs 1 more for div */
@@ -4071,6 +4093,7 @@ VpAlloc(size_t mx, const char *szVal)
         }
     }
     else {
+      return_zero:
         /* necessary to be able to store */
         /* at least mx digits. */
         /* szVal==NULL ==> allocate zero value. */
@@ -4081,105 +4104,166 @@ VpAlloc(size_t mx, const char *szVal)
         return vp;
     }
 
-    /* Skip all '_' after digit: 2006-6-30 */
-    ni = 0;
+    /* Check on Inf & NaN */
+    if ((vp = rmpd_parse_special_string(szVal)) != NULL) {
+        return vp;
+    }
+
+    /* Scanning digits */
+
+    /* A buffer for keeping scanned digits */
     buf = rb_str_tmp_new(strlen(szVal) + 1);
     psz = RSTRING_PTR(buf);
-    i   = 0;
-    ipn = 0;
-    while ((psz[i] = szVal[ipn]) != 0) {
-        if (ISSPACE(psz[i])) {
-            psz[i] = 0;
+
+    /* cursor: i for psz, and j for szVal */
+    i = j = 0;
+
+    /* Scanning: sign part */
+    v = psz[i] = szVal[j];
+    if ((v == '-') || (v == '+')) {
+        sign = -(v == '-');
+        ++i;
+        ++j;
+    }
+
+    /* Scanning: integer part */
+    ni  = 0; /* number of digits in the integer part */
+    while ((v = psz[i] = szVal[j]) != '\0') {
+        if (!strict_p && ISSPACE(v)) {
+            v = psz[i] = '\0';
             break;
         }
-        if (ISDIGIT(psz[i])) ++ni;
-        if (psz[i] == '_') {
+        if (v == '_') {
             if (ni > 0) {
-                ipn++;
-                continue;
+                v = szVal[j+1];
+                if (v == '\0' || ISSPACE(v) || ISDIGIT(v)) {
+                    ++j;
+                    continue;
+                }
+                if (!strict_p) {
+                    v = psz[i] = '\0';
+                    break;
+                }
             }
-            psz[i] = 0;
+            goto invalid_value;
+        }
+        if (!ISDIGIT(v)) {
             break;
         }
+        ++ni;
         ++i;
-        ++ipn;
+        ++j;
     }
-    szVal = psz;
 
-    /* Check on Inf & NaN */
-    if (StrCmp(szVal, SZ_PINF) == 0 || StrCmp(szVal, SZ_INF) == 0 ) {
-        vp = VpAllocReal(1);
-        vp->MaxPrec = 1;    /* set max precision */
-        VpSetPosInf(vp);
-        return vp;
-    }
-    if (StrCmp(szVal, SZ_NINF) == 0) {
-        vp = VpAllocReal(1);
-        vp->MaxPrec = 1;    /* set max precision */
-        VpSetNegInf(vp);
-        return vp;
-    }
-    if (StrCmp(szVal, SZ_NaN) == 0) {
-        vp = VpAllocReal(1);
-        vp->MaxPrec = 1;    /* set max precision */
-        VpSetNaN(vp);
-        return vp;
-    }
-
-    /* check on number szVal[] */
-    ipn = i = 0;
-    if      (szVal[i] == '-') { sign=-1; ++i; }
-    else if (szVal[i] == '+')            ++i;
-    /* Skip digits */
-    ni = 0;            /* digits in mantissa */
-    while ((v = szVal[i]) != 0) {
-        if (!ISDIGIT(v)) break;
-        ++i;
-        ++ni;
-    }
-    nf  = 0;
-    ipf = 0;
-    ipe = 0;
-    ne  = 0;
+    /* Scanning: fractional part */
+    nf  = 0; /* number of digits in the fractional part */
+    ne  = 0; /* number of digits in the exponential part */
+    ipf = 0; /* index of the beginning of the fractional part */
+    ipe = 0; /* index of the beginning of the exponential part */
     dot_seen = 0;
     exp_seen = 0;
-    if (v) {
-        /* other than digit nor \0 */
-        if (szVal[i] == '.') {    /* xxx. */
+
+    if (v != '\0') {
+        /* Scanning fractional part */
+        if ((psz[i] = szVal[j]) == '.') {
             dot_seen = 1;
             ++i;
+            ++j;
             ipf = i;
-            while ((v = szVal[i]) != 0) {    /* get fraction part. */
+            while ((v = psz[i] = szVal[j]) != '\0') {
+                if (!strict_p && ISSPACE(v)) {
+                    v = psz[i] = '\0';
+                    break;
+                }
+                if (v == '_') {
+                    if (nf > 0 && ISDIGIT(szVal[j+1])) {
+                        ++j;
+                        continue;
+                    }
+                    if (!strict_p) {
+                        v = psz[i] = '\0';
+                        if (nf == 0) {
+                            dot_seen = 0;
+                        }
+                        break;
+                    }
+                    goto invalid_value;
+                }
                 if (!ISDIGIT(v)) break;
                 ++i;
+                ++j;
                 ++nf;
             }
         }
-        ipe = 0;        /* Exponent */
-
-        switch (szVal[i]) {
-            case '\0':
-                break;
-            case 'e': case 'E':
-            case 'd': case 'D':
-                exp_seen = 1;
-                ++i;
-                ipe = i;
-                v = szVal[i];
-                if ((v == '-') || (v == '+')) ++i;
-                while ((v=szVal[i]) != 0) {
-                    if (!ISDIGIT(v)) break;
+
+        /* Scanning exponential part */
+        if (v != '\0') {
+            switch ((psz[i] = szVal[j])) {
+                case '\0':
+                    break;
+                case 'e': case 'E':
+                case 'd': case 'D':
+                    exp_seen = 1;
                     ++i;
-                    ++ne;
-                }
-                break;
-            default:
-                break;
+                    ++j;
+                    ipe = i;
+                    v = psz[i] = szVal[j];
+                    if ((v == '-') || (v == '+')) {
+                        ++i;
+                        ++j;
+                    }
+                    while ((v = psz[i] = szVal[j]) != '\0') {
+                        if (!strict_p && ISSPACE(v)) {
+                            v = psz[i] = '\0';
+                            break;
+                        }
+                        if (v == '_') {
+                            if (ne > 0 && ISDIGIT(szVal[j+1])) {
+                                ++j;
+                                continue;
+                            }
+                            if (!strict_p) {
+                                v = psz[i] = '\0';
+                                if (ne == 0) {
+                                    exp_seen = 0;
+                                }
+                                break;
+                            }
+                            goto invalid_value;
+                        }
+                        if (!ISDIGIT(v)) break;
+                        ++i;
+                        ++j;
+                        ++ne;
+                    }
+                    break;
+                default:
+                    break;
+            }
+        }
+
+        if (v != '\0') {
+            /* Scanning trailing spaces */
+            while (ISSPACE(szVal[j])) ++j;
+
+            /* Invalid character */
+            if (szVal[j]) {
+                goto invalid_value;
+            }
         }
     }
+
+    psz[i] = '\0';
+
     if (((ni == 0 || dot_seen) && nf == 0) || (exp_seen && ne == 0)) {
-	VALUE str = rb_str_new2(orig_szVal);
-	rb_raise(rb_eArgError, "invalid value for BigDecimal(): \"%"PRIsVALUE"\"", str);
+        VALUE str;
+      invalid_value:
+        if (!strict_p) {
+            goto return_zero;
+        }
+
+        str = rb_str_new2(orig_szVal);
+        rb_raise(rb_eArgError, "invalid value for BigDecimal(): \"%"PRIsVALUE"\"", str);
     }
 
     nalloc = (ni + nf + BASE_FIG - 1) / BASE_FIG + 1;    /* set effective allocation  */
@@ -4191,7 +4275,7 @@ VpAlloc(size_t mx, const char *szVal)
     /* xmalloc() alway returns(or throw interruption) */
     vp->MaxPrec = mx;        /* set max precision */
     VpSetZero(vp, sign);
-    VpCtoV(vp, &szVal[ipn], ni, &szVal[ipf], nf, &szVal[ipe], ne);
+    VpCtoV(vp, psz, ni, psz + ipf, nf, psz + ipe, ne);
     rb_str_resize(buf, 0);
     return vp;
 }
@@ -4754,7 +4838,7 @@ VpMult(Real *c, Real *a, Real *b)
 
     if (MxIndC < MxIndAB) {    /* The Max. prec. of c < Prec(a)+Prec(b) */
 	w = c;
-	c = VpAlloc((size_t)((MxIndAB + 1) * BASE_FIG), "#0");
+	c = VpAlloc((size_t)((MxIndAB + 1) * BASE_FIG), "#0", 1);
 	MxIndC = MxIndAB;
     }
 
@@ -5922,8 +6006,8 @@ VpSqrt(Real *y, Real *x)
     if (x->MaxPrec > (size_t)n) n = (ssize_t)x->MaxPrec;
 
     /* allocate temporally variables  */
-    f = VpAlloc(y->MaxPrec * (BASE_FIG + 2), "#1");
-    r = VpAlloc((n + n) * (BASE_FIG + 2), "#1");
+    f = VpAlloc(y->MaxPrec * (BASE_FIG + 2), "#1", 1);
+    r = VpAlloc((n + n) * (BASE_FIG + 2), "#1", 1);
 
     nr = 0;
     y_prec = y->MaxPrec;
@@ -6375,8 +6459,8 @@ VpPower(Real *y, Real *x, SIGNED_VALUE n)
 
     /* Allocate working variables  */
 
-    w1 = VpAlloc((y->MaxPrec + 2) * BASE_FIG, "#0");
-    w2 = VpAlloc((w1->MaxPrec * 2 + 1) * BASE_FIG, "#0");
+    w1 = VpAlloc((y->MaxPrec + 2) * BASE_FIG, "#0", 1);
+    w2 = VpAlloc((w1->MaxPrec * 2 + 1) * BASE_FIG, "#0", 1);
     /* calculation start */
 
     VpAsgn(y, x, 1);
diff --git a/ext/bigdecimal/bigdecimal.def b/ext/bigdecimal/bigdecimal.def
new file mode 100644
index 0000000000..615bf72e20
--- /dev/null
+++ b/ext/bigdecimal/bigdecimal.def
@@ -0,0 +1,3 @@
+EXPORTS
+rmpd_util_str_to_d
+Init_bigdecimal
diff --git a/ext/bigdecimal/bigdecimal.gemspec b/ext/bigdecimal/bigdecimal.gemspec
index c8c90870ea..bc0cbac6b5 100644
--- a/ext/bigdecimal/bigdecimal.gemspec
+++ b/ext/bigdecimal/bigdecimal.gemspec
@@ -1,6 +1,6 @@
 # coding: utf-8
 
-bigdecimal_version = '1.3.4'
+bigdecimal_version = '1.4.0.pre.20181121a'
 
 Gem::Specification.new do |s|
   s.name          = "bigdecimal"
@@ -14,13 +14,15 @@ Gem::Specification.new do |s|
   s.license       = "ruby"
 
   s.require_paths = %w[lib]
-  s.extensions    = %w[ext/bigdecimal/extconf.rb]
+  s.extensions    = %w[ext/bigdecimal/extconf.rb ext/bigdecimal/util/extconf.rb]
   s.files         = %w[
     bigdecimal.gemspec
     ext/bigdecimal/bigdecimal.c
     ext/bigdecimal/bigdecimal.h
     ext/bigdecimal/depend
     ext/bigdecimal/extconf.rb
+    ext/bigdecimal/util/extconf.rb
+    ext/bigdecimal/util/util.c
     lib/bigdecimal/jacobian.rb
     lib/bigdecimal/ludcmp.rb
     lib/bigdecimal/math.rb
@@ -31,9 +33,11 @@ Gem::Specification.new do |s|
     sample/pi.rb
   ]
 
+  s.required_ruby_version = Gem::Requirement.new(">= 2.3.0".freeze)
+
   s.add_development_dependency "rake", "~> 10.0"
   s.add_development_dependency "rake-compiler", ">= 0.9"
   s.add_development_dependency "rake-compiler-dock", ">= 0.6.1"
-  s.add_development_dependency "minitest", "~> 4.7.5"
+  s.add_development_dependency "minitest", "< 5.0.0"
   s.add_development_dependency "pry"
 end
diff --git a/ext/bigdecimal/bigdecimal.h b/ext/bigdecimal/bigdecimal.h
index 3297f18867..b7b0abe28c 100644
--- a/ext/bigdecimal/bigdecimal.h
+++ b/ext/bigdecimal/bigdecimal.h
@@ -308,7 +308,7 @@ VP_EXPORT size_t VpInit(BDIGIT BaseVal);
 VP_EXPORT void *VpMemAlloc(size_t mb);
 VP_EXPORT void *VpMemRealloc(void *ptr, size_t mb);
 VP_EXPORT void VpFree(Real *pv);
-VP_EXPORT Real *VpAlloc(size_t mx, const char *szVal);
+VP_EXPORT Real *VpAlloc(size_t mx, const char *szVal, int strict_p);
 VP_EXPORT size_t VpAsgn(Real *c, Real *a, int isw);
 VP_EXPORT size_t VpAddSub(Real *c,Real *a,Real *b,int operation);
 VP_EXPORT size_t VpMult(Real *c,Real *a,Real *b);
diff --git a/ext/bigdecimal/extconf.rb b/ext/bigdecimal/extconf.rb
index e565da891a..602c7d6162 100644
--- a/ext/bigdecimal/extconf.rb
+++ b/ext/bigdecimal/extconf.rb
@@ -1,6 +1,10 @@
 # frozen_string_literal: false
 require 'mkmf'
 
+def windows_platform?
+  /cygwin|mingw|mswin/ === RUBY_PLATFORM
+end
+
 gemspec_name = gemspec_path = nil
 unless ['', '../../'].any? {|dir|
          gemspec_name = "#{dir}bigdecimal.gemspec"
@@ -17,8 +21,6 @@ bigdecimal_version =
 
 $defs << %Q[-DRUBY_BIGDECIMAL_VERSION=\\"#{bigdecimal_version}\\"]
 
-alias __have_macro__ have_macro
-
 have_func("labs", "stdlib.h")
 have_func("llabs", "stdlib.h")
 have_func("finite", "math.h")
@@ -30,6 +32,38 @@ have_func("rb_rational_den", "ruby.h")
 have_func("rb_array_const_ptr", "ruby.h")
 have_func("rb_sym2str", "ruby.h")
 
+if windows_platform?
+  library_base_name = "ruby-bigdecimal"
+  case RUBY_PLATFORM
+  when /cygwin|mingw/
+    import_library_name = "libruby-bigdecimal.a"
+  when /mswin/
+    import_library_name = "bigdecimal-$(arch).lib"
+  end
+end
+
+checking_for(checking_message("Windows")) do
+  if windows_platform?
+    case RUBY_PLATFORM
+    when /cygwin|mingw/
+      $DLDFLAGS << " $(srcdir)/bigdecimal.def"
+      $DLDFLAGS << " -Wl,--out-implib=$(TARGET_SO_DIR)#{import_library_name}"
+    when /mswin/
+      $DLDFLAGS << " /DEF:$(srcdir)/bigdecimal.def"
+    end
+    $cleanfiles << import_library_name
+    true
+  else
+    false
+  end
+end
+
 create_makefile('bigdecimal') {|mf|
   mf << "\nall:\n\nextconf.h: $(srcdir)/#{gemspec_name}\n"
+  case RUBY_PLATFORM
+  when /mswin/
+    mf << "\nall:\n\tdir $(TARGET_SO_DIR)"
+  else
+    mf << "\nall:\n\tls $(TARGET_SO_DIR)"
+  end
 }
diff --git a/ext/bigdecimal/lib/bigdecimal/jacobian.rb b/ext/bigdecimal/lib/bigdecimal/jacobian.rb
index 9cad06c09b..84c50248b7 100644
--- a/ext/bigdecimal/lib/bigdecimal/jacobian.rb
+++ b/ext/bigdecimal/lib/bigdecimal/jacobian.rb
@@ -21,6 +21,9 @@
 #
 # fx is f.values(x).
 #
+
+require 'bigdecimal'
+
 module Jacobian
   module_function
 
diff --git a/ext/bigdecimal/lib/bigdecimal/util.rb b/ext/bigdecimal/lib/bigdecimal/util.rb
index 911fa6fe3a..88f490cb45 100644
--- a/ext/bigdecimal/lib/bigdecimal/util.rb
+++ b/ext/bigdecimal/lib/bigdecimal/util.rb
@@ -5,6 +5,8 @@
 # and provides BigDecimal#to_d and BigDecimal#to_digits.
 #++
 
+require 'bigdecimal'
+require 'bigdecimal/util.so'
 
 class Integer < Numeric
   # call-seq:
@@ -42,8 +44,8 @@ class Float < Numeric
   #
   # See also BigDecimal::new.
   #
-  def to_d(precision=nil)
-    BigDecimal(self, precision || Float::DIG)
+  def to_d(precision=Float::DIG)
+    BigDecimal(self, precision)
   end
 end
 
@@ -64,13 +66,6 @@ class String
   #
   # See also BigDecimal::new.
   #
-  def to_d
-    begin
-      BigDecimal(self)
-    rescue ArgumentError
-      BigDecimal(0)
-    end
-  end
 end
 
 
@@ -132,3 +127,20 @@ class Rational < Numeric
     BigDecimal(self, precision)
   end
 end
+
+
+class NilClass
+  # call-seq:
+  #     nil.to_d -> bigdecimal
+  #
+  # Returns nil represented as a BigDecimal.
+  #
+  #     require 'bigdecimal'
+  #     require 'bigdecimal/util'
+  #
+  #     nil.to_d   # => 0.0
+  #
+  def to_d
+    BigDecimal(0)
+  end
+end
diff --git a/ext/bigdecimal/sample/linear.rb b/ext/bigdecimal/sample/linear.rb
index 3b23269f8a..516c2473be 100644
--- a/ext/bigdecimal/sample/linear.rb
+++ b/ext/bigdecimal/sample/linear.rb
@@ -28,8 +28,8 @@ def rd_order(na)
 end
 
 na   = ARGV.size
-zero = BigDecimal.new("0.0")
-one  = BigDecimal.new("1.0")
+zero = BigDecimal("0.0")
+one  = BigDecimal("1.0")
 
 while (n=rd_order(na))>0
   a = []
@@ -37,27 +37,28 @@ while (n=rd_order(na))>0
   b = []
   if na <= 0
      # Read data from console.
-     printf("\nEnter coefficient matrix element A[i,j]\n");
+     printf("\nEnter coefficient matrix element A[i,j]\n")
      for i in 0...n do
        for j in 0...n do
          printf("A[%d,%d]? ",i,j); s = ARGF.gets
-         a  << BigDecimal.new(s);
-         as << BigDecimal.new(s);
+         a  << BigDecimal(s)
+         as << BigDecimal(s)
        end
-       printf("Contatant vector element b[%d] ? ",i); b << BigDecimal.new(ARGF.gets);
+       printf("Contatant vector element b[%d] ? ",i)
+       b << BigDecimal(ARGF.gets)
      end
   else
      # Read data from specified file.
-     printf("Coefficient matrix and constant vector.\n");
+     printf("Coefficient matrix and constant vector.\n")
      for i in 0...n do
        s = ARGF.gets
        printf("%d) %s",i,s)
        s = s.split
        for j in 0...n do
-         a  << BigDecimal.new(s[j]);
-         as << BigDecimal.new(s[j]);
+         a  << BigDecimal(s[j])
+         as << BigDecimal(s[j])
        end
-       b << BigDecimal.new(s[n]);
+       b << BigDecimal(s[n])
      end
   end
   x = lusolve(a,b,ludecomp(a,n,zero,one),zero)
diff --git a/ext/bigdecimal/sample/nlsolve.rb b/ext/bigdecimal/sample/nlsolve.rb
index b1dd08e0a3..c2227dac73 100644
--- a/ext/bigdecimal/sample/nlsolve.rb
+++ b/ext/bigdecimal/sample/nlsolve.rb
@@ -12,11 +12,11 @@ include Newton
 
 class Function # :nodoc: all
   def initialize()
-    @zero = BigDecimal.new("0.0")
-    @one  = BigDecimal.new("1.0")
-    @two  = BigDecimal.new("2.0")
-    @ten  = BigDecimal.new("10.0")
-    @eps  = BigDecimal.new("1.0e-16")
+    @zero = BigDecimal("0.0")
+    @one  = BigDecimal("1.0")
+    @two  = BigDecimal("2.0")
+    @ten  = BigDecimal("10.0")
+    @eps  = BigDecimal("1.0e-16")
   end
   def zero;@zero;end
   def one ;@one ;end
diff --git a/ext/bigdecimal/util/extconf.rb b/ext/bigdecimal/util/extconf.rb
new file mode 100644
index 0000000000..023b19ba7e
--- /dev/null
+++ b/ext/bigdecimal/util/extconf.rb
@@ -0,0 +1,40 @@
+# frozen_string_literal: false
+require 'mkmf'
+
+def windows_platform?
+  /cygwin|mingw|mswin/ === RUBY_PLATFORM
+end
+
+if windows_platform?
+  library_base_name = "ruby-bigdecimal"
+  case RUBY_PLATFORM
+  when /cygwin|mingw/
+    import_library_name = "libruby-bigdecimal.a"
+  when /mswin/
+    import_library_name = "bigdecimal-$(arch).lib"
+  end
+end
+
+checking_for(checking_message("Windows")) do
+  if windows_platform?
+    if defined?($extlist)
+      build_dir = "$(TARGET_SO_DIR)../"
+    else
+      base_dir = File.expand_path('../../../..', __FILE__)
+      build_dir = File.join(base_dir, "tmp", RUBY_PLATFORM, "bigdecimal", RUBY_VERSION)
+    end
+    case RUBY_PLATFORM
+    when /cygwin|mingw/
+      $LDFLAGS << " -L#{build_dir} -L.. -L .."
+      $libs << " -l#{library_base_name}"
+    when /mswin/
+      $DLDFLAGS << " /libpath:#{build_dir} /libpath:.."
+      $libs << " #{import_library_name}"
+    end
+    true
+  else
+    false
+  end
+end
+
+create_makefile('bigdecimal/util')
diff --git a/ext/bigdecimal/util/util.c b/ext/bigdecimal/util/util.c
new file mode 100644
index 0000000000..8d38d87852
--- /dev/null
+++ b/ext/bigdecimal/util/util.c
@@ -0,0 +1,9 @@
+#include "ruby.h"
+
+RUBY_EXTERN VALUE rmpd_util_str_to_d(VALUE str);
+
+void
+Init_util(void)
+{
+  rb_define_method(rb_cString, "to_d", rmpd_util_str_to_d, 0);
+}