Remove unneeded casts

Most of these casts are redundant, and automatically dealt with by either the C++/C standard, or operations done onto said numbers, such as & 1 operation making casts to int pointless.

include/boost/math/cstdfloat/cstdfloat_iostream.hpp

@@ -448,7 +448,7 @@
       eval_floor(t, t);
       eval_convert_to(&expon, t);
 
-      if(-expon > std::numeric_limits<float_type>::max_exponent10 - 3)
+      if(expon < -std::numeric_limits<float_type>::max_exponent10 + 3)
       {
         int e = -expon / 2;
 
@@ -518,15 +518,14 @@
         if((cdigit == 5) && (t == 0))
         {
           // Use simple bankers rounding.
-
-          if((static_cast<int>(*result.rbegin() - '0') & 1) != 0)
+          if(((*result.rbegin() - '0') & 1) != 0)
           {
             round_string_up_at(result, static_cast<int>(result.size() - 1U), expon);
           }
         }
         else if(cdigit >= 5)
         {
-          round_string_up_at(result, static_cast<int>(result.size() - 1), expon);
+          round_string_up_at(result, static_cast<int>(result.size() - 1U), expon);
         }
       }
     }
@@ -569,9 +568,9 @@
   {
     value = 0;
 
-    if((p == static_cast<const char*>(0U)) || (*p == static_cast<char>(0)))
+    if((p == nullptr) || (*p == '\0'))
     {
-      return;
+      return false;
     }
 
     bool is_neg       = false;
@@ -584,11 +583,11 @@
 
     constexpr int max_digits = std::numeric_limits<float_type>::max_digits10 + 1;
 
-    if(*p == static_cast<char>('+'))
+    if(*p == '+')
     {
       ++p;
     }
-    else if(*p == static_cast<char>('-'))
+    else if(*p == '-')
     {
       is_neg = true;
       ++p;
@@ -632,24 +631,20 @@
       ++digits_seen;
     }
 
-    if(*p == static_cast<char>('.'))
+    if(*p == '.')
     {
       // Grab everything after the point, stop when we've seen
       // enough digits, even if there are actually more available.
 
       ++p;
 
-      while(std::isdigit(*p))
+      while(std::isdigit(*p) && digits_seen <= max_digits)
       {
         eval_multiply(value, ten);
         eval_add(value, static_cast<int>(*p - '0'));
         ++p;
         --expon;
-
-        if(++digits_seen > max_digits)
-        {
-          break;
-        }
+        ++digits_seen;
       }
 
       while(std::isdigit(*p))
@@ -659,15 +654,15 @@
     }
 
     // Parse the exponent.
-    if((*p == static_cast<char>('e')) || (*p == static_cast<char>('E')))
+    if((*p == 'e') || (*p == 'E'))
     {
       ++p;
 
-      if(*p == static_cast<char>('+'))
+      if(*p == '+')
       {
         ++p;
       }
-      else if(*p == static_cast<char>('-'))
+      else if(*p == '-')
       {
         is_neg_expon = true;
         ++p;
@@ -678,7 +673,7 @@
       while(std::isdigit(*p))
       {
         e2 *= 10;
-        e2 += (*p - '0');
+        e2 += static_cast<int>(*p - '0');
         ++p;
       }
 
@@ -718,7 +713,7 @@
       value = -value;
     }
 
-    return (*p == static_cast<char>(0));
+    return (*p == '\0');
   }
   } } } } // boost::math::cstdfloat::detail