Backport of WL #2934: Make/find library for doing float/double

                      to string conversions and vice versa" 
Initial import of the dtoa.c code and custom wrappers around it 
to allow its usage from the server code. 
 
Conversion of FLOAT/DOUBLE values to DECIMAL ones or strings 
and vice versa has been significantly reworked. As the new 
algoritms are more precise than the older ones, results of such 
conversions may not always match those obtained from older 
server versions. This in turn may break compatibility for some 
applications. 
 
This patch also fixes the following bugs: 
- bug #12860 "Difference in zero padding of exponent between 
Unix and Windows" 
- bug #21497 "DOUBLE truncated to unusable value" 
- bug #26788 "mysqld (debug) aborts when inserting specific 
numbers into char fields" 
- bug #24541 "Data truncated..." on decimal type columns 
without any good reason" 

1 files changed, 7 insertions, 73 deletions

diff --git a/client/sql_string.cc b/client/sql_string.cc
index 3292bc7e6f2..ccbc8977e7f 100644
--- a/client/sql_string.cc
+++ b/client/sql_string.cc
@@ -24,10 +24,6 @@
 #include <m_string.h>
 #include <m_ctype.h>
 #include <mysql_com.h>
-#ifdef HAVE_FCONVERT
-#include <floatingpoint.h>
-#endif
-
 /*
   The following extern declarations are ok as these are interface functions
   required by the string function
@@ -117,82 +113,19 @@ bool String::set(ulonglong num, CHARSET_INFO *cs)
 
 bool String::set(double num,uint decimals, CHARSET_INFO *cs)
 {
-  char buff[331];
+  char buff[FLOATING_POINT_BUFFER];
   uint dummy_errors;
+  size_t len;
 
   str_charset=cs;
   if (decimals >= NOT_FIXED_DEC)
   {
-    uint32 len= my_sprintf(buff,(buff, "%.15g",num));// Enough for a DATETIME
+    len= my_gcvt(num, MY_GCVT_ARG_DOUBLE, sizeof(buff) - 1, buff, NULL);
     return copy(buff, len, &my_charset_latin1, cs, &dummy_errors);
   }
-#ifdef HAVE_FCONVERT
-  int decpt,sign;
-  char *pos,*to;
-
-  (void) fconvert(num,(int) decimals,&decpt,&sign,buff+1);
-  if (!my_isdigit(&my_charset_latin1, buff[1]))
-  {						// Nan or Inf
-    pos=buff+1;
-    if (sign)
-    {
-      buff[0]='-';
-      pos=buff;
-    }
-    uint dummy_errors;
-    return copy(pos,(uint32) strlen(pos), &my_charset_latin1, cs, &dummy_errors);
-  }
-  if (alloc((uint32) ((uint32) decpt+3+decimals)))
-    return TRUE;
-  to=Ptr;
-  if (sign)
-    *to++='-';
-
-  pos=buff+1;
-  if (decpt < 0)
-  {					/* value is < 0 */
-    *to++='0';
-    if (!decimals)
-      goto end;
-    *to++='.';
-    if ((uint32) -decpt > decimals)
-      decpt= - (int) decimals;
-    decimals=(uint32) ((int) decimals+decpt);
-    while (decpt++ < 0)
-      *to++='0';
-  }
-  else if (decpt == 0)
-  {
-    *to++= '0';
-    if (!decimals)
-      goto end;
-    *to++='.';
-  }
-  else
-  {
-    while (decpt-- > 0)
-      *to++= *pos++;
-    if (!decimals)
-      goto end;
-    *to++='.';
-  }
-  while (decimals--)
-    *to++= *pos++;
-
-end:
-  *to=0;
-  str_length=(uint32) (to-Ptr);
-  return FALSE;
-#else
-#ifdef HAVE_SNPRINTF
-  buff[sizeof(buff)-1]=0;			// Safety
-  snprintf(buff,sizeof(buff)-1, "%.*f",(int) decimals,num);
-#else
-  sprintf(buff,"%.*f",(int) decimals,num);
-#endif
-  return copy(buff,(uint32) strlen(buff), &my_charset_latin1, cs,
+  len= my_fcvt(num, decimals, buff, NULL);
+  return copy(buff, (uint32) len, &my_charset_latin1, cs,
               &dummy_errors);
-#endif
 }
 
 
@@ -675,7 +608,8 @@ void String::qs_append(const char *str, uint32 len)
 void String::qs_append(double d)
 {
   char *buff = Ptr + str_length;
-  str_length+= my_sprintf(buff, (buff, "%.15g", d));
+  str_length+= my_gcvt(d, MY_GCVT_ARG_DOUBLE, FLOATING_POINT_BUFFER - 1, buff,
+                       NULL);
 }
 
 void String::qs_append(double *d)