9
0
Fork 0

Fix a floating point presentation error

git-svn-id: https://nuttx.svn.sourceforge.net/svnroot/nuttx/trunk@5012 7fd9a85b-ad96-42d3-883c-3090e2eb8679
This commit is contained in:
patacongo 2012-08-06 02:14:36 +00:00
parent 55a1ed3e20
commit 171dfd3f7f
3 changed files with 111 additions and 226 deletions

View File

@ -3125,3 +3125,8 @@
* lib/stdio/lib_libvsprintf.c: Fieldwidth and justification were not
supported for the %s format. As a result, %s, %12s, and %-12s all
produced the same output.
* lib/stdio/lib_libdtoa.c: Fix several issues with presenting floating
point numbers (conversions are fine, but presentation was bad). This
is a critical bug fix if you use printf or sprintf to deal with floating
point numbers.

View File

@ -15,7 +15,7 @@ nuttx/
(5) Binary loaders (binfmt/)
(17) Network (net/, drivers/net)
(3) USB (drivers/usbdev, drivers/usbhost)
(9) Libraries (lib/)
(10) Libraries (lib/)
(10) File system/Generic drivers (fs/, drivers/)
(5) Graphics subystem (graphics/)
(1) Pascal add-on (pcode/)
@ -685,6 +685,12 @@ o Libraries (lib/)
Status: Open
Priority: Low -- more of a roadmap
Title: FLOATING POINT FORMATS
Description: Only the %f floating point format is supported. Others are accepted
but treated like %f.
Status: Open
Priority: Medium (this might important to someone.
o File system / Generic drivers (fs/, drivers/)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

View File

@ -48,7 +48,13 @@
* Pre-processor Definitions
****************************************************************************/
#define MAXEXP 308
#ifndef MIN
# define MIN(a,b) (a < b ? a : b)
#endif
#ifndef MAX
# define MAX(a,b) (a > b ? a : b)
#endif
/****************************************************************************
* Private Type Declarations
@ -58,10 +64,6 @@
* Private Function Prototypes
****************************************************************************/
static char* cvt(double value, int ndigits, int flags, char *sign,
int *decpt, int ch, int *length);
static int exponent(char *p0, int exp, int fmtch);
/****************************************************************************
* Global Constant Data
****************************************************************************/
@ -78,269 +80,123 @@ static int exponent(char *p0, int exp, int fmtch);
* Private Variables
****************************************************************************/
/****************************************************************************
* Name: zeroes
*
* Description:
* Print the specified number of zeres
*
****************************************************************************/
static void zeroes(FAR struct lib_outstream_s *obj, int nzeroes)
{
int i;
for (i = nzeroes; i > 0; i--)
{
obj->put(obj, '0');
}
}
/****************************************************************************
* Private Functions
****************************************************************************/
/****************************************************************************
* Name: cvt
****************************************************************************/
static char* cvt(double value, int ndigits, int flags, char *sign,
int *decpt, int ch, int *length)
{
int mode, dsgn;
char *digits, *bp, *rve;
if (ch == 'f')
{
mode = 3; /* ndigits after the decimal point */
}
else
{
/* To obtain ndigits after the decimal point for the 'e' and 'E'
* formats, round to ndigits + 1 significant figures.
*/
if (ch == 'e' || ch == 'E')
{
ndigits++;
}
mode = 2; /* ndigits significant digits */
}
if (value < 0)
{
value = -value;
*sign = '-';
}
else
{
*sign = '\000';
}
digits = __dtoa(value, mode, ndigits, decpt, &dsgn, &rve);
if ((ch != 'g' && ch != 'G') || IS_ALTFORM(flags))
{
/* Print trailing zeros */
bp = digits + ndigits;
if (ch == 'f')
{
if (*digits == '0' && value)
{
*decpt = -ndigits + 1;
}
bp += *decpt;
}
if (value == 0)
{
/* kludge for __dtoa irregularity */
rve = bp;
}
while (rve < bp)
{
*rve++ = '0';
}
}
*length = rve - digits;
return digits;
}
/****************************************************************************
* Name: exponent
****************************************************************************/
static int exponent(FAR char *p0, int exp, int fmtch)
{
FAR char *p;
FAR char *t;
char expbuf[MAXEXP];
p = p0;
*p++ = fmtch;
if (exp < 0)
{
exp = -exp;
*p++ = '-';
}
else
{
*p++ = '+';
}
t = expbuf + MAXEXP;
if (exp > 9)
{
do
{
*--t = (exp % 10) + '0';
}
while ((exp /= 10) > 9);
*--t = exp + '0';
for (; t < expbuf + MAXEXP; *p++ = *t++);
}
else
{
*p++ = '0';
*p++ = exp + '0';
}
return (p - p0);
}
/****************************************************************************
* Name: lib_dtoa
*
* Description:
* This is part of lib_vsprintf(). It handles the floating point formats.
* This version supports only the &f (with precision).
*
****************************************************************************/
static void lib_dtoa(FAR struct lib_outstream_s *obj, int ch, int prec,
uint8_t flags, double _double)
static void lib_dtoa(FAR struct lib_outstream_s *obj, int fmt, int prec,
uint8_t flags, double value)
{
FAR char *cp; /* Handy char pointer (short term usage) */
FAR char *cp_free = NULL; /* BIONIC: copy of cp to be freed after usage */
char expstr[7]; /* Buffer for exponent string */
char sign; /* Temporary negative sign for floats */
int expt; /* Integer value of exponent */
int expsize = 0; /* Character count for expstr */
int ndig; /* Actual number of digits returned by cvt */
int size; /* Size of converted field or string */
FAR char *digits; /* String returned by __dtoa */
FAR char *digalloc; /* Copy of digits to be freed after usage */
FAR char *rve; /* Points to the end of the return value */
char sign; /* Temporary negative sign for floats */
int expt; /* Integer value of exponent */
int numlen; /* Actual number of digits returned by cvt */
int nchars; /* Number of characters to print */
int dsgn; /* Unused sign indicator */
int i;
cp = cvt(_double, prec, flags, &sign, &expt, ch, &ndig);
cp_free = cp;
/* Non-zero... positive or negative */
if (ch == 'g' || ch == 'G')
if (value < 0)
{
/* 'g' or 'G' fmt */
if (expt <= -4 || expt > prec)
{
ch = (ch == 'g') ? 'e' : 'E';
}
else
{
ch = 'g';
}
}
if (ch <= 'e')
{
/* 'e' or 'E' fmt */
--expt;
expsize = exponent(expstr, expt, ch);
size = expsize + ndig;
if (ndig > 1 || IS_ALTFORM(flags))
{
++size;
}
}
else if (ch == 'f')
{
/* f fmt */
if (expt > 0)
{
size = expt;
if (prec || IS_ALTFORM(flags))
{
size += prec + 1;
}
}
else /* "0.X" */
{
size = prec + 2;
}
}
else if (expt >= ndig)
{
/* fixed g fmt */
size = expt;
if (IS_ALTFORM(flags))
{
++size;
}
value = -value;
sign = '-';
}
else
{
size = ndig + (expt > 0 ? 1 : 2 - expt);
sign = '\0';
}
/* Perform the conversion */
digits = __dtoa(value, 3, prec, &expt, &dsgn, &rve);
digalloc = digits;
numlen = rve - digits;
if (sign)
{
obj->put(obj, '-');
}
if (_double == 0)
/* Always print at least one digit to the right of the decimal point. */
prec = MAX(1, prec);
/* Special case exact zero or the case where the number is smaller than
* the print precision.
*/
if (value == 0 || expt < -prec)
{
/* kludge for __dtoa irregularity */
obj->put(obj, '0');
if (expt < ndig || IS_ALTFORM(flags))
{
obj->put(obj, '.');
i = ndig - 1;
while (i > 0)
{
obj->put(obj, '0');
i--;
}
}
obj->put(obj, '.');
}
else if (expt <= 0)
{
obj->put(obj, '0');
obj->put(obj, '.');
i = ndig;
while (i > 0)
/* Print leading zeros */
if (expt < 0)
{
obj->put(obj, *cp);
i--;
cp++;
nchars = MIN(-expt, prec);
zeroes(obj, nchars);
prec -= nchars;
}
}
else if (expt >= ndig)
{
i = ndig;
while (i > 0)
/* Print the significant digits */
nchars = MIN(numlen, prec);
for (i = nchars; i > 0; i--)
{
obj->put(obj, *cp);
i--;
cp++;
obj->put(obj, *digits);
digits++;
}
i = expt - ndig;
while (i > 0)
{
obj->put(obj, '0');
i--;
}
/* Decremnt to get the number of trailing zeroes to print */
if (IS_ALTFORM(flags))
{
obj->put(obj, '.');
}
prec -= nchars;
}
else
{
/* Print the integer */
/* Print the integer part */
i = expt;
while (i > 0)
for (i = expt; i > 0; i--)
{
obj->put(obj, *cp);
i--;
cp++;
obj->put(obj, *digits);
digits++;
}
/* Print the decimal place */
@ -349,14 +205,32 @@ static void lib_dtoa(FAR struct lib_outstream_s *obj, int ch, int prec,
/* Print the decimal */
i = ndig - expt;
while (i > 0)
numlen -= expt;
nchars = MIN(numlen, prec);
for (i = nchars; i > 0; i--)
{
obj->put(obj, *cp);
i--;
cp++;
obj->put(obj, *digits);
digits++;
}
/* Decremnt to get the number of trailing zeroes to print */
prec -= nchars;
}
/* Finally, print any trailing zeroes */
zeroes(obj, prec);
/* Is this memory supposed to be freed or not? */
#if 0
if (digalloc)
{
free(digalloc);
}
#endif
}
/****************************************************************************