Nyaaori/rockbox
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
rockbox/firmware/common/vuprintf.c

1441 lines
44 KiB
C
Raw Normal View History

Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
/***************************************************************************
* __________ __ ___.
* Open \______ \ ____ ____ | | _\_ |__ _______ ___
* Source | _// _ \_/ ___\| |/ /| __ \ / _ \ \/ /
* Jukebox | | ( <_> ) \___| < | \_\ ( <_> > < <
* Firmware |____|_ /\____/ \___ >__|_ \|___ /\____/__/\_ \
* \/ \/ \/ \/ \/
* $Id$
*
* Copyright (C) 2002 by Gary Czvitkovicz
* Copyright (C) 2017 by Michael A. Sevakis
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
* KIND, either express or implied.
*
****************************************************************************/
#include <sys/types.h>
#include <limits.h>
#include <string.h>
Port of Duke Nukem 3D This ports Fabien Sanglard's Chocolate Duke to run on a version of SDL for Rockbox. Change-Id: I8f2c4c78af19de10c1633ed7bb7a997b43256dd9
2017-01-21 20:18:31 +00:00
#include <stddef.h>
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
#include <stdlib.h>
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
#include "system.h"
#include "vuprintf.h"
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
#include "ap_int.h"
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
Don't enable full vuprintf file in bootloader. Saves ~2KB packed space when patching clip+ OF. Change-Id: Ie03714778918bfb4a1087f60534d60ff349a8330
2018-05-22 23:07:57 +00:00
#ifndef BOOTLOADER
Limit float formatting to only targets building Quake The Quake plugin is the only code that actually relies on float formatting. Because Quake only runs on targets with huge memory anyway, limiting their formatting to just those targets will minimize the increased memory use in the core. Change-Id: Icdbe26ec6ede564861cc01ac4add76a12b00ecd5
2019-07-20 03:20:56 +00:00
Enable float formatting in printf Needed for g#3415, and if the comment is to be believed, this needed updating for recent MIPS targets anyway. Just blanket enable everything for all targets with >= 8 MiB of RAM. The only targets with less than this are the Sansa Clip, c200v2, and m200v4, which all have 2 MB of RAM. The added code size is around 4 KiB to 7 KiB, depending on the target. Change-Id: I5773482a13543dabb1f93f713a21f1382a9c5a22
2021-06-11 18:31:12 +00:00
/* Turn everything on if we have enough RAM. */
#if MEMORYSIZE >= 8
# define FMT_LENMOD (0xffffffff)
# define FMT_RADIX (0xffffffff)
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
#endif
Limit float formatting to only targets building Quake The Quake plugin is the only code that actually relies on float formatting. Because Quake only runs on targets with huge memory anyway, limiting their formatting to just those targets will minimize the increased memory use in the core. Change-Id: Icdbe26ec6ede564861cc01ac4add76a12b00ecd5
2019-07-20 03:20:56 +00:00
#endif
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
/* these are the defaults if no other preference is given */
#ifndef FMT_LENMOD
#define FMT_LENMOD (FMT_LENMOD_l | \
FMT_LENMOD_z)
#endif /* FMT_LENMOD */
#ifndef FMT_RADIX
#define FMT_RADIX (FMT_RADIX_c | \
FMT_RADIX_d | \
FMT_RADIX_p | \
FMT_RADIX_s | \
FMT_RADIX_u | \
FMT_RADIX_x)
#endif /* FMT_RADIX */
/** Length modifier and radix flags **/
/* compulsory length modifiers: NONE
* however a compatible 'l' or 'll' must be defined if another requires it */
#define FMT_LENMOD_h 0x001 /* signed/unsigned short (%h<radix>) */
#define FMT_LENMOD_hh 0x002 /* signed/unsigned char (%hh<radix>) */
#define FMT_LENMOD_j 0x004 /* intmax_t/uintmax_t (%j<radix>) */
#define FMT_LENMOD_l 0x008 /* signed/unsigned long (%l<radix>) */
#define FMT_LENMOD_ll 0x010 /* signed/unsigned long long (%ll<radix>) */
#define FMT_LENMOD_t 0x020 /* signed/unsigned ptrdiff_t (%t<radix>) */
#define FMT_LENMOD_z 0x040 /* size_t/ssize_t (%z<radix>) */
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
#if 0
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
#define FMT_LENMOD_L 0x080 /* long double (instead of double) */
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
#else
#define FMT_LENMOD_L 0x000
#endif
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
/* compulsory radixes: c, d, i, u, s */
#define FMT_RADIX_c 0x001 /* single character (%c) */
#define FMT_RADIX_d 0x002 /* signed integer type, decimal (%d %i) */
#define FMT_RADIX_n 0x004 /* bytes output so far (%n) */
#define FMT_RADIX_o 0x008 /* unsigned integer type, octal (%o) */
#define FMT_RADIX_p 0x010 /* pointer (%p %P) */
#define FMT_RADIX_s 0x020 /* string (%s) */
#define FMT_RADIX_u 0x040 /* unsigned integer type, decimal (%u) */
#define FMT_RADIX_x 0x080 /* unsigned integer type, hex (%x %X) */
#define FMT_RADIX_a 0x100 /* hex floating point "[-]0xh.hhhhp<68>d" */
#define FMT_RADIX_e 0x200 /* floating point with exponent "[-]d.ddde<64>dd" */
#define FMT_RADIX_f 0x400 /* floating point "[-]ddd.ddd" */
#define FMT_RADIX_g 0x800 /* floating point exponent or decimal depending
upon value and precision */
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
/* TODO: 'a' 'A' */
#define FMT_RADIX_floats (FMT_RADIX_e|FMT_RADIX_f|FMT_RADIX_g)
#if (FMT_RADIX & FMT_RADIX_floats)
/* Assumes IEEE 754 double-precision, native-endian; replace to parse and init
for some other format */
#define parse_double parse_ieee754_double
#define init_double_chunks init_ieee754_double_chunks
#define format_double_int10 format_ap_int10
#define format_double_frac10 format_ap_frac10
#endif
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
/* avoid defining redundant functions if two or more types can use the same
* something not getting a macro means it gets assigned its own value and
* formatter */
/* l */
#if LONG_MIN == INT_MIN && LONG_MAX == INT_MAX
#define val_ld val_d
#define format_ld format_d
#define branch_fmt_ld branch_fmt_d
#elif !(FMT_LENMOD & FMT_LENMOD_l) /* unique */
#define val_ld
#endif /* LONG_ */
#if ULONG_MAX == UINT_MAX
#define val_lu val_u
#define format_lu format_u
#define branch_fmt_lu branch_fmt_u
#elif !(FMT_LENMOD & FMT_LENMOD_l) /* unique */
#define val_lu
#endif /* ULONG_ */
/* ll */
#if LLONG_MIN == INT_MIN && LLONG_MAX == INT_MAX
#define val_lld val_d
#define format_lld format_d
#define branch_fmt_lld branch_fmt_d
#elif LLONG_MIN == LONG_MIN && LLONG_MAX == LONG_MAX
#define val_lld val_ld
#define format_lld format_ld
#define branch_fmt_lld branch_fmt_ld
#elif !(FMT_LENMOD & FMT_LENMOD_ll) /* unique */
#define val_lld
#endif /* LLONG_ */
#if ULLONG_MAX == UINT_MAX
#define val_llu val_u
#define format_llu format_u
#define branch_fmt_llu branch_fmt_u
#elif ULLONG_MAX == ULONG_MAX
#define val_llu val_lu
#define format_llu format_lu
#define branch_fmt_llu branch_fmt_lu
#elif !(FMT_LENMOD & FMT_LENMOD_ll) /* unique */
#define val_llu
#endif /* ULLONG_ */
/* char/short parameter type promotions */
#define SCHAR_INT_ARG int
#define UCHAR_INT_ARG int
#define SSHRT_INT_ARG int
#if USHRT_MAX == UINT_MAX
#define USHRT_INT_ARG unsigned int
#else
#define USHRT_INT_ARG int
#endif
/* some macros to have conditional work inside macros */
#if (FMT_LENMOD & FMT_LENMOD_l)
#define IF_FMT_LENMOD_l(...) __VA_ARGS__
#else
#define IF_FMT_LENMOD_l(...)
#endif
#if (FMT_LENMOD & FMT_LENMOD_ll)
#define IF_FMT_LENMOD_ll(...) __VA_ARGS__
#else
#define IF_FMT_LENMOD_ll(...)
#endif
#if (FMT_RADIX & FMT_RADIX_o)
#define IF_FMT_RADIX_o(...) __VA_ARGS__
#else
#define IF_FMT_RADIX_o(...)
#endif
#if (FMT_RADIX & FMT_RADIX_x)
#define IF_FMT_RADIX_x(...) __VA_ARGS__
#else
#define IF_FMT_RADIX_x(...)
#endif
/* synthesize multicharacter constant */
#define LENMOD2(cv, ch) \
(((cv) << CHAR_BIT) | (ch))
#define LENMOD_NONE 0
#if (FMT_LENMOD & FMT_LENMOD_h)
#define LENMOD_h 'h'
#endif
#if (FMT_LENMOD & FMT_LENMOD_hh)
#define LENMOD_hh LENMOD2('h', 'h') /* 'hh' */
#endif
#if (FMT_LENMOD & FMT_LENMOD_j)
#define LENMOD_j 'j'
#endif
#if (FMT_LENMOD & FMT_LENMOD_l)
#define LENMOD_l 'l'
#endif
#if (FMT_LENMOD & FMT_LENMOD_ll)
#undef FMT_MAX_L
#define LENMOD_ll LENMOD2('l', 'l') /* 'll' */
#endif
#if (FMT_LENMOD & FMT_LENMOD_t)
#define LENMOD_t 't'
#endif
#if (FMT_LENMOD & FMT_LENMOD_z)
#define LENMOD_z 'z'
#endif
/* select type-compatible length modifier
* (a bit hacky; it should be range-based) */
#define LENMOD_INTCOMPAT_SEL(type, signd) \
({ int __lenmod; \
size_t __size = sizeof (type); \
if (__size == ((signd) ? sizeof (int) : \
sizeof (unsigned int))) { \
__lenmod = LENMOD_NONE; \
} \
else if (__size == ((signd) ? sizeof (long) : \
sizeof (unsigned long))) { \
IF_FMT_LENMOD_l(__lenmod = LENMOD_l;) \
} \
else if (__size == ((signd) ? sizeof (long long) : \
sizeof (unsigned long long))) { \
IF_FMT_LENMOD_ll(__lenmod = LENMOD_ll;) \
} \
__lenmod; })
/* call formatting function for the compatible integer type */
#define LENMOD_INTCOMPAT_CALL(inteqv, val, fmt_buf, x, signd) \
({ const char *__buf; \
switch (inteqv) { \
case LENMOD_NONE: \
__buf = (signd) ? \
format_d((val), (fmt_buf), (x)) : \
format_u((val), (fmt_buf), (x)); \
break; \
IF_FMT_LENMOD_l( \
case LENMOD_l: \
__buf = (signd) ? \
format_ld((val), (fmt_buf), (x)) : \
format_lu((val), (fmt_buf), (x)); \
break; \
) \
IF_FMT_LENMOD_ll( \
case LENMOD_ll: \
__buf = (signd) ? \
format_lld((val), (fmt_buf), (x)) : \
format_llu((val), (fmt_buf), (x)); \
break; \
) \
} \
__buf; \
})
/* execute formatting branch for the compatible integer type */
#define LENMOD_INTCOMPAT_BRANCH(inteqv, val, signd) \
({ switch (inteqv) { \
case LENMOD_NONE: \
if (signd) { \
val_d = (val); \
goto branch_fmt_d; \
} \
else { \
val_u = (val); \
goto branch_fmt_u; \
} \
IF_FMT_LENMOD_l( \
case LENMOD_l: \
if (signd) { \
val_ld = (val); \
goto branch_fmt_ld; \
} \
else { \
val_lu = (val); \
goto branch_fmt_lu; \
} \
) \
IF_FMT_LENMOD_ll( \
case LENMOD_ll: \
if (signd) { \
val_lld = (val); \
goto branch_fmt_lld; \
} \
else { \
val_llu = (val); \
goto branch_fmt_llu; \
} \
) \
} \
})
#define CONVERT_RADIX_10_SIGN(val, fmt_buf, p, signchar, type) \
do { \
if (val) { \
unsigned type v; \
\
if (val < 0) { \
v = (typeof (v))-(val + 1) + 1; \
signchar = '-'; \
} \
else { \
v = val; \
} \
\
do { \
*--p = (v % 10) + '0'; \
v /= 10; \
} while (v); \
} \
\
if (signchar) { \
p[-1] = signchar; \
fmt_buf->length = 1; \
break; \
} \
\
fmt_buf->length = 0; \
} while (0)
#define CONVERT_RADIX_8(val, fmt_buf, p) \
do { \
if (val) { \
typeof (val) v = val; \
\
do { \
*--p = (v % 010) + '0'; \
v /= 010; \
} while (v); \
} \
\
if (fmt_buf->length) { \
*--p = '0'; \
fmt_buf->length = 0; \
} \
} while (0)
#define CONVERT_RADIX_10(val, fmt_buf, p) \
do { \
if (val) { \
typeof (val) v = val; \
\
do { \
*--p = (v % 10) + '0'; \
v /= 10; \
} while (v); \
} \
\
fmt_buf->length = 0; \
} while (0)
#define CONVERT_RADIX_16(val, fmt_buf, p, x) \
do { \
if (val) { \
const int h = x - 'X' - 0xA \
+ 'A' - '0'; \
typeof (val) v = val; \
\
do { \
unsigned int d = v % 0x10; \
if (d >= 0xA) { \
d += h; \
} \
*--p = d + '0'; \
v /= 0x10; \
} while (v); \
\
if (fmt_buf->length) { \
p[-1] = x; \
p[-2] = '0'; \
fmt_buf->length = 2; \
break; \
} \
} \
\
fmt_buf->length = 0; \
} while (0)
#define CONVERT_RADIX_NOSIGN(val, fmt_buf, p, x) \
switch (x) \
{ \
IF_FMT_RADIX_o( case 'o': \
CONVERT_RADIX_8(val, fmt_buf, p); \
break; ) \
case 'u': \
CONVERT_RADIX_10(val, fmt_buf, p); \
break; \
IF_FMT_RADIX_x( default: \
CONVERT_RADIX_16(val, fmt_buf, p, x); \
break; ) \
}
struct fmt_buf {
const char *fmt_start; /* second character of formatter after '%' */
size_t length; /* length of formatted text (non-numeric)
or prefix (numeric) */
char buf[24]; /* work buffer */
char bufend[1]; /* buffer end marker and guard '0' */
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
#if (FMT_RADIX & FMT_RADIX_floats)
int lenmod;
int radixchar;
int signchar;
int alignchar;
int width;
int precision;
char *p;
#endif
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
};
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
#define PUSHCHAR(ch) \
({ int __rc = push(userp, (ch)); \
count += __rc >= 0; \
if (__rc <= 0) { \
goto done; \
} })
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
/* %d %i */
static inline const char * format_d(int val,
struct fmt_buf *fmt_buf,
int signchar)
{
char *p = fmt_buf->bufend;
CONVERT_RADIX_10_SIGN(val, fmt_buf, p, signchar, int);
return p;
}
/* %o %u %x %X */
static inline const char * format_u(unsigned int val,
struct fmt_buf *fmt_buf,
int radixchar)
{
char *p = fmt_buf->bufend;
CONVERT_RADIX_NOSIGN(val, fmt_buf, p, radixchar);
return p;
}
#if (FMT_LENMOD & FMT_LENMOD_l)
#ifndef format_ld
/* %ld %li */
static inline const char * format_ld(long val,
struct fmt_buf *fmt_buf,
int signchar)
{
char *p = fmt_buf->bufend;
CONVERT_RADIX_10_SIGN(val, fmt_buf, p, signchar, long);
return p;
}
#endif /* format_ld */
#ifndef format_lu
/* %lo %lu %lx %lX */
static inline const char * format_lu(unsigned long val,
struct fmt_buf *fmt_buf,
int radixchar)
{
char *p = fmt_buf->bufend;
CONVERT_RADIX_NOSIGN(val, fmt_buf, p, radixchar);
return p;
}
#endif /* format_lu */
#endif /* FMT_LENMOD_l */
#if (FMT_LENMOD & FMT_LENMOD_ll)
#ifndef format_lld
/* %lld %lli */
static inline const char * format_lld(long long val,
struct fmt_buf *fmt_buf,
int signchar)
{
char *p = fmt_buf->bufend;
CONVERT_RADIX_10_SIGN(val, fmt_buf, p, signchar, long long);
return p;
}
#endif /* format_lld */
#ifndef format_llu
/* %llo %llu %llx %llX */
static inline const char * format_llu(unsigned long long val,
struct fmt_buf *fmt_buf,
int radixchar)
{
char *p = fmt_buf->bufend;
CONVERT_RADIX_NOSIGN(val, fmt_buf, p, radixchar);
return p;
}
#endif /* format_llu */
#endif /* FMT_LENMOD_ll */
/* %c */
static inline const char * format_c(int c,
struct fmt_buf *fmt_buf,
int lenmod)
{
if (lenmod != LENMOD_NONE) {
return NULL; /* wchar_t support for now */
}
char *p = fmt_buf->bufend;
fmt_buf->length = 1;
*--p = (unsigned char)c;
return p;
}
/* %s */
static inline const char * format_s(const void *str,
struct fmt_buf *fmt_buf,
int precision,
int lenmod)
{
if (lenmod != LENMOD_NONE) {
return NULL; /* wchar_t support for now */
}
Fix vuprintf fix possible %s buffer over-read when precision is not specified memchr recieved -1 for count count is unsigned so it looks in a potentially very large area for a terminator and returns this whole area if \0 is not found Instead we should use memchr when precision is specified and if precision is not specified use strlen Fixes 60+Mb Config.cfg files Change-Id: Ic4d1439334588f999c9071235430c42df2af5cc4
2018-11-06 17:33:38 +00:00
const char *s = str;
size_t len;
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
/* string length may be specified by precision instead of \0-
terminated; however, don't go past a \0 if one is there */
Fix vuprintf fix possible %s buffer over-read when precision is not specified memchr recieved -1 for count count is unsigned so it looks in a potentially very large area for a terminator and returns this whole area if \0 is not found Instead we should use memchr when precision is specified and if precision is not specified use strlen Fixes 60+Mb Config.cfg files Change-Id: Ic4d1439334588f999c9071235430c42df2af5cc4
2018-11-06 17:33:38 +00:00
if (precision >= 0) {
const char *nil = memchr(s, '\0', (size_t) precision);
if (nil != NULL && (nil - s) < precision)
len = nil - s;
else
len = precision;
}
else
len = strlen(s);
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
fmt_buf->length = len;
return s;
}
#if (FMT_RADIX & FMT_RADIX_n)
/* %n */
static inline bool format_n(void *np,
int count,
int lenmod)
{
if (lenmod != LENMOD_NONE) {
return false; /* int only for now */
}
*(int *)np = count;
return true;
}
#endif /* FMT_RADIX_n */
#if (FMT_RADIX & FMT_RADIX_p)
/* %p %P */
static inline const char * format_p(const void *p,
struct fmt_buf *fmt_buf,
int radixchar,
bool *numericp)
{
if (p) {
/* format as %#x or %#X */
*numericp = true;
radixchar -= 'P' - 'X';
fmt_buf->length = 2;
return LENMOD_INTCOMPAT_CALL(LENMOD_INTCOMPAT_SEL(uintptr_t, false),
(uintptr_t)p, fmt_buf, radixchar, false);
}
else {
/* format as %s */
fmt_buf->length = 5;
return "(nil)";
}
}
#endif /* FMT_RADIX_p */
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
#if (FMT_RADIX & FMT_RADIX_floats)
/* find out how many uint32_t chunks need to be allocated, if any
* if none are needed, finish the init for the number here */
static long parse_ieee754_double(double f,
struct ap_int *ia,
struct ap_int *fa,
struct fmt_buf *fmt_buf)
{
long rc = 0;
union {
double f;
uint64_t f64;
} u = { .f = f };
int e = ((int)(u.f64 >> 52) & 0x7ff) - 1023; /* -1023..1024 */
uint64_t mantissa = u.f64 & 0x000fffffffffffffull;
if (u.f64 >> 63) {
fmt_buf->signchar = '-';
}
if (LIKELY(e >= -8 && e <= 63)) { /* -8 to +63 */
/* integer, fraction and manipulations fit in uint64_t */
mantissa |= 0x0010000000000000ull;
ia->numchunks = 0;
ia->shift = 0;
fa->numchunks = 0;
if (e < 0) { /* -8 to -1 - fraction */
long fracbits = 52 - e;
/* int - none */
ia->len = 0;
ia->val = 0;
/* frac */
fa->len = fracbits - __builtin_ctzll(mantissa);
fa->shift = fracbits;
fa->val = mantissa;
}
else if (e <= 51) { /* 0 to +51 - integer|fraction */
long fracbits = 52 - e;
/* int */
ia->len = base10exp(e) + 2; /* go up + possibly 1 longer */
ia->val = mantissa >> fracbits;
/* frac */
fa->shift = fracbits;
fa->val = mantissa ^ (ia->val << fracbits);
fa->len = fa->val ? fracbits - __builtin_ctzll(mantissa) : 0;
}
else { /* +52 to +63 - integer */
/* int */
ia->len = base10exp(e) + 2;
ia->val = mantissa << (e - 52);
/* frac - none */
fa->len = 0;
fa->shift = 0;
fa->val = 0;
}
}
else if (e < 0) { /* -1023 to -9 - fraction */
/* int - none */
ia->numchunks = 0;
ia->len = 0;
ia->shift = 0;
ia->val = 0;
/* frac - left-justify on bit 31 of the chunk of the MSb */
if (e >= -1022) { /* normal */
mantissa |= 0x0010000000000000ull;
}
else { /* subnormal (including zero) */
e = -1022;
}
if (mantissa) {
long fracbits = 52 - e;
fa->len = fracbits - __builtin_ctzll(mantissa);
fa->shift = 31 - ((51 - e) % 32);
fa->val = mantissa;
fa->basechunk = (fa->shift + 52) / 32;
fa->numchunks = (51 - e + fa->shift) / 32 + 1;
rc = fa->numchunks;
}
else { /* zero */
fa->numchunks = 0;
fa->len = 0;
fa->shift = 0;
fa->val = 0;
}
}
else if (e <= 1023) { /* +64 to +1023 - integer */
/* int - right-justify on bit 0 of the first chunk */
ia->val = mantissa | 0x0010000000000000ull;
ia->len = base10exp(e) + 2;
ia->shift = (e - 52) % 32;
ia->basechunk = e / 32;
ia->numchunks = ia->basechunk + 1;
rc = ia->numchunks;
/* frac - none */
fa->numchunks = 0;
fa->len = 0;
fa->shift = 0;
fa->val = 0;
}
else { /* +1024: INF, NAN */
rc = -1 - !!mantissa;
}
return rc;
}
/* construct the arbitrary-precision value in the provided allocation */
static void init_ieee754_double_chunks(struct ap_int *a,
uint32_t *a_chunks)
{
long basechunk = a->basechunk;
long shift = a->shift;
uint64_t val = a->val;
a->chunks = a_chunks;
memset(a_chunks, 0, a->numchunks*sizeof (uint32_t));
if (shift < 12) {
a_chunks[basechunk - 1] = val << shift;
a_chunks[basechunk - 0] = val >> (32 - shift);
}
else {
a_chunks[basechunk - 2] = val << shift;
a_chunks[basechunk - 1] = val >> (32 - shift);
a_chunks[basechunk - 0] = val >> (64 - shift);
}
}
/* format inf, nan strings */
static void format_inf_nan(struct fmt_buf *fmt_buf, long type)
{
/* certain special values */
static const char text[2][2][3] =
{
{ { 'I', 'N', 'F' }, { 'i', 'n', 'f' } },
{ { 'N', 'A', 'N' }, { 'n', 'a', 'n' } },
};
char *p = fmt_buf->buf;
fmt_buf->p = p;
fmt_buf->length = 3;
/* they also have a sign */
if (fmt_buf->signchar) {
*p++ = fmt_buf->signchar;
fmt_buf->length++;
}
memcpy(p, &text[type][(fmt_buf->radixchar >> 5) & 0x1], 3);
}
/* %e %E %f %F %g %G */
static int format_double_radix(double f,
struct fmt_buf *fmt_buf,
vuprintf_push_cb push,
void *userp)
{
struct ap_int ia, fa;
long rc = parse_double(f, &ia, &fa, fmt_buf);
if (UNLIKELY(rc < 0)) {
format_inf_nan(fmt_buf, -rc - 1);
return 0;
}
int count = 0;
/* default precision is 6 for all formats */
int prec_rem = fmt_buf->precision < 0 ? 6 : fmt_buf->precision;
int exp = exp;
int explen = 0;
switch (fmt_buf->radixchar & 0x3)
{
case 3: /* %g, %G */
fmt_buf->precision = prec_rem;
if (prec_rem) {
prec_rem--;
}
Document intentional fallthroughs + fix harmless unintended ones Change-Id: I1ca5b1027ec30cbf61093bab35b980196ed14e6b
2021-07-24 14:39:01 +00:00
/* fallthrough */
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
case 1: /* %e, %E */
explen = 2;
break;
default:
break;
}
if (rc > 0 && ia.numchunks > 0) {
/* large integer required */
init_double_chunks(&ia, alloca(rc*sizeof(*ia.chunks)));
rc = 0;
}
const int bufoffs = 6; /* log rollover + round rollover + leading zeros (%g) */
long f_prec = MIN(fa.len, prec_rem + 1);
char buf[bufoffs + ia.len + f_prec + 1];
char *p_last = &buf[bufoffs + ia.len];
char *p_dec = p_last;
char *p_first = format_double_int10(&ia, p_last);
if (explen) {
if (!ia.val && fa.len) {
p_first = p_last = &buf[bufoffs];
f_prec = -f_prec - 1; /* no lead zeros */
}
else { /* handles 0e+0 too */
exp = ia.len - 1;
if (exp) {
prec_rem -= exp;
if (prec_rem < 0) {
p_last += prec_rem + 1;
f_prec = 0;
}
else {
f_prec = MIN(fa.len, prec_rem + 1);
}
}
}
p_dec = p_first + 1;
}
if (f_prec) {
if (rc > 0) {
/* large integer required */
init_double_chunks(&fa, alloca(rc*sizeof(*fa.chunks)));
}
p_last = format_double_frac10(&fa, p_last, f_prec);
if (f_prec < 0) {
f_prec = -f_prec - 1;
exp = f_prec - fa.len;
}
prec_rem -= f_prec;
}
if (prec_rem < 0) {
prec_rem = 0;
p_last--;
if (round_number_string10(p_last, p_last - p_first)) {
/* carried left */
p_first--;
if (explen) {
/* slide everything left by 1 */
exp++;
p_dec--;
p_last--;
}
}
}
if (explen) {
if ((fmt_buf->radixchar & 0x3) == 0x3) { /* g, G */
/* 'g' is some weird crap */
/* now that the final exponent is known and everything rounded,
it is possible to decide whether to format similarly to
'e' or 'f' */
if (fmt_buf->precision > exp && exp >= -4) { /* P > X >= -4 */
if (exp >= 0) {
/* integer digits will be in the buffer */
p_dec = p_first + exp + 1;
}
else {
/* we didn't keep leading zeros and need to regenerate
them; space was reserved just in case */
p_first = memset(p_dec + exp - 1, '0', -exp);
p_dec = p_first + 1;
}
/* suppress exponent */
explen = 0;
}
if (!fmt_buf->length) {
/* strip any trailing zeros from the fraction */
while (p_last > p_dec && p_last[-1] == '0') {
p_last--;
}
/* suppress trailing precision fill */
prec_rem = 0;
}
}
if (explen) {
/* build exponent string: 'e<>dd' */
char *p = fmt_buf->bufend;
int signchar = '+';
if (exp < 0) {
signchar = '-';
exp = -exp;
}
while (exp || explen < 4) {
*--p = exp % 10 + '0';
exp /= 10;
explen++;
}
*--p = signchar;
*--p = fmt_buf->radixchar & ~0x2;
}
}
int width = fmt_buf->width;
int point = p_last > p_dec || prec_rem || fmt_buf->length;
int length = p_last - p_first + !!fmt_buf->signchar + point + explen;
if (width) {
if (width - length <= prec_rem) {
width = 0;
}
else {
width -= length + prec_rem;
}
}
rc = -1;
/* left padding */
if (fmt_buf->alignchar > '0') {
/* space-padded width -- before sign */
while (width > 0) {
PUSHCHAR(' ');
width--;
}
}
if (fmt_buf->signchar) {
PUSHCHAR(fmt_buf->signchar);
}
if (fmt_buf->alignchar == '0') {
/* zero-padded width -- after sign */
while (width > 0) {
PUSHCHAR('0');
width--;
}
}
/* integer part */
while (p_first < p_dec) {
PUSHCHAR(*p_first++);
}
/* decimal point */
if (point) {
PUSHCHAR('.');
}
/* fractional part */
while (p_first < p_last) {
PUSHCHAR(*p_first++);
}
/* precision 0-padding */
while (prec_rem > 0) {
PUSHCHAR('0');
prec_rem--;
}
/* exponent */
if (explen > 0) {
char *p = fmt_buf->bufend;
while (explen > 0) {
PUSHCHAR(p[-explen--]);
}
}
/* right padding */
while (width > 0) {
PUSHCHAR(' ');
width--;
}
rc = 1;
done:
fmt_buf->length = count;
return rc;
}
#endif /* FMT_RADIX_floats */
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
/* parse fixed width or precision field */
static const char * parse_number_spec(const char *fmt,
int ch,
int *out)
{
int i = ch - '0';
while (1) {
ch = *fmt - '0';
if (ch < 0 || ch > 9 || i > INT_MAX / 10 ||
(i == INT_MAX / 10 && ch > INT_MAX % 10)) {
break;
}
i = i * 10 + ch;
fmt++;
}
*out = i;
return fmt;
}
int vuprintf(vuprintf_push_cb push, /* call 'push()' for each output letter */
void *userp,
const char *fmt,
va_list ap)
{
int count = 0;
int ch;
/* macrofied identifiers share a variable with another */
unsigned int val_d;
unsigned int val_u;
#ifndef val_ld
unsigned long val_ld;
#endif
#ifndef val_lu
unsigned long val_lu;
#endif
#ifndef val_lld
unsigned long long val_lld;
#endif
#ifndef val_llu
unsigned long long val_llu;
#endif
struct fmt_buf fmt_buf;
fmt_buf.bufend[0] = '0';
while (1) {
while (1) {
if ((ch = *fmt++) == '\0') {
goto done;
}
if (ch == '%' && (ch = *fmt++) != '%') {
break;
}
PUSHCHAR(ch);
}
/* set to defaults */
fmt_buf.fmt_start = fmt;
int signchar = 0;
unsigned int width = 0;
int lenmod = LENMOD_NONE;
size_t length = 0;
size_t pfxlen = 0;
bool numeric = false;
int alignchar = '0' + 1;
int precision = -1;
const char *buf = NULL;
/*** flags ***/
while (1) {
switch (ch)
{
case ' ': /* <space> before non-negative value (signed conversion) */
case '+': /* '+' before non-negative value (signed conversion) */
/* '+' overrides ' ' */
if (ch > signchar) {
signchar = ch;
}
break;
case '-': /* left-justify in field */
case '0': /* zero-pad to fill field */
/* '-' overrides '0' */
if (ch < alignchar) {
alignchar = ch;
}
break;
case '#': /* number prefix (nonzero %o:'0' %x/%X:'0x') */
/* indicate; formatter updates with actual length */
pfxlen = 1;
break;
#if 0
case '\'': /* digit grouping (non-monetary) */
break;
#endif
default:
goto flags_done;
}
ch = *fmt++;
}
flags_done:
/*** width ***/
if (ch == '*') {
/* variable width */
int w = va_arg(ap, int);
if (w < 0) {
/* negative width is width with implied '-' */
width = (unsigned int)-(w + 1) + 1;
alignchar = '-';
}
else {
width = w;
}
ch = *fmt++;
}
else if (ch >= '1' && ch <= '9') {
/* fixed width */
fmt = parse_number_spec(fmt, ch, &width);
ch = *fmt++;
}
/*** precision ***/
if (ch == '.') {
ch = *fmt++;
if (ch == '*') {
/* variable precision; negative precision is ignored */
precision = va_arg (ap, int);
ch = *fmt++;
}
else if (ch >= '0' && ch <= '9') {
/* fixed precision */
fmt = parse_number_spec(fmt, ch, &precision);
ch = *fmt++;
}
}
/*** length modifier ***/
#if FMT_LENMOD
switch (ch)
{
#if (FMT_LENMOD & (FMT_LENMOD_h | FMT_LENMOD_hh))
case 'h':
#endif
#if (FMT_LENMOD & FMT_LENMOD_j)
case 'j':
#endif
#if (FMT_LENMOD & (FMT_LENMOD_l | FMT_LENMOD_ll))
case 'l':
#endif
#if (FMT_LENMOD & FMT_LENMOD_t)
case 't':
#endif
#if (FMT_LENMOD & FMT_LENMOD_z)
case 'z':
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
#endif
#if (FMT_LENMOD & FMT_LENMOD_L)
case 'L':
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
#endif
lenmod = ch;
ch = *fmt++;
#if (FMT_LENMOD & (FMT_LENMOD_hh | FMT_LENMOD_ll))
/* doesn't matter if jj, tt or zz happen; they will be rejected
by the radix handler */
if (ch == lenmod) {
lenmod = LENMOD2(lenmod, ch);
ch = *fmt++;
}
#endif
}
#endif /* FMT_LENMOD */
/*** radix ***/
switch (ch)
{
/** non-numeric **/
case 'c':
buf = format_c(va_arg(ap, int), &fmt_buf, lenmod);
break;
#if (FMT_RADIX & FMT_RADIX_n)
case 'n':
if (format_n(va_arg(ap, void *), count, lenmod)) {
continue; /* no output */
}
break;
#endif
case 's':
buf = format_s(va_arg(ap, const void *), &fmt_buf,
precision, lenmod);
break;
/** non-integer **/
#if (FMT_RADIX & FMT_RADIX_p)
case 'p':
case 'P':
buf = format_p(va_arg(ap, void *), &fmt_buf, ch,
&numeric);
break;
#endif
Add proper float formatting to vuprintf Wanted to see how gnarly it is to do. Big number handling could be done with better algorithms since it can get a bit slow with large integers or tiny fractions with many lead zeros when only a few digits are needed. Anyway, it supports %e, %E, %f, %F, %g and %G. No %a or long double support seems warranted at the moment. Assumes IEEE 754 double format but it's laid out to be able to replace a function to handle others if needed. Tested in a driver program that has a duplicate vuprintf and the content was pasted in once it looked sound enough to put up a patch. Change-Id: I6dae8624d3208e644c88e36e6a17d8fc9144f988
2018-01-06 12:17:04 +00:00
#if (FMT_RADIX & FMT_RADIX_floats)
/* any floats gets all of them (except with 'L' and %a, %A for now) */
case 'e':
case 'E':
case 'f':
case 'F':
case 'g':
case 'G':
/* LENMOD_L isn't supported for now and will be rejected automatically */
/* floating point has very different spec interpretations to other
formats and requires special handling */
fmt_buf.length = pfxlen;
fmt_buf.lenmod = lenmod;
fmt_buf.radixchar = ch;
fmt_buf.signchar = signchar;
fmt_buf.alignchar = alignchar;
fmt_buf.width = width;
fmt_buf.precision = precision;
ch = format_double_radix(va_arg(ap, double), &fmt_buf, push, userp);
if (ch) {
count += fmt_buf.length;
if (ch > 0) {
continue;
}
goto done;
}
buf = fmt_buf.p;
break;
#endif
Implement a much more capable vuprintf() New support as well as some buggy support fixed. Still no floating point support if ever that would be desired. Support (*): * Flags: '-', '+', ' ', '#', '0' * Width and precision: 'n', '.n', '*' and '.*' * Length modifiers: 'hh', 'h', 'j', 'l', 'll', 't', 'z' * Radix: 'c', 'd', 'i', 'n', 'o', 'p/P', 's', 'u', 'x/X' (*) Provision exists to switch lesser-used stuff on or off or when certain functionality isn't desired (bootloader?). The compulsory radixes are everything but 'o', 'n', 'p/P' and 'x/X' with length modifiers being optional. The default setup is 'l', 'z', 'c', 'd', 'p/P', 's', 'u', 'x/X'. * Move fdprintf() to its own file. It was in a strange place. * Make callers compatible and fix a couple snprintf() bugs while at it. Could smush it down in size but I'm gonna get over the binsize neurosis and just the let optimizer do its thing. Change-Id: Ibdc613a9b6775802c188b29b9dd46c568c94f7c3
2017-09-08 23:28:02 +00:00
/** signed integer **/
case 'd':
case 'i':
fmt_buf.length = pfxlen;
switch (lenmod)
{
case LENMOD_NONE:
val_d = va_arg(ap, signed int);
goto branch_fmt_d;
#if (FMT_LENMOD & FMT_LENMOD_h)
case LENMOD_h:
val_d = (signed short)va_arg(ap, SSHRT_INT_ARG);
goto branch_fmt_d;
#endif
#if (FMT_LENMOD & FMT_LENMOD_hh)
case LENMOD_hh:
val_d = (signed char)va_arg(ap, SCHAR_INT_ARG);
goto branch_fmt_d;
#endif
#if (FMT_LENMOD & FMT_LENMOD_j)
case LENMOD_j:
LENMOD_INTCOMPAT_BRANCH(LENMOD_INTCOMPAT_SEL(intmax_t, true),
va_arg(ap, intmax_t), true);
#endif
#if (FMT_LENMOD & FMT_LENMOD_l)
case LENMOD_l:
val_ld = va_arg(ap, signed long);
goto branch_fmt_ld;
#endif
#if (FMT_LENMOD & FMT_LENMOD_ll)
case LENMOD_ll:
val_lld = va_arg(ap, signed long long);
goto branch_fmt_lld;
#endif
#if (FMT_LENMOD & FMT_LENMOD_t)
case LENMOD_t:
LENMOD_INTCOMPAT_BRANCH(LENMOD_INTCOMPAT_SEL(ptrdiff_t, true),
va_arg(ap, ptrdiff_t), true);
#endif
#if (FMT_LENMOD & FMT_LENMOD_z)
case LENMOD_z:
LENMOD_INTCOMPAT_BRANCH(LENMOD_INTCOMPAT_SEL(ssize_t, true),
va_arg(ap, ssize_t), true);
#endif
}
/* macrofied labels share a formatter with another */
if (0) {
branch_fmt_d:
buf = format_d(val_d, &fmt_buf, signchar);
} else if (0) {
#ifndef val_ld
branch_fmt_ld:
buf = format_ld(val_ld, &fmt_buf, signchar);
#endif
} else if (0) {
#ifndef val_lld
branch_fmt_lld:
buf = format_lld(val_lld, &fmt_buf, signchar);
#endif
}
numeric = true;
break;
/** unsigned integer **/
#if (FMT_RADIX & FMT_RADIX_o)
case 'o':
#endif
case 'u':
#if (FMT_RADIX & FMT_RADIX_x)
case 'x':
case 'X':
#endif
fmt_buf.length = pfxlen;
switch (lenmod)
{
case LENMOD_NONE:
val_u = va_arg(ap, unsigned int);
goto branch_fmt_u;
#if (FMT_LENMOD & FMT_LENMOD_h)
case LENMOD_h:
val_u = (unsigned short)va_arg(ap, USHRT_INT_ARG);
goto branch_fmt_u;
#endif
#if (FMT_LENMOD & FMT_LENMOD_hh)
case LENMOD_hh:
val_u = (unsigned char)va_arg(ap, UCHAR_INT_ARG);
goto branch_fmt_u;
#endif
#if (FMT_LENMOD & FMT_LENMOD_j)
case LENMOD_j:
LENMOD_INTCOMPAT_BRANCH(LENMOD_INTCOMPAT_SEL(uintmax_t, false),
va_arg(ap, uintmax_t), false);
#endif
#if (FMT_LENMOD & FMT_LENMOD_l)
case LENMOD_l:
val_lu = va_arg(ap, unsigned long);
goto branch_fmt_lu;
#endif
#if (FMT_LENMOD & FMT_LENMOD_ll)
case LENMOD_ll:
val_llu = va_arg(ap, unsigned long long);
goto branch_fmt_llu;
#endif
#if (FMT_LENMOD & (FMT_LENMOD_t | FMT_LENMOD_z))
/* format "uptrdiff_t" as size_t (unless it becomes standard) */
#if (FMT_LENMOD & FMT_LENMOD_t)
case LENMOD_t:
#endif
#if (FMT_LENMOD & FMT_LENMOD_z)
case LENMOD_z:
#endif
LENMOD_INTCOMPAT_BRANCH(LENMOD_INTCOMPAT_SEL(size_t, false),
va_arg(ap, size_t), false);
#endif
}
/* macrofied labels share a formatter with another */
if (0) {
branch_fmt_u:
buf = format_u(val_u, &fmt_buf, ch);
} else if (0) {
#ifndef val_lu
branch_fmt_lu:
buf = format_lu(val_lu, &fmt_buf, ch);
#endif
} else if (0) {
#ifndef val_llu
branch_fmt_llu:
buf = format_llu(val_llu, &fmt_buf, ch);
#endif
}
numeric = true;
break;
}
if (buf) {
/** padding **/
if (numeric) {
/* numeric formats into fmt_buf.buf */
pfxlen = fmt_buf.length;
length = fmt_buf.bufend - buf;
size_t size = pfxlen + length;
if (precision >= 0) {
/* explicit precision */
precision -= (int)length;
if (precision > 0) {
size += precision;
}
width -= MIN(width, size);
}
else {
/* default precision */
if (!length) {
length = 1;
size++;
}
width -= MIN(width, size);
if (alignchar == '0') {
/* width zero-fill */
precision = width;
width = 0;
}
}
}
else {
/* non-numeric: supress prefix and precision; keep length and
width */
pfxlen = 0;
precision = 0;
length = fmt_buf.length;
width -= MIN(width, length);
}
}
else {
/* format not accepted; print it literally */
buf = fmt_buf.fmt_start - 2;
length = fmt - buf;
width = 0;
pfxlen = 0;
precision = 0;
}
/** push all the stuff **/
if (alignchar != '-') {
/* left padding */
while (width > 0) {
PUSHCHAR(' ');
width--;
}
}
/* prefix */
while (pfxlen > 0) {
PUSHCHAR(buf[-pfxlen]);
pfxlen--;
}
/* 0-padding */
while (precision > 0) {
PUSHCHAR('0');
precision--;
}
/* field */
while (length > 0) {
PUSHCHAR(*buf++);
length--;
}
/* right padding */
while (width > 0) {
PUSHCHAR(' ');
width--;
}
}
done:
return count;
}
Reference in a new issue Copy permalink