diff options
| author | Zack Weinberg <zackw@panix.com> | 2018-03-07 14:31:58 -0500 |
|---|---|---|
| committer | Gabriel F. T. Gomes <gabriel@inconstante.eti.br> | 2018-12-05 18:15:42 -0200 |
| commit | 349718d4d7841df46bcc36df9bc2baef4c40d6f5 (patch) | |
| tree | 1c6ae071a4a6d063e551ae6577b04d952262daea /stdio-common/vfscanf.c | |
| parent | 72b8692d7e640eb85ea0fb7de6d5e797512691c1 (diff) | |
| download | glibc-349718d4d7841df46bcc36df9bc2baef4c40d6f5.tar.xz glibc-349718d4d7841df46bcc36df9bc2baef4c40d6f5.zip | |
Add __vfscanf_internal and __vfwscanf_internal with flags arguments.
There are two flags currently defined: SCANF_LDBL_IS_DBL is the mode
used by __nldbl_ scanf variants, and SCANF_ISOC99_A is the mode used
by __isoc99_ scanf variants. In this patch, the new functions honor
these flag bits if they're set, but they still also look at the
corresponding bits of environmental state, and callers all pass zero.
The new functions do *not* have the "errp" argument possessed by
_IO_vfscanf and _IO_vfwscanf. All internal callers passed NULL for
that argument. External callers could theoretically exist, so I
preserved wrappers, but they are flagged as compat symbols and they
don't preserve the three-way distinction among types of errors that
was formerly exposed. These functions probably should have been in
the list of deprecated _IO_ symbols in 2.27 NEWS -- they're not just
aliases for vfscanf and vfwscanf.
(It was necessary to introduce ldbl_compat_symbol for _IO_vfscanf.
Please check that part of the patch very carefully, I am still not
confident I understand all of the details of ldbl-opt.)
This patch also introduces helper inlines in libio/strfile.h that
encapsulate the process of initializing an _IO_strfile object for
reading. This allows us to call __vfscanf_internal directly from
sscanf, and __vfwscanf_internal directly from swscanf, without
duplicating the initialization code. (Previously, they called their
v-counterparts, but that won't work if we want to control *both* C99
mode and ldbl-is-dbl mode using the flags argument to__vfscanf_internal.)
It's still a little awkward, especially for wide strfiles, but it's
much better than what we had.
Tested for powerpc and powerpc64le.
Diffstat (limited to 'stdio-common/vfscanf.c')
| -rw-r--r-- | stdio-common/vfscanf.c | 3042 |
1 files changed, 1 insertions, 3041 deletions
diff --git a/stdio-common/vfscanf.c b/stdio-common/vfscanf.c index 1ce836a324..5eedca8340 100644 --- a/stdio-common/vfscanf.c +++ b/stdio-common/vfscanf.c @@ -15,3053 +15,13 @@ License along with the GNU C Library; if not, see <http://www.gnu.org/licenses/>. */ -#include <assert.h> -#include <errno.h> -#include <limits.h> -#include <ctype.h> -#include <stdarg.h> -#include <stdbool.h> -#include <stdio.h> -#include <stdint.h> -#include <stdlib.h> -#include <string.h> -#include <wchar.h> -#include <wctype.h> -#include <libc-diag.h> -#include <libc-lock.h> -#include <locale/localeinfo.h> -#include <scratch_buffer.h> - -#ifdef __GNUC__ -# define HAVE_LONGLONG -# define LONGLONG long long -#else -# define LONGLONG long -#endif - -/* Determine whether we have to handle `long long' at all. */ -#if LONG_MAX == LONG_LONG_MAX -# define need_longlong 0 -#else -# define need_longlong 1 -#endif - -/* Determine whether we have to handle `long'. */ -#if INT_MAX == LONG_MAX -# define need_long 0 -#else -# define need_long 1 -#endif - -/* Those are flags in the conversion format. */ -#define LONG 0x0001 /* l: long or double */ -#define LONGDBL 0x0002 /* L: long long or long double */ -#define SHORT 0x0004 /* h: short */ -#define SUPPRESS 0x0008 /* *: suppress assignment */ -#define POINTER 0x0010 /* weird %p pointer (`fake hex') */ -#define NOSKIP 0x0020 /* do not skip blanks */ -#define NUMBER_SIGNED 0x0040 /* signed integer */ -#define GROUP 0x0080 /* ': group numbers */ -#define GNU_MALLOC 0x0100 /* a: malloc strings */ -#define CHAR 0x0200 /* hh: char */ -#define I18N 0x0400 /* I: use locale's digits */ -#define HEXA_FLOAT 0x0800 /* hexadecimal float */ -#define READ_POINTER 0x1000 /* this is a pointer value */ -#define POSIX_MALLOC 0x2000 /* m: malloc strings */ -#define MALLOC (GNU_MALLOC | POSIX_MALLOC) - -#include <locale/localeinfo.h> #include <libioP.h> -#ifdef COMPILE_WSCANF -# define ungetc(c, s) ((void) (c == WEOF \ - || (--read_in, \ - _IO_sputbackwc (s, c)))) -# define ungetc_not_eof(c, s) ((void) (--read_in, \ - _IO_sputbackwc (s, c))) -# define inchar() (c == WEOF ? ((errno = inchar_errno), WEOF) \ - : ((c = _IO_getwc_unlocked (s)), \ - (void) (c != WEOF \ - ? ++read_in \ - : (size_t) (inchar_errno = errno)), c)) - -# define ISSPACE(Ch) iswspace (Ch) -# define ISDIGIT(Ch) iswdigit (Ch) -# define ISXDIGIT(Ch) iswxdigit (Ch) -# define TOLOWER(Ch) towlower (Ch) -# define ORIENT if (_IO_fwide (s, 1) != 1) return WEOF -# define __strtoll_internal __wcstoll_internal -# define __strtoull_internal __wcstoull_internal -# define __strtol_internal __wcstol_internal -# define __strtoul_internal __wcstoul_internal -# define __strtold_internal __wcstold_internal -# define __strtod_internal __wcstod_internal -# define __strtof_internal __wcstof_internal - -# define L_(Str) L##Str -# define CHAR_T wchar_t -# define UCHAR_T unsigned int -# define WINT_T wint_t -# undef EOF -# define EOF WEOF -#else -# define ungetc(c, s) ((void) ((int) c == EOF \ - || (--read_in, \ - _IO_sputbackc (s, (unsigned char) c)))) -# define ungetc_not_eof(c, s) ((void) (--read_in, \ - _IO_sputbackc (s, (unsigned char) c))) -# define inchar() (c == EOF ? ((errno = inchar_errno), EOF) \ - : ((c = _IO_getc_unlocked (s)), \ - (void) (c != EOF \ - ? ++read_in \ - : (size_t) (inchar_errno = errno)), c)) -# define ISSPACE(Ch) __isspace_l (Ch, loc) -# define ISDIGIT(Ch) __isdigit_l (Ch, loc) -# define ISXDIGIT(Ch) __isxdigit_l (Ch, loc) -# define TOLOWER(Ch) __tolower_l ((unsigned char) (Ch), loc) -# define ORIENT if (_IO_vtable_offset (s) == 0 \ - && _IO_fwide (s, -1) != -1) \ - return EOF - -# define L_(Str) Str -# define CHAR_T char -# define UCHAR_T unsigned char -# define WINT_T int -#endif - -#include "printf-parse.h" /* Use read_int. */ - -#define encode_error() do { \ - errval = 4; \ - __set_errno (EILSEQ); \ - goto errout; \ - } while (0) -#define conv_error() do { \ - errval = 2; \ - goto errout; \ - } while (0) -#define input_error() do { \ - errval = 1; \ - if (done == 0) done = EOF; \ - goto errout; \ - } while (0) -#define add_ptr_to_free(ptr) \ - do \ - { \ - if (ptrs_to_free == NULL \ - || ptrs_to_free->count == (sizeof (ptrs_to_free->ptrs) \ - / sizeof (ptrs_to_free->ptrs[0]))) \ - { \ - struct ptrs_to_free *new_ptrs = alloca (sizeof (*ptrs_to_free)); \ - new_ptrs->count = 0; \ - new_ptrs->next = ptrs_to_free; \ - ptrs_to_free = new_ptrs; \ - } \ - ptrs_to_free->ptrs[ptrs_to_free->count++] = (ptr); \ - } \ - while (0) -#define ARGCHECK(s, format) \ - do \ - { \ - /* Check file argument for consistence. */ \ - CHECK_FILE (s, EOF); \ - if (s->_flags & _IO_NO_READS) \ - { \ - __set_errno (EBADF); \ - return EOF; \ - } \ - else if (format == NULL) \ - { \ - __set_errno (EINVAL); \ - return EOF; \ - } \ - } while (0) -#define LOCK_STREAM(S) \ - __libc_cleanup_region_start (1, (void (*) (void *)) &_IO_funlockfile, (S)); \ - _IO_flockfile (S) -#define UNLOCK_STREAM(S) \ - _IO_funlockfile (S); \ - __libc_cleanup_region_end (0) - -struct ptrs_to_free -{ - size_t count; - struct ptrs_to_free *next; - char **ptrs[32]; -}; - -struct char_buffer { - CHAR_T *current; - CHAR_T *end; - struct scratch_buffer scratch; -}; - -/* Returns a pointer to the first CHAR_T object in the buffer. Only - valid if char_buffer_add (BUFFER, CH) has been called and - char_buffer_error (BUFFER) is false. */ -static inline CHAR_T * -char_buffer_start (const struct char_buffer *buffer) -{ - return (CHAR_T *) buffer->scratch.data; -} - -/* Returns the number of CHAR_T objects in the buffer. Only valid if - char_buffer_error (BUFFER) is false. */ -static inline size_t -char_buffer_size (const struct char_buffer *buffer) -{ - return buffer->current - char_buffer_start (buffer); -} - -/* Reinitializes BUFFER->current and BUFFER->end to cover the entire - scratch buffer. */ -static inline void -char_buffer_rewind (struct char_buffer *buffer) -{ - buffer->current = char_buffer_start (buffer); - buffer->end = buffer->current + buffer->scratch.length / sizeof (CHAR_T); -} - -/* Returns true if a previous call to char_buffer_add (BUFFER, CH) - failed. */ -static inline bool -char_buffer_error (const struct char_buffer *buffer) -{ - return __glibc_unlikely (buffer->current == NULL); -} - -/* Slow path for char_buffer_add. */ -static void -char_buffer_add_slow (struct char_buffer *buffer, CHAR_T ch) -{ - if (char_buffer_error (buffer)) - return; - size_t offset = buffer->end - (CHAR_T *) buffer->scratch.data; - if (!scratch_buffer_grow_preserve (&buffer->scratch)) - { - buffer->current = NULL; - buffer->end = NULL; - return; - } - char_buffer_rewind (buffer); - buffer->current += offset; - *buffer->current++ = ch; -} - -/* Adds CH to BUFFER. This function does not report any errors, check - for them with char_buffer_error. */ -static inline void -char_buffer_add (struct char_buffer *buffer, CHAR_T ch) - __attribute__ ((always_inline)); -static inline void -char_buffer_add (struct char_buffer *buffer, CHAR_T ch) -{ - if (__glibc_unlikely (buffer->current == buffer->end)) - char_buffer_add_slow (buffer, ch); - else - *buffer->current++ = ch; -} - -/* Read formatted input from S according to the format string - FORMAT, using the argument list in ARG. - Return the number of assignments made, or -1 for an input error. */ -#ifdef COMPILE_WSCANF -int -_IO_vfwscanf (FILE *s, const wchar_t *format, va_list argptr, - int *errp) -#else -int -_IO_vfscanf_internal (FILE *s, const char *format, va_list argptr, - int *errp) -#endif -{ - va_list arg; - const CHAR_T *f = format; - UCHAR_T fc; /* Current character of the format. */ - WINT_T done = 0; /* Assignments done. */ - size_t read_in = 0; /* Chars read in. */ - WINT_T c = 0; /* Last char read. */ - int width; /* Maximum field width. */ - int flags; /* Modifiers for current format element. */ - int errval = 0; -#ifndef COMPILE_WSCANF - locale_t loc = _NL_CURRENT_LOCALE; - struct __locale_data *const curctype = loc->__locales[LC_CTYPE]; -#endif - - /* Errno of last failed inchar call. */ - int inchar_errno = 0; - /* Status for reading F-P nums. */ - char got_digit, got_dot, got_e, got_sign; - /* If a [...] is a [^...]. */ - CHAR_T not_in; -#define exp_char not_in - /* Base for integral numbers. */ - int base; - /* Decimal point character. */ -#ifdef COMPILE_WSCANF - wint_t decimal; -#else - const char *decimal; -#endif - /* The thousands character of the current locale. */ -#ifdef COMPILE_WSCANF - wint_t thousands; -#else - const char *thousands; -#endif - struct ptrs_to_free *ptrs_to_free = NULL; - /* State for the conversions. */ - mbstate_t state; - /* Integral holding variables. */ - union - { - long long int q; - unsigned long long int uq; - long int l; - unsigned long int ul; - } num; - /* Character-buffer pointer. */ - char *str = NULL; - wchar_t *wstr = NULL; - char **strptr = NULL; - ssize_t strsize = 0; - /* We must not react on white spaces immediately because they can - possibly be matched even if in the input stream no character is - available anymore. */ - int skip_space = 0; - /* Workspace. */ - CHAR_T *tw; /* Temporary pointer. */ - struct char_buffer charbuf; - scratch_buffer_init (&charbuf.scratch); - -#ifdef __va_copy - __va_copy (arg, argptr); -#else - arg = (va_list) argptr; -#endif - -#ifdef ORIENT - ORIENT; -#endif - - ARGCHECK (s, format); - - { -#ifndef COMPILE_WSCANF - struct __locale_data *const curnumeric = loc->__locales[LC_NUMERIC]; -#endif - - /* Figure out the decimal point character. */ -#ifdef COMPILE_WSCANF - decimal = _NL_CURRENT_WORD (LC_NUMERIC, _NL_NUMERIC_DECIMAL_POINT_WC); -#else - decimal = curnumeric->values[_NL_ITEM_INDEX (DECIMAL_POINT)].string; -#endif - /* Figure out the thousands separator character. */ -#ifdef COMPILE_WSCANF - thousands = _NL_CURRENT_WORD (LC_NUMERIC, _NL_NUMERIC_THOUSANDS_SEP_WC); -#else - thousands = curnumeric->values[_NL_ITEM_INDEX (THOUSANDS_SEP)].string; - if (*thousands == '\0') - thousands = NULL; -#endif - } - - /* Lock the stream. */ - LOCK_STREAM (s); - - -#ifndef COMPILE_WSCANF - /* From now on we use `state' to convert the format string. */ - memset (&state, '\0', sizeof (state)); -#endif - - /* Run through the format string. */ - while (*f != '\0') - { - unsigned int argpos; - /* Extract the next argument, which is of type TYPE. - For a %N$... spec, this is the Nth argument from the beginning; - otherwise it is the next argument after the state now in ARG. */ -#ifdef __va_copy -# define ARG(type) (argpos == 0 ? va_arg (arg, type) : \ - ({ unsigned int pos = argpos; \ - va_list arg; \ - __va_copy (arg, argptr); \ - while (--pos > 0) \ - (void) va_arg (arg, void *); \ - va_arg (arg, type); \ - })) -#else -# if 0 - /* XXX Possible optimization. */ -# define ARG(type) (argpos == 0 ? va_arg (arg, type) : \ - ({ va_list arg = (va_list) argptr; \ - arg = (va_list) ((char *) arg \ - + (argpos - 1) \ - * __va_rounded_size (void *)); \ - va_arg (arg, type); \ - })) -# else -# define ARG(type) (argpos == 0 ? va_arg (arg, type) : \ - ({ unsigned int pos = argpos; \ - va_list arg = (va_list) argptr; \ - while (--pos > 0) \ - (void) va_arg (arg, void *); \ - va_arg (arg, type); \ - })) -# endif -#endif - -#ifndef COMPILE_WSCANF - if (!isascii ((unsigned char) *f)) - { - /* Non-ASCII, may be a multibyte. */ - int len = __mbrlen (f, strlen (f), &state); - if (len > 0) - { - do - { - c = inchar (); - if (__glibc_unlikely (c == EOF)) - input_error (); - else if (c != (unsigned char) *f++) - { - ungetc_not_eof (c, s); - conv_error (); - } - } - while (--len > 0); - continue; - } - } -#endif - - fc = *f++; - if (fc != '%') - { - /* Remember to skip spaces. */ - if (ISSPACE (fc)) - { - skip_space = 1; - continue; - } - - /* Read a character. */ - c = inchar (); - - /* Characters other than format specs must just match. */ - if (__glibc_unlikely (c == EOF)) - input_error (); - - /* We saw white space char as the last character in the format - string. Now it's time to skip all leading white space. */ - if (skip_space) - { - while (ISSPACE (c)) - if (__glibc_unlikely (inchar () == EOF)) - input_error (); - skip_space = 0; - } - - if (__glibc_unlikely (c != fc)) - { - ungetc (c, s); - conv_error (); - } - - continue; - } - - /* This is the start of the conversion string. */ - flags = 0; - - /* Initialize state of modifiers. */ - argpos = 0; - - /* Prepare temporary buffer. */ - char_buffer_rewind (&charbuf); - - /* Check for a positional parameter specification. */ - if (ISDIGIT ((UCHAR_T) *f)) - { - argpos = read_int ((const UCHAR_T **) &f); - if (*f == L_('$')) - ++f; - else - { - /* Oops; that was actually the field width. */ - width = argpos; - argpos = 0; - goto got_width; - } - } - - /* Check for the assignment-suppressing, the number grouping flag, - and the signal to use the locale's digit representation. */ - while (*f == L_('*') || *f == L_('\'') || *f == L_('I')) - switch (*f++) - { - case L_('*'): - flags |= SUPPRESS; - break; - case L_('\''): -#ifdef COMPILE_WSCANF - if (thousands != L'\0') -#else - if (thousands != NULL) -#endif - flags |= GROUP; - break; - case L_('I'): - flags |= I18N; - break; - } - - /* Find the maximum field width. */ - width = 0; - if (ISDIGIT ((UCHAR_T) *f)) - width = read_int ((const UCHAR_T **) &f); - got_width: - if (width == 0) - width = -1; - - /* Check for type modifiers. */ - switch (*f++) - { - case L_('h'): - /* ints are short ints or chars. */ - if (*f == L_('h')) - { - ++f; - flags |= CHAR; - } - else - flags |= SHORT; - break; - case L_('l'): - if (*f == L_('l')) - { - /* A double `l' is equivalent to an `L'. */ - ++f; - flags |= LONGDBL | LONG; - } - else - /* ints are long ints. */ - flags |= LONG; - break; - case L_('q'): - case L_('L'): - /* doubles are long doubles, and ints are long long ints. */ - flags |= LONGDBL | LONG; - break; - case L_('a'): - /* The `a' is used as a flag only if followed by `s', `S' or - `['. */ - if (*f != L_('s') && *f != L_('S') && *f != L_('[')) - { - --f; - break; - } - /* In __isoc99_*scanf %as, %aS and %a[ extension is not - supported at all. */ - if (s->_flags2 & _IO_FLAGS2_SCANF_STD) - { - --f; - break; - } - /* String conversions (%s, %[) take a `char **' - arg and fill it in with a malloc'd pointer. */ - flags |= GNU_MALLOC; - break; - case L_('m'): - flags |= POSIX_MALLOC; - if (*f == L_('l')) - { - ++f; - flags |= LONG; - } - break; - case L_('z'): - if (need_longlong && sizeof (size_t) > sizeof (unsigned long int)) - flags |= LONGDBL; - else if (sizeof (size_t) > sizeof (unsigned int)) - flags |= LONG; - break; - case L_('j'): - if (need_longlong && sizeof (uintmax_t) > sizeof (unsigned long int)) - flags |= LONGDBL; - else if (sizeof (uintmax_t) > sizeof (unsigned int)) - flags |= LONG; - break; - case L_('t'): - if (need_longlong && sizeof (ptrdiff_t) > sizeof (long int)) - flags |= LONGDBL; - else if (sizeof (ptrdiff_t) > sizeof (int)) - flags |= LONG; - break; - default: - /* Not a recognized modifier. Backup. */ - --f; - break; - } - - /* End of the format string? */ - if (__glibc_unlikely (*f == L_('\0'))) - conv_error (); - - /* Find the conversion specifier. */ - fc = *f++; - if (skip_space || (fc != L_('[') && fc != L_('c') - && fc != L_('C') && fc != L_('n'))) - { - /* Eat whitespace. */ - int save_errno = errno; - __set_errno (0); - do - /* We add the additional test for EOF here since otherwise - inchar will restore the old errno value which might be - EINTR but does not indicate an interrupt since nothing - was read at this time. */ - if (__builtin_expect ((c == EOF || inchar () == EOF) - && errno == EINTR, 0)) - input_error (); - while (ISSPACE (c)); - __set_errno (save_errno); - ungetc (c, s); - skip_space = 0; - } - - switch (fc) - { - case L_('%'): /* Must match a literal '%'. */ - c = inchar (); - if (__glibc_unlikely (c == EOF)) - input_error (); - if (__glibc_unlikely (c != fc)) - { - ungetc_not_eof (c, s); - conv_error (); - } - break; - - case L_('n'): /* Answer number of assignments done. */ - /* Corrigendum 1 to ISO C 1990 describes the allowed flags - with the 'n' conversion specifier. */ - if (!(flags & SUPPRESS)) - { - /* Don't count the read-ahead. */ - if (need_longlong && (flags & LONGDBL)) - *ARG (long long int *) = read_in; - else if (need_long && (flags & LONG)) - *ARG (long int *) = read_in; - else if (flags & SHORT) - *ARG (short int *) = read_in; - else if (!(flags & CHAR)) - *ARG (int *) = read_in; - else - *ARG (char *) = read_in; - -#ifdef NO_BUG_IN_ISO_C_CORRIGENDUM_1 - /* We have a severe problem here. The ISO C standard - contradicts itself in explaining the effect of the %n - format in `scanf'. While in ISO C:1990 and the ISO C - Amendement 1:1995 the result is described as - - Execution of a %n directive does not effect the - assignment count returned at the completion of - execution of the f(w)scanf function. - - in ISO C Corrigendum 1:1994 the following was added: - - Subclause 7.9.6.2 - Add the following fourth example: - In: - #include <stdio.h> - int d1, d2, n1, n2, i; - i = sscanf("123", "%d%n%n%d", &d1, &n1, &n2, &d2); - the value 123 is assigned to d1 and the value3 to n1. - Because %n can never get an input failure the value - of 3 is also assigned to n2. The value of d2 is not - affected. The value 3 is assigned to i. - - We go for now with the historically correct code from ISO C, - i.e., we don't count the %n assignments. When it ever - should proof to be wrong just remove the #ifdef above. */ - ++done; -#endif - } - break; - - case L_('c'): /* Match characters. */ - if ((flags & LONG) == 0) - { - if (width == -1) - width = 1; - -#define STRING_ARG(Str, Type, Width) \ - do if (!(flags & SUPPRESS)) \ - { \ - if (flags & MALLOC) \ - { \ - /* The string is to be stored in a malloc'd buffer. */ \ - /* For %mS using char ** is actually wrong, but \ - shouldn't make a difference on any arch glibc \ - supports and would unnecessarily complicate \ - things. */ \ - strptr = ARG (char **); \ - if (strptr == NULL) \ - conv_error (); \ - /* Allocate an initial buffer. */ \ - strsize = Width; \ - *strptr = (char *) malloc (strsize * sizeof (Type)); \ - Str = (Type *) *strptr; \ - if (Str != NULL) \ - add_ptr_to_free (strptr); \ - else if (flags & POSIX_MALLOC) \ - { \ - done = EOF; \ - goto errout; \ - } \ - } \ - |
