代码:

---

#include    <stdio.h>

main()
{
      char  buf[1024];
      while(1) {
               scanf("%s[ a-zA-Z0-9]", buf);
               printf("%s\n", buf);
      }
}

---

代码:

---

scanf("%[ a-zA-Z0-9]", buf);

---

代码:

---

#include <stdio.h>
main(){
  char  buf[1024];
  while(1) {
        printf("[a-zA-Z0-9]");
        scanf("%s", buf);
        printf("%s\n", buf);
  }
}

---

代码:

---

#include <stdio.h>
int main(void)
{
        int ch;
        while((ch = getchar()) != EOF)
                putchar(ch);
        return 0;
}

---

代码:

---

#include <stdio.h>
main(){
  char  buf[1024]={0};
  while(1) {
        printf("[a-zA-Z0-9]");
        scanf("%1023s", buf);
        printf("%s\n", buf);
  }
}

---

引用:

---

To read strings not delimited by space characters, a set of characters in brackets ([ ]) can be substituted for the s (string) type character. The corresponding input field is read up to the first character that does not appear in the bracketed character set. If the first character in the set is a caret (^), the effect is reversed: The input field is read up to the first character that does appear in the rest of the character set.

Note that %[a-z] and %[z-a] are interpreted as equivalent to %[abcde...z]. This is a common scanf function extension, but note that the ANSI standard does not require it.

---

代码:

---

SCANF(3)               FreeBSD Library Functions Manual               SCANF(3)

NAME
     scanf, fscanf, sscanf, vscanf, vsscanf, vfscanf -- input format conver-
     sion

LIBRARY
     Standard C Library (libc, -lc)

SYNOPSIS
     #include <stdio.h>

     int
     scanf(const char *format, ...);

     int
     fscanf(FILE *stream, const char *format, ...);

     int
     sscanf(const char *str, const char *format, ...);

     #include <stdarg.h>

     int
     vscanf(const char *format, va_list ap);

     int
     vsscanf(const char *str, const char *format, va_list ap);

     int
     vfscanf(FILE *stream, const char *format, va_list ap);

DESCRIPTION
     The scanf() family of functions scans input according to a format as
     described below.  This format may contain conversion specifiers; the
     results from such conversions, if any, are stored through the pointer
     arguments.  The scanf() function reads input from the standard input
     stream stdin, fscanf() reads input from the stream pointer stream, and
     sscanf() reads its input from the character string pointed to by str.
     The vfscanf() function is analogous to vfprintf(3) and reads input from
     the stream pointer stream using a variable argument list of pointers (see
     stdarg(3)).  The vscanf() function scans a variable argument list from
     the standard input and the vsscanf() function scans it from a string;
     these are analogous to the vprintf() and vsprintf() functions respec-
     tively.  Each successive pointer argument must correspond properly with
     each successive conversion specifier (but see the * conversion below).
     All conversions are introduced by the % (percent sign) character.  The
     format string may also contain other characters.  White space (such as
     blanks, tabs, or newlines) in the format string match any amount of white
     space, including none, in the input.  Everything else matches only
     itself.  Scanning stops when an input character does not match such a
     format character.  Scanning also stops when an input conversion cannot be
     made (see below).

CONVERSIONS
     Following the % character introducing a conversion there may be a number
     of flag characters, as follows:

     *       Suppresses assignment.  The conversion that follows occurs as
             usual, but no pointer is used; the result of the conversion is
             simply discarded.

     h       Indicates that the conversion will be one of dioux or n and the
             next pointer is a pointer to a short int (rather than int).

     l       Indicates either that the conversion will be one of dioux or n
             and the next pointer is a pointer to a long int (rather than
             int), or that the conversion will be one of efg and the next
             pointer is a pointer to double (rather than float).

     L       Indicates that the conversion will be efg and the next pointer is
             a pointer to long double.  (This type is not implemented; the L
             flag is currently ignored.)

     q       Indicates either that the conversion will be one of dioux or n
             and the next pointer is a pointer to a long long int (rather than
             int),

     In addition to these flags, there may be an optional maximum field width,
     expressed as a decimal integer, between the % and the conversion.  If no
     width is given, a default of `infinity' is used (with one exception,
     below); otherwise at most this many characters are scanned in processing
     the conversion.  Before conversion begins, most conversions skip white
     space; this white space is not counted against the field width.

     The following conversions are available:

     %     Matches a literal `%'.  That is, `%%' in the format string matches
           a single input `%' character.  No conversion is done, and assign-
           ment does not occur.

     d     Matches an optionally signed decimal integer; the next pointer must
           be a pointer to int.

     D     Equivalent to ld; this exists only for backwards compatibility.

     i     Matches an optionally signed integer; the next pointer must be a
           pointer to int.  The integer is read in base 16 if it begins with
           `0x' or `0X', in base 8 if it begins with `0', and in base 10 oth-
           erwise.  Only characters that correspond to the base are used.

     o     Matches an octal integer; the next pointer must be a pointer to
           unsigned int.

     O     Equivalent to lo; this exists for backwards compatibility.

     u     Matches an optionally signed decimal integer; the next pointer must
           be a pointer to unsigned int.

     x     Matches an optionally signed hexadecimal integer; the next pointer
           must be a pointer to unsigned int.

     X     Equivalent to lx; this violates the ISO/IEC 9899:1990
           (``ISO C90''), but is backwards compatible with previous UNIX sys-
           tems.

     f     Matches an optionally signed floating-point number; the next
           pointer must be a pointer to float.

     e     Equivalent to f.

     g     Equivalent to f.

     E     Equivalent to lf; this violates the ISO/IEC 9899:1990
           (``ISO C90''), but is backwards compatible with previous UNIX sys-
           tems.

     F     Equivalent to lf; this exists only for backwards compatibility.

     s     Matches a sequence of non-white-space characters; the next pointer
           must be a pointer to char, and the array must be large enough to
           accept all the sequence and the terminating NUL character.  The
           input string stops at white space or at the maximum field width,
           whichever occurs first.

     c     Matches a sequence of width count characters (default 1); the next
           pointer must be a pointer to char, and there must be enough room
           for all the characters (no terminating NUL is added).  The usual
           skip of leading white space is suppressed.  To skip white space
           first, use an explicit space in the format.

     [     Matches a nonempty sequence of characters from the specified set of
           accepted characters; the next pointer must be a pointer to char,
           and there must be enough room for all the characters in the string,
           plus a terminating NUL character.  The usual skip of leading white
           space is suppressed.  The string is to be made up of characters in
           (or not in) a particular set; the set is defined by the characters
           between the open bracket [ character and a close bracket ] charac-
           ter.  The set excludes those characters if the first character
           after the open bracket is a circumflex ^.  To include a close
           bracket in the set, make it the first character after the open
           bracket or the circumflex; any other position will end the set.
           The hyphen character - is also special; when placed between two
           other characters, it adds all intervening characters to the set.
           To include a hyphen, make it the last character before the final
           close bracket.  For instance, `[^]0-9-]' means the set `everything
           except close bracket, zero through nine, and hyphen'.  The string
           ends with the appearance of a character not in the (or, with a cir-
           cumflex, in) set or when the field width runs out.

     p     Matches a pointer value (as printed by `%p' in printf(3)); the next
           pointer must be a pointer to void.

     n     Nothing is expected; instead, the number of characters consumed
           thus far from the input is stored through the next pointer, which
           must be a pointer to int.  This is not a conversion, although it
           can be suppressed with the * flag.

     For backwards compatibility, other conversion characters (except `\0')
     are taken as if they were `%d' or, if uppercase, `%ld', and a `conver-
     sion' of `%\0' causes an immediate return of EOF.  The F and X conver-
     sions will be changed in the future to conform to the ANSI C standard,
     after which they will act like f and x respectively.

RETURN VALUES
     These functions return the number of input items assigned, which can be
     fewer than provided for, or even zero, in the event of a matching fail-
     ure.  Zero indicates that, while there was input available, no conver-
     sions were assigned; typically this is due to an invalid input character,
     such as an alphabetic character for a `%d' conversion.  The value EOF is
     returned if an input failure occurs before any conversion such as an end-
     of-file occurs.  If an error or end-of-file occurs after conversion has
     begun, the number of conversions which were successfully completed is
     returned.

SEE ALSO
     getc(3), printf(3), strtod(3), strtol(3), strtoul(3)

STANDARDS
     The functions fscanf(), scanf(), and sscanf() conform to ISO/IEC
     9899:1990 (``ISO C90'').

BUGS
     The current situation with %F and %X conversions is unfortunate.

     All of the backwards compatibility formats will be removed in the future.

     Numerical strings are truncated to 512 characters; for example, %f and %d
     are implicitly %512f and %512d.

FreeBSD 4.11                   December 11, 1993                  FreeBSD 4.11

---

引用:

---

顺便问一句，如果想不以空格作为分隔符，一行全部读入，用ANSI C应该怎么做呢？

---

引用:

---

方括号中可以出现一个或多个字符。它说明读入的是一个字符串，所以与s格式类似。但输入的字符串中字符必须是方括号中出现的字符。当遇到不是方括号中的字符输入时，终止相应的输入。当方括号的第一个字符为"^"时，则正好相反：即输入字符串时，遇到属于方括号中出现的字符时就结束。

---

引用:

---

对scanf("%s[..]",..)很困惑，scanf()为什么要有这么BT的格式控制符呢

---