Classification

The classification table is created with one or more entries of the form:

  ctype[character1 - character2] =
     ctype_classification |
     ctype_classification | ...
     ctype[character] = ctype_classification |
     ctype_classification | ...
  

where character, character1 and character2 are characters represented according to the rules for Strings and Characters in Locale Data Files. The characters may appear as normal characters:

  ctype[a - z]
  ctype['a' - 'z']
  

or as octal, hexadecimal or universal:

  ctype['\101']
  ctype['\x41']
  ctype['\u41']
  

The usual escape sequences are also recognized: \n, \t, \a, \\, \' and so on.

On the right hand side of the equal sign, ctype_classification is one of:

• alpha
• blank
• cntrl
• digit
• graph
• lower
• print
• punct
• space
• upper
• xdigit

An | can be used to assign a character, or range of characters, more than one classification. These keywords correspond to the names of the enum ctype_base::mask, except that alnum is not present. To get alnum simply specifiy "alpha | digit". The keyword blank is introduced, motivated by C99's isblank function.

Each of these keywords represent one bit in the ctype_base::mask. Thus for each entry into the ctype table, one must specify all attributes that apply. For example, in the "C" locale a-z are represented as:

  ctype['a' - 'z'] = 

  

        xdigit | lower | alpha | graph | print