Character Classes

This module contains character classification functions, very similar to the macros from "ctype.h". Differences are as follows:

• My version allows national characters (such as "äöü") whereas this is at least not very portable between current implementations of "ctype.h". Since I want that the input language understood by my program is portable, I had to program it myself.
• My version defines an enumeration type ch_t of character classes, and a function classify which returns the class of a character. This makes the main switch in the lexical scanner more elegant and efficient. Furthermore, all characters are classified in this module, the scanner works only with character classes. So this is the central place where special characters can be changed or synonyms defined.
• This module also defines function for returning a sort sequence number for characters (including the correct treatment of national characters).

Currently, this module is realized as a class ch_c with some class functions. It is not possible to create instances of this class.

Note that the class method init must be called before any of the other functions.

The enumeration type ch_t defines the possible character classes:

• CH_NULL: The NULL character ('\0').
• CH_EOF: The "end of file" character (Crtl+D and Crtl+Z. Currently, both characters are defined no matter what the operating system is. If that should lead to problems, it can easily be changed in ch_c::init.
• CH_ATOM: A lowercase character (indicating the start of an atom).
• CH_VAR: A uppercase character (or an underscore), indicating a variable.
• CH_MINUS: The minus character ("-").
• CH_DIGIT: A digit ("0123456789").
• CH_FULLSTOP: A fullstop character (".").
• CH_COMMA: A comma (",).
• CH_COLON: A colon (":").
• CH_SEMICOLON: A semicolon (";").
• CH_LPAREN: A left parenthesis ("(").
• CH_RPAREN: A right parenthesis (")").
• CH_NOT: A negation character (snake symbol): "~".
• CH_AND: A "logical and" character ("&").
• CH_OR: A "logical or" character ("|").
• CH_LESS: A "less than" character ("<").
• CH_GREATER: A "greater than" character (">").
• CH_EQUALS: An "is equal to" character ("=").
• CH_QUEST: A question mark character ("?").
• CH_CMD: A dollar sign ("$"). It is used as first character of commands.
• CH_UNUSED: An unused character, which has no defined meaning in SLP. Currently, the characters "!@#^*+\{}[]"`/" are classified in this way. There sometime has been a bug in my program where it would have been useful if I had made such characters explicit (like it is done now).
• CH_UNDEFINED: Characters which have not yet been assigned a character class.

init:

void init(void);
This function must be called before any other function of this module. It initializes the class variables.

classify:

static ch_t classify(char c);
This function returns the character class of an input character.

is_alnum:

static bool_t is_alnum(char_c);
This function returns BOOL_TRUE iff the input character is a letter (including national ones), a digit, or the underscore character ("_"). Otherwise, BOOL_FALSE is returned.

is_space:

static bool_t is_space(char c);
This function returns BOOL_TRUE iff the input character is a "white space" character, i.e. a space, a tabulator, or a carriage return. Note that a linefeed is not treated as white space here.

is_comment:

static bool_t is_comment(char c);
This function returns BOOL_TRUE if the input character indicates a begin of a comment. Currently, this is the character "%".

is_newline:

static bool_t is_newline(char c);
This function returns BOOL_TRUE iff the input character is the linefeed character.

is_quote:

static bool_t is_quote(char c);
This function returns BOOL_TRUE if the input character is a quotation character ("'") used in quoted atoms.

is_undersc:

static bool_t is_undersc(char c);
This function returns BOOL_TRUE iff the input character is the underscore character ("_"). The underscore character is special because of the anonymous variable.

ord:

static int ord(char c);
This function returns the position of the input character in some "alphabet". The absolute value has no significance, but comparing the ord-values of two characters should give the correct order. Uppercase and lowercase characters return the same value (whereas, if the ASCII sequence would be used, all uppercase characters come before all lowercase characters). Also, national characters with accents (like "äöü") return the same values as the corresponding normal characters (i.e. "aou").

xord:

static int xord(char c);
Some national characters must be sorted like a two character sequence. For instance, the German letter "ß" must be treated like "ss" for correct alphabetical sorting. In this case, ord returns the position of the first character and xord returns the position of the second character. For normal characters, xord returns 0.

• Header File (ch.h)
• Source File (ch.cxx)