xyntax

xyntax -- Xolatile-style “header-only” library for syntax definition control.

• Primary focus of this library is for syntax highlighting, hence the name...
• Important note: Regular expressions are more robust, this is simple solution for simple problems.
• Everything related to my libraries is clean of all warning options on Clang, GCC and Valgrind.

Installing:

$ sudo sh install.sh

Using:

#include <xolatile/xyntax.h>
#include <xolatile/xyntax.c> /* Instead of '#define BLA_BLA_IMPLEMENTATION'. */
...
int symbols = syntax_insert (1, 0, ".,:;<=>+-*/%!&~^?|()[]{}", "", '\0', 0, 0);
...
int select = syntax_select (& buffer [offset], & length);
/* Variable 'select' will become the index of syntax rule you defined previously, or 'syntax_count' if there is no match. */
...
syntax_delete ();
/* Free used memory, after this you can define some other syntax rules... */

Xolatile-style “header-only” library is my take on ‘stb’ header-only libraries. There are a lot of ideas that came from Ada, which is my second language. Main idea behind them is to avoid standard library and macros in programs. Also, I like to avoid C-style “namespaces” and bad function names...

It can be used for:

• syntax highlighting in terminal or graphical text editors...
• source code processing, parsing, tokenization...
• counting source code elements such as keywords, literals, brackets...

For example, your can make simple ANSI C syntax highlight like this:

char * separators = ".,:;<=>+-*/%!&~^?|()[]{}'\" \t\r\n";

char * keywords [32] = {
	"register", "volatile", "auto",     "const",    "static",   "extern",   "if",       "else",
	"do",       "while",    "for",      "continue", "switch",   "case",     "default",  "break",
	"enum",     "union",    "struct",   "typedef",  "goto",     "void",     "return",   "sizeof",
	"char",     "short",    "int",      "long",     "signed",   "unsigned", "float",    "double"
};

With using helper ‘syntax_define_*’ functions:

syntax_define_separators (separators);

syntax_define_default (COLOUR_RED, EFFECT_NORMAL, COLOUR_CYAN, EFFECT_BOLD);

syntax_define_range ("/*", "*/", '\0', COLOUR_GREY, EFFECT_BOLD);
syntax_define_range ("//", "\n", '\0', COLOUR_GREY, EFFECT_BOLD);
syntax_define_range ("#",  "\n", '\\', COLOUR_PINK, EFFECT_NORMAL);
syntax_define_range ("'",  "'",  '\\', COLOUR_PINK, EFFECT_BOLD);

syntax_define_operators (".,:;<=>+*-/%!&~^?|()[]{}", COLOUR_BLUE,  EFFECT_BOLD);

syntax_define_words (keywords, 32, COLOUR_BLUE, EFFECT_NORMAL);

Or using core ‘syntax_push’ functions:

(void) syntax_insert (1, 0, " \t\r\n",                  "",   '\0', COLOUR_WHITE, EFFECT_NORMAL);
(void) syntax_insert (0, 0, "#",                        "\n", '\\', COLOUR_PINK,  EFFECT_BOLD);
(void) syntax_insert (0, 0, "//",                       "\n", '\0', COLOUR_GREY,  EFFECT_BOLD);
(void) syntax_insert (0, 0, "/*",                       "*/", '\0', COLOUR_GREY,  EFFECT_BOLD);
(void) syntax_insert (0, 0, "'",                        "'",  '\\', COLOUR_RED,   EFFECT_NORMAL);
(void) syntax_insert (0, 0, "\"",                       "\"", '\\', COLOUR_RED,   EFFECT_BOLD);
(void) syntax_insert (1, 0, ".,:;<=>+-*/%!&~^?|()[]{}", "",   '\0', COLOUR_CYAN,  EFFECT_NORMAL);

for (word = 0; word != 32; ++word) {
	(void) syntax_insert (0, 1, c_keywords [word], separator, '\0', COLOUR_BLUE, EFFECT_BOLD);
}

(void) syntax_insert (1, 1, "0123456789",                 separator, '\0', COLOUR_CYAN,  EFFECT_BOLD);
(void) syntax_insert (1, 1, "ABCDEFGHIJKLMNOPQRSTUVWXYZ", separator, '\0', COLOUR_PINK,  EFFECT_ITALIC);
(void) syntax_insert (1, 1, "abcdefghijklmnopqrstuvwxyz", separator, '\0', COLOUR_WHITE, EFFECT_ITALIC);
(void) syntax_insert (0, 1, "_",                          separator, '\0', COLOUR_PINK,  EFFECT_BOLD);

If you want to do parsing, counting, tokenization, you can use return value of ‘syntax_push’...