1 files changed, 103 insertions, 0 deletions
diff --git a/parse.c b/parse.c
new file mode 100644
index 0000000..e48adda
--- /dev/null
+++ b/parse.c
@@ -0,0 +1,103 @@
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <stdio.h>
+#include "parse.h"
+#include "hashtable.h"
+
+void symbol_add_arg(sym_t* symbol, sym_t* arg) {
+	symbol->argCount += 1;
+	if (symbol->argCount == 1) {
+		symbol->args = malloc(sizeof(sym_t*));
+	}
+	else {
+		symbol->args = realloc(symbol->args, symbol->argCount * (sizeof(sym_t*)));
+	}
+	(symbol->args)[symbol->argCount - 1] = arg;
+}
+void symset_add_symbol(symset_t* symbolSet, sym_t* symbol) {
+	symbolSet->count += 1;
+	if (symbolSet->count == 1) {
+		symbolSet->symbols = malloc(sizeof(sym_t*));
+	}
+	else {
+		symbolSet->symbols = realloc(symbolSet->symbols, symbolSet->count * (sizeof(sym_t*)));
+	}
+	(symbolSet->symbols)[symbolSet->count - 1] = symbol;
+}
+symset_t* parse_symbols(char** str) {
+	symset_t* symbolSet = malloc(sizeof(symset_t));
+	symbolSet->count = 0;
+	while (1) {
+		/* clear whitespace */
+		while (!isalnum(**str) && **str != '(' && **str != ')' && **str != '\0') {
+			(*str)++;
+		}
+		if (**str == ')') {
+			(*str)++;
+			if (**str == ';') {
+				(*str)++;
+				return symbolSet;
+			}
+		}
+		else if (**str == '\0') {
+			return symbolSet;
+		}
+		char* symbolName = calloc(30, sizeof(char));
+		int i = 0;
+		while (1) {
+			symbolName[i] = **str;
+			if (**str == ',' || **str == ';') {
+				(*str)++;
+				symbolName[i] = '\0';
+				sym_t* symbol = malloc(sizeof(sym_t));
+				symbol->province = LITERAL;
+				symbol->name = symbolName;
+				symbol->argCount = 0;
+				symset_add_symbol(symbolSet, symbol);
+				break;
+			}
+			else if (**str == '(') {
+				(*str)++;
+				symbolName[i] = '\0';
+				sym_t* symbol = malloc(sizeof(sym_t));
+				symbol->province = KEY_REF;
+				if(i > 1) {
+					symbol->name = symbolName;
+				}
+				else {
+					symbol->name = "fork";
+				}
+				symset_t* symbols = parse_symbols(str);
+				for (int s = 0; s < symbols->count; s++) {
+					symbol_add_arg(symbol, symbols->symbols[s]);
+				}
+				symset_add_symbol(symbolSet, symbol);
+				break;
+			}
+			i++;
+			(*str)++;
+		}
+	}
+}
+sym_t* parse_symbol_final(char* str) {
+	char** strp = malloc(sizeof(char**));
+	*strp = str;
+	symset_t* finalSymset = parse_symbols(strp);
+	sym_t* root = malloc(sizeof(sym_t));
+	root->name = "root";
+	root->province = KEY_REF;
+	for (int s = 0; s < finalSymset->count; s++) {
+		symbol_add_arg(root, finalSymset->symbols[s]);
+	}
+	return root;
+}
+void write_symbol_visual(sym_t* symbol, int offset) {
+	for (int o = 0; o < offset; o++) {
+		printf(" ");
+	}
+	printf("%s (%s)\n", (char*)(symbol->name), symbol->province == 1 ? "KEY" : "LITERAL");
+	for (int i = 0; i < symbol->argCount; i++) {
+		write_symbol_visual(symbol->args[i], offset + 4);
+	}
+}