Created a small utility program to generate JSON files for testing the main input.

json_generator/tokenizer.py

@@ -0,0 +1,78 @@
+import token
+
+def get_tokens_from_expression_string(expression_string):
+
+	#Takes user input of an actuarial formula and parses the formula to id its components.
+	#:return: tokens (List of objects of CToken class).
+	
+	tokens = []  # List of objects of CToken class
+	tmp = ""
+	parsing_number = False
+	symbols_dic = {}  # Dictionary to keep track of the no. of times each variable appears.
+
+	for i in range(0, len(expression_string)):
+		c = expression_string[i]
+		if c == '.' or c.isdigit():
+			if tmp.count('.') == 1:
+				raise SyntaxError("Invalid grammar, to many periods in number. Found at position " + str (i) + ".")
+			if len(tokens) > 1:
+				lookBehind = tokens[-1]
+				if lookBehind.type == token.TokenType.exp_end:
+					tokens.append(token.CToken("^", token.TokenType.power))
+			if (i + 1) == len(expression_string):
+				tokens.append(token.CToken(tmp + c, token.TokenType.constant))
+			parsing_number = True
+			tmp = tmp + c
+			continue
+		elif c == '(':
+			if parsing_number:
+				tokens.append(token.CToken(tmp, token.TokenType.constant))
+				tokens.append(token.CToken("*", token.TokenType.multiply))
+			tokens.append(token.CToken("(", token.TokenType.exp_start))
+			parsing_number = False
+			tmp = ""
+			continue
+		elif c == ')':
+			if parsing_number:				
+				tokens.append(token.CToken(tmp, token.TokenType.constant))
+			tokens.append(token.CToken(")", token.TokenType.exp_end))
+			parsing_number = False
+			tmp = ""
+			continue
+		elif token.TokenType.get_operator(c) != token.TokenType.unknown:
+			if parsing_number:
+				tokens.append(token.CToken(tmp, token.TokenType.constant))
+			tokens.append(token.CToken(c, token.TokenType.get_operator(c)))
+			parsing_number = False
+			tmp = ""
+			continue
+		elif c != ' ':
+			if len(tokens) > 1:
+				lookBehind = tokens[-1]
+				if lookBehind.type == token.TokenType.exp_end:
+					tokens.append(token.CToken("^", token.TokenType.power))
+			if (i + 1) < len(expression_string):
+				lookAHead = expression_string[i + 1]
+				if lookAHead == '(':
+					tokens.append(token.CToken(c, token.TokenType.variable))
+					tokens.append(token.CToken("*", token.TokenType.multiply))
+					tmp = ""
+					parsing_number = False
+					continue
+			if parsing_number:
+				tokens.append(token.CToken(tmp, token.TokenType.constant))
+				tokens.append(token.CToken("*", token.TokenType.multiply))
+			if c in symbols_dic.keys():
+				symbols_dic[c] += 1
+			else:
+				symbols_dic[c] = 1
+			tokens.append(token.CToken(c + str(symbols_dic[c]), token.TokenType.variable))
+			parsing_number = False
+			tmp = ""
+	#When building this list we've effectively ordered the tokens in reverse order from how we want to
+	#process them. So, reverse the order of the list before returning them to the caller, we don't want
+	#the caller to have to worry about such a minor detail.
+	tokens.reverse()
+	
+	return tokens
+