dragonfly/tools/sqf_validator.py

#!/usr/bin/env python3

import fnmatch
import os
import re
import sys

def valid_keyword_after_code(content, index):
	for word in ["for", "do", "count", "each", "forEach", "else", "and", "not", "isEqualTo", "isNotEqualTo", "in", "call", "spawn", "execVM", "catch", "param", "select", "apply", "findIf", "remoteExec"]:
		if content.find(word, index, index + len(word)) != -1:
			return True

	return False

def check_sqf(filepath):
	errors = []

	with open(filepath, "r", encoding = "utf-8", errors = "ignore") as file:
		content = file.read()

		# Store all brackets we find in this file, so we can validate everything on the end
		brackets = []

		# Used in case we are in a line comment (//)
		ignore_till_eol = False

		# To check if we are in a comment block
		in_comment_block = False
		check_if_comment = False

		# Used in case we are in a comment block (/* */)
		# This is true if we detect a * inside a comment block
		# If the next character is a /, it means we end our comment block
		check_if_closing = False

		# We ignore everything inside a string
		in_string = False

		# Used to store the starting type of a string, so we can match that to the end of a string
		string_type = ""

		# Used to check for semicolon after code blocks
		last_is_curly_brace = False
		check_for_semicolon = False

		# Extra information so we know what line we find errors at
		line_number = 1

		char_index = 0

		for c in content:
			if last_is_curly_brace:
				last_is_curly_brace = False

				# Test generates false positives with binary commands that take CODE as 2nd arg (e.g. findIf)
				check_for_semicolon = not re.search("findIf", content, re.IGNORECASE)

			# Keep track of current line number
			if c == "\n":
				line_number += 1

			# While we are in a string, we can ignore everything else, except the end of the string
			if in_string:
				if c == string_type:
					in_string = False

			# Look for the end of this comment block
			elif in_comment_block:
				if c == "*":
					check_if_closing = True
				elif check_if_closing:
					if c == "/":
						in_comment_block = False
					elif c != "*":
						check_if_closing = False

			# If we are not in a comment block, we will check if we are at the start of one or count the () {} and []
			else:
				# This means we have encountered a /, so we are now checking if this is an inline comment or a comment block
				if check_if_comment:
					check_if_comment = False

					# If the next character after / is a *, we are at the start of a comment block
					if c == "*":
						in_comment_block = True

					# Otherwise, check if we are in an line comment, / followed by another / (//)
					elif c == "/":
						ignore_till_eol = True

				if not in_comment_block:
					if ignore_till_eol:
						# We are in a line comment, just continue going through the characters until we find an end of line
						if c == "\n":
								ignore_till_eol = False
					else:
						if c == '"' or c == "'":
							in_string = True
							string_type = c
						elif c == "/":
							check_if_comment = True
						elif c == "\t":
							errors.append("  ERROR: Found a tab on line {}.".format(line_number))
						elif c in ["(", "[", "{"]:
							brackets.append(c)
						elif c == ")":
							if not brackets or brackets[-1] in ["[", "{"]:
								errors.append("  ERROR: Missing parenthesis '(' on line {}.".format(line_number))
							brackets.append(c)
						elif c == "]":
							if not brackets or brackets[-1] in ["(", "{"]:
								errors.append("  ERROR: Missing square bracket '[' on line {}.".format(line_number))
							brackets.append(c)
						elif c == "}":
							last_is_curly_brace = True

							if not brackets or brackets[-1] in ["(", "["]:
								errors.append("  ERROR: Missing curly brace '{{' on line {}.".format(line_number))
							brackets.append(c)

							if check_for_semicolon:
								# Keep reading until no white space or comments
								if c not in [" ", "\t", "\n", "/"]:
									check_for_semicolon = False
									if c not in ["]", ")", "}", ";", ",", "&", "!", "|", "="] and not valid_keyword_after_code(content, char_index):
										errors.append("  ERROR: Possible missing semicolon ';' on line {}.".format(line_number))

			char_index += 1

		# Compare opening and closing bracket counts
		if brackets.count("(") != brackets.count(")"):
				errors.append("  ERROR: Unequal number of parentheses, '(' = {}, ')' = {}.".format(brackets.count("("), brackets.count(")")))

		if brackets.count("[") != brackets.count("]"):
				errors.append("  ERROR: Unequal number of square brackets, '[' = {}, ']' = {}.".format(brackets.count("["), brackets.count("]")))

		if brackets.count("{") != brackets.count("}"):
				errors.append("  ERROR: Unequal number of curly braces, '{{' = {}, '}}' = {}.".format(brackets.count("{"), brackets.count("}")))

		# Ensure includes are before block comments
		if re.compile('\s*(/\*[\s\S]+?\*/)\s*#include').match(content):
				errors.append("  ERROR: Found an #include after a block comment.")

	return errors

def main():
	print("Validating SQF")
	print("--------------")

	# Allow running from root directory and tools directory
	root_dir = ".."
	if os.path.exists("addons"):
		root_dir = "."

	# Check all SQF files in the project directory
	sqf_files = []

	for root, _, files in os.walk(root_dir):
		for file in fnmatch.filter(files, "*.sqf"):
			sqf_files.append(os.path.join(root, file))

		sqf_files.sort()

		bad_count = 0

		for filepath in sqf_files:
			errors = check_sqf(filepath)

			if errors:
				print("\nFound {} error(s) in {}:".format(len(errors), os.path.relpath(filepath, root_dir)))

				for error in errors:
					print(error)

				bad_count += 1

		print("\nChecked {} files, found errors in {}.".format(len(sqf_files), bad_count))

		if bad_count == 0:
			print("SQF Validation PASSED")
		else:
			print("SQF Validation FAILED")

		return bad_count

if __name__ == "__main__":
	sys.exit(main())