1. 去除c语言中的注释

注释是为了让人更容易理解代码，但是机器不需要，所以在要去除这部分的注释，方便后面做处理。

用的python做实现。

from copy import deepcopy
import sys 

define_debug = False

def is_alnum(c):
    if ('0' <= c and c <= '9') or ('a' <= c and c <= 'z') or ('A' <= c and c <= 'Z') or c in ['_', '#', '\\', '%']:
        return True
    else:
        return  False
    pass


def de_comment(c_line):
    is_mcomment = False                  # 是否是多行注释
    is_scomment = False                  # 是否是单行注释
    is_str = False                       # 是否是字符串
    is_char = False                      # 是否是字符
    is_escape = False                    # 是否是转义符

    de_comment_line = []
    for l in c_line:
        l = l.strip()
        len_w = len(l)
        if len_w == 0:
            continue
        is_scomment = False
        w_t = ''
        c_i = 0
        c_o = 0
        while(c_i < len_w):
            # 字符串开始判断
            if not is_scomment and not is_mcomment and not is_char and not is_str and l[c_i] == '"':
                if define_debug:
                    print('a:', l[c_i:], is_str)
                is_str = True
            # 字符串结束判断
            elif is_str and not is_escape and l[c_i] == '"':
                if define_debug:
                    print('b:', l[c_i:])
                is_str = False
            # 字符开始判断
            if not is_scomment and not is_mcomment and not is_char and not is_str and l[c_i] == '\'':
                if define_debug:
                    print('c:', l[c_i:])
                is_char = True
            # 字符结束判断
            elif is_char and not is_escape and l[c_i] == '\'':
                if define_debug:
                    print('d:', l[c_i:])
                is_char = False
            # 转义符开始判断
            elif is_str and not is_escape and l[c_i] == '\\':
                if define_debug:
                    print('e:', l[c_i:])
                is_escape = True
            #多行注释开始判断
            elif c_i + 2 <= len_w and l[c_i: c_i + 2] == '/*' and not is_scomment and not is_str:
                if define_debug:
                    print('g:', l[c_i:])
                c_i += 2
                is_mcomment = True
            #多行注释结束判断
            elif c_i + 2 <= len_w and l[c_i: c_i + 2] == '*/' and is_mcomment:
                if define_debug:
                    print('h:', l[c_i:])
                c_i += 2
                is_mcomment = False
                continue
            #单行注释开始判断
            elif c_i + 2 <= len_w and l[c_i: c_i + 2] == '//' and not is_mcomment and not is_str:
                if define_debug:
                    print('i:', l[c_i:], is_str)
                is_scomment = True
                c_i = len_w

            if not is_scomment and not is_mcomment:
                if c_i >= len_w:
                    if define_debug:
                        print(c_i, len_w)
                    continue

                if is_escape and c_i != len_w-1:
                    w_t += l[c_i:c_i+2]
                    c_i += 2
                    is_escape = False
                else:
                    w_t += l[c_i]

            if c_o == c_i:
                c_o += 1
                c_i += 1
            else:
                c_o = c_i
        w_t = w_t.strip()
        if len(w_t) != 0:
            de_comment_line.append(w_t)
    return de_comment_line
    
if __name__ == "__main__":
    c_file = open('x265.cpp.test', 'r') # 最后测试是拿的x265的开源代码
    c_line = c_file.readlines()
    c_file.close()
    
    de_comment_line = de_comment(c_line)
    for l in de_comment_line:
        print(l)

---

这个代码的思想其实就是严格控制状态，开始的时候我以为就是简单的找到C语言的注释符号，后来发现有字符串这些里面可能也有双斜杠啥的，于是代码就慢慢变得复杂起来了。

下一步应该是将代码各个部分结构化。