【Python】简单Web框架从零开始（四）：300行代码搞定模板渲染
三、源码
这个模板语言模块是在Python2.7上开发使用的，如果要在Python3+上使用需要对str和bytes进行一些处理即可，由于没有引用任何其他模块，可以很好的独立使用。
# -*- coding:utf-8 -*- """ 模板语言""" # TOKEN相关的定义 TOKEN_S_BRACE = "{" TOKEN_S_BLOCK = "%" TOKEN_EXPRESSION_L = "{{" TOKEN_EXPRESSION_R = "}}" TOKEN_BLOCK_L = "{%" TOKEN_BLOCK_R = "%}" TOKEN_KEY_SET = "set" TOKEN_KEY_RAW = "raw" TOKEN_KEY_IF = "if" TOKEN_KEY_ELIF = "elif" TOKEN_KEY_ELSE = "else" TOKEN_KEY_FOR = "for" TOKEN_KEY_WHILE = "while" TOKEN_KEY_END = "end" TOKEN_KEY_BREAK = "break" TOKEN_KEY_CONTINUE = "continue" TOKEN_SPACE = " " TOKEN_COLON = ":" # Token标记 {{}} {% %} TOKEN_FLAG_SET = {TOKEN_S_BRACE, TOKEN_S_BLOCK} # 简单的语句 TOKEN_KEY_SET_SIMPLE_EXPRESSION = {TOKEN_KEY_SET, TOKEN_KEY_RAW} # 前置条件 TOKEN_KEY_PRE_CONDITION = { # end 必须在if/elif/else/for/while 后面 TOKEN_KEY_END: {TOKEN_KEY_IF, TOKEN_KEY_ELIF, TOKEN_KEY_ELSE, TOKEN_KEY_FOR, TOKEN_KEY_WHILE}, # elif 必须在if 后面 TOKEN_KEY_ELIF: {TOKEN_KEY_IF}, # else 必须在if/elif 后面 TOKEN_KEY_ELSE: {TOKEN_KEY_IF, TOKEN_KEY_ELIF, TOKEN_KEY_FOR, TOKEN_KEY_WHILE}, } # 循环语句 TOKEN_KEY_LOOP = {TOKEN_KEY_WHILE, TOKEN_KEY_FOR} # 循环的控制break continue TOKEN_KEY_LOOP_CTRL = {TOKEN_KEY_BREAK, TOKEN_KEY_CONTINUE} class ParseException(Exception): pass class TemplateCode(object): def __init__(self): self.codeTrees = {"parent": None, "nodes": []} self.cursor = self.codeTrees self.compiled_code = None def create_code(self): """创建一个代码子块""" child_codes = {"parent": self.cursor, "nodes": []} self.cursor["nodes"].append(child_codes) self.cursor = child_codes def close_code(self): """ 关闭一个代码子块 """ assert self.cursor["parent"] is not None, "overflow" self.cursor = self.cursor["parent"] def append_text(self, text): """ 添加文本 """ # 排除空行 self.cursor["nodes"].append("_add(%r)" % text) def append_express(self, express, raw=False): """ 表达式 """ if raw: temp_exp = "_t_exp = _str_(%s)" % express else: temp_exp = "_t_exp = _esc_(%s)" % express self.cursor["nodes"].append(temp_exp) self.cursor["nodes"].append("_add(_t_exp)") def append_statement(self, statement): """ 语句 """ temp_statement = "%s" % statement self.cursor["nodes"].append(temp_statement) def reset(self): self.codeTrees = {"parent": None, "nodes": []} self.cursor = self.codeTrees self.compiled_code = None def build_code(self, filename): temp_code_buff = [] self.write_buff_with_indent(temp_code_buff, "def _template_render():", 0) self.write_buff_with_indent(temp_code_buff, "_codes = []", 4) self.write_buff_with_indent(temp_code_buff, "_add = _codes.append", 4) self.write_codes(temp_code_buff, self.codeTrees, 4) self.write_buff_with_indent(temp_code_buff, "return ''.join(_codes)", 4) temp_code = "".join(temp_code_buff) self.compiled_code = compile(temp_code,filename, "exec", dont_inherit=True) def write_codes(self, code_buff, codes, indent): for node in codes.get("nodes", []): if isinstance(node, dict): self.write_codes(code_buff, node, indent+4) else: self.write_buff_with_indent(code_buff, node, indent) def generate(self, **kwargs): temp_namespace = {} temp_namespace['_str_'] = self.to_utf8 temp_namespace['_esc_'] = self.to_safe_utf8 temp_namespace.update(kwargs) exec(self.compiled_code, temp_namespace) return temp_namespace['_template_render']() @staticmethod def write_buff_with_indent(code_buff, raw_str, indent): """""" temp = (" " * indent) + raw_str + "\n" code_buff.append(temp) @staticmethod def to_utf8(raw_str): """ 转换 """ if isinstance(raw_str, str): return raw_str elif isinstance(raw_str, bytes): return raw_str.decode() return str(raw_str) @staticmethod def to_safe_utf8(raw_str): """ 过滤html转义 """ text = TemplateCode.to_utf8(raw_str) return text.replace("&", "&").replace("<", "<").replace(">", ">") class Template(object): """模板类""" def __init__(self, input_obj,filename="<string>", **namespace): """模板初始化""" self.namespace = {} self.namespace.update(namespace) # 将数据丢进去解析生成编译代码 self.lexer = TemplateLexer(input_obj, filename) def render(self, **kwargs): """渲染模板 """ temp_name_space = {} temp_name_space.update(self.namespace) temp_name_space.update(kwargs) # 执行渲染 return self.lexer.render(**kwargs) class TemplateLexer(object): """模板语法分析器 """ def __init__(self, input_obb, filename="<string>"): if hasattr(input_obb, "read"): self.raw_string = input_obb.read() else: self.raw_string = input_obb self.filename = filename # 记录当前的位置 self.pos = 0 # 记录原始数据的总长度 self.raw_str_len = len(self.raw_string) # 记录解析的数据 self.code_data = TemplateCode() # 开始解析 self.parse_template() def match(self, keyword, pos=None): return self.raw_string.find(keyword, pos if pos is not None else self.pos) def cut(self, size=-1): """剪取数据 size切割数据的大小，-1表示全部""" if size == -1: new_pos = self.raw_str_len else: new_pos = self.pos + size s = self.raw_string[self.pos: new_pos] self.pos = new_pos return s def remaining(self): """获取剩余大小 """ return self.raw_str_len - self.pos def function_brace(self): """ 获取{{ / {% """ skip_index = self.pos while True: index = self.match(TOKEN_S_BRACE, skip_index) # {% {{ # 没找到 if index == -1: return None, -1 # 末尾 if index >= self.raw_str_len: return None, -1 # 匹配类型 next_value = self.raw_string[index + 1:index + 2] if next_value not in TOKEN_FLAG_SET: skip_index = index + 1 # 说明不是关键类型 continue brace = self.raw_string[index: index + 2] return brace, index return None, -1 def read_content_with_token(self, index, begin_token, end_token): """ 读取匹配token的内容 """ end_index = self.match(end_token) if end_index == -1: return ParseException("{0} missing end token {1}".format(begin_token, end_token)) # 过滤 begin_token self.pos = index + len(begin_token) content = self.cut(end_index - self.pos) # 去除末尾 end_token self.cut(len(end_token)) return content def add_simple_block_statement(self, operator, suffix): if not suffix: raise ParseException("{0} missing content".format(operator)) if operator == TOKEN_KEY_SET: self.code_data.append_statement(suffix) elif operator == TOKEN_KEY_RAW: self.code_data.append_express(suffix, True) else: raise ParseException("{0} is undefined".format(operator)) def parse_template(self): """解析模板 """ # TODO 检查模板文件是否更改过，如果没有则不需要重新解析 self.code_data.reset() # 解析模板原文件 self.__parse() # 生成编译code self.__compiled_code() def render(self, **kwargs): return self.code_data.generate(**kwargs) def __parse(self, control_operator=None, in_loop=False): """开始解析""" while True: if self.remaining() <= 0: if control_operator or in_loop: raise ParseException("%s missing {%% end %%}" % control_operator) break # 读取 {{ {% brace, index = self.function_brace() # 说明没有找到 if not brace: text = self.cut(index) self.code_data.append_text(text) continue else: text = self.cut(index - self.pos) if text: self.code_data.append_text(text) if brace == TOKEN_EXPRESSION_L: content = self.read_content_with_token(index, TOKEN_EXPRESSION_L, TOKEN_EXPRESSION_R).strip() if not content: raise ParseException("Empty Express") self.code_data.append_express(content) continue elif brace == TOKEN_BLOCK_L: content = self.read_content_with_token(index, TOKEN_BLOCK_L, TOKEN_BLOCK_R).strip() if not content: raise ParseException("Empty block") # 得到表达式 for x in x ; if x ; elif x ; else ; end ; set ; while x ; operator, _, suffix = content.partition(TOKEN_SPACE) if not operator: raise ParseException("block missing operator") suffix = suffix.strip() # 简单语句，set / raw if operator in TOKEN_KEY_SET_SIMPLE_EXPRESSION: self.add_simple_block_statement(operator, suffix) elif operator in TOKEN_KEY_LOOP_CTRL: if not in_loop: raise ParseException("{0} must in loop block".format(operator)) self.code_data.append_statement(operator) else: # 控制语句 检查匹配if 后面可以跟elif/else pre_condition = TOKEN_KEY_PRE_CONDITION.get(operator, None) if pre_condition: # 里面就是elif/else/end if control_operator not in pre_condition: raise ParseException("{0} must behind with {1}".format(operator, pre_condition)) elif operator == TOKEN_KEY_END: # 遇到{% end %}则结束 self.code_data.close_code() return else: # 由于是依据if 进入 来计算elif ，因此elif与if是同级的 self.code_data.close_code() self.code_data.append_statement(content + TOKEN_COLON) self.code_data.create_code() self.__parse(operator, in_loop or (operator in TOKEN_KEY_LOOP)) break # 添加控制语句及内部语句体 if for while self.code_data.append_statement(content + TOKEN_COLON) self.code_data.create_code() self.__parse(operator, in_loop or (operator in TOKEN_KEY_LOOP)) else: raise ParseException("Unkown brace") return def __compiled_code(self): """生成 编译code """ self.code_data.build_code(self.filename) if __name__ == "__main__": t = Template("<html>{{hello}}</html>") t.render(hello="你好")
