文件名称：python 最牛的解析html的方法
文件大小：15KB
文件格式：PY
更新时间：2014-11-20 03:26:29
python html 解析
class MyHTMLParser(HTMLParser.HTMLParser): def __init__(self): self.selected = ['html', 'body', 'table', 'head', 'meta', 'style', 'tr', 'td'] self.reset() self._level_stack = [] self.flag = True global style_list, br_list, tr_list, td_list, td_class_list style_list, br_list, tr_list, td_list, td_class_list = [], [], [], [], [] def handle_starttag(self, tag, attrs): if tag in self.selected: self._level_stack.append(tag) if tag == 'td': if len(attrs)==0: if self.flag: td_class_list.append('') else: for(varviable,value) in attrs: if self.flag: td_class_list.append(value) elif tag == 'tr': if len(attrs) != 0: for(varviable,value) in attrs: if varviable == 'class' and value == 'SectionGap': self.flag = False else: self.flag = True tr_list.append(value) def handle_endtag(self, tag): if self._level_stack and tag in self.selected and tag == self._level_stack[-1]: self._level_stack.pop() def handle_data(self, data): for key in codemap.keys(): data = data.replace(codemap[key], key) if '/'.join(self._level_stack) == 'html/head/meta/body/table/tr/td' and self.flag: td_list.append(data) elif '/'.join(self._level_stack) == 'html/head/meta/body': br_list.append(data) elif '/'.join(self._level_stack) == 'html/head/meta/style': style_list.append(data)