XML即可扩展标记语言(Extensible Markup Language), 其被设计为传输和存储数据,其焦点是数据的内容。 而熟悉的HTML则是 被设计用来显示数据,其焦点是数据的外观。
python里面操作XML有两种方法:DOM和SAX。
DOM会把整个XML读入内存,解析为树,因此占用内存大,解析慢,优点是可以任意遍历树的节点。
SAX是流模式,边读边解析,占用内存小,解析快,缺点是我们需要自己处理事件。
但是由于DOM太占内存所以一般采用SAX。
用SAX解析XML文件的时候我们通常只用关注三个件:start_element(读取标签头),end_element(读取标签尾)以及char_data(标签中间的内容)
比如要从这段xml代码里面获取到获取当天和第二天的天气
data = r'''<?xml version="1.0" encoding="UTF-8" standalone="yes" ?> <rss version="2.0" xmlns:yweather="http://xml.weather.yahoo.com/ns/rss/1.0" xmlns:geo="http://www.w3.org/2003/01/geo/wgs84_pos#"> <channel> <title>Yahoo! Weather - Beijing, CN</title> <lastBuildDate>Wed, 27 May 2015 11:00 am CST</lastBuildDate> <yweather:location city="Beijing" region="" country="China"/> <yweather:units temperature="C" distance="km" pressure="mb" speed="km/h"/> <yweather:wind chill="28" direction="180" speed="14.48" /> <yweather:atmosphere humidity="53" visibility="2.61" pressure="1006.1" rising="0" /> <yweather:astronomy sunrise="4:51 am" sunset="7:32 pm"/> <item> <geo:lat>39.91</geo:lat> <geo:long>116.39</geo:long> <pubDate>Wed, 27 May 2015 11:00 am CST</pubDate> <yweather:condition text="Haze" code="21" temp="28" date="Wed, 27 May 2015 11:00 am CST" /> <yweather:forecast day="Wed" date="27 May 2015" low="20" high="33" text="Partly Cloudy" code="30" /> <yweather:forecast day="Thu" date="28 May 2015" low="21" high="34" text="Sunny" code="32" /> <yweather:forecast day="Fri" date="29 May 2015" low="18" high="25" text="AM Showers" code="39" /> <yweather:forecast day="Sat" date="30 May 2015" low="18" high="32" text="Sunny" code="32" /> <yweather:forecast day="Sun" date="31 May 2015" low="20" high="37" text="Sunny" code="32" /> </item> </channel> </rss> ''' from xml.parsers.expat import ParserCreate class A(object): def __init__(self): self._data = {} self._count = 0 def get_data(self): return self._data def start_element(self,name,attrs): if name == 'yweather:location': self._data['city'] = attrs['city'] self._data['country'] = attrs['country'] elif name == 'yweather:forecast': if self._count == 0: self._data['today'] = {} self._data['today']['text'] = attrs['text'] self._data['today']['low'] = attrs['low'] self._data['today']['high'] = attrs['high'] self._count += 1 elif self._count ==1: self._data['tomorrow'] = {} self._data['tomorrow']['text'] = attrs['text'] self._data['tomorrow']['low'] = attrs['low'] self._data['tomorrow']['high'] = attrs['high'] self._count += 1 def end_element(self,name): pass def char_data(self,text): pass def parse_weather(xml): p = ParserCreate() a = A() p.StartElementHandler = a.start_element p.EndElementHandler = a.end_element p.CharacterDataHandler = a.char_data p.Parse(xml) return a.get_data() weather = parse_weather(data) assert weather['city'] == 'Beijing', weather['city'] #这些assert语句是一个断点,如果城市不是北京就报错 assert weather['country'] == 'China', weather['country'] assert weather['today']['text'] == 'Partly Cloudy', weather['today']['text'] assert weather['today']['low'] == '20', weather['today']['low'] assert weather['today']['high'] == '33', weather['today']['high'] assert weather['tomorrow']['text'] == 'Sunny', weather['tomorrow']['text'] assert weather['tomorrow']['low'] == '21', weather['tomorrow']['low'] assert weather['tomorrow']['high'] == '34', weather['tomorrow']['high'] print('weather:',str(weather)) ''' 输出: weather: {'country': 'China', 'today': {'text': 'Partly Cloudy', 'low': '20', 'high': '33'}, 'tomorrow': {'text': 'Sunny', 'low': '21', 'high': '34'}, 'city': 'Beijing'} '''从这一段xml代码里面我们发现所需要的内容全部都是在标签头里面,所以只需要在start_element方法里面协商合适的判断条件去获得所需要的内容即可。