天气小工具 <三>

之前我们能得到省份，市，区的唯一ID，则我们可以根据这个ID获得当前想要查看的地区的天气状况。
继续使用我们的HttpWatch分析。例如我们希望知道江苏省南京市南京市区的天气状况。通过之前的城市ID分析，我们知道江苏省ID为10119，南京市ID01，南京市区ID为01，则查询ID为101190101，天气查询URL为：http://www.weather.com.cn/weather/101190101.shtml

这个返回的信息很符合HTML的标准。
我们可以使用HTMLParser提取信息，当然也可以使用正则表达式。

// 制定得到天气的Filter
			NodeFilter prosFilter = new TagNameFilter("table");
			NodeFilter prosAttrFilter = new HasAttributeFilter("class",
					"yuBaoTable");
			AndFilter filters = new AndFilter(prosFilter, prosAttrFilter);
			// 得到对应的天气
			NodeList nodes = parser.extractAllNodesThatMatch(filters);
			for (int i = 0; i < nodes.size(); i++) {
				Node node = nodes.elementAt(i);
				String weatherInfo = node.toPlainTextString();
				weatherInfo = weatherInfo.replaceAll("\\s", "");
}

因为提取出来的信息会有很多的空格符，就将其替换为空，weatherInfo.replaceAll("\\s", "");

最终得到的信息为：
15日星期三白天小雨高温23℃东风3-4级夜间阴低温19℃东风3-4级15日,星期三,白天:小雨,高温:高温23℃,风:东风,风力3-4级,夜间:阴,温度:低温19℃,风:东风,风力:3-4级
16日星期四白天阴高温28℃东风3-4级夜间多云低温21℃东南风3-4级16日,星期四,白天:阴,高温:高温28℃,风:东风,风力3-4级,夜间:多云,温度:低温21℃,风:东南风,风力:3-4级
17日星期五白天多云高温31℃东南风3-4级夜间阴低温23℃东南风3-4级17日,星期五,白天:多云,高温:高温31℃,风:东南风,风力3-4级,夜间:阴,温度:低温23℃,风:东南风,风力:3-4级
18日星期六白天雷阵雨高温28℃东风3-4级夜间中雨低温22℃东北风3-4级18日,星期六,白天:雷阵雨,高温:高温28℃,风:东风,风力3-4级,夜间:中雨,温度:低温22℃,风:东北风,风力:3-4级
19日星期日白天中雨高温25℃东北风3-4级夜间阵雨低温21℃东北风3-4级19日,星期日,白天:中雨,高温:高温25℃,风:东北风,风力3-4级,夜间:阵雨,温度:低温21℃,风:东北风,风力:3-4级
20日星期一白天阴高温29℃北风3-4级夜间多云低温22℃东风3-4级20日,星期一,白天:阴,高温:高温29℃,风:北风,风力3-4级,夜间:多云,温度:低温22℃,风:东风,风力:3-4级
21日星期二白天阴高温31℃东风3-4级21日,星期二,白天:阴,高温:高温31℃,风:东风,风力3-4级


为了方便信息，我们将每天的天气信息创建一个类。
类的属性有：日期，星期，白天天气，白天最高温，白天风向，白天风力，夜间天气，夜间最低温，夜间风向，夜间风力。

但是比较不爽的是，不是每天都会有白天信息或者夜间信息。例如21日。
所以在使用正则表达式提取上诉信息的时候分为三种情况：
1.白天和夜间都有
2.只有白天
3.只有夜间

不要抓狂，抓狂的来了。
不是所有的天气状况都是有风向和风力的。
我们得到了如下的天气状况：
15日星期三白天雷阵雨高温32℃无持续风向微风夜间中雨低温19℃无持续风向微风
竟然出现了什么无持续风向和微风。
好吧~使用正则表达式提取信息的时候，我们又得分出一种情况了！

代码如下：
参数content为城市天气信息页面返回的信息；

	public static Vector<Day> GetWeather(String content) {
		Vector<Day> days = new Vector<Day>();
		// 根据所获得的天气网页信息得到未来七天温度信息
		Parser parser;
		try {
			parser = new Parser(content);
			// 制定得到天气的Filter
			NodeFilter prosFilter = new TagNameFilter("table");
			NodeFilter prosAttrFilter = new HasAttributeFilter("class",
					"yuBaoTable");
			AndFilter filters = new AndFilter(prosFilter, prosAttrFilter);
			// 得到对应的天气
			NodeList nodes = parser.extractAllNodesThatMatch(filters);
			for (int i = 0; i < nodes.size(); i++) {
				Node node = nodes.elementAt(i);
				String weatherInfo = node.toPlainTextString();
				weatherInfo = weatherInfo.replaceAll("\\s", "");
//				System.out.print(weatherInfo);
				// 得到的天气信息有两种,一种白天和夜间都有,一种只有白天
				if (weatherInfo.indexOf("夜") != -1
						&& weatherInfo.indexOf("白") != -1) {
					String day = "", week = "", h_weather = "", high = "", h_wind = "", h_wind_level = "", l_weather = "", low = "", l_wind = "", l_wind_level = "";
					if (weatherInfo.indexOf("风向") != -1) {
						Pattern adayPattern = Pattern
								.compile(MyRegex.WeatherRegexNoWind);
						Matcher adayMatcher = adayPattern.matcher(weatherInfo);
						while (adayMatcher.find()) {
							day = adayMatcher.group(1);
							week = adayMatcher.group(2);
							h_weather = adayMatcher.group(3);
							high = "高温" + adayMatcher.group(4) + "℃";
							h_wind = adayMatcher.group(5) + "风向"; // 白天的风向
							h_wind_level = adayMatcher.group(6) + "风"; // 风力
							l_weather = adayMatcher.group(7);
							low = "低温" + adayMatcher.group(8) + "℃";
							l_wind = adayMatcher.group(9) + "风向";
							l_wind_level = adayMatcher.group(10) + "风";
						}
					} else {
						Pattern adayPattern = Pattern
								.compile(MyRegex.WeatherRegex);
						Matcher adayMatcher = adayPattern.matcher(weatherInfo);
						while (adayMatcher.find()) {
							day = adayMatcher.group(1);
							week = adayMatcher.group(2);
							h_weather = adayMatcher.group(3);
							high = "高温" + adayMatcher.group(4) + "℃";
							h_wind = adayMatcher.group(5) + "风"; // 白天的风向
							h_wind_level = adayMatcher.group(6) + "级"; // 风力
							l_weather = adayMatcher.group(7);
							low = "低温" + adayMatcher.group(8) + "℃";
							l_wind = adayMatcher.group(9) + "风";
							l_wind_level = adayMatcher.group(10) + "级";
						}
					}
					Day aday = new Day(day, week, h_weather, high, h_wind,
							h_wind_level, l_weather, low, l_wind, l_wind_level);
					days.add(aday);
//					System.out.println(aday);

				} else if (weatherInfo.indexOf("白") != -1) {
					String day = "", week = "", h_weather = "", high = "", h_wind = "", h_wind_level = "";
					if (weatherInfo.indexOf("风向") != -1) {
						Pattern adayPattern = Pattern
								.compile(MyRegex.WeatherDayRegexNoWind);
						Matcher adayMatcher = adayPattern.matcher(weatherInfo);
						while (adayMatcher.find()) {
							day = adayMatcher.group(1);
							week = adayMatcher.group(2);
							h_weather = adayMatcher.group(3);
							high = "高温" + adayMatcher.group(4) + "℃";
							h_wind = adayMatcher.group(5) + "风向"; // 白天的风向
							h_wind_level = adayMatcher.group(6) + "风"; // 风力
						}
					} else {
						Pattern adayPattern = Pattern
								.compile(MyRegex.WeatherDayRegex);
						Matcher adayMatcher = adayPattern.matcher(weatherInfo);
						while (adayMatcher.find()) {
							day = adayMatcher.group(1);
							week = adayMatcher.group(2);
							h_weather = adayMatcher.group(3);
							high = "高温" + adayMatcher.group(4) + "℃";
							h_wind = adayMatcher.group(5) + "风"; // 白天的风向
							h_wind_level = adayMatcher.group(6) + "级"; // 风力
						}
					}
					Day aday = new Day(day, week, h_weather, high, h_wind,
							h_wind_level, "", "", "", "");
					days.add(aday);
//					System.out.println(aday);
				} else {
					String day = "", week = "", l_weather = "", low = "", l_wind = "", l_wind_level = "";
					if (weatherInfo.indexOf("风向") != -1) {
						Pattern adayPattern = Pattern
								.compile(MyRegex.WeatherNightRegexNoWind);
						Matcher adayMatcher = adayPattern.matcher(weatherInfo);
						while (adayMatcher.find()) {
							day = adayMatcher.group(1);
							week = adayMatcher.group(2);
							l_weather = adayMatcher.group(3);
							low = "低温" + adayMatcher.group(4) + "℃";
							l_wind = adayMatcher.group(5) + "风向";
							l_wind_level = adayMatcher.group(6) + "风";
						}
					} else {
						Pattern adayPattern = Pattern
								.compile(MyRegex.WeatherNightRegex);
						Matcher adayMatcher = adayPattern.matcher(weatherInfo);
						while (adayMatcher.find()) {
							day = adayMatcher.group(1);
							week = adayMatcher.group(2);
							l_weather = adayMatcher.group(3);
							low = "低温" + adayMatcher.group(4) + "℃";
							l_wind = adayMatcher.group(5) + "风";
							l_wind_level = adayMatcher.group(6) + "级";

						}
						Day aday = new Day(day, week, "", "", "", "",
								l_weather, low, l_wind, l_wind_level);
						days.add(aday);
//						System.out.println(aday);
					}
				}
			}
		} catch (ParserException e) {
			e.printStackTrace();
		}
		return days;
	}

这部分其实就是信息的提取和分析。


目前测试使用的界面
[img]http://dl.iteye.com/upload/picture/pic/92366/b9a87d38-9076-36a4-a9fe-510366908c69.bmp[/img]