使用第三方类库对html进行解析

一时失言乱红尘 2021-11-22 10:00 291阅读 0赞

html解析最重要的就是看清楚节点，看是用DIV取还是用class，搞清楚结构之后，解析规范的网页都不是什么问题。

如果网页不规范，则要看具体情况而定了

把NSData转成NSString类型的数据

NSString * str = [[NSString alloc]initWithData:data encoding:NSUTF8StringEncoding];

把NSString类型的数据转成DocumentRoot类型的文件（DocumentRoot是第三方类库提供的类，把数据转成这种类型才能用第三方类库进一步解析）

DocumentRoot * document = [Element parseHTML:str];

取出所有的DIV:

NSArray * childSecond = [childEl selectElements:@"div"];

取class为item的div块做进一步解析

for (Element* element in elements) {
            if ([[element attribute:@"class"] isEqualToString:@"item"]) {
            }
        }

该块DIV为

<div class="item">
    					<div class="pic"><a href="http://www.weiphone.com/iPhone/news/2013-08-02/Come_the_Blackberry_company_to_send_BBM_for_iOS_beta_invites_560440.shtml"><img src="http://resource.weiphone.com/resource/h027/h73/img201308021306430.jpg" alt="" height="100" width="158" /></a></div>
    					<div class="head">
    						<h3><a href="http://www.weiphone.com/iPhone/news/2013-08-02/Come_the_Blackberry_company_to_send_BBM_for_iOS_beta_invites_560440.shtml">快来了 黑莓公司发送BBM for iOS测试邀请</a></h3>
    						<div class="meta">
    							<span class="timer" title="发表时间">2013/08/02 13:05</span>
    							<span class="line">|</span>
    							<a href="http://bbs.weiphone.com/u.php?uid=798198" class="author" title="作者"> 黄晓闷</a>							<span class="line">|</span>
    							<a href="javascript:void(0);" class="link" title="文章来源">weiphone</a>
    							<div class="funs">
    								<span class="view" title="浏览次数">2689</span>
    								<span class="line">|</span>
    								<a href="http://www.weiphone.com/iPhone/news/2013-08-02/Come_the_Blackberry_company_to_send_BBM_for_iOS_beta_invites_560440.shtml#comment" class="cmt" title="评论次数">5</a>
    							</div>
    						</div>
    					</div>
    					<div class="desc">
    						<p>威锋网 8 月 2 日消息，黑莓公司日前向 iOS 用户发送了 BBM 的测试邀请，暗示着该服务的正式登陆已经进入最后阶段。</p>
    					</div>
    				</div>

把class为item的DIV的下级DIV全部取出来放入childSecond

NSArray * childSecond = [childEl selectElements:@"div"];

取标签包围的内容

Element * child = [secondEl selectElement:@"a"];

取标签尖括号里的内容

new.detailURL = [child.attributes objectForKey:@"href"];

以下是一个完整的解析方法，解析的网页为http://www.weiphone.com/iPhone/news/index\_0.shtml的class为item的DIV

这个方法是在下载完成调用的，传递一个NSData类型的参数进去：

数据模型:

//
    //  News.h
    //  LookNewsProject
    //
    //  Created by ibokan on 13-08-01.
    //  Copyright (c) 2013年 laomaoshiba. All rights reserved.
    //
    
    #import <Foundation/Foundation.h>
    
    @interface News : NSObject
    
    //标题,发布时间,详情链接,图片链接,浏览次数,评价,类别,作者,简介,来源
    @property(copy,nonatomic)NSString * title, * publishTime, * detailURL, * imgURL, * viewTimes, * evaluateTimes, * category, * author, * intro, * origin;
    @end

解析方法：

-(void) analyNews:(NSData *)data
    {
        //中文转码
        NSString * str = [[NSString alloc]initWithData:data encoding:NSUTF8StringEncoding];
        //NSLog(@"%@",str);
        //html解析
        DocumentRoot * document = [Element parseHTML:str];
        //以div分割
        NSArray * elements= [document selectElements:@"div"];
        //创建存储数组
        NSMutableArray * newArr = [[NSMutableArray alloc]init];
        //循环解析
        for (Element* element in elements){
            if ([[element attribute:@"class"] isEqualToString:@"item"]){
                NSArray * childElement = [element childElements];
                //创建新闻实体
                News * new = [[News alloc]init];
                int i=0;
                for(Element* childEl in childElement){
                    i++;
                    NSArray * childSecond = [childEl selectElements:@"div"];
                    for(Element * secondEl in childSecond){
                        if([[secondEl attribute:@"class"] isEqualToString:@"pic"]){
                            Element * child = [secondEl selectElement:@"a"];
                            //获取详细信息
                            //NSLog(@"详细链接：%@",[child.attributes objectForKey:@"href"]);
                            new.detailURL = [child.attributes objectForKey:@"href"];
                            //获取图片链接
                            //NSLog(@"图片链接：%@",[[child selectElement:@"img"].attributes objectForKey:@"src"]);
                            new.imgURL = [[child selectElement:@"img"].attributes objectForKey:@"src"];
                        }
                        else if([[secondEl attribute:@"class"] isEqualToString:@"head"]){
                            //获取新闻标题
                            //NSLog(@"标题：%@",[[secondEl selectElement:@"a"] contentsSource]);
                            new.title = [[secondEl selectElement:@"a"] contentsSource];
                        }
                        else if([[secondEl attribute:@"class"] isEqualToString:@"meta"]){
                            //获取作者
                            //NSLog(@"test-------->%@",[[secondEl selectElement:@"div"] contentsSource]);
                            //NSLog(@"作者：%@",[[secondEl selectElement:@"a"] contentsSource]);
                            new.author = [[secondEl selectElement:@"a"] contentsSource];
                            //获取发表时间
                            //NSLog(@"发表时间：%@",[[secondEl selectElement:@"span"] contentsSource]);
                            new.publishTime = [[secondEl selectElement:@"span"] contentsSource];
                            //获取来源
                            Element * originEl = [[secondEl selectElements:@"a"] objectAtIndex:1] ;
                            //NSLog(@"来源：%@",[originEl contentsSource]);
                            new.origin = [originEl contentsSource];
                        }
                        else if([[secondEl attribute:@"class"] isEqualToString:@"funs"]){
                            //获取浏览
                            //NSLog(@"浏览次数：%@",[[secondEl selectElement:@"span"] contentsSource]);
                            new.viewTimes = [[secondEl selectElement:@"span"] contentsSource];
                            //获取评价次数
                            //NSLog(@"评价次数：%@",[[secondEl selectElement:@"a"] contentsSource]);
                            new.evaluateTimes = [[secondEl selectElement:@"a"] contentsSource];
                        }
                        else if([[secondEl attribute:@"class"] isEqualToString:@"desc"]){
                            //获取简介
                            //NSLog(@"简介：%@",[[secondEl selectElement:@"p"] contentsSource]);
                            new.intro = [[secondEl selectElement:@"p"] contentsSource];
                        }
                    }
                }
                
                [newArr addObject:new];
            }
        }
        [str release];
        [newArr release];
        
    }