基于爬虫实现技术文章爬虫存储到数据库

一、需求分析

网站：https://www.cnblogs.com/

分析：

需要的数据：标题、摘要、原文地址、发布时间

存储数据库

 

二、设计数据库

标题、摘要、原文地址、发布时间

文章表：id主键 title标题summary摘要detailurl详细地址pubtime发布时间ctime创建时间

SQL脚本：

create database db_data1906;
use db_data1906;
create  table t_bkyarticle(id int primary key auto_increment,title varchar(100),summary text,detailurl varchar(200),pubtime date,ctime date);

三、实现编码

技术栈：SpringBoot

1.新建项目

SpringBoot

2、依赖jar

3、逐层编写代码

实体层

@TableName("t_bkyarticle")
@Data
public class BkyArticle {
    @TableId(type = IdType.AUTO)
    private Integer id;
    private String title;
    private String summary;
    private String detailurl;
    private Date pubtime;
    private Date ctime;
}

持久层

public interface BkyArticleDao extends BaseMapper<BkyArticle> {
    @Insert("insert into t_bkyarticle(title,summary,detailurl,pubtime,ctime) values(#{title},#{summary},#{detailurl},#{pubtime},now())")
    int save(BkyArticle article);
}

业务逻辑层

public interface BkyArticleService extends IService<BkyArticle> {
    boolean saveEntity(BkyArticle article);
}

@Service
public class BkyArticleServiceImpl extends ServiceImpl<BkyArticleDao, BkyArticle> implements BkyArticleService {
​
    @Override
    public boolean saveEntity(BkyArticle article) {
        return getBaseMapper().save(article)>0;
    }
}

4、编写爬虫核心代码

自定义页面处理器

@Service
public class BkyArticlePage implements PageProcessor {
​
    private String baseUrl="https://www.cnblogs.com/";
    @Override
    public void process(Page page) {
        //1、解析当前页面的内容
        List<String> titles=page.getHtml().xpath("div[@id='post_list']/div[@class='post_item']/div[@class='post_item_body']/h3/a/text()").all();
        List<String> urls=page.getHtml().xpath("div[@id='post_list']/div[@class='post_item']/div[@class='post_item_body']/h3/a/@href").all();
        List<String> infos=page.getHtml().xpath("div[@id='post_list']/div[@class='post_item']/div[@class='post_item_body']/p[@class='post_item_summary']/text()").all();
        List<String> times=page.getHtml().xpath("div[@id='post_list']/div[@class='post_item']/div[@class='post_item_body']/div[@class='post_item_foot']/a/text()").all();
        //2、组装解析的结果
        List<BkyArticle> articles=new ArrayList<>();
        for(int i=0;i<titles.size();i++){
            BkyArticle article=new BkyArticle();
            article.setTitle(titles.get(i));
            article.setSummary(infos.get(i));
            article.setDetailurl(urls.get(i));
            article.setPubtime(parseTime(getTimeStr(times.get(i))));
            articles.add(article);
        }
        //3、传递给了结果处理器
        page.putField("list",articles);
​
        //4、分页查询 获取分页的路径并标记继续爬取
        if(page.getUrl().get().equals(baseUrl)){
            //计算所有的分页请路径
            List<String> pageurls=new ArrayList<>();
            List<String>allpages=page.getHtml().xpath("div[@id='paging_block']/div[@class='pager']/a/text()").all();
            int maxPage=Integer.parseInt(allpages.get(allpages.size()-2));
            for(int i=2;i<=maxPage;i++){
                pageurls.add(baseUrl+"/sitehome/p/"+i);
            }
            //设置继续爬取的网页
            page.addTargetRequests(pageurls);
        }
    }
    private String getTimeStr(String s){
        String s1=s.trim();
        if(s1.indexOf(" ")>0){
            return s.substring(s.indexOf(' ')+1);
        }else {
            return null;
        }
    }
    private Date parseTime(String time){
        if(time!=null) {
            SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm");
            try {
                return sdf.parse(time);
            } catch (ParseException e) {
                e.printStackTrace();
                return new Date();
            }
        }else {
            return new Date();
        }
    }
    private Site site=Site.me().setTimeOut(6000).setSleepTime(2000);
​
    @Override
    public Site getSite() {
        return site;
    }
}

结果处理器

@Repository
public class BkyArticPipeline implements Pipeline {
    @Autowired
    private BkyArticleDao bkyArticleDao;
    @Override
    public void process(ResultItems resultItems, Task task) {
        List<BkyArticle> articleList=resultItems.get("list");
        System.out.println("爬取数据："+articleList.size());
        for(BkyArticle a:articleList){
            bkyArticleDao.save(a);
        }
    }
}

5、编写启动接口

控制器 实现爬取的运行

@Api
@RestController
public class BkyArticController {
    @Autowired
    private BkyArticleService bkyArticleService;
    @Autowired
    private BkyArticlePage page;
    @Autowired
    private BkyArticPipeline pipeline;
    //启动爬虫
    @GetMapping("/api/spider/start.do")
    public R start(){
        Spider.create(page).addPipeline(pipeline).addUrl("https://www.cnblogs.com/").thread(5).run();
        return R.ok("爬取已经启动");
    }
    //查询爬取数据
    @GetMapping("api/bkyartic/all.do")
    public R all(){
        return R.ok(bkyArticleService.list());
    }
}

6、配置Swagger

@Configuration //配置文件
public class SwaggerConfig {
    //创建文档说明
    public ApiInfo createAI(){
        ApiInfo apiInfo=new ApiInfoBuilder().title("文章接口").description("实现一款基于爬虫实现的数据接口").contact(new Contact("Feri","http://www.17feri.top","[email protected]")).build();
        return apiInfo;
    }
    //创建Swagger扫描信息
    @Bean
    public Docket createD(){
        return new Docket(DocumentationType.SWAGGER_2).apiInfo(createAI()).select().
                apis(RequestHandlerSelectors.basePackage("com.feri.point.controller")).build();
    }
}

7、启动测试