美文网首页
Java抓取公众号(服务号)所有文章

Java抓取公众号(服务号)所有文章

作者: jsjack_wang | 来源:发表于2018-02-05 15:33 被阅读0次

    1.添加Maven依赖

    <dependency>
        <groupId>org.seleniumhq.selenium</groupId>
        <artifactId>selenium-server</artifactId>
        <version>3.0.1</version>
    </dependency>
    

    2.下载chromedriver

    http://npm.taobao.org/mirrors/chromedriver/
    

    注意:chrome和chromedriver版本一定要一致,不然会出现一些小问题

    3.小栗子

    private static String webDriver = "webdriver.chrome.driver";
    private static String webDriverPath = "C:\\MineProjects\\JavaDemo\\chromedriver_win32\\chromedriver.exe";
    private static String targetPath = "https://mp.weixin.qq.com";
    private static String searchPath = "https://mp.weixin.qq.com/cgi-bin/searchbiz";
    private static String appmsgPath = "https://mp.weixin.qq.com/cgi-bin/appmsg";
    private static Random random = new Random(1);
    private static  Gson gson = new Gson();
    
    private static String sourceName = "渔愉鱼";   // 要爬的公众号名称(准确名称)
    
    private static String username = null;
    private static String password = null;
    
    static {
        ResourceBundle rb = ResourceBundle.getBundle("reptile");
        username = rb.getString("reptile.username");
        password = rb.getString("reptile.password");
    }
    
    public static void main(String[] args) {
        System.setProperty(webDriver, webDriverPath);
        WebDriver driver = null;
        try {
            driver = new ChromeDriver();
    
            weixinLogin(driver);
    
            String token = getToken(driver);
    
            Platform platform = getPlatform(driver, token);
            if (Objects.isNull(platform)) {
                throw new Exception("不存在" + sourceName + "公众号");
            }
            InfoResult infoResult = getInfoResult(driver, token, platform.getFakeId(), 0, 5);
            System.out.println(infoResult);
        } catch (Exception e) {
            e.printStackTrace();
        } finally {
            if (Objects.nonNull(driver)) {
                driver.close();
            }
        }
    }
    
    /**
     * 获取公众号某一页数据
     */
    private static InfoResult getInfoResult(WebDriver driver, String token, String fakeId,
                                            Integer begin, Integer count) throws Exception {
        Map<String, String> queryInfoParams = new HashMap<>();
        queryInfoParams.put("token", token);
        queryInfoParams.put("lang", "zh_CN");
        queryInfoParams.put("f", "json");
        queryInfoParams.put("ajax", "1");
        queryInfoParams.put("random", random.nextDouble() + "");
        queryInfoParams.put("action", "list_ex");
        queryInfoParams.put("query", "");
        queryInfoParams.put("type", "9");
        queryInfoParams.put("fakeid", fakeId);
    
        queryInfoParams.put("begin", begin + "");
        queryInfoParams.put("count", count + "");
    
        appmsgPath = HttpUtils.setParams(appmsgPath, queryInfoParams);
        driver.get(appmsgPath);
        Document infoDocument = Jsoup.parse(driver.getPageSource());
        Elements infoList = infoDocument.select("pre");
        if (Objects.isNull(infoList)) {
            throw new Exception("获取公众号文章错误");
        }
        return gson.fromJson(infoList.text(), InfoResult.class);
    }
    
    /**
     * 获取公众号信息
     */
    private static Platform getPlatform(WebDriver driver, String token) throws Exception {
        Map<String, String> searchNameParams = new HashMap<>();
        searchNameParams.put("action", "search_biz");
        searchNameParams.put("token", token);
        searchNameParams.put("lang", "zh_CN");
        searchNameParams.put("f", "json");
        searchNameParams.put("ajax", "1");
        searchNameParams.put("random", random.nextDouble() + "");
        searchNameParams.put("query", sourceName);
        searchNameParams.put("begin", "0");
        searchNameParams.put("count", "5");
    
        searchPath = HttpUtils.setParams(searchPath, searchNameParams);
        driver.get(searchPath);
    
        Document preDocument = Jsoup.parse(driver.getPageSource());
        Elements preList = preDocument.select("pre");
        if (Objects.isNull(preList)) {
            throw new Exception("获取公众号错误");
        }
        PlatformResult result = gson.fromJson(preList.text(), PlatformResult.class);
    
        Platform platform = null;
        for (int index = 0; index < result.getList().size(); index ++) {
            Platform item = result.getList().get(index);
            if (sourceName.equals(item.getNickname())) {
                platform = item;
            }
        }
        return platform;
    }
    
    /**
     * 获取token
     */
    private static String getToken(WebDriver driver) throws Exception {
        String current = driver.getCurrentUrl();
        if (StringUtils.isBlank(current)) {
            throw new Exception("获取token链接有误");
        }
        String token = current.split("token=")[1];
        if (StringUtils.isBlank(token)) {
            throw new Exception("token错误");
        }
        return token;
    }
    
    /**
     * 登录模块
     */
    private static void weixinLogin(WebDriver driver) throws Exception {
        driver.get(targetPath);
        WebElement usernameWebElement = driver.findElement(By.name("account"));
        usernameWebElement.clear();
        usernameWebElement.sendKeys(username);
    
        WebElement passwordWebElement = driver.findElement(By.name("password"));
        passwordWebElement.clear();
        passwordWebElement.sendKeys(password);
    
        WebElement helpWebElement = driver.findElement(By.className("icon_checkbox"));
        helpWebElement.click();
    
        WebElement btnWebElement = driver.findElement(By.className("btn_login"));
        btnWebElement.click();
    
        System.out.println("请用手机微信扫码二维码登录公众号");
        Thread.sleep(15000);
    }
    

    4.总结

    以前没有用过selenium,这2天研究了一下,还是挺好玩的。抢票什么的,用这个应该可以做。这里贴的只是部分源码,可以直接到码云上,看我写的ReptileDemo。
    

    源码地址:https://gitee.com/jsjack_wang/JavaDemo

    相关文章

      网友评论

          本文标题:Java抓取公众号(服务号)所有文章

          本文链接:https://www.haomeiwen.com/subject/uzfizxtx.html