以产品的思维来做项目或者有产品思维

##  爬取一个电影网站详情页

> 匹配 标题、描述、缩略图、详情、下载链接

[示例](http://mrdoc.tool115.com/media/202303/2023-03-09_213206.png "示例")

1. 实现一

```php
public function parseFileds($document)
{
        // 匹配标题
        // 匹配缩略图
        // 匹配下载地址
        // 匹配描述
        return [
            'title',
            'thumb',
            'detail',
            'download_url',
        ];
}
```

2. 实现二

```php
public function parseTitle($document, $rule)
{
	return $title;
}

public function parseThumb($document, $rule)
{
	return $thumb;
}

public function parseDetail($document, $rule)
{
	return $detail;
}
```

3. 实现三

```php

// 处理器接口
interface InterfaceProcessor
{
    public function run($document, $rule);
}

// 处理器抽象类
abstract class AbstractProcessor
{
    // 公用属性
    protected $filedKey;

// 其它公用方法
}

// 处理器链
class ProcessorChain
{
    protected $processors = [];

public function addProcessor(AbstractProcessor $processor)
    {
        $this->processors[] = $processor;
    }

public function runProcessor($document, $rule)
    {
        $data = [];
        foreach ($this->processors as $processor) {
            $data = array_merge($data, $processor->run($document, $rule));
        }
        return $data;
    }
}

// 处理标题
class ProcessorTitle extends AbstractProcessor implements InterfaceProcessor
{
    protected $filedKey = 'title';

public function run($document, $rule)
    {
        return [
            $this->filedKey => 'xxxx',
        ];
    }
}

// 处理缩略图
class ProcessorThumb extends AbstractProcessor implements InterfaceProcessor
{
    protected $filedKey = 'thumb';

public function run($document, $rule)
    {
        return [
            $this->filedKey => 'xxxx',
        ];
    }
}

$processorChain = new ProcessorChain();
$processorChain->addProcessor(new ProcessorTitle());
$processorChain->addProcessor(new ProcessorThumb());

$data = $processorChain->runProcessor($document, $rule);

//...
```

##  序列化新闻列表

[示例](http://mrdoc.tool115.com/media/202303/7361678369326_pic_hd_1678369698.jpg "示例")

> 三图、广告、大图、视频、大鱼号

1. 实现一

```php
   // 获取序列化后的列表
    public function lists()
    {
        $data = [];
        foreach ($rows as $row) {
            switch ($row['type']) {
                case 'article':
                    $data[] = $this->parseArticle($row);
                    break;
                case 'video':
                    $data[] = $this->parseVideo($row);
                    break;
                case 'ad':
                    $data[] = $this->parseAd($row);
                    break;
            }
        }
        return $data;
    }

public function parseArticle($content)
    {
        return $article;
    }

public function parseVideo($content)
    {
        return $video;
    }

public function parseAd($content)
    {
        return $ad;
    }
```

2. 实现二

```php

// 处理器接口
interface InterfaceProcessor
{
    public function run($row);
}

// 处理器抽象类
abstract class AbstractProcessor
{
    // 公用属性

// 其它公用方法
}

// 处理器链
class ProcessorChain
{
    protected $processors = [];

public function addProcessor(AbstractProcessor $processor)
    {
        $this->processors[] = $processor;
    }

public function runProcessor($row)
    {
        foreach ($this->processors as $processor) {
            $row = $processor->run($row);
        }
        return $row;
    }
}

class ProcessorArticle extends AbstractProcessor implements InterfaceProcessor
{
    public function run($row)
    {

}
}

class ProcessorVideo extends AbstractProcessor implements InterfaceProcessor
{
    public function run($row)
    {

}
}

class ProcessorAd extends AbstractProcessor implements InterfaceProcessor
{
    public function run($row)
    {

}
}

//... 更多类型数据  例如：图片、图集

$processorChain = new ProcessorChain();
$processorChain->addProcessor(new ProcessorArticle());
$processorChain->addProcessor(new ProcessorVideo());
$processorChain->addProcessor(new ProcessorAd());

$data = [];
foreach ($rows as $row) {
    $data[] = $processorChain->runProcessor($row);
}

```

- 多记录日志，尤其是与三方对接的场景，记录请求、响应
- 适当的留`后门`  便于排查问题 
- 多写幂等的业务代码    -  批量消息推送