Category: PHP

基于 PHP 和 word2vec 的简单分类器，用于文章、新闻等内容自动分类，项目包含样本训练、识别代码，

分词组件用的是 PhpAnalysis，简单灵活。欢迎大家一起优化并完善。

项目地址：

码云: https://gitee.com/mz/classifier4php

Github: https://github.com/djunny/classifier4php

/**
 * 删除指定标签
 * @param array  $tags    删除的标签
 * @param string $str     html字符串
 * @param bool   $type    是否保留标签的内容
 * @return mixed
 * @usage
 *  $str = 'Hello</p><p><img src="http://www.baidu.com" /></b><Video>';
 *
 *      // echo strip_tags($str, '<img');
 *      echo $this->stripHtmlTags(array('b','p'),$str,true);
 */
public static function stripHtmlTags($str,$tags=['p','b'], $content=true)
{
    $html = [];
    if($content){
        foreach ($tags as $tag) {
            $html[] = "/(<(?:\/" .$tag. "|" .$tag. ")[^>]*>)/is";
        }
    }else{
        foreach ($tags as $tag) {
            $html[] = '/<' .$tag. '.*?>[\s|\S]*?<\/' .$tag. '>/is';
            $html[] = '/<' .$tag. '.*?>/is';
        }
    }
    $data = preg_replace($html, '', $str);
    return $data;
}

$filetype = array(".docx",".doc",".xlsx",".xls",".pptx",".ppt",".jpg",".png",".pdf"); //文件类型
$tempFile = "/uploads/".$log["attachment"];                 //$log["attachment"]为文件地址;
$url = str_replace($filetype,"",$tempFile).".pdf";             //替换文件后缀
//header('Location: '.$url);
//die();
$tempFile = "/tmp/".basename($log["attachment"]);             //临时文件地址
copy("www.xxx.com/".$log["attachment"],$tempFile);             //移动文件
exec("unoconv -f pdf ".$tempFile);                           //文件转pdf
$pdf = str_replace($filetype,"",$tempFile).".pdf";           
header("Content-type:application/pdf");
// 文件将被称为 downloaded.pdf
//header("Content-Disposition:attachment;filename=downloaded.pdf");
// PDF 源在 original.pdf 中
readfile($pdf);
//$url = "/uploads/".$log["attachment"];
die();

原文链接：https://blog.csdn.net/leesin2011/article/details/53317362

一、composer安装phpWord

composer require phpoffice/phpword

传送门：https://packagist.org/packages/phpoffice/phpword

二、phpWord 读取 docx 文档（注意是docx格式，doc格式不行）

如果你的文件是doc格式，直接另存为一个docx就行了；如果你的doc文档较多，可以下一个批量转换工具：http://www.batchwork.com/en/doc2doc/download.htm

如果你还没配置自动加载，则先配置一下：

require ‘./vendor/autoload.php’;

加载文档：

$dir = str_replace('\', '/', DIR) . '/';

$source = $dir . 'test.docx';

$phpWord = \PhpOffice\PhpWord\IOFactory::load($source);

三、关键点

1）对齐方式：PhpOffice\PhpWord\Style\Paragraph -> getAlignment()

2）字体名称：\PhpOffice\PhpWord\Style\Font -> getName()

3）字体大小：\PhpOffice\PhpWord\Style\Font -> getSize()

4）是否加粗：\PhpOffice\PhpWord\Style\Font -> isBold()

5）读取图片：\PhpOffice\PhpWord\Element\Image -> getImageStringData()

6）ba64格式图片数据保存为图片：file_put_contents($imageSrc, base64_decode($imageData))

四、完整代码

 require './vendor/autoload.php';
 function docx2html($source)
 {
     $phpWord = \PhpOffice\PhpWord\IOFactory::load($source);
     $html = '';
     foreach ($phpWord->getSections() as $section) {
         foreach ($section->getElements() as $ele1) {
             $paragraphStyle = $ele1->getParagraphStyle();
             if ($paragraphStyle) {
                 $html .= '
             } else {
                 $html .= '
';             }             if ($ele1 instanceof \PhpOffice\PhpWord\Element\TextRun) {                 foreach ($ele1->getElements() as $ele2) {                     if ($ele2 instanceof \PhpOffice\PhpWord\Element\Text) {                         $style = $ele2->getFontStyle();                         $fontFamily = mb_convert_encoding($style->getName(), 'GBK', 'UTF-8');                         $fontSize = $style->getSize();                         $isBold = $style->isBold();                         $styleString = '';                         $fontFamily && $styleString .= "font-family:{$fontFamily};";                         $fontSize && $styleString .= "font-size:{$fontSize}px;";                         $isBold && $styleString .= "font-weight:bold;";                         $html .= sprintf('%s',                             $styleString,                             mb_convert_encoding($ele2->getText(), 'GBK', 'UTF-8')                         );                     } elseif ($ele2 instanceof \PhpOffice\PhpWord\Element\Image) {                         $imageSrc = 'images/' . md5($ele2->getSource()) . '.' . $ele2->getImageExtension();                         $imageData = $ele2->getImageStringData(true);                         // $imageData = 'data:' . $ele2->getImageType() . ';base64,' . $imageData;                         file_put_contents($imageSrc, base64_decode($imageData));                         $html .= '';                     }                 }             }             $html .= '
';
         }
     }
 return mb_convert_encoding($html, 'UTF-8', 'GBK');
 }
 $dir = str_replace('\', '/', DIR) . '/';
 $source = $dir . 'test.docx';
 echo docx2html($source);     

五、补充

很明显，这是一个简陋的word读取示例，只读取了段落的对齐方式，文字的字体、大小、是否加粗及图片等信息，其他例如文字颜色、行高。。。等等信息都忽悠了。需要的话，请自行查看phpWord源码，看\PhpOffice\PhpWord\Style\xxx 和 \PhpOffice\PhpWord\Element\xxx 等类里有什么读取方法就可以了

六、2020-07-21 补充

可以用以下方法直接获取到完整的html

$phpWord = \PhpOffice\PhpWord\IOFactory::load('xxx.docx');

$xmlWriter = \PhpOffice\PhpWord\IOFactory::createWriter($phpWord, "HTML");

$html = $xmlWriter->getContent();

注：html内容里包含了head部分，如果只需要style和body的话，需要自己处理一下；然后图片是base64的，要保存的话，也需要自己处理一下

base64数据保存为图片请参考上面代码

如果只想获取body里的内容，可以参考 \PhpOffice\PhpWord\Writer\HTML\Part\Body 里的 write 方法
复制代码

$phpWord = \PhpOffice\PhpWord\IOFactory::load('xxxx.docx');

$htmlWriter = \PhpOffice\PhpWord\IOFactory::createWriter($phpWord, "HTML");

$content = '';

foreach ($phpWord->getSections() as $section) {

    $writer = new \PhpOffice\PhpWord\Writer\HTML\Element\Container($htmlWriter, $section);

    $content .= $writer->write();

}

echo $content;exit;

图片的处理的话，暂时没有好办法能在不修改源码的情况下处理好，改源码的话，相关代码在 \PhpOffice\PhpWord\Writer\HTML\Element\Image 里

public function write()
 {
     if (!$this->element instanceof ImageElement) {
         return '';
     }
     $content = '';
     $imageData = $this->element->getImageStringData(true);
     if ($imageData !== null) {
         $styleWriter = new ImageStyleWriter($this->element->getStyle());
         $style = $styleWriter->write();
         // $imageData = 'data:' . $this->element->getImageType() . ';base64,' . $imageData;
         $imageSrc = 'images/' . md5($this->element->getSource()) . '.' . $this->element->getImageExtension();
         // 这里可以自己处理，上传oss之类的
         file_put_contents($imageSrc, base64_decode($imageData));
     $content .= $this->writeOpening();     $content .= "<img border=\"0\" style=\"{$style}\" src=\"{$imageSrc}\"/>";     $content .= $this->writeClosing(); } return $content;
 }

//Word转HTML
$phpWord = \PhpOffice\PhpWord\IOFactory::load('./word/hello.docx');
$xmlWriter = \PhpOffice\PhpWord\IOFactory::createWriter($phpWord, "HTML");
$xmlWriter->save('./html/hello.html');