如何使用 php 将 docx 文档转换为 html？
Posted
技术标签:
【中文标题】如何使用 php 将 docx 文档转换为 html？【英文标题】：How can I convert a docx document to html using php? 【发布时间】：2011-06-03 00:00:09 【问题描述】：我希望能够上传 MS Word 文档并将其导出到我的站点中的页面。
有没有办法做到这一点？
【问题讨论】：
我对 php 不是很熟悉 - 但也许这可以帮助你？ phpLiveDocx -Convert DOCX to html in PHP 您可以使用phpLiveDocx。 你的方法是使用LiveDocx，你需要一个account。然后关注这个guide，或者自己学习如何使用Zend_Service_LiveDocx。 phpLiveDocx 似乎有点矫枉过正......而且它的服务似乎非常有限（没有动态表格或图表） 【参考方案1】：//FUNCTION :: read a docx file and return the string
function readDocx($filePath)
// Create new ZIP archive
$zip = new ZipArchive;
$dataFile = 'word/document.xml';
// Open received archive file
if (true === $zip->open($filePath))
// If done, search for the data file in the archive
if (($index = $zip->locateName($dataFile)) !== false)
// If found, read it to the string
$data = $zip->getFromIndex($index);
// Close archive file
$zip->close();
// Load XML from a string
// Skip errors and warnings
$xml = DOMDocument::loadXML($data, LIBXML_NOENT | LIBXML_XINCLUDE | LIBXML_NOERROR | LIBXML_NOWARNING);
// Return data without XML formatting tags
$contents = explode('\n',strip_tags($xml->saveXML()));
$text = '';
foreach($contents as $i=>$content)
$text .= $contents[$i];
return $text;
$zip->close();
// In case of failure return empty string
return "";
ZipArchive 和 DOMDocument 都在 PHP 中，因此您无需安装/包含/需要其他库。
【讨论】：
谢谢，太好了，但是有没有办法保持格式，例如粗体和斜体字 谢谢...它正在返回整个文档。但是有什么方法可以单独获取页面文本！ 这个答案没有提供将 .docx 转换为 HTML 的解决方案——正如代码strip_tags() 中所显示的那样——OP 专门询问如何转换为 HTML【参考方案2】：
可以使用PHPDocX。
它支持几乎所有的 HTML CSS 样式。此外，您可以使用模板通过replaceTemplateVariableByHTML 为您的 HTML 添加额外的格式。
PHPDocX 的 HTML 方法还允许直接使用 Word 样式。你可以使用这样的东西：
$docx->embedHTML($myHTML, array('tableStyle' => 'MediumGrid3-accent5PHPDOCX'));
如果您希望所有表格都使用 MediumGrid3-accent5 Word 样式。 embedHTML 方法及其模板版本 (replaceTemplateVariableByHTML) 保留了继承性，这意味着您可以使用预定义的 Word 样式并用 CSS 覆盖其任何属性。
您还可以使用“JQuery 类型”选择器提取 HTML 的选定部分。
【讨论】：
不得不说它不是免费的！至少没有了。低至 399.00 美元。 建议：让我们向 *** 引入一个“商业”徽章/标记，以使此类内容可见【参考方案3】：您可以使用 Print2flash 库将 Word docx 文档转换为 html。这是我客户网站的 PHP 摘录，它将文档转换为 html：
include("const.php");
$p2fServ = new COM("Print2Flash4.Server2");
$p2fServ->DefaultProfile->DocumentType=HTML5;
$p2fServ->ConvertFile($wordfile,$htmlFile);
它将在 $wordfile 变量中指定路径的文档转换为由 $htmlFile 变量指定的 html 页面文件。保留所有格式、超链接和图表。您可以从Print2flash SDK 获得所需的 const.php 文件以及更完整的示例。
【讨论】：
【参考方案4】：这是基于 David Lin 上述回答的解决方法 删除 docx 的 xml 标签中的“w:”会留下 Html 之类的标签
function readDocx($filePath)
// Create new ZIP archive
$zip = new ZipArchive;
$dataFile = 'word/document.xml';
// Open received archive file
if (true === $zip->open($filePath))
// If done, search for the data file in the archive
if (($index = $zip->locateName($dataFile)) !== false)
// If found, read it to the string
$data = $zip->getFromIndex($index);
// Close archive file
$zip->close();
// Load XML from a string
// Skip errors and warnings
$xml = new DOMDocument("1.0", "utf-8");
$xml->loadXML($data, LIBXML_NOENT | LIBXML_XINCLUDE | LIBXML_NOERROR | LIBXML_NOWARNING|LIBXML_PARSEHUGE);
$xml->encoding = "utf-8";
// Return data without XML formatting tags
$output = $xml->saveXML();
$output = str_replace("w:","",$output);
return $output;
$zip->close();
// In case of failure return empty string
return "";
【讨论】：
【参考方案5】：如果你不拒绝 REST API，那么你可以使用：
Apache Tika。是久经考验的文本提取 OSS 领导者 如果您不想麻烦配置并想要现成的解决方案，您可以使用RawText，但它不是免费的。RawText 的示例代码：
$result = $rawText -> parse($your_file)
【讨论】：
【参考方案6】：好的，我来晚了，但我想我会发布这个来节省大家的时间。 这是我整理的一些 php 代码，不仅用于从 docx 读取文本，还用于读取图像，目前它不支持浮动图像/文本，但到目前为止我所做的是向已经发布的内容迈出了一大步此处 - 请注意您需要将 https://sharinggodslove.uk 更新为您的域名。
<?php
class Docx_ws_imglnk
public $originalpath = '';
public $extractedpath = '';
class Docx_ws_rel
public $Id = '';
public $Target = '';
class Docx_ws_def
public $styleId = '';
public $type = '';
public $color = '000000';
class Docx_p_def
public $data = array();
public $text = "";
class Docx_p_item
public $name = "";
public $value = "";
public $innerstyle = "";
public $type = "text";
class Docx_reader
private $fileData = false;
private $errors = array();
public $rels = array();
public $imglnks = array();
public $styles = array();
public $document = null;
public $paragraphs = array();
public $path = '';
private $saveimgpath = 'docimages';
public function __construct()
private function load($file)
if (file_exists($file))
$zip = new ZipArchive();
$openedZip = $zip->open($file);
if ($openedZip === true)
$this->path = $file;
//read and save images
for ( $i = 0; $i < $zip->numFiles; $i ++ )
$zip_element = $zip->statIndex( $i );
if ( preg_match( "([^\s]+(\.(?i)(jpg|jpeg|png|gif|bmp))$)", $zip_element['name'] ) )
$imglnk = new Docx_ws_imglnk;
$imglnk->originalpath = $zip_element['name'];
$imagename = explode( '/', $zip_element['name'] );
$imagename = end( $imagename );
$imglnk->extractedpath = dirname( __FILE__ ) . '/' . $this->savepath . $imagename;
$putres = file_put_contents( $imglnk->extractedpath, $zip->getFromIndex( $i ));
$imglnk->extractedpath = str_replace('var/www/', 'https://sharinggodslove.uk/', $imglnk->extractedpath);
$imglnk->extractedpath = substr($imglnk->extractedpath, 1);
array_push($this->imglnks, $imglnk);
//read relationships
if (($styleIndex = $zip->locateName('word/_rels/document.xml.rels')) !== false)
$stylesRels = $zip->getFromIndex($styleIndex);
$xml = simplexml_load_string($stylesRels);
$XMLTEXT = $xml->saveXML();
$doc = new DOMDocument();
$doc->loadXML($XMLTEXT);
foreach($doc->documentElement->childNodes as $childnode)
$nodename = $childnode->nodeName;
if($childnode->hasAttributes())
$rel = new Docx_ws_rel;
for ($a = 0; $a < $childnode->attributes->count(); $a++)
$attrNode = $childnode->attributes->item($a);
if (strcmp( $attrNode->nodeName, 'Id') == 0)
$rel->Id = $attrNode->nodeValue;
if (strcmp( $attrNode->nodeName, 'Target') == 0)
$rel->Target = $attrNode->nodeValue;
array_push($this->rels, $rel);
//attempt to load styles:
if (($styleIndex = $zip->locateName('word/styles.xml')) !== false)
$stylesXml = $zip->getFromIndex($styleIndex);
$xml = simplexml_load_string($stylesXml);
$XMLTEXT = $xml->saveXML();
$doc = new DOMDocument();
$doc->loadXML($XMLTEXT);
foreach($doc->documentElement->childNodes as $childnode)
$nodename = $childnode->nodeName;
//get style
if (strcmp($nodename, "w:style") == 0)
$ws_def = new Docx_ws_def;
for ($a=0; $a < $childnode->attributes->count(); $a++ )
$item = $childnode->attributes->item($a);
//style id
if (strcmp($item->nodeName, "w:styleId") == 0)
$ws_def->styleId = $item->nodeValue;
//style type
if (strcmp($item->nodeName, "w:type") == 0)
$ws_def->type = $item->nodeValue;
//push style to the array of styles
if (strcmp($ws_def->styleId, "") != 0 && strcmp($ws_def->type, "") != 0)
array_push($this->styles, $ws_def);
if (($index = $zip->locateName('word/document.xml')) !== false)
$stylesDoc = $zip->getFromIndex($index);
$xml = simplexml_load_string($stylesDoc);
$XMLTEXT = $xml->saveXML();
$this->document = new DOMDocument();
$this->document->loadXML($XMLTEXT);
$zip->close();
else
switch($openedZip)
case ZipArchive::ER_EXISTS:
$this->errors[] = 'File exists.';
break;
case ZipArchive::ER_INCONS:
$this->errors[] = 'Inconsistent zip file.';
break;
case ZipArchive::ER_MEMORY:
$this->errors[] = 'Malloc failure.';
break;
case ZipArchive::ER_NOENT:
$this->errors[] = 'No such file.';
break;
case ZipArchive::ER_NOZIP:
$this->errors[] = 'File is not a zip archive.';
break;
case ZipArchive::ER_OPEN:
$this->errors[] = 'Could not open file.';
break;
case ZipArchive::ER_READ:
$this->errors[] = 'Read error.';
break;
case ZipArchive::ER_SEEK:
$this->errors[] = 'Seek error.';
break;
else
$this->errors[] = 'File does not exist.';
public function setFile($path)
$this->fileData = $this->load($path);
public function to_plain_text()
if ($this->fileData)
return strip_tags($this->fileData);
else
return false;
public function processDocument()
$html = '';
foreach($this->document->documentElement->childNodes as $childnode)
$nodename = $childnode->nodeName;
//get the body of the document
if (strcmp($nodename, "w:body") == 0)
foreach($childnode->childNodes as $subchildnode)
$pnodename = $subchildnode->nodeName;
//process every paragraph
if (strcmp($pnodename, "w:p") == 0)
$pdef = new Docx_p_def;
foreach($subchildnode->childNodes as $pchildnode)
//process any inner children
if (strcmp($pchildnode, "w:pPr") == 0)
foreach($pchildnode->childNodes as $prchildnode)
//process text alignment
if (strcmp($prchildnode->nodeName, "w:pStyle") == 0)
$pitem = new Docx_p_item;
$pitem->name = 'styleId';
$pitem->value = $prchildnode->attributes->getNamedItem('val')->nodeValue;
array_push($pdef->data, $pitem);
//process text alignment
if (strcmp($prchildnode->nodeName, "w:jc") == 0)
$pitem = new Docx_p_item;
$pitem->name = 'align';
$pitem->value = $prchildnode->attributes->getNamedItem('val')->nodeValue;
if (strcmp($pitem->value, "left") == 0)
$pitem->innerstyle .= "text-align:" . $pitem->value . ";";
if (strcmp($pitem->value, "center") == 0)
$pitem->innerstyle .= "text-align:" . $pitem->value . ";";
if (strcmp($pitem->value, "right") == 0)
$pitem->innerstyle .= "text-align:" . $pitem->value . ";";
if (strcmp($pitem->value, "both") == 0)
$pitem->innerstyle .= "word-spacing:" . 10 . "px;";
array_push($pdef->data, $pitem);
//process drawing
if (strcmp($prchildnode->nodeName, "w:drawing") == 0)
$pitem = new Docx_p_item;
$pitem->name = 'drawing';
$pitem->value = '';
$pitem->type = 'graphic';
$extents = $prchildnode->getElementsByTagName('extent')[0];
$cx = $extents->attributes->getNamedItem('cx')->nodeValue;
$cy = $extents->attributes->getNamedItem('cy')->nodeValue;
$pcx = (int)$cx / 9525;
$pcy = (int)$cy / 9525;
$pitem->innerstyle .= "width:" . $pcx . "px;";
$pitem->innerstyle .= "height:" . $pcy . "px;";
$blip = $prchildnode->getElementsByTagName('blip')[0];
$pitem->value = $blip->attributes->getNamedItem('embed')->nodeValue;
array_push($pdef->data, $pitem);
//process spacing
if (strcmp($prchildnode->nodeName, "w:spacing") == 0)
$pitem = new Docx_p_item;
$pitem->name = 'paragraphSpacing';
$bval = $prchildnode->attributes->getNamedItem('before')->nodeValue;
if (strcmp($bval, '') == 0)
$bval = 0;
$pitem->innerstyle .= "padding-top:" . $bval . "px;";
$aval = $prchildnode->attributes->getNamedItem('after')->nodeValue;
if (strcmp($aval, '') == 0)
$aval = 0;
$pitem->innerstyle .= "padding-bottom:" . $aval . "px;";
array_push($pdef->data, $pitem);
if (strcmp($pchildnode, "w:r") == 0)
foreach($pchildnode->childNodes as $rchildnode)
//process text
if (strcmp($rchildnode->nodeName, "w:t") == 0)
$pdef->text .= $rchildnode->nodeValue;
if (count($pdef->data) == 0)
$pitem = new Docx_p_item;
$pitem->name = 'styleId';
$pitem->value = '';
array_push($pdef->data, $pitem);
if (strcmp($rchildnode->nodeName, "w:rPr") == 0)
foreach($rchildnode->childNodes as $rPrchildnode)
if (strcmp($rPrchildnode->nodeName, "w:b") == 0 )
$pitem = new Docx_p_item;
$pitem->name = 'textBold';
$pitem->value = '';
$pitem->innerstyle .= "text-weight: 500;";
array_push($pdef->data, $pitem);
if (strcmp($rPrchildnode->nodeName, "w:i") == 0 )
$pitem = new Docx_p_item;
$pitem->name = 'textItalic';
$pitem->value = '';
$pitem->innerstyle .= "text-style: italic;";
array_push($pdef->data, $pitem);
if (strcmp($rPrchildnode->nodeName, "w:u") == 0 )
$pitem = new Docx_p_item;
$pitem->name = 'textUnderline';
$pitem->value = '';
$pitem->innerstyle .= "text-decoration: underline;";
array_push($pdef->data, $pitem);
if (strcmp($rPrchildnode->nodeName, "w:sz") == 0 )
$pitem = new Docx_p_item;
$pitem->name = 'textSize';
$sz = $rPrchildnode->attributes->getNamedItem('val')->nodeValue;
if ($sz == '')
$sz=0;
$pitem->value = $sz;
array_push($pdef->data, $pitem);
array_push($this->paragraphs, $pdef);
public function to_html()
$html = '';
foreach($this->paragraphs as $para)
$styleselect = null;
$type = 'text';
$content = $para->text;
$sz = 0;
$extent = '';
$embedid = '';
$pinnerstylesid = '';
$pinnerstylesunderline = '';
$pinnerstylessz = '';
if (count($para->data) > 0)
foreach($para->data as $node)
if (strcmp($node->name, "styleId") == 0)
$type = $node->type;
$pinnerstylesid = $node->innerstyle;
foreach($this->styles as $style)
if (strcmp ($node->value, $style->styleId) == 0)
$styleselect = $style;
if (strcmp($node->name, "align") == 0)
$pinnerstylesid .= $node->innerstyle. ";";
if (strcmp($node->name, "drawing") == 0)
$type = $node->type;
$extent = $node->innerstyle;
$embedid = $node->value;
if (strcmp($node->name, "textSize") == 0)
$sz = $node->value;
if (strcmp($node->name, "textUnderline") == 0)
$pinnerstylesunderline = $node->innerstyle;
if (strcmp($type, 'text') == 0)
//echo "has valid para";
//echo "<br>";
if ($styleselect != null)
//echo "has valid style";
//echo "<br>";
if (strcmp($styleselect->color, '') != 0)
$pinnerstylesid .= "color:#" . $styleselect->color. ";";
if ($sz != 0)
$pinnerstylesid .= 'font-size:' . $sz . 'px;';
//echo "sz<br>";
$span = "<p style='". $pinnerstylesid . $pinnerstylesunderline ."'>";
$span .= $content;
$span .= "</p>";
//echo $span;
$html .= $span;
if (strcmp($type, 'graphic') == 0)
$imglnk = '';
foreach($this->rels as $rel)
if(strcmp($embedid, '') != 0 && strcmp($rel->Id, $embedid) == 0)
foreach($this->imglnks as $imgpathdef)
if (strpos($imgpathdef->extractedpath, $rel->Target) >= 0)
$imglnk = $imgpathdef->extractedpath;
//echo "has img link<br>";
//echo $imglnk . "<br>";
if ($styleselect != null)
//echo "has valid style";
//echo "<br>";
if (strcmp($styleselect->color, '') != 0)
$pinnerstylesid .= "color:#" . $styleselect->color. ";";
if ($sz != 0)
$pinnerstylesid .= 'font-size:' . $sz . 'px;';
//echo "sz<br>";
$span = "<p style='". $pinnerstylesid . $pinnerstylesunderline ."'>";
$span .= "<img style='". $extent ."' alt='image coming soon' src ='". $imglnk ."'/>";
$span .= "</p>";
//echo $span;
$html .= $span;
return $html;
public function get_errors()
return $this->errors;
private function getStyles()
function getDocX($path)
//echo $path;
$doc = new Docx_reader();
$doc->setFile($path);
if(!$doc->get_errors())
$doc->processDocument();
$html = $doc->to_html();
echo $html;
return "";
?>
【讨论】：
【参考方案7】：现在更常见的方法是使用 composer package phpoffice/phpword，一个纯 php 库，可以将任何办公文档转换为 html，反之亦然，无需依赖。
【讨论】：
以上是关于如何使用 php 将 docx 文档转换为 html？的主要内容，如果未能解决你的问题，请参考以下文章
PHP - Laravel - 将 Docx 转换为 PDF