I have a docx file consisting of  images .

Help me out in getting images from the [document.xml] file. I am using xsl to parse the docx file to html.

The part of the code (xsl) is wriiten down here.

<xsl:for-each select="//w:sdt/w:sdtContent">

      <xsl:for-each select="w:p/w:r">

         <p><xsl:value-of select="w:t"/></p>

         <xsl:for-each select="w:p/w:r" >

                  <xsl:for-each select="w:drawing/a:graphic/a:graphicData/pic:pic">

                           <xsl:for-each select="pic:nvPicPr/pic:cNvPr">

                                 <img alt="" src="{$imgSrc}"/>

                           </xsl:for-each>

                  </xsl:for-each>

      </xsl:for-each>

</xsl:for-each>

<p><xsl:value-of select="w:r/w:t"/></p>

</xsl:for-each>

I am passing the [imgSrc] in <xsl:param name="imgSrc"/>

This code works for the text part but the issue is that I am not able to get the image at all.

Please help me out.

Hi,
Image parts are not stored in the document.xml.
Extract files from your document and check in /word/media.
All your images are here.
What you can get from the document.xml file is the RID of the image you want:

<pic:blipFill>

thanx Kirikou
I am trying to generate a html file exactly the same as of what is being there in docx. I am using java and xsl to serve the purpose. Everything is working as par requirement but the issue is still with images which I want to put there in the html. Please do help me once again.
Regards,
broshni

Hi,

I have implemented this. So if you have already notfigures this out, I can tell you how I did t. But I am working with C#, not with JAVA.

 

By the way, if you have designed the XSLT conversion of bullets and numbering, please do share with me, coz that is the point where I am stuck with currently.

/Suralk

Hi folks,
the /word/_rels/document.xml.rels contains the relationship for the image I am trying extract.


now the issue is in extracting the Relationship Id="rId4" which is actually image part I am trying to put in the html to be generated.
please do suggest me how to extract the document.xml.rels so I can iterate though the file to extract the required target.
any comment would be highly appreciative.
regards,
broshni

hi,

there is no such thing as extracting the document.xml.rels. What u can do is, u can extract the packageRelationship corresponding to that and write it into a folder as an image file. then u can create a reference o that from your html

are you using C3 .Net? If so, can give you a code snippet to do that.

 

/Suralk

Im not sure if this is the same for Word '07, but in Word '03 xml many pictures were stored inside <w:binData> with Base64Endconding. To pull these out (in Java), parse the xml document and call these two methods:

/**
     *Extracts all images from a parsed MS-XML Document
     *
     *@param xmlDoc the MS-XML-Document (before transform)
     */
    public void extractImages(Document xmlDoc)
    {
        NodeList binDataList = xmlDoc.getElementsByTagName("w:binData");
      
        String fileName = "";
        Node currentNode;
        for(int i = 0; i < binDataList.getLength(); i++)
        {
            currentNode = binDataList.item(i);
            if(currentNode.getNodeType() == Node.ELEMENT_NODE && ((Element)currentNode).hasAttribute("w:name"))
            {               
                File newImageFile = new File(picDirectory, ((Element)currentNode).getAttribute("w:name").replaceFirst("wordml://", ""));
                if(newImageFile.exists())
                {
                  
                }
                else
                {
                    if(writeImage(newImageFile, currentNode))
                    {
                        //Print some success message
                    }
                }
            }
        }
    }
   
    /**
     *Writes the content of the w:bin node to the specified file
     *
     *@param toFile File to write bin data to
     *@param binNode the node containing the bin data
     */
    public boolean writeImage(File toFile, Node binNode)
    {
        FileOutputStream dataOut = null;
        try
        {
            dataOut = new FileOutputStream(toFile);
           
            ByteArrayInputStream bis = new ByteArrayInputStream(binNode.toString().replaceAll("<[^<]+?>", "").getBytes());
            BASE64Decoder decoder = new BASE64Decoder();
            decoder.decodeBuffer(bis, dataOut);
            dataOut.flush();
        }
        catch(Exception e)
        {
           
            return false;
        }
        finally
        {
            dataOut.close();
        }
        return true;
    }

The problem with the above code is extracting .wmz, .emz, and .mso data. These are compressed, but when trying to use java.util.zip, etc. it does not like the encoding/compression. I have tried decoding with the BASE64Decoder (sun.misc.BASE64Decoder) then decompressing it, however it does not like this either....

You can use docx4j to manipulate the images in Java.  docx4j is open source (ASL).  There are several posts on our forums about how to do this.

Hi

 

You can also use Aspose.Words for extracting pictures from the Word document.

http://www.aspose.com/documentation/file-format-components/aspose.words-for-.net-and-java/extract-images-from-a-document.html

 

There are .NET and Java version of this tool.

 

WBR,

Alex