如何在word文档apache.poi中读取图像的替代文本

Question

如何使用 apache.poi 读取/编辑 word 文档图像上的 alt 文本？

我试图找到正确的功能，但 XWPFPictureData 对象内部没有任何内容。

Answer 1

XWPFDocument.getAllPictures 仅提供对XWPFPictureData 的访问，它仅表示嵌入图片的二进制文件数据。这不包含文档中图片形状的元数据。

还有XWPFRun.getEmbeddedPictures 得到XWPFPicture。这是CTPicture 元素的表示，它包含图片形状的一些元数据，但不是全部。

Word 文档中的图片存储在包含内联或锚点的图形绘图中。与文本对齐的图片存储在行内。锚定在文档中某处并具有文本流的图片存储在锚点中。内联或锚点包含形状的所有元数据。

所以需要获取图片的内联或锚点。拥有这个可以得到CTNonVisualDrawingProps，这是图片的替代文本数据的表示。

直接使用apache poi 可以获得的最接近的元素是XWPFPicture。拥有那个可以尝试获取包含找到的图片的内联或锚点。

以下完整示例显示了这一点。方法org.apache.xmlbeans.XmlObject getInlineOrAnchor(XWPFRun run, XWPFPicture picture) 是从XWPFRun run 中包含的XWPFPicture picture 中获取内联或锚点。然后方法org.openxmlformats.schemas.drawingml.x2006.main.CTNonVisualDrawingProps getNonVisualDrawingProps(org.apache.xmlbeans.XmlObject inlineOrAnchor) 从中获取CTNonVisualDrawingProps。 CTNonVisualDrawingProps 提供 getId、getName、getTitle 和 getDescr。 getDescr 返回 Word 显示为替代文本的文本。

import java.io.FileInputStream;

import org.apache.poi.xwpf.usermodel.*;

import java.util.List;

public class WordReadAllContent {
    
 static org.apache.xmlbeans.XmlObject getInlineOrAnchor(org.openxmlformats.schemas.drawingml.x2006.picture.CTPicture ctPictureToFind, org.apache.xmlbeans.XmlObject inlineOrAnchor) {
  String declareNameSpaces = "declare namespace pic='http://schemas.openxmlformats.org/drawingml/2006/picture'; ";
  org.apache.xmlbeans.XmlObject[] selectedObjects = inlineOrAnchor.selectPath(
   declareNameSpaces 
   + "$this//pic:pic");
  for (org.apache.xmlbeans.XmlObject selectedObject : selectedObjects) {
   if (selectedObject instanceof org.openxmlformats.schemas.drawingml.x2006.picture.CTPicture) {
    org.openxmlformats.schemas.drawingml.x2006.picture.CTPicture ctPicture = (org.openxmlformats.schemas.drawingml.x2006.picture.CTPicture)selectedObject;
    if (ctPictureToFind.equals(ctPicture)) {
     // this is the inlineOrAnchor for that picture   
     return inlineOrAnchor;
    }        
   }          
  }
  return null;
 }
                
 static org.apache.xmlbeans.XmlObject getInlineOrAnchor(XWPFRun run, XWPFPicture picture) {
  org.openxmlformats.schemas.drawingml.x2006.picture.CTPicture ctPictureToFind = picture.getCTPicture();
  for (org.openxmlformats.schemas.wordprocessingml.x2006.main.CTDrawing drawing : run.getCTR().getDrawingList()) {
   for (org.openxmlformats.schemas.drawingml.x2006.wordprocessingDrawing.CTInline inline : drawing.getInlineList()) {
    org.apache.xmlbeans.XmlObject inlineOrAnchor = getInlineOrAnchor(ctPictureToFind, inline);
    // if inlineOrAnchor is not null, then this is the inline for that picture
    if (inlineOrAnchor != null) return inlineOrAnchor;
   }
   for (org.openxmlformats.schemas.drawingml.x2006.wordprocessingDrawing.CTAnchor anchor : drawing.getAnchorList()) {
    org.apache.xmlbeans.XmlObject inlineOrAnchor = getInlineOrAnchor(ctPictureToFind, anchor);
    // if inlineOrAnchor is not null, then this is the anchor for that picture
    if (inlineOrAnchor != null) return inlineOrAnchor;
   }
  }
  return null;
 }

 static org.openxmlformats.schemas.drawingml.x2006.main.CTNonVisualDrawingProps getNonVisualDrawingProps(org.apache.xmlbeans.XmlObject inlineOrAnchor) {
  if (inlineOrAnchor == null) return null;
  if (inlineOrAnchor instanceof org.openxmlformats.schemas.drawingml.x2006.wordprocessingDrawing.CTInline) {
   org.openxmlformats.schemas.drawingml.x2006.wordprocessingDrawing.CTInline inline = (org.openxmlformats.schemas.drawingml.x2006.wordprocessingDrawing.CTInline)inlineOrAnchor;
   return inline.getDocPr();    
  } else if (inlineOrAnchor instanceof org.openxmlformats.schemas.drawingml.x2006.wordprocessingDrawing.CTAnchor) {
   org.openxmlformats.schemas.drawingml.x2006.wordprocessingDrawing.CTAnchor anchor = (org.openxmlformats.schemas.drawingml.x2006.wordprocessingDrawing.CTAnchor)inlineOrAnchor;
   return anchor.getDocPr();
  }
  return null;
 }

 static String getSummary(org.openxmlformats.schemas.drawingml.x2006.main.CTNonVisualDrawingProps nonVisualDrawingProps) {
  if (nonVisualDrawingProps == null) return "";
  String summary = "Id:=" + nonVisualDrawingProps.getId();
  summary += " Name:=" + nonVisualDrawingProps.getName();
  summary += " Title:=" + nonVisualDrawingProps.getTitle();
  summary += " Descr:=" + nonVisualDrawingProps.getDescr();
  return summary;
 }
 
 static void traversePictures(XWPFRun run, List<XWPFPicture> pictures) throws Exception {
  for (XWPFPicture picture : pictures) {
   //System.out.println(picture);
   
   String picturesNonVisualDrawingProps = getSummary(getNonVisualDrawingProps(getInlineOrAnchor(run, picture)));
   System.out.println(picturesNonVisualDrawingProps);
   
   XWPFPictureData pictureData = picture.getPictureData();
   //System.out.println(pictureData);
  }
 }

 static void traverseRunElements(List<IRunElement> runElements) throws Exception {
  for (IRunElement runElement : runElements) {
   if (runElement instanceof XWPFFieldRun) {
    XWPFFieldRun fieldRun = (XWPFFieldRun)runElement;
    //System.out.println(fieldRun.getClass().getName());
    //System.out.println(fieldRun);
    traversePictures(fieldRun, fieldRun.getEmbeddedPictures());
   } else if (runElement instanceof XWPFHyperlinkRun) {
    XWPFHyperlinkRun hyperlinkRun = (XWPFHyperlinkRun)runElement;
    //System.out.println(hyperlinkRun.getClass().getName());
    //System.out.println(hyperlinkRun);
    traversePictures(hyperlinkRun, hyperlinkRun.getEmbeddedPictures());
   } else if (runElement instanceof XWPFRun) {
    XWPFRun run = (XWPFRun)runElement;
    //System.out.println(run.getClass().getName());
    //System.out.println(run);
    traversePictures(run, run.getEmbeddedPictures());
   } else if (runElement instanceof XWPFSDT) {
    XWPFSDT sDT = (XWPFSDT)runElement;
    //System.out.println(sDT);
    //System.out.println(sDT.getContent());
    //ToDo: The SDT may have traversable content too.
   }
  }
 }

 static void traverseTableCells(List<ICell> tableICells) throws Exception {
  for (ICell tableICell : tableICells) {
   if (tableICell instanceof XWPFSDTCell) {
    XWPFSDTCell sDTCell = (XWPFSDTCell)tableICell;
    //System.out.println(sDTCell);
    //ToDo: The SDTCell may have traversable content too.
   } else if (tableICell instanceof XWPFTableCell) {
    XWPFTableCell tableCell = (XWPFTableCell)tableICell;
    //System.out.println(tableCell);
    traverseBodyElements(tableCell.getBodyElements());
   }
  }
 }

 static void traverseTableRows(List<XWPFTableRow> tableRows) throws Exception {
  for (XWPFTableRow tableRow : tableRows) {
   //System.out.println(tableRow);
   traverseTableCells(tableRow.getTableICells());
  }
 }

 static void traverseBodyElements(List<IBodyElement> bodyElements) throws Exception {
  for (IBodyElement bodyElement : bodyElements) {
   if (bodyElement instanceof XWPFParagraph) {
    XWPFParagraph paragraph = (XWPFParagraph)bodyElement;
    //System.out.println(paragraph);
    traverseRunElements(paragraph.getIRuns());
   } else if (bodyElement instanceof XWPFSDT) {
    XWPFSDT sDT = (XWPFSDT)bodyElement;
    //System.out.println(sDT);
    //System.out.println(sDT.getContent());
    //ToDo: The SDT may have traversable content too.
   } else if (bodyElement instanceof XWPFTable) {
    XWPFTable table = (XWPFTable)bodyElement;
    //System.out.println(table);
    traverseTableRows(table.getRows());
   }
  }
 }

 public static void main(String[] args) throws Exception {
  String inFilePath = "./WordDocument.docx";
  XWPFDocument document = new XWPFDocument(new FileInputStream(inFilePath));
  traverseBodyElements(document.getBodyElements());
  document.close();
 }

}

Answer 2

我们从XWPFPicture 的属性中找到了这个。像这样。

for (XWPFParagraph paragraph : docx.getParagraphs()) {
    for (XWPFRun run : paragraph.getRuns()) {
        for (XWPFPicture embeddedPicture : run.getEmbeddedPictures()) {
            String description = embeddedPicture.getDescription();
            // Something
        }
    }
}