java实现查找PDF关键字所在页码及其坐标

1、因为最近有这方面的需求，用过之后记录一下。

2、此功能跟PDF中Ctrl+F性质一样，如果PDF中为图片形式的不支持定位到关键字。

import com.itextpdf.awt.geom.Rectangle2D.Float;
import com.itextpdf.text.pdf.PdfDictionary;
import com.itextpdf.text.pdf.PdfName;
import com.itextpdf.text.pdf.PdfReader;
import com.itextpdf.text.pdf.parser.*;
import java.io.File;
import java.io.FileInputStream;
import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
/**
 * 消失的太阳
 */
public class MyTest {
 public static void main(String[] args) throws IOException {
  //1.给定文件
  File pdfFile = new File("D://test.pdf");
  //2.定义一个byte数组，长度为文件的长度
  byte[] pdfData = new byte[(int) pdfFile.length()];
  //3.IO流读取文件内容到byte数组
  FileInputStream inputStream = null;
  try {
   inputStream = new FileInputStream(pdfFile);
   inputStream.read(pdfData);
  } catch (IOException e) {
   throw e;
  } finally {
   if (inputStream != null) {
    try {
     inputStream.close();
    } catch (IOException e) {
    }
   }
  }
  //4.指定关键字
  String keyword = "消失的太阳：";
  //5.调用方法，给定关键字和文件
  List positions = findKeywordPostions(pdfData, keyword);
  //6.返回值类型是 List 每个list元素代表一个匹配的位置，分别为 float[0]所在页码 float[1]所在x轴 float[2]所在y轴
  System.out.println("total:" + positions.size());
  if (positions != null && positions.size() > 0) {
   for (float[] position : positions) {
    System.out.print("pageNum: " + (int) position[0]);
    System.out.print("\tx: " + position[1]);
    System.out.println("\ty: " + position[2]);
   }
  }
 }
 /**
  * findKeywordPostions
  * @param pdfData  通过IO流 PDF文件转化的byte数组
  * @param keyword  关键字
  * @return List : float[0]:pageNum float[1]:x float[2]:y
  * @throws IOException
  */
 public static List findKeywordPostions(byte[] pdfData, String keyword) throws IOException {
  List result = new ArrayList<>();
  List pdfPageContentPositions = getPdfContentPostionsList(pdfData);
  for (PdfPageContentPositions pdfPageContentPosition : pdfPageContentPositions) {
   List charPositions = findPositions(keyword, pdfPageContentPosition);
   if (charPositions == null || charPositions.size() < 1) {
    continue;
   }
   result.addAll(charPositions);
  }
  return result;
 }
 private static List getPdfContentPostionsList(byte[] pdfData) throws IOException {
  PdfReader reader = new PdfReader(pdfData);
  List result = new ArrayList<>();
  int pages = reader.getNumberOfPages();
  for (int pageNum = 1; pageNum <= pages; pageNum++) {
   float width = reader.getPageSize(pageNum).getWidth();
   float height = reader.getPageSize(pageNum).getHeight();
   PdfRenderListener pdfRenderListener = new PdfRenderListener(pageNum, width, height);
   //解析pdf，定位位置
   PdfContentStreamProcessor processor = new PdfContentStreamProcessor(pdfRenderListener);
   PdfDictionary pageDic = reader.getPageN(pageNum);
   PdfDictionary resourcesDic = pageDic.getAsDict(PdfName.RESOURCES);
   try {
    processor.processContent(ContentByteUtils.getContentBytesForPage(reader, pageNum), resourcesDic);
   } catch (IOException e) {
    reader.close();
    throw e;
   }
   String content = pdfRenderListener.getContent();
   List charPositions = pdfRenderListener.getcharPositions();
   List positionsList = new ArrayList<>();
   for (CharPosition charPosition : charPositions) {
    float[] positions = new float[]{charPosition.getPageNum(), charPosition.getX(), charPosition.getY()};
    positionsList.add(positions);
   }
   PdfPageContentPositions pdfPageContentPositions = new PdfPageContentPositions();
   pdfPageContentPositions.setContent(content);
   pdfPageContentPositions.setPostions(positionsList);
   result.add(pdfPageContentPositions);
  }
  reader.close();
  return result;
 }
 private static List findPositions(String keyword, PdfPageContentPositions pdfPageContentPositions) {
  List result = new ArrayList<>();
  String content = pdfPageContentPositions.getContent();
  List charPositions = pdfPageContentPositions.getPositions();
  for (int pos = 0; pos < content.length(); ) {
   int positionIndex = content.indexOf(keyword, pos);
   if (positionIndex == -1) {
    break;
   }
   float[] postions = charPositions.get(positionIndex);
   result.add(postions);
   pos = positionIndex + 1;
  }
  return result;
 }
 private static class PdfPageContentPositions {
  private String content;
  private List positions;
  public String getContent() {
   return content;
  }
  public void setContent(String content) {
   this.content = content;
  }
  public List getPositions() {
   return positions;
  }
  public void setPostions(List positions) {
   this.positions = positions;
  }
 }
 private static class PdfRenderListener implements RenderListener {
  private int pageNum;
  private float pageWidth;
  private float pageHeight;
  private StringBuilder contentBuilder = new StringBuilder();
  private List charPositions = new ArrayList<>();
  public PdfRenderListener(int pageNum, float pageWidth, float pageHeight) {
   this.pageNum = pageNum;
   this.pageWidth = pageWidth;
   this.pageHeight = pageHeight;
  }
  public void beginTextBlock() {
  }
  public void renderText(TextRenderInfo renderInfo) {
   List characterRenderInfos = renderInfo.getCharacterRenderInfos();
   for (TextRenderInfo textRenderInfo : characterRenderInfos) {
    String word = textRenderInfo.getText();
    if (word.length() > 1) {
     word = word.substring(word.length() - 1, word.length());
    }
    Float rectangle = textRenderInfo.getAscentLine().getBoundingRectange();
    float x = (float)rectangle.getX();
    float y = (float)rectangle.getY();
//    float x = (float)rectangle.getCenterX();
//    float y = (float)rectangle.getCenterY();
//    double x = rectangle.getMinX();
//    double y = rectangle.getMaxY();
    //这两个是关键字在所在页面的XY轴的百分比
    float xPercent = Math.round(x / pageWidth * 10000) / 10000f;
    float yPercent = Math.round((1 - y / pageHeight) * 10000) / 10000f;
//    CharPosition charPosition = new CharPosition(pageNum, xPercent, yPercent);
    CharPosition charPosition = new CharPosition(pageNum, (float)x, (float)y);
    charPositions.add(charPosition);
    contentBuilder.append(word);
   }
  }
  public void endTextBlock() {
  }
  public void renderImage(ImageRenderInfo renderInfo) {
  }
  public String getContent() {
   return contentBuilder.toString();
  }
  public List getcharPositions() {
   return charPositions;
  }
 }
 private static class CharPosition {
  private int pageNum = 0;
  private float x = 0;
  private float y = 0;
  public CharPosition(int pageNum, float x, float y) {
   this.pageNum = pageNum;
   this.x = x;
   this.y = y;
  }
  public int getPageNum() {
   return pageNum;
  }
  public float getX() {
   return x;
  }
  public float getY() {
   return y;
  }
  @Override
  public String toString() {
   return "[pageNum=" + this.pageNum + ",x=" + this.x + ",y=" + this.y + "]";
  }
 }
}

总结

以上所述是小编给大家介绍的java实现查找PDF关键字所在页码及其坐标,希望对大家有所帮助，如果大家有任何疑问请给我留言，小编会及时回复大家的。在此也非常感谢大家对脚本之家网站的支持！
如果你觉得本文对你有帮助，欢迎转载，烦请注明出处，谢谢！