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！

 
packagecom.alphajuns.util;importcom.itextpdf.text.pdf.PdfReader;importcom.itextpdf.text.pdf.parser.PdfReaderContentParser;importjava.io.IOException;/*** @ClassName PdfHelper
 
* @Description Pdf帮助类
 
* @Author AlphaJunS
 
* @Date 2020/3/7 17:40
 
* @Version 1.0*/
 
public classPdfHelper {/*** @Author AlphaJunS
 
* @Date 18:24 2020/3/7
 
* @Description 用于供外部类调用获取关键字所在PDF文件坐标
 
*@paramfilepath
 
*@paramkeyWords
 
*@returnfloat[]*/
 
public static float[] getKeyWordsByPath(String filepath, String keyWords) {float[] coordinate = null;try{
 
PdfReader pdfReader= newPdfReader(filepath);
 
coordinate=getKeyWords(pdfReader, keyWords);
 
}catch(IOException e) {
 
e.printStackTrace();
 
}returncoordinate;
 
}/*** @Author AlphaJunS
 
* @Date 18:26 2020/3/7
 
* @Description 获取关键字所在PDF坐标
 
*@parampdfReader
 
*@paramkeyWords
 
*@returnfloat[]*/
 
private static float[] getKeyWords(PdfReader pdfReader, String keyWords) {float[] coordinate = null;int page = 0;try{int pageNum =pdfReader.getNumberOfPages();
 
PdfReaderContentParser pdfReaderContentParser= newPdfReaderContentParser(pdfReader);
 
CustomRenderListener renderListener= newCustomRenderListener();
 
renderListener.setKeyWord(keyWords);for (page = 1; page <= pageNum; page++) {
 
renderListener.setPage(page);
 
pdfReaderContentParser.processContent(page, renderListener);
 
coordinate=renderListener.getPcoordinate();if (coordinate != null) break;
 
}
 
}catch(IOException e) {
 
e.printStackTrace();
 
}returncoordinate;
 
}
 
}

 
根据指定关键字获取坐标：
 
1.使用的是 itextpdf ，所用依赖：
 
com.itextpdf
 
itextpdf
 
5.5.6
 
2.获取坐标代码：
 
package test;
 
import java.io.IOException;
 
import java.util.ArrayList;
 
import java.util.List;
 
import com.itextpdf.awt.geom.Rectangle2D.Float;
 
import com.itextpdf.text.pdf.PdfReader;
 
import com.itextpdf.text.pdf.parser.ImageRenderInfo;
 
import com.itextpdf.text.pdf.parser.PdfReaderContentParser;
 
import com.itextpdf.text.pdf.parser.RenderListener;
 
import com.itextpdf.text.pdf.parser.TextRenderInfo;
 
public class Test {
 
// 指定关键字
 
public static String KEY_WORD = "总计:";
 
// PDF当前页数
 
public static int curPage = 0;
 
// 集合里放了多少条数据
 
public static int count = 0;
 
// 坐标信息集合
 
public static List arrays = new ArrayList();
 
// 关键字字符数组
 
public static char[] charArray = KEY_WORD.toCharArray();
 
/**
 
* 获取指定PDF文件中指定关键字的坐标
 
*
 
* @param filePath
 
* @return
 
* @throws Exception
 
*/
 
public static List getKeyWordCoordinate(String filePath) throws Exception {
 
if (charArray.length < 2) {
 
throw new Exception("关键字长度不能小于2");
 
}
 
PdfReader pdfReader = null;
 
try {
 
pdfReader = new PdfReader(filePath);
 
int pageNum = pdfReader.getNumberOfPages();
 
PdfReaderContentParser pdfReaderContentParser = new PdfReaderContentParser(pdfReader);
 
for (curPage = 1; curPage <= pageNum; curPage++) {
 
pdfReaderContentParser.processContent(curPage, new RenderListener() {
 
// 关键字标志
 
boolean isKeyWord = false;
 
// PDF读出来的文字与截取关键字的第index位对比，如果在关键字的长度里
 
/**
 
* 这里从PDF中是一个一个读取文字的，把读取出来的PDF文字先和关键字的第一个字比较看是否相同：
 
* 如果相同，则假设已经找到关键字(将isKeyWord设为true)，把这个关键字的坐标存到集合里面，并计算集合里存放了多少个数据(count++)、设置进入到一个循环(loop
 
* = true一个循环即为关键字从第一个字到最后一个字依次和读取的PDF文字比较的过程)；
 
* 如果在这个比较过程中发现有的字符和关键字不匹配，则移除与之对应的集合里的数据。
 
*/
 
// 关键字的第几位
 
int index = 0;
 
// 对比关键字是否进入到一个循环
 
boolean loop = false;
 
// 处理数据
 
public void renderText(TextRenderInfo textRenderInfo) {
 
String text = textRenderInfo.getText();
 
if (text.equals(new String(charArray, 0, 1)) && text != null) {
 
// 把当前从PDF文件中读取的文字横坐标、纵坐标和当前页存到集合里面
 
Float boundingRectange = textRenderInfo.getBaseline().getBoundingRectange();
 
float[] resu = new float[3];
 
resu[0] = boundingRectange.x;
 
resu[1] = boundingRectange.y;
 
resu[2] = curPage;
 
arrays.add(resu);
 
count++;
 
isKeyWord = true;
 
loop = true;
 
return;
 
}
 
if (isKeyWord) {
 
// 获取关键字中下一个字符和读取文字比较
 
index++;
 
if (index < charArray.length) {
 
if (text.equalsIgnoreCase(new String(charArray, index, 1)) && text != null) {
 
isKeyWord = true;
 
return;
 
} else {
 
isKeyWord = false;
 
// 如果在一个循环中比较时，发现有字符和关键字不匹配，则移除与之对应的集合里的数据
 
if (loop) {
 
count--;
 
arrays.remove(count);
 
}
 
}
 
} else {
 
// 如果index大于关键字字符数组的长度，则该循环结束
 
loop = false;
 
}
 
}
 
index = 0;
 
}
 
public void renderImage(ImageRenderInfo arg0) {
 
}
 
public void endTextBlock() {
 
}
 
public void beginTextBlock() {
 
}
 
});
 
}
 
} catch (IOException e) {
 
e.printStackTrace();
 
} finally {
 
if (pdfReader != null) {
 
pdfReader.close();
 
}
 
}
 
return arrays;
 
}
 
//测试
 
public static void main(String[] args) throws Exception {
 
String filePath = "F:/test.pdf";
 
List list = getKeyWordCoordinate(filePath);
 
for (float[] f : list) {
 
System.out.println("x坐标：" + f[0] + "，y坐标：" + f[1] + "， 关键字所在页：" + f[2]);
 
}
 
}
 
}
 
以上代码执行结果示例：
 
需要注意的是：这里获取坐标的原点是PDF文件纵向显示时的左下角

 
importjava.util.ArrayList;importjava.util.List;importorg.apache.log4j.Logger;importorg.drools.util.StringUtils;importcom.itextpdf.awt.geom.Rectangle2D;importcom.itextpdf.text.Rectangle;importcom.itextpdf.text.pdf.parser.ImageRenderInfo;importcom.itextpdf.text.pdf.parser.RenderListener;importcom.itextpdf.text.pdf.parser.TextRenderInfo;public class KeyWordPositionListener implementsRenderListener {private static Logger logger = Logger.getLogger(KeyWordPositionListener.class);private List matches = new ArrayList();private List allItems = new ArrayList();privateRectangle curPageSize;/*** 匹配的关键字*/
 
privateString keyword;/*** 匹配的当前页*/
 
privateInteger pageNumber;public voidbeginTextBlock() {//do nothing
 
}public voidrenderText(TextRenderInfo renderInfo) {
 
String content=renderInfo.getText();
 
content= content.replace("
 
.replace(">", "").replace("》", "").replace(")", "").replace(")", "").replace("、", "").replace(".", "")
 
.replace("：", "").replace(":", "").replace(" ", "");
 
Rectangle2D.Float textRectangle=renderInfo.getDescentLine().getBoundingRectange();
 
MatchItem item= newMatchItem();
 
item.setContent(content);
 
item.setPageNum(pageNumber);
 
item.setPageWidth(curPageSize.getWidth());
 
item.setPageHeight(curPageSize.getHeight());
 
item.setX((float)textRectangle.getX());
 
item.setY((float)textRectangle.getY());if(!StringUtils.isEmpty(content)){if(content.equalsIgnoreCase(keyword)) {
 
matches.add(item);
 
}
 
}else{
 
item.setContent("空字符串");
 
}
 
allItems.add(item);//先保存所有的项
 
}public voidendTextBlock() {//do nothing
 
}public voidrenderImage(ImageRenderInfo renderInfo) {//do nothing
 
}/*** 设置需要匹配的当前页
 
*@parampageNumber*/
 
public voidsetPageNumber(Integer pageNumber) {this.pageNumber =pageNumber;
 
}/*** 设置需要匹配的关键字，忽略大小写
 
*@paramkeyword*/
 
public voidsetKeyword(String keyword) {this.keyword =keyword;
 
}/*** 返回匹配的结果列表
 
*@return
 
*/
 
public ListgetMatches() {returnmatches;
 
}voidsetCurPageSize(Rectangle rect) {this.curPageSize =rect;
 
}public ListgetAllItems() {returnallItems;
 
}public void setAllItems(ListallItems) {this.allItems =allItems;
 
}
 
}