spire.pdf 使用學(xué)習(xí)記錄

• • 背景
  • 簡(jiǎn)介
  • 相似產(chǎn)品及特點(diǎn)
  • 官網(wǎng)地址
  • 功能實(shí)現(xiàn)
  • • 切割pdf
    • • 按每頁(yè)切割
      • 按指定頁(yè)切割
    • 將pdf轉(zhuǎn)成txt
    • 識(shí)別pdf中的篇章節(jié)生成對(duì)應(yīng)文本
    • 去除水印
    • 獲取書簽
    • 獲取每頁(yè)中圖片
    • 項(xiàng)目地址

背景

通過(guò)打印機(jī)將一本紙質(zhì)書轉(zhuǎn)為pdf的格式，以下所有操作都是在這個(gè)基礎(chǔ)上操作，最終目的生成n篇txt文件，每個(gè)txt名字規(guī)則是:起始頁(yè)_篇&章&節(jié) 內(nèi)容:是對(duì)應(yīng)切割的內(nèi)容；

簡(jiǎn)介

這是一款基于OCR框架的解析工具，擁有比價(jià)完整的Java 類庫(kù)，及完善的API文檔，不但具備pdf的讀寫，還支持文本和圖片的提取，水印的添加，書簽的增刪改，表格的操作，同時(shí)還支持，將pdf轉(zhuǎn)化成word、HTML、XPS、SVG、等多種方式；當(dāng)然目前這類工具市面上有很多，不過(guò)通過(guò)比較最后選擇了spire.pdf ;目前這款項(xiàng)目共有兩個(gè)版本，一個(gè)是免費(fèi)版本一個(gè)是付費(fèi)版本，免費(fèi)版本如果只是處理簡(jiǎn)單的pdf是沒問題的，但是如果涉及到輸出為pdf則會(huì)只顯示前10頁(yè)，第十一頁(yè)則是預(yù)定的購(gòu)買頁(yè)介紹，不過(guò)介于spire.pdf的完善性我最后還是選擇了他，至于10頁(yè)的問題，后面會(huì)拿出我的處理辦法,如果你是其他語(yǔ)言如.NET、Android、也可以使用這款產(chǎn)品;

相似產(chǎn)品及特點(diǎn)

• PDFBox
• itext
• 百度
• Tesseract 只能用于識(shí)別圖片，如果需要先將pdf轉(zhuǎn)為圖片

官網(wǎng)地址

https://www.e-iceblue.cn/Introduce/Free-Spire-PDF-JAVA.html

功能實(shí)現(xiàn)

前往官網(wǎng)下載jar包或者直接在maven上通過(guò)坐標(biāo)也可實(shí)現(xiàn)，不過(guò)maven上的肯定沒有網(wǎng)頁(yè)下載的新；

<!-- 事先配置maven倉(cāng)庫(kù)路徑--> <repositories><repository><id>com.e-iceblue</id><url>http://repo.e-iceblue.cn/repository/maven-public/</url></repository></repositories><!--maven中 spire.pdf鏡像依賴--> <dependencies><dependency><groupId> e-iceblue </groupId><artifactId>spire.pdf.free</artifactId><version>3.11.6</version></dependency> </dependencies>

切割pdf

按每頁(yè)切割

/***每一頁(yè)生成一個(gè)pdf文檔*/public static void splitPdfOneByOne(){PdfDocument pdf = new PdfDocument();int count = pdf.getPages().getCount();System.out.println(count);pdf.loadFromFile("C:\\Users\\wangchenchen\\Desktop\\boot-structure\\book.pdf");pdf.split("C:\\Users\\wangchenchen\\Desktop\\boot-structure\\output\\surgery_{0}.pdf",0);pdf.close();}

按指定頁(yè)切割

package com.wangcc.wangccocrdemo001.ocr;import com.spire.pdf.PdfDocument; import com.spire.pdf.PdfPageBase; import com.spire.pdf.graphics.PdfMargins; import com.wangcc.wangccocrdemo001.util.FileUtil;import java.awt.geom.Point2D; import java.io.File;/*** 拆分文檔* 拆成每頁(yè)一個(gè)* 拆成每九頁(yè)一個(gè)(免費(fèi)版最多只能處理九頁(yè))* @author wangcc* @createTime 2021年08月31日 23:25:00*/ public class SubPDF {/*** 每九頁(yè)生成一個(gè)pdf**/public static void splitPdfMoreByOne(){String fileName = "C:\\Users\\wangchenchen\\Desktop\\boot-structure\\book.pdf";String outPath = "C:\\Users\\wangchenchen\\Desktop\\boot-structure\\outFile\\outPDFByMore";PdfDocument pdf = new PdfDocument();pdf.loadFromFile(fileName);int totalCount = pdf.getPages().getCount();PdfPageBase pageBase;PdfDocument document = new PdfDocument();int count = 1;for(int i = 41; i< 822;i++){System.out.println(i+"/"+822);pageBase = document.getPages().add(pdf.getPages().get(i).getSize(),new PdfMargins(0));pdf.getPages().get(i).createTemplate().draw(pageBase, new Point2D.Float(0,0));if(count % 9 == 0){String path = "\\splitPdf-"+i+".pdf";document.saveToFile(outPath+path);document = new PdfDocument();}count++;}if(document.getPages().getCount() >0){String path = "\\splitPdf-999999.pdf";document.saveToFile(outPath+path);}}/*** @Description //* @return void**/public static void splitPdfByNumber(Integer begin, Integer end,String filePath,String pdfOutPath){if(begin.equals("") || end.equals("") || filePath.equals("") || pdfOutPath.equals("")){System.out.println("傳入?yún)?shù)有空.......");return;}if(begin >= end){System.out.println("截止頁(yè)數(shù)不能小于或等于開始頁(yè)數(shù).......");return;}if(end-begin > 9){System.out.println("操作頁(yè)數(shù)最多為9頁(yè)");return;}File file = new File(pdfOutPath);if (file.exists()){FileUtil fileUtil = new FileUtil();fileUtil.DeleteFolder(pdfOutPath);}PdfDocument pdf = new PdfDocument();pdf.loadFromFile(filePath);int totalCount = pdf.getPages().getCount();PdfPageBase pageBase;PdfDocument document = new PdfDocument();for(int i = begin; i< end;i++){System.out.println(i+"/"+end);pageBase = document.getPages().add(pdf.getPages().get(i).getSize(),new PdfMargins(0));pdf.getPages().get(i).createTemplate().draw(pageBase, new Point2D.Float(0,0));String path = "\\surgery_"+i+".pdf";document.saveToFile(pdfOutPath+path);document = new PdfDocument();}} }

將pdf轉(zhuǎn)成txt

package com.wangcc.wangccocrdemo001.ocr;import com.spire.pdf.PdfDocument; import com.spire.pdf.PdfPageBase;import java.io.*; import java.util.*; import java.util.regex.Matcher; import java.util.regex.Pattern;/*** 讀取所有拆分文件生成txt文件* @author wangcc* @createTime 2021年08月31日 23:51:00*/ public class ReadAllSplitFile {public static String fileName = "C:\\Users\\wangchenchen\\Desktop\\boot-structure\\outFile\\outPDFByMore\\";public static String outPath = "C:\\Users\\wangchenchen\\Desktop\\boot-structure\\outFile\\readPdfFile.txt";public static void main(String[] args) {List<File> fileList = readAllFile();List<String> pdfFileNameList = new ArrayList<>();for (File file:fileList) {pdfFileNameList.add(file.getName());}Collections.sort(pdfFileNameList, new Comparator<String>() {@Overridepublic int compare(String o1, String o2) {int n1 = extractNumber(o1);int n2 = extractNumber(o2);return n1 - n2;}});File file = new File(outPath);if(file.exists()){file.delete();}for (String s:pdfFileNameList) {try {readFile(s);} catch (IOException e) {e.printStackTrace();}}}public static List<File> readAllFile(){String filePath = "C:\\Users\\wangchenchen\\Desktop\\boot-structure\\outFile\\outPDFByMore";ArrayList<File> fileList = new ArrayList<>();File file = new File(filePath);File[] files = file.listFiles();if(Objects.isNull(files)){return null;}for (File f:files) {if(f.isFile()){fileList.add(f);}}return fileList;}/*** @Param orderStr 排序：asc,des,不區(qū)分大小寫**/public static List<File> sortFileByName(List<File> fileList, final String orderStr){if(!orderStr.equalsIgnoreCase("asc") && !orderStr.equalsIgnoreCase("desc")){return fileList;}File[] files = fileList.toArray(new File[0]);Arrays.sort(files, new Comparator<File>() {@Overridepublic int compare(File o1, File o2) {int n1 = extractNumber(o1.getName());int n2 = extractNumber(o2.getName());if(orderStr == null || orderStr.length() < 1 || orderStr.equalsIgnoreCase("asc")){return n1 - n2;}else {//降序return n2 - n1;}}});return new ArrayList<File>(Arrays.asList(files));}public static int extractNumber(String name){int i;try {String s = name.replaceAll("[^\\d]", "");i = Integer.parseInt(s);}catch (Exception e){i = 0;}return i;}public static void readFile(String path) throws IOException {PdfDocument pdf = new PdfDocument();pdf.loadFromFile(fileName+path);PdfPageBase page;StringBuilder sb = new StringBuilder();Pattern pattern = Pattern.compile("(^(\\s*)第)(.{1,9})[章節(jié)卷集部篇回](\\s{1,10})(.{1,20})(\\s{1,10})");Pattern pattern1 = Pattern.compile("(\\s{0,10})([0-9][0-9]?[0-9]?[0-9]?)");//遍歷PDF頁(yè)面，獲取每個(gè)頁(yè)面的文本并添加到StringBuilder對(duì)象for(int i = 0;i < pdf.getPages().getCount();i++) {//System.out.println("循環(huán)遍歷pdf頁(yè)數(shù)：當(dāng)前" + i + "頁(yè)/" + pdf.getPages().getCount() + "頁(yè)");page = pdf.getPages().get(i);int count = 0;String extractText = null;BufferedReader br = new BufferedReader(new InputStreamReader(new ByteArrayInputStream(page.extractText(true).getBytes())));while ((extractText = br.readLine())!= null){Matcher matcher = pattern.matcher(extractText);Matcher matcher1 = pattern1.matcher(extractText);/*末尾包含數(shù)字的*/if (count != 0 || matcher.find()){//System.out.println(extractText);if(!extractText.equals("") && !matcher1.find()){String s = extractText.replaceAll("\\s{5,9}", " ");sb.append(s+"\n");}}count++;}br.close();}FileWriter writer;try {//將StringBuilder對(duì)象中的文本寫入到文本文件writer = new FileWriter(outPath,true);System.out.println(sb.toString());writer.write(sb.toString());writer.flush();writer.close();sb.delete(0,sb.length());} catch (IOException e) {e.printStackTrace();}pdf.close();} }

識(shí)別pdf中的篇章節(jié)生成對(duì)應(yīng)文本

去除水印

package com.wangcc.wangccocrdemo001.ocr;import com.spire.pdf.PdfDocument;import javax.imageio.ImageIO; import java.awt.image.BufferedImage; import java.io.File; import java.io.IOException;/*** 去除水印（未必都有用，主要是看水印的類型）* 水印有兩種* 1.文字，放在對(duì)應(yīng)的位置* 2.大圖片，png 背景透明* 一些特殊的水印原則上是去不掉的* @author wangcc* @createTime 2021年08月31日 23:46:00*/ public class ClearWaterMark {public static void main(String[] args) throws IOException {PdfDocument pdf = new PdfDocument();pdf.loadFromFile("C:\\Users\\wangchenchen\\Desktop\\boot-structure\\out\\splitPdf-18.pdf");BufferedImage bufferedImage = null;for (int i = 0; i<pdf.getPages().getCount();i++){bufferedImage = pdf.saveAsImage(i);bufferedImage.getSubimage(bufferedImage.getMinX(),15,bufferedImage.getWidth(),bufferedImage.getHeight()-15);File saveFile = new File("C:\\Users\\wangchenchen\\Desktop\\boot-structure\\outImg\\" + i + ".png");if(!saveFile.exists()){saveFile.mkdirs();}ImageIO.write(bufferedImage,"PNG",saveFile);bufferedImage.flush();}} }

獲取書簽

package com.wangcc.wangccocrdemo001.ocr;import com.spire.pdf.PdfDocument; import com.spire.pdf.bookmarks.PdfBookmark; import com.spire.pdf.bookmarks.PdfBookmarkCollection;import java.io.FileWriter; import java.io.IOException;/*** 獲取書簽* @author wangcc* @createTime 2021年08月31日 23:47:00*/ public class getBookMake {public static void main(String[] args) {PdfDocument pdf = new PdfDocument();pdf.loadFromFile("C:\\Users\\wangchenchen\\Desktop\\boot-structure\\out\\splitPdf-18.pdf");PdfBookmarkCollection bookmarkCollection = pdf.getBookmarks();StringBuilder stringBuilder = new StringBuilder();/*獲取書簽*/GetBookMakeTitle(bookmarkCollection,stringBuilder);FileWriter writer;try {writer = new FileWriter("C:\\Users\\wangchenchen\\Desktop\\boot-structure\\讀取的書簽.txt");writer.write(stringBuilder.toString());writer.flush();} catch (IOException e) {e.printStackTrace();}pdf.dispose();}public static void GetBookMakeTitle(PdfBookmarkCollection bookmarkCollection,StringBuilder stringBuilder){if(bookmarkCollection.getCount() > 0){for (int i = 0; i < bookmarkCollection.getCount(); i++){PdfBookmark bookmark = bookmarkCollection.get(i);stringBuilder.append(bookmark.getTitle());GetBookMakeTitle(bookmark,stringBuilder);}}} }

獲取每頁(yè)中圖片

package com.wangcc.wangccocrdemo001.ocr;import com.spire.pdf.PdfDocument; import com.spire.pdf.PdfPageBase;import javax.imageio.ImageIO; import java.awt.image.BufferedImage; import java.io.BufferedInputStream; import java.io.File; import java.io.FileInputStream; import java.io.InputStream; import java.util.Objects;/*** 提取每頁(yè)pdf中的圖片* @author wangcc* @createTime 2021年08月31日 23:41:00*/ public class ExtractImg {public static void main(String[] args) throws Exception{//加載測(cè)試文檔InputStream inputStream = new BufferedInputStream(new FileInputStream(new File("C:\\Users\\wangchenchen\\Desktop\\boot-structure\\demo.pdf")));PdfDocument pdf = new PdfDocument();//pdf.loadFromFile("C:\\Users\\wangchenchen\\Desktop\\boot-structure\\demo.pdf");pdf.loadFromStream(inputStream);//定義一個(gè)int型變量int index = 0;//遍歷PDF每一頁(yè)for (int i= 0;i< pdf.getPages().getCount(); i ++){System.out.println(i+"/"+pdf.getPages().getCount());//獲取PDF頁(yè)面PdfPageBase page = pdf.getPages().get(i);//使用extractImages方法獲取頁(yè)面上圖片for (BufferedImage image : page.extractImages()) {if(Objects.nonNull(image)){//指定輸出圖片名稱File output = new File( String.format("C:\\Users\\wangchenchen\\Desktop\\boot-structure\\Img\\Image_%d.pdf", index++));//將圖片保存為PNG格式文件ImageIO.write(image, "PNG", output);}}}} }

項(xiàng)目地址

spire.pdf 下載

總結(jié)

以上是生活随笔為你收集整理的Spire.pdf-使用学习记录的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。