diff --git a/src/main/java/com/docus/sw/fenpan/FenpanService.java b/src/main/java/com/docus/sw/fenpan/FenpanService.java index eae64a8..998f3da 100644 --- a/src/main/java/com/docus/sw/fenpan/FenpanService.java +++ b/src/main/java/com/docus/sw/fenpan/FenpanService.java @@ -4,6 +4,7 @@ import com.alibaba.excel.util.FileUtils; import com.docus.sw.Config; import com.docus.sw.word.GetPicsDocx; import com.docus.sw.word.PdfBoxUtils; +import com.docus.sw.word.ReadImgDoc; import lombok.extern.slf4j.Slf4j; import org.apache.commons.imaging.ImageInfo; import org.apache.commons.imaging.ImageReadException; @@ -111,7 +112,7 @@ public class FenpanService { file.mkdirs(); } - GetPicsDocx.getPics(piece.getAbsolutePath(),file.getAbsolutePath()); + ReadImgDoc.readPicture(piece.getAbsolutePath(),file.getAbsolutePath()); List documentList = new ArrayList<>(); try { PdfBoxUtils.pdf2image(piece.getAbsolutePath(),file.getAbsolutePath()); diff --git a/src/main/java/com/docus/sw/souyin/SuoyinService.java b/src/main/java/com/docus/sw/souyin/SuoyinService.java index 240ac7b..936eb66 100644 --- a/src/main/java/com/docus/sw/souyin/SuoyinService.java +++ b/src/main/java/com/docus/sw/souyin/SuoyinService.java @@ -7,6 +7,9 @@ import com.docus.sw.fenpan.Document; import com.docus.sw.fenpan.FileTypeEnum; import com.docus.sw.fenpan.Pieces; import com.docus.sw.fenpan.Roll; +import com.docus.sw.word.GetPicsDocx; +import com.docus.sw.word.PdfBoxUtils; +import com.docus.sw.word.ReadImgDoc; import com.drew.tools.FileUtil; import com.google.gson.Gson; import org.apache.commons.imaging.ImageInfo; @@ -72,50 +75,69 @@ public class SuoyinService { for(File piece:piecesFile){ //区分word和pdf + if(piece.getName().endsWith(".doc")){ + Pieces pieces = new Pieces(FileTypeEnum.DOC, piece.getAbsolutePath(), piece.getName()); - List documentList = new ArrayList<>(); - Pieces pieces = new Pieces(FileTypeEnum.JPG, piece.getAbsolutePath(), piece.getName()); - for (File docfile : piece.listFiles()) { + File temp = new File("temp"); + if(!temp.exists()){ + temp.mkdirs(); + } + + ReadImgDoc.readPicture(piece.getAbsolutePath(),temp.getAbsolutePath()); + try { + PdfBoxUtils.pdf2image(piece.getAbsolutePath(),temp.getAbsolutePath()); + File[] listFiles = temp.listFiles(); + genPieces(pieces, listFiles); + + } catch (IOException e) { + + } + //删除对应的temp 文件 + file.delete(); + }else if(piece.getName().endsWith(".docx")){ + Pieces pieces = new Pieces(FileTypeEnum.DOC, piece.getAbsolutePath(), piece.getName()); + + File temp = new File("temp"); + if(!temp.exists()){ + temp.mkdirs(); + } + + GetPicsDocx.getPics(piece.getAbsolutePath(),temp.getAbsolutePath()); + try { + PdfBoxUtils.pdf2image(piece.getAbsolutePath(),temp.getAbsolutePath()); + File[] listFiles = temp.listFiles(); + genPieces(pieces, listFiles); + + } catch (IOException e) { - //非图片模式，跳过。 - if(!(docfile.getName().endsWith(".jpg") || docfile.getName().endsWith(".png") - || docfile.getName().endsWith(".jpeg") || docfile.getName().endsWith(".tif") - || docfile.getName().endsWith(".tiff")|| file.getName().endsWith(".jp2") || file.getName().endsWith(".jpm"))|| file.getName().endsWith(".gif")){ - continue; } - if (file.getName().endsWith(".jp2") || file.getName().endsWith(".jpm")) { - // 读取 JPEG 2000 图像文件 - - try { - BufferedImage image = ImageIO.read(file); - int height = image.getHeight(); - int width = image.getWidth(); - Document document = new Document(width, height, 300); - documentList.add(document); - } catch (IOException e) { - throw new RuntimeException(e); - } - - } else { - try { - ImageInfo imageInfo = Imaging.getImageInfo(docfile); - int height = imageInfo.getHeight(); - int width = imageInfo.getWidth(); - int physicalHeightDpi = imageInfo.getPhysicalHeightDpi(); - Document document = new Document(width, height, physicalHeightDpi); - documentList.add(document); - } catch (IOException e) { - throw new RuntimeException("非图片格式", e); - } catch (ImageReadException e) { - throw new RuntimeException(e); - } catch (IllegalArgumentException e) { - throw new RuntimeException(e); - } + //删除对应的temp 文件 + file.delete(); + }else if(piece.getName().endsWith(".pdf")){ + Pieces pieces = new Pieces(FileTypeEnum.DOC, piece.getAbsolutePath(), piece.getName()); + + File temp = new File("temp"); + if(!temp.exists()){ + temp.mkdirs(); } + GetPicsDocx.getPics(piece.getAbsolutePath(),temp.getAbsolutePath()); + try { + PdfBoxUtils.pdf2image(piece.getAbsolutePath(),temp.getAbsolutePath()); + File[] listFiles = temp.listFiles(); + genPieces(pieces, listFiles); + + } catch (IOException e) { + + } + //删除对应的temp 文件 + file.delete(); + }else{ + Pieces pieces = new Pieces(FileTypeEnum.JPG, piece.getAbsolutePath(), piece.getName()); + File[] listFiles = piece.listFiles(); + genPieces(pieces, listFiles); + pieceList.add(pieces); } - pieces.put(documentList); - pieceList.add(pieces); } @@ -230,6 +252,48 @@ public class SuoyinService { // } + private static void genPieces(Pieces pieces, File[] listFiles) { + List documentList = new ArrayList<>(); + for (File docfile : listFiles) { + //非图片模式，跳过。 + if(!(docfile.getName().endsWith(".jpg") || docfile.getName().endsWith(".png") + || docfile.getName().endsWith(".jpeg") || docfile.getName().endsWith(".tif") + || docfile.getName().endsWith(".tiff")|| docfile.getName().endsWith(".jp2") || docfile.getName().endsWith(".jpm"))|| docfile.getName().endsWith(".gif")){ + continue; + } + if (docfile.getName().endsWith(".jp2") || docfile.getName().endsWith(".jpm")) { + // 读取 JPEG 2000 图像文件 + + try { + BufferedImage image = ImageIO.read(docfile); + int height = image.getHeight(); + int width = image.getWidth(); + Document document = new Document(width, height, 300); + documentList.add(document); + } catch (IOException e) { + throw new RuntimeException(e); + } + + } else { + try { + ImageInfo imageInfo = Imaging.getImageInfo(docfile); + int height = imageInfo.getHeight(); + int width = imageInfo.getWidth(); + int physicalHeightDpi = imageInfo.getPhysicalHeightDpi(); + Document document = new Document(width, height, physicalHeightDpi); + documentList.add(document); + } catch (IOException e) { + throw new RuntimeException("非图片格式", e); + } catch (ImageReadException e) { + throw new RuntimeException(e); + } catch (IllegalArgumentException e) { + throw new RuntimeException(e); + } + } + pieces.put(documentList); + } + } + private List indexGen(IndexPlate indexPlate, List rollList, Integer pianPageNum, Integer totalSize) { List indexPageRows = new ArrayList<>(); int start = 1; diff --git a/src/main/java/com/docus/sw/word/ReadImgDoc.java b/src/main/java/com/docus/sw/word/ReadImgDoc.java index e5e9f32..5cdb4d0 100644 --- a/src/main/java/com/docus/sw/word/ReadImgDoc.java +++ b/src/main/java/com/docus/sw/word/ReadImgDoc.java @@ -6,10 +6,7 @@ import org.apache.poi.hwpf.usermodel.CharacterRun; import org.apache.poi.hwpf.usermodel.Picture; import org.apache.poi.hwpf.usermodel.Range; -import java.io.File; -import java.io.FileInputStream; -import java.io.FileOutputStream; -import java.io.OutputStream; +import java.io.*; import java.util.UUID; public class ReadImgDoc { @@ -43,23 +40,52 @@ public class ReadImgDoc { } - private static void readPicture(String path, String toPath) throws Exception { - FileInputStream in = new FileInputStream(new File(path)); - HWPFDocument doc = new HWPFDocument(in); - int length = doc.characterLength(); - PicturesTable pTable = doc.getPicturesTable(); - // int TitleLength=doc.getSummaryInformation().getTitle().length(); + public static void readPicture(String path, String toPath) { + FileInputStream in = null; + try { + in = new FileInputStream(new File(path)); + HWPFDocument doc = new HWPFDocument(in); + int length = doc.characterLength(); + PicturesTable pTable = doc.getPicturesTable(); + // int TitleLength=doc.getSummaryInformation().getTitle().length(); - // System.out.println(TitleLength); - // System.out.println(length); - for (int i = 0; i < length; i++) { - Range range = new Range(i, i + 1, doc); - CharacterRun cr = range.getCharacterRun(0); - if (pTable.hasPicture(cr)) { - Picture pic = pTable.extractPicture(cr, false); - String afileName = pic.suggestFullFileName(); - OutputStream out = new FileOutputStream(new File(toPath + i + afileName)); - pic.writeImageContent(out); + // System.out.println(TitleLength); + // System.out.println(length); + for (int i = 0; i < length; i++) { + Range range = new Range(i, i + 1, doc); + CharacterRun cr = range.getCharacterRun(0); + if (pTable.hasPicture(cr)) { + Picture pic = pTable.extractPicture(cr, false); + String afileName = pic.suggestFullFileName(); + OutputStream out = null; + try { + out = new FileOutputStream(new File(toPath + i + afileName)); + pic.writeImageContent(out); + } catch (FileNotFoundException e) { + e.getMessage(); + } catch (IOException e) { + e.getMessage(); + }finally { + if(out!= null){ + try { + out.close(); + } catch (IOException e) { + throw new RuntimeException(e); + } + } + } + } + } + + } catch (IOException e) { + e.getMessage(); + }finally { + if(in!=null){ + try { + in.close(); + } catch (IOException e) { + throw new RuntimeException(e); + } } }