/**
* Returns the format for the file passed as a parameter.
*
* <p>We will ask the mimetype registry service to sniff its mimetype.
*
* @return DocumentFormat for the given file
*/
private static DocumentFormat getSourceFormat(
OfficeDocumentConverter documentConverter, File file) {
MimetypeRegistry mimetypeRegistry = Framework.getService(MimetypeRegistry.class);
String mimetypeStr = mimetypeRegistry.getMimetypeFromFile(file);
DocumentFormat format = documentConverter.getFormatRegistry().getFormatByMediaType(mimetypeStr);
return format;
}
private boolean commonOfficeDoc2Pdf(File inputFile, File outputFile) {
try {
DocumentFormat opf = converter.getFormatRegistry().getFormatByExtension("pdf");
logger.info("start converting...");
converter.convert(inputFile, outputFile, opf);
return true;
} catch (Exception cex) {
cex.printStackTrace();
return false;
} finally {
// close the connection
if (officeManager != null) {
logger.info("converting processing has done: " + inputFile.getName());
// officeManager.stop();
}
}
}
@Override
public void startConvertion() {
OfficeDocumentConverter converter =
new OfficeDocumentConverter(ConverterFactoryOffice.getOfficeManager());
_logger.debug("Start converting " + source.getName());
long startTime = System.currentTimeMillis();
try {
converter.convert(source, target);
setModificationDate();
target.copyTo(target.replaceExtention(Constants.TXT));
target.deleteOnExit();
success = true;
} catch (OfficeException e) {
close();
_logger.error(e.getMessage());
}
_logger.debug("convertion time: " + (System.currentTimeMillis() - startTime) + " ms");
}
/**
* Returns the destination format for the given plugin.
*
* <p>It takes the actual destination mimetype from the plugin configuration.
*
* @param sourceFormat the source format
* @param pdfa1 true if PDF/A-1 is required
*/
protected DocumentFormat getDestinationFormat(
OfficeDocumentConverter documentConverter, DocumentFormat sourceFormat, boolean pdfa1) {
String mimeType = getDestinationMimeType();
DocumentFormat destinationFormat =
documentConverter.getFormatRegistry().getFormatByMediaType(mimeType);
if ("application/pdf".equals(mimeType)) {
destinationFormat = extendPDFFormat(sourceFormat, destinationFormat, pdfa1);
}
return destinationFormat;
}
/**
* 将Office文档转换为PDF. 运行该函数需要用到OpenOffice, OpenOffice
*
* @param sourceFile 源文件,绝对路径. 可以是Office2003-2007全部格式的文档, Office2010的没测试. 包括.doc, .docx, .xls,
* .xlsx, .ppt, .pptx等.
* @param destFile 目标文件.绝对路径.
*/
public static void word2pdf(String inputFilePath) {
DefaultOfficeManagerConfiguration config = new DefaultOfficeManagerConfiguration();
String officeHome = getOfficeHome();
config.setOfficeHome(officeHome);
OfficeManager officeManager = config.buildOfficeManager();
officeManager.start();
OfficeDocumentConverter converter = new OfficeDocumentConverter(officeManager);
String outputFilePath = getOutputFilePath(inputFilePath);
File inputFile = new File(inputFilePath);
if (inputFile.exists()) { // 找不到源文件, 则返回
File outputFile = new File(outputFilePath);
if (!outputFile.getParentFile().exists()) { // 假如目标路径不存在, 则新建该路径
outputFile.getParentFile().mkdirs();
}
converter.convert(inputFile, outputFile);
}
officeManager.stop();
}
@Override
public void afterPropertiesSet() throws Exception {
System.out.println("******************");
// 启动libreoffice进程 Windows平台需要启动，linux不需要启动
if (PlatformUtils.isWindows()) {
String command =
"cmd /c soffice --headless --accept=\"socket,host=127.0.0.1,port=8100;urp\" --nofirststartwizard";
Runtime runtime = Runtime.getRuntime();
runtime.exec(command);
}
officeManager = configuration.buildOfficeManager();
converter = new OfficeDocumentConverter(officeManager);
customDocumentFormatRegistry = (SimpleDocumentFormatRegistry) converter.getFormatRegistry();
/*
* // TODO：这里可以增加OpenOffice支持的文件类型 异常json
* customDocumentFormatRegistry.addFormat(new
* DocumentFormat("OpenOffice.org 1.0 Template", "sxd",
* "application/vnd.sun.xml.draw"));
* customDocumentFormatRegistry.addFormat(new
* DocumentFormat("OpenOffice.org 1.0 Template", "odf",
* "application/vnd.oasis.opendocument.formula"));
*/
DocumentFormat txt = new DocumentFormat("Java", "java", "text/plain");
txt.setInputFamily(DocumentFamily.TEXT);
Map<String, Object> txtLoadAndStoreProperties = new LinkedHashMap<String, Object>();
txtLoadAndStoreProperties.put("FilterName", "Text (encoded)");
txtLoadAndStoreProperties.put("FilterOptions", "utf8");
txt.setLoadProperties(txtLoadAndStoreProperties);
txt.setStoreProperties(DocumentFamily.TEXT, txtLoadAndStoreProperties);
customDocumentFormatRegistry.addFormat(txt);
// 添加wps格式
customDocumentFormatRegistry.addFormat(
new DocumentFormat("wps文字", "wps", "application/msword")); // wps文字
customDocumentFormatRegistry.addFormat(
new DocumentFormat("wps表格", "et", "application/msword")); // wps表格
customDocumentFormatRegistry.addFormat(
new DocumentFormat("wps演示", "dps", "application/msword")); // wps演示
// 启动officeManager，保持在Bean的生命周期中一直在启动状态
officeManager.start();
}
@Override
public BlobHolder convert(BlobHolder blobHolder, Map<String, Serializable> parameters)
throws ConversionException {
blobHolder = new UTF8CharsetConverter().convert(blobHolder, parameters);
Blob inputBlob = blobHolder.getBlob();
String blobPath = blobHolder.getFilePath();
if (inputBlob == null) {
return null;
}
OfficeDocumentConverter documentConverter = newDocumentConverter();
// This plugin do deal only with one input source.
String sourceMimetype = inputBlob.getMimeType();
boolean pdfa1 = parameters != null && Boolean.TRUE.equals(parameters.get(PDFA1_PARAM));
File sourceFile = null;
File outFile = null;
File[] files = null;
try {
// If the input blob has the HTML mime type, make sure the
// charset meta is present, add it if not
if ("text/html".equals(sourceMimetype)) {
inputBlob = checkCharsetMeta(inputBlob);
}
// Get original file extension
String ext = inputBlob.getFilename();
int dotPosition = ext.lastIndexOf('.');
if (dotPosition == -1) {
ext = ".bin";
} else {
ext = ext.substring(dotPosition);
}
// Copy in a file to be able to read it several time
sourceFile = Framework.createTempFile("NXJOOoConverterDocumentIn", ext);
InputStream stream = inputBlob.getStream();
FileUtils.copyToFile(stream, sourceFile);
stream.close();
DocumentFormat sourceFormat = null;
if (sourceMimetype != null) {
// Try to fetch it from the registry.
sourceFormat = getSourceFormat(documentConverter, sourceMimetype);
}
// If not found in the registry or not given as a parameter.
// Try to sniff ! What does that smell ? :)
if (sourceFormat == null) {
sourceFormat = getSourceFormat(documentConverter, sourceFile);
}
// From plugin settings because we know the destination
// mimetype.
DocumentFormat destinationFormat =
getDestinationFormat(documentConverter, sourceFormat, pdfa1);
// allow HTML2PDF filtering
List<Blob> blobs = new ArrayList<>();
if (descriptor.getDestinationMimeType().equals("text/html")) {
String tmpDirPath = getTmpDirectory();
File myTmpDir = new File(tmpDirPath + "/JODConv_" + System.currentTimeMillis());
boolean created = myTmpDir.mkdir();
if (!created) {
throw new IOException("Unable to create temp dir");
}
outFile =
new File(
myTmpDir.getAbsolutePath()
+ "/"
+ "NXJOOoConverterDocumentOut."
+ destinationFormat.getExtension());
created = outFile.createNewFile();
if (!created) {
throw new IOException("Unable to create temp file");
}
log.debug("Input File = " + outFile.getAbsolutePath());
// Perform the actual conversion.
documentConverter.convert(sourceFile, outFile, destinationFormat);
files = myTmpDir.listFiles();
for (File file : files) {
// copy the files to a new tmp location, as we'll delete them
Blob blob;
try (FileInputStream in = new FileInputStream(file)) {
blob = Blobs.createBlob(in);
}
blob.setFilename(file.getName());
blobs.add(blob);
// add a blob for the index
if (file.getName().equals(outFile.getName())) {
Blob indexBlob;
try (FileInputStream in = new FileInputStream(file)) {
indexBlob = Blobs.createBlob(in);
}
indexBlob.setFilename("index.html");
blobs.add(0, indexBlob);
}
}
} else {
outFile =
Framework.createTempFile(
"NXJOOoConverterDocumentOut", '.' + destinationFormat.getExtension());
// Perform the actual conversion.
documentConverter.convert(sourceFile, outFile, destinationFormat, parameters);
Blob blob;
try (FileInputStream in = new FileInputStream(outFile)) {
blob = Blobs.createBlob(in, getDestinationMimeType());
}
blobs.add(blob);
}
return new SimpleCachableBlobHolder(blobs);
} catch (IOException e) {
String msg =
String.format(
"An error occurred trying to convert file %s to from %s to %s",
blobPath, sourceMimetype, getDestinationMimeType());
throw new ConversionException(msg, e);
} finally {
if (sourceFile != null) {
sourceFile.delete();
}
if (outFile != null) {
outFile.delete();
}
if (files != null) {
for (File file : files) {
if (file.exists()) {
file.delete();
}
}
}
}
}
/**
* Returns the DocumentFormat for the given mimetype.
*
* @return DocumentFormat for the given mimetype
*/
private static DocumentFormat getSourceFormat(
OfficeDocumentConverter documentConverter, String mimetype) {
return documentConverter.getFormatRegistry().getFormatByMediaType(mimetype);
}