Java压缩和解压缩

压缩需求:压缩文件夹下所有具有统一文件名规范的普通文件,同时添加文件名与压缩包相同的MANIFEST文件,包含所有被压缩文件的文件名及大小列表。
Java文件压缩涉及到的类均在java.uti.zip包下:
  • ZipOutputStream: Zip包的输出流,利用putNextEntry()向包内添加一个文件。
  • ZipEntry: 将被压缩到包内的文件实例,在文件名后添加文件分隔符可以压缩目录。

代码如下:
public class FileZipper {

	private static String INPUT_FPATH = "C:/temp/fileZipper/DropZone/Input_Min";
	private static String OUTPUT_ZIP_PATH = "C:/temp/fileZipper/DropZone/Input_Zip";
	private static String OUTPUT_ZIP_TSPATH = "/SRE/apps/ican505/sandbox/jerry/filezipper/Output_TS";
	private static String ARCHIEVE_ZONE = "C:/temp/fileZipper/ArchiveZone";
	private static String INPUT_DELIM = ":";
	private static String ZIP_FPATTERN = "I2001_GLB";
	private static int AVG_FILENAME_LEN = 45; // Default to length of each entry in MANIFEST
	private static int BUF_SIZE = 1024;
	private static int STD_ZIP_ENTRIES = 20;

	public String zipFilesByJava(String strZipFilePattern, String strFilePath, String strOutputFileList,	String strErrorZonePath) {
			System.out.println("***** Inside zipFilesByJava method  *****");
			System.out.println("***** Source Location of files: " + strFilePath + ", Zip File Pattern: " + strZipFilePattern + " *****");
			String strZipFName = "";
			String strEntryFName = "";
			String strManifestFName = "";
			SimpleDateFormat dateFormatter = new SimpleDateFormat("yyyyMMddHHmmssSS");
			StringTokenizer strFileList = new StringTokenizer(strOutputFileList, "|");
			if (strFileList == null || strFileList.countTokens() == 0) {
				System.out.println("[ERROR] Input file list is NULL or EMPTY!");
				return null;
			}
			FileInputStream fin = null;
			ZipOutputStream zos = null;
			ZipEntry zipEntry = null;
			StringBuffer sbZippedEntryList = null;
			if (STD_ZIP_ENTRIES <= 0) {
				System.out.println("[WARN] The property STD_ZIP_ENTRIES can NOT be negative or zero! Reset to 100");
				STD_ZIP_ENTRIES = 100;
			}
			if (BUF_SIZE <= 0) {
				System.out.println("[WARN] The property BUF_SIZE can NOT be negative or zero! Reset to 1024");
				BUF_SIZE = 1024;
			}
			byte[] bInputBuf = new byte[BUF_SIZE];
			int iTotalFiles = strFileList.countTokens();
			int iTotalZips = iTotalFiles / STD_ZIP_ENTRIES == 0 ? 1	: iTotalFiles / STD_ZIP_ENTRIES;
			int iEntryListLen = (iTotalZips == 1) ? AVG_FILENAME_LEN : STD_ZIP_ENTRIES * AVG_FILENAME_LEN;
			int AVG_FILENAME_LEN = 45; // Length of entry in MANIFEST which contains standard filename and file size. // i.e, I1001_GWM_7360_MAN_20100426000000.txt|1111
			StringBuffer sbZipFNameList = new StringBuffer(iTotalZips);
			String strTS = dateFormatter.format(new Date());
			try {
				System.out.println("[INFO] Current Time in Millis: " + System.currentTimeMillis());
				for (int iCntZip = 0; iCntZip < iTotalZips; iCntZip++) {
					if (iTotalFiles == 1) {
						strZipFName = strOutputFileList;
						strZipFName = OUTPUT_ZIP_PATH + File.separatorChar + strZipFName.replaceAll(".txt$", ".zip");
					} else {
						strZipFName = OUTPUT_ZIP_PATH + File.separatorChar + strZipFilePattern + "_" + strTS + ".zip";
					}
					zos = new ZipOutputStream(new CheckedOutputStream(new FileOutputStream(strZipFName), new CRC32()));
					if (zos == null) {
						System.out.println("[ERROR] ZIPOutputStream for output zip file is NULL!");
						return null;
					}
					if (iCntZip != 0 && iCntZip == iTotalZips - 1)
						iEntryListLen = STD_ZIP_ENTRIES * 2 * AVG_FILENAME_LEN;
					sbZippedEntryList = new StringBuffer(iEntryListLen);
					int iCntEntry = 0;
					int iReadSize = 0;
					int iFileSize = 0;
					while ((iCntZip != iTotalZips - 1 && iCntEntry < STD_ZIP_ENTRIES) || (iCntZip == iTotalZips - 1 && iCntEntry < 2 * STD_ZIP_ENTRIES) && strFileList.hasMoreTokens()) {
						strEntryFName = strFileList.nextToken();
						if (!(new File(strFilePath + File.separatorChar + strEntryFName).exists())) {
							System.out.println("[ERROR] The input file " + strFilePath + File.separatorChar + strEntryFName + " does NOT exist!");
							return null;
						}
						zipEntry = new ZipEntry(strEntryFName);
						zos.putNextEntry(zipEntry);
						fin = new FileInputStream(strFilePath + File.separatorChar + strEntryFName);
						while ((iReadSize = fin.read(bInputBuf)) != -1) {
							zos.write(bInputBuf);
							iFileSize += iReadSize;
							bInputBuf = new byte[BUF_SIZE];
						}
						iCntEntry++;
						sbZippedEntryList.append(strEntryFName + "|" + iFileSize + "\n");
						iReadSize = 0;
						iFileSize = 0;
						System.out.println("[INFO] Entry No. " + iCntEntry + ": " + strEntryFName);
						fin.close();
						if(new File(strFilePath + File.separatorChar + strEntryFName).renameTo(new File(ARCHIEVE_ZONE + File.separatorChar + strEntryFName)))
							System.out.println("[INFO] Entry file " + strEntryFName + " is already moved to Archieve Zone!");
					}
					strManifestFName = strZipFilePattern + "_" + strTS + ".MANIFEST";
					zipEntry = new ZipEntry(strManifestFName);
					zos.putNextEntry(zipEntry);
					zos.write(sbZippedEntryList.toString().getBytes());
					zos.close();
					System.out.println("---------------------------\n[TOTAL] Zip No. "
						+ (iCntZip + 1) + ": " + iCntEntry + " files compressed into zip "
						+ strZipFName + "!\n---------------------------\n");
					sbZipFNameList.append(strZipFName);
				}
				System.out.println("[END] Finished reading and zipping!");
				System.out.println("[INFO] Current Time in Millis: " + System.currentTimeMillis());
			} catch(ZipException zipE){
				System.out.println("[ERROR] ZipException thrown!");
				zipE.printStackTrace();
			} catch (IOException ioE) {
				System.out.println("[ERROR] IOException thrown!");
				ioE.printStackTrace();
			}
			return sbZipFNameList.toString();
	}



解压缩需求:解压缩一个zip文件,读取MANIFEST文件,一一验证MANIFEST文件中所列的文件名和大小,只有当所有文件和大小均匹配时解压zip包至特定目录。
Java文件解压缩需要用到的类包括ZipFile(用文件名实例化一个压缩包)和ZipEntry,ZipEntry实例被从ZipFile实例获取。 利用ZipFile的getOutputStream()(zipEntry实例作为参数)可以获取对zipEntry的输入流,由此读入数据并写出到文件系统。
代码如下:
public class FileUnzipper {
	public static Logger logger = Logger.getLogger(FileUnzipper.class.getName());
	public String strErrMsg = ""; 
	void unzip(String strZipFilePath, String strTargetFilePath, String strErrorZone, String strZipPattern, String strClaimsFilePattern,
			String strArchivePath) {
		if (!new File(strZipFilePath).exists() || !new File(strTargetFilePath).exists() || strZipPattern == null || strZipPattern.trim().length() <= 0
				|| strClaimsFilePattern == null || strClaimsFilePattern.trim().length() <= 0){
			strErrMsg = "The mandatory location/filename for unzipping is(are) NOT valid!";
			logger.error("[" + PropSet.GIFInterfaceId + "] " + strErrMsg);
			return;
		}
		File[] fZipArray = new File(strZipFilePath).listFiles(new FileListFilter(strZipPattern, "zip"));
		if (fZipArray == null || fZipArray.length == 0){
			strErrMsg = "No Zip file with pattern " + strZipPattern + " exist in the source path: " + strZipFilePath;
			logger.error("[" + PropSet.GIFInterfaceId + "] " + strErrMsg);
			return;
		}
		ZipFile zipFile = null;
		ZipEntry zipEntry = null;
		Enumeration zipAllEntries = null;
		BufferedReader bufManifestReader = null;
		Map mapFNameToSize = new HashMap();
		InputStream isEntryFile = null;
		OutputStream osEntryFile = null;
		OutputStream osEntryArchive = null;
		String strZipFName = "";
		String strManifestFName = "";
		String strReadLine = "";
		String strEntryFName = "";
		long lEntryFSize = 0L;
		int iEntryCounter = 0;
		byte[] bReadBuf = new byte[PropSet.FILE_WRITE_BUFSIZE];
		boolean bEntryMismatched = false;
		try{
			for (int iZip = 0; iZip < fZipArray.length; iEntryCounter = 0, iZip++){
				strZipFName = fZipArray[iZip].getName();
				strManifestFName = strZipFName.substring(0, strZipFName.indexOf(".zip")) + ".MANIFEST";
				if(!fZipArray[iZip].isFile()){
					logger.warn("The zip file " + strZipFName + " is directory or NOT a normal zip file! Moving to ErrorZone...");
					if(fZipArray[iZip].renameTo(new File(strErrorZone + File.separator + strZipFName)))
						logger.info("Successfully moved to ErrorZone!");
					else logger.error("Failed to move to ErrorZone!");
					continue;
				}
				zipFile = new ZipFile(fZipArray[iZip]);
				
				// process manifest entry
				zipEntry = zipFile.getEntry(strManifestFName);
				if(zipEntry == null){
					strErrMsg = "The zip file " + strZipFName + " does NOT contain MANIFEST file! Moving the zip file to ErrorZone...";
					logger.error("[" + PropSet.GIFInterfaceId + "] " + strErrMsg);
					if(!fZipArray[iZip].renameTo(new File(strErrorZone + File.separator + strZipFName)))
						logger.info("Successfully moved to ErrorZone!");
					else logger.error("Failed to move to ErrorZone!");
					continue;
				}
				bufManifestReader = new BufferedReader(new InputStreamReader(zipFile.getInputStream(zipEntry)));
				while((strReadLine = bufManifestReader.readLine()) != null){
					try{
						mapFNameToSize.put(strReadLine.split("[|]")[0], strReadLine.split("[|]")[1]);
					}catch(Exception exc){
						throw new Exception("Failed to add [filename --> filesize] mapping to hashmap from manifest file!	Line supposed to be mapped: " + strReadLine);
					}
				}
				mapFNameToSize.put(strManifestFName, Long.toString(zipEntry.getSize()));
				bufManifestReader.close();
				
				// process data entries
				zipAllEntries = zipFile.entries();
				while(zipAllEntries.hasMoreElements()){
					bEntryMismatched = false;
					zipEntry = (ZipEntry)zipAllEntries.nextElement();
					strEntryFName = zipEntry.getName().trim();
					lEntryFSize = zipEntry.getSize();
					if(zipEntry.isDirectory()){
						logger.warn("The entry " + strEntryFName + " IS directory, to ignore it.");
						continue;
					}
					if(!(strEntryFName.startsWith(strClaimsFilePattern) || strEntryFName.trim().equals(strManifestFName))){
						bEntryMismatched = true;
						strErrMsg = "Invalid Zip Entry Name " + strEntryFName + ", Decompressing to ErrorZone...";
						logger.error("[" + PropSet.GIFInterfaceId + "] " + strErrMsg);
					}else if(!mapFNameToSize.containsKey(strEntryFName) || !mapFNameToSize.get(strEntryFName).equals(Long.toString(lEntryFSize))){
						bEntryMismatched = true;
						strErrMsg = "Entry-filename: " + strEntryFName + ", Entry-filesize: " + lEntryFSize
							+ " mismatch with the value " + mapFNameToSize.get(strEntryFName) + " in Manifest file! Decompressing to ErrorZone...";
						logger.error("[" + PropSet.GIFInterfaceId + "] " + strErrMsg);
					}
					isEntryFile = zipFile.getInputStream(zipEntry);
					osEntryFile = bEntryMismatched ? new FileOutputStream(strErrorZone + File.separator + strEntryFName) : new FileOutputStream(strTargetFilePath + File.separator + strEntryFName);
					osEntryArchive = new FileOutputStream(strArchivePath + File.separator + strEntryFName);
					while(isEntryFile.read(bReadBuf) != -1){
						osEntryFile.write(bReadBuf);
						osEntryArchive.write(bReadBuf);
					}
					isEntryFile.close();
					osEntryFile.close();
					osEntryArchive.close();
					if(!bEntryMismatched){
						iEntryCounter++;
						logger.info("Matched Entry No. " + iEntryCounter + " | Entry Name: " + strEntryFName +", Entry File Size: " + lEntryFSize);
					}else logger.info("The mismatched Entry file was Decompressed to ErrorZone and Archived!");
				}
				logger.info("Zip No. " + iZip + " | Zip Filename: " + strZipFName + ", Entries contained: " + zipFile.size() + ", Entries decompressed: " + iEntryCounter);
				zipFile.close();
				if(fZipArray[iZip].delete())
					logger.info("Deleted the zip file!");
				else logger.warn("Failed to delete the zip file!");
			}
			logger.info("Finished process the zip file(s) in location " + strZipFilePath);
		}catch(IOException ioE){
			strErrMsg = "IOException happned with msg: " + ioE.getMessage();
			logger.error("[" + PropSet.GIFInterfaceId + "] " + strErrMsg);
			if(fZipArray != null && fZipArray.length > 0){
				for(int iZip = 0; iZip < fZipArray.length; iZip++){
					if(!fZipArray[iZip].renameTo(new File(strErrorZone + File.separator + fZipArray[iZip].getName()))){
						logger.warn("Failed to move the left zip file " + fZipArray[iZip].getName() + " to ErrorZone " + strErrorZone);
					}else logger.info("Moved the zip file " + fZipArray[iZip].getName() + " to ErrorZone " + strErrorZone);
				}
			}
		}catch(Exception exc){
			strErrMsg = "Exception happened with msg: " + exc.getMessage();
			logger.error("[" + PropSet.GIFInterfaceId + "] " + strErrMsg);
		}
	}
    
class FileListFilter implements FilenameFilter {

	private String name;

	private String extension;

	public FileListFilter(String name, String extension) {
		this.name = name;
		this.extension = extension;
	}

	public boolean accept(File directory, String filename) {
		boolean fileOK = true;
		if (name != null) {
			fileOK &= filename.startsWith(name);
		}
		if (extension != null) {
			fileOK &= filename.endsWith('.' + extension);
		}
		return fileOK;
	}

}

class PropSet {
	static int FILE_WRITE_BUFSIZE = 1024;
	static String GIFInterfaceId = "I2001_GLB";
}


由于需求特定,代码实现有所局限。
看到 Snowolf对带有压缩算法的压缩/解压缩的研究,在实际应用中需要面对不同的问题和场景,如文件读写同步/目录压缩及解压缩/中文编码等。故本文章也会适时更新。

你可能感兴趣的:(java,算法,Blog)