cn.hutool.extra.compress.extractor.StreamExtractor Maven / Gradle / Ivy
The newest version!
package cn.hutool.extra.compress.extractor;
import cn.hutool.core.io.FileUtil;
import cn.hutool.core.io.IORuntimeException;
import cn.hutool.core.io.IoUtil;
import cn.hutool.core.lang.Assert;
import cn.hutool.core.lang.Filter;
import cn.hutool.core.util.StrUtil;
import cn.hutool.extra.compress.CompressException;
import org.apache.commons.compress.archivers.ArchiveEntry;
import org.apache.commons.compress.archivers.ArchiveException;
import org.apache.commons.compress.archivers.ArchiveInputStream;
import org.apache.commons.compress.archivers.ArchiveStreamFactory;
import org.apache.commons.compress.archivers.tar.TarArchiveInputStream;
import org.apache.commons.compress.compressors.gzip.GzipCompressorInputStream;
import java.io.File;
import java.io.IOException;
import java.io.InputStream;
import java.nio.charset.Charset;
/**
* 数据解压器,即将归档打包的数据释放
*
* @author looly
* @since 5.5.0
*/
public class StreamExtractor implements Extractor {
private final ArchiveInputStream in;
/**
* 构造
*
* @param charset 编码
* @param file 包文件
*/
public StreamExtractor(Charset charset, File file) {
this(charset, null, file);
}
/**
* 构造
*
* @param charset 编码
* @param archiverName 归档包格式,null表示自动检测
* @param file 包文件
*/
public StreamExtractor(Charset charset, String archiverName, File file) {
this(charset, archiverName, FileUtil.getInputStream(file));
}
/**
* 构造
*
* @param charset 编码
* @param in 包流
*/
public StreamExtractor(Charset charset, InputStream in) {
this(charset, null, in);
}
/**
* 构造
* 如果抛出异常,则提供的流将被关闭
*
* @param charset 编码
* @param archiverName 归档包格式,null表示自动检测
* @param in 包流
*/
public StreamExtractor(Charset charset, String archiverName, InputStream in) {
// issue#2736 自定义ArchiveInputStream
if (in instanceof ArchiveInputStream) {
this.in = (ArchiveInputStream) in;
return;
}
final ArchiveStreamFactory factory = new ArchiveStreamFactory(charset.name());
try {
in = IoUtil.toBuffered(in);
if (StrUtil.isBlank(archiverName)) {
this.in = factory.createArchiveInputStream(in);
} else if ("tgz".equalsIgnoreCase(archiverName) || "tar.gz".equalsIgnoreCase(archiverName)) {
//issue#I5J33E,支持tgz格式解压
try {
this.in = new TarArchiveInputStream(new GzipCompressorInputStream(in));
} catch (IOException e) {
throw new IORuntimeException(e);
}
} else {
this.in = factory.createArchiveInputStream(archiverName, in);
}
} catch (ArchiveException e) {
// issue#2384,如果报错可能持有文件句柄,导致无法删除文件
IoUtil.close(in);
throw new CompressException(e);
}
}
/**
* 释放(解压)到指定目录,结束后自动关闭流,此方法只能调用一次
*
* @param targetDir 目标目录
* @param filter 解压文件过滤器,用于指定需要释放的文件,null表示不过滤。当{@link Filter#accept(Object)}为true时释放。
*/
@Override
public void extract(File targetDir, int stripComponents, Filter filter) {
try {
extractInternal(targetDir, stripComponents, filter);
} catch (IOException e) {
throw new IORuntimeException(e);
} finally {
close();
}
}
/**
* 释放(解压)到指定目录
*
* @param targetDir 目标目录
* @param stripComponents 清除(剥离)压缩包里面的 n 级文件夹名
* @param filter 解压文件过滤器,用于指定需要释放的文件,null表示不过滤。当{@link Filter#accept(Object)}为true时释放。
* @throws IOException IO异常
*/
private void extractInternal(File targetDir, int stripComponents, Filter filter) throws IOException {
Assert.isTrue(null != targetDir && ((false == targetDir.exists()) || targetDir.isDirectory()), "target must be dir.");
final ArchiveInputStream in = this.in;
ArchiveEntry entry;
File outItemFile;
while (null != (entry = in.getNextEntry())) {
if (null != filter && false == filter.accept(entry)) {
continue;
}
if (false == in.canReadEntryData(entry)) {
// 无法读取的文件直接跳过
continue;
}
String entryName = this.stripName(entry.getName(), stripComponents);
if (entryName == null) {
// 剥离文件夹层级
continue;
}
outItemFile = FileUtil.file(targetDir, entryName);
if (entry.isDirectory()) {
// 创建对应目录
//noinspection ResultOfMethodCallIgnored
outItemFile.mkdirs();
} else {
FileUtil.writeFromStream(in, outItemFile, false);
}
}
}
@Override
public void close() {
IoUtil.close(this.in);
}
}