|
| 1 | +/** |
| 2 | + * Licensed to the Apache Software Foundation (ASF) under one |
| 3 | + * or more contributor license agreements. See the NOTICE file |
| 4 | + * distributed with this work for additional information |
| 5 | + * regarding copyright ownership. The ASF licenses this file |
| 6 | + * to you under the Apache License, Version 2.0 (the |
| 7 | + * "License"); you may not use this file except in compliance |
| 8 | + * with the License. You may obtain a copy of the License at |
| 9 | + * <p> |
| 10 | + * http://www.apache.org/licenses/LICENSE-2.0 |
| 11 | + * <p> |
| 12 | + * Unless required by applicable law or agreed to in writing, software |
| 13 | + * distributed under the License is distributed on an "AS IS" BASIS, |
| 14 | + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
| 15 | + * See the License for the specific language governing permissions and |
| 16 | + * limitations under the License. |
| 17 | + */ |
| 18 | + |
| 19 | +package com.qlangtech.tis.plugin.datax.aliyunoss; |
| 20 | + |
| 21 | +import com.aliyun.oss.OSS; |
| 22 | +import com.aliyun.oss.model.GenericResult; |
| 23 | +import com.aliyun.oss.model.ListObjectsRequest; |
| 24 | +import com.aliyun.oss.model.OSSObject; |
| 25 | +import com.aliyun.oss.model.OSSObjectSummary; |
| 26 | +import com.aliyun.oss.model.ObjectListing; |
| 27 | +import com.aliyun.oss.model.PutObjectRequest; |
| 28 | +import com.google.common.collect.Lists; |
| 29 | +import com.google.common.collect.Sets; |
| 30 | +import com.qlangtech.tis.fs.IPath; |
| 31 | +import com.qlangtech.tis.plugin.tdfs.ITDFSSession; |
| 32 | +import org.apache.commons.collections.CollectionUtils; |
| 33 | +import org.apache.commons.lang.StringUtils; |
| 34 | +import org.slf4j.Logger; |
| 35 | +import org.slf4j.LoggerFactory; |
| 36 | + |
| 37 | +import java.io.File; |
| 38 | +import java.io.FilterOutputStream; |
| 39 | +import java.io.IOException; |
| 40 | +import java.io.InputStream; |
| 41 | +import java.io.OutputStream; |
| 42 | +import java.io.PipedInputStream; |
| 43 | +import java.io.PipedOutputStream; |
| 44 | +import java.util.Collections; |
| 45 | +import java.util.HashSet; |
| 46 | +import java.util.List; |
| 47 | +import java.util.Set; |
| 48 | +import java.util.concurrent.CountDownLatch; |
| 49 | +import java.util.concurrent.ExecutorService; |
| 50 | +import java.util.concurrent.Executors; |
| 51 | +import java.util.concurrent.TimeUnit; |
| 52 | + |
| 53 | +/** |
| 54 | + * @author: 百岁（baisui@qlangtech.com） |
| 55 | + * @create: 2023-08-04 13:43 |
| 56 | + **/ |
| 57 | +public class OSSSession implements ITDFSSession { |
| 58 | + private static final Logger logger = LoggerFactory.getLogger(OSSSession.class); |
| 59 | + private final AliyunOSSTDFSLinker ossLinker; |
| 60 | + private OSS oss; |
| 61 | + private static final ExecutorService ossPutExecutor = Executors.newCachedThreadPool(); |
| 62 | + |
| 63 | + public OSSSession(AliyunOSSTDFSLinker ossLinker) { |
| 64 | + this.ossLinker = ossLinker; |
| 65 | + } |
| 66 | + |
| 67 | + private OSS getOSS() { |
| 68 | + if (oss == null) { |
| 69 | + oss = ossLinker.createOSSClient(); |
| 70 | + } |
| 71 | + return oss; |
| 72 | + } |
| 73 | + |
| 74 | + @Override |
| 75 | + public String getRootPath() { |
| 76 | + return this.ossLinker.getRootPath(); |
| 77 | + } |
| 78 | + |
| 79 | + @Override |
| 80 | + public boolean isDirExist(String dir) { |
| 81 | + return true; |
| 82 | + } |
| 83 | + |
| 84 | + @Override |
| 85 | + public void mkDirRecursive(String directoryPath) { |
| 86 | + |
| 87 | + } |
| 88 | + |
| 89 | + @Override |
| 90 | + public Set<String> getAllFilesInDir(String path, String prefixFileName) { |
| 91 | + ObjectListing objectList = this.getOSS().listObjects(ossLinker.bucket, path + prefixFileName); |
| 92 | + List<OSSObjectSummary> objectSummaries = objectList.getObjectSummaries(); |
| 93 | + Set<String> result = Sets.newHashSet(); |
| 94 | + for (OSSObjectSummary obj : objectSummaries) { |
| 95 | + result.add(obj.getKey()); |
| 96 | + } |
| 97 | + return result; |
| 98 | + } |
| 99 | + |
| 100 | + @Override |
| 101 | + public void deleteFiles(Set<String> filesToDelete) { |
| 102 | + if (CollectionUtils.isEmpty(filesToDelete)) { |
| 103 | + throw new IllegalArgumentException("param filesToDelete can not be empty"); |
| 104 | + } |
| 105 | + for (String df : filesToDelete) { |
| 106 | + this.getOSS().deleteObject(ossLinker.bucket, df); |
| 107 | + } |
| 108 | + } |
| 109 | + |
| 110 | + @Override |
| 111 | + public HashSet<Res> getAllFiles(List<String> srcPaths, int parentLevel, int maxTraversalLevel) { |
| 112 | + HashSet<Res> allFiles = Sets.newHashSet(); |
| 113 | + for (String path : srcPaths) { |
| 114 | + getListFiles(allFiles, path, path, Collections.emptyList(), parentLevel, maxTraversalLevel); |
| 115 | + } |
| 116 | + return allFiles; |
| 117 | + } |
| 118 | + |
| 119 | + @Override |
| 120 | + public HashSet<Res> getListFiles(String directoryPath, int parentLevel, int maxTraversalLevel) { |
| 121 | + HashSet<Res> allFiles = Sets.newHashSet(); |
| 122 | + getListFiles(allFiles, directoryPath, directoryPath, Collections.emptyList(), parentLevel, maxTraversalLevel); |
| 123 | + return allFiles; |
| 124 | + } |
| 125 | + |
| 126 | + private void getListFiles(HashSet<Res> listFiles, final String rootDir, String directoryPath, List<String> relevantPath, final int parentLevel, final int maxTraversalLevel) { |
| 127 | + if (StringUtils.startsWith(directoryPath, File.separator)) { |
| 128 | + throw new IllegalArgumentException("path:" + directoryPath + " can not start with '" + File.separator + "'"); |
| 129 | + } |
| 130 | + if (parentLevel >= maxTraversalLevel) { |
| 131 | + //超出最大递归层数 |
| 132 | + String message = String.format("获取path：[%s] 下文件列表时超出最大层数,请确认路径[%s]下不存在软连接文件", directoryPath, directoryPath); |
| 133 | + throw new IllegalStateException(message); |
| 134 | + } else { |
| 135 | + |
| 136 | + ListObjectsRequest listReq = new ListObjectsRequest(); |
| 137 | + listReq.setDelimiter("/"); |
| 138 | + listReq.setBucketName(ossLinker.bucket); |
| 139 | + // IPath.pathConcat() 处理很重要，因为OSS取子文件一定要以 '/'结尾 ，才能取到子文件列表 |
| 140 | + if (StringUtils.isNotEmpty(directoryPath)) { |
| 141 | + listReq.setPrefix(IPath.pathConcat(directoryPath, StringUtils.EMPTY)); |
| 142 | + } else { |
| 143 | + listReq.setPrefix(StringUtils.EMPTY); |
| 144 | + } |
| 145 | + |
| 146 | + |
| 147 | + ObjectListing children = this.getOSS().listObjects(listReq); |
| 148 | + // 文件列表 |
| 149 | + String fileKey; |
| 150 | + for (OSSObjectSummary obj : children.getObjectSummaries()) { |
| 151 | + fileKey = obj.getKey(); |
| 152 | + if (StringUtils.endsWith(fileKey, File.separator)) { |
| 153 | + continue; |
| 154 | + } |
| 155 | + listFiles.add(new Res(fileKey, Res.buildRelevantPath(relevantPath, StringUtils.substringAfter(fileKey, directoryPath)))); |
| 156 | + } |
| 157 | + if (parentLevel + 1 >= maxTraversalLevel) { |
| 158 | + return; |
| 159 | + } |
| 160 | + |
| 161 | + // 文件夹列表 |
| 162 | + for (String commonPrefix : children.getCommonPrefixes()) { |
| 163 | + getListFiles(listFiles, rootDir, commonPrefix, Lists.newArrayList(StringUtils.split(StringUtils.removeStart(commonPrefix, rootDir), File.separatorChar)), parentLevel + 1, maxTraversalLevel); |
| 164 | + } |
| 165 | + |
| 166 | + // getListFiles(listFiles, null, parentLevel + 1, maxTraversalLevel); |
| 167 | + } |
| 168 | + } |
| 169 | + |
| 170 | + @Override |
| 171 | + public OutputStream getOutputStream(String filePath, boolean append) { |
| 172 | + PipedOutputStream output = new PipedOutputStream(); |
| 173 | + |
| 174 | + CountDownLatch countDown = new CountDownLatch(1); |
| 175 | + |
| 176 | + final CountDownLatch over = new CountDownLatch(1); |
| 177 | + |
| 178 | + ossPutExecutor.execute(() -> { |
| 179 | + try { |
| 180 | + OSS oss = this.getOSS(); |
| 181 | + |
| 182 | + try (PipedInputStream input = new PipedInputStream(output)) { |
| 183 | + countDown.countDown(); |
| 184 | + |
| 185 | + GenericResult result = null; |
| 186 | + if (append) { |
| 187 | + throw new UnsupportedOperationException("not support with append mode"); |
| 188 | + // result = oss.appendObject(new AppendObjectRequest(ossLinker.bucket, filePath, input)); |
| 189 | + } else { |
| 190 | + result = oss.putObject(new PutObjectRequest(ossLinker.bucket, filePath, input)); |
| 191 | + } |
| 192 | + |
| 193 | + } |
| 194 | + over.countDown(); |
| 195 | + } catch (Exception e) { |
| 196 | + logger.error("wirte path faild,path:" + filePath, e); |
| 197 | + Thread.currentThread().interrupt(); |
| 198 | + } |
| 199 | + });// execute over |
| 200 | + try { |
| 201 | + countDown.await(20, TimeUnit.SECONDS); |
| 202 | + } catch (InterruptedException e) { |
| 203 | + throw new RuntimeException(e); |
| 204 | + } |
| 205 | + return new FilterOutputStream(output) { |
| 206 | + @Override |
| 207 | + public void close() throws IOException { |
| 208 | + super.close(); |
| 209 | + try { |
| 210 | + over.await(2, TimeUnit.HOURS); |
| 211 | + } catch (InterruptedException e) { |
| 212 | + throw new IOException(e); |
| 213 | + } |
| 214 | + } |
| 215 | + }; |
| 216 | + |
| 217 | + |
| 218 | + } |
| 219 | + |
| 220 | + @Override |
| 221 | + public InputStream getInputStream(String filePath) { |
| 222 | + OSSObject ossObject = this.getOSS().getObject(ossLinker.bucket, filePath); |
| 223 | + return ossObject.getObjectContent(); |
| 224 | + } |
| 225 | + |
| 226 | + @Override |
| 227 | + public void close() throws Exception { |
| 228 | + getOSS().shutdown(); |
| 229 | + } |
| 230 | +} |
0 commit comments