init datamate

This commit is contained in:
Dallas98
2025-10-21 23:00:48 +08:00
commit 1c97afed7d
692 changed files with 135442 additions and 0 deletions

View File

@@ -0,0 +1,35 @@
<assembly
xmlns="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/plugins/maven-assembly-plugin/assembly/1.1.0 http://maven.apache.org/xsd/assembly-1.1.0.xsd">
<id></id>
<formats>
<format>dir</format>
</formats>
<includeBaseDirectory>false</includeBaseDirectory>
<fileSets>
<fileSet>
<directory>src/main/resources</directory>
<includes>
<include>plugin.json</include>
<include>plugin_job_template.json</include>
</includes>
<outputDirectory>plugin/writer/nfswriter</outputDirectory>
</fileSet>
<fileSet>
<directory>target/</directory>
<includes>
<include>nfswriter-0.0.1-SNAPSHOT.jar</include>
</includes>
<outputDirectory>plugin/writer/nfswriter</outputDirectory>
</fileSet>
</fileSets>
<dependencySets>
<dependencySet>
<useProjectArtifact>false</useProjectArtifact>
<outputDirectory>plugin/writer/nfswriter/libs</outputDirectory>
<scope>runtime</scope>
</dependencySet>
</dependencySets>
</assembly>

View File

@@ -0,0 +1,121 @@
package com.modelengine.edatamate.plugin.writer.nfswriter;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.nio.file.DirectoryNotEmptyException;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.Paths;
import java.util.List;
/**
* 一个简单的 Linux NAS 挂载工具类
* 仅适用于 Linux,需具备 sudo 权限或 root。
*/
public final class MountUtil {
private static final Logger LOG = LoggerFactory.getLogger(MountUtil.class);
private MountUtil() {
}
/**
* 挂载远程目录
*
* @param remote 远程地址,如 192.168.1.1:/test
* @param mountPoint 本地挂载点,如 /mnt/nas
* @param type 文件系统类型:nfs、cifs ...
* @param options 额外挂载参数,如 ro,vers=3 或 username=xxx,password=xxx
*/
public static void mount(String remote, String mountPoint, String type, String options) {
try {
Path mp = Paths.get(mountPoint);
if (isMounted(mountPoint)) {
throw new IOException("Already mounted: " + mountPoint);
}
Files.createDirectories(mp);
ProcessBuilder pb = new ProcessBuilder();
if (options == null || options.isEmpty()) {
pb.command("mount", "-t", type, remote, mountPoint);
} else {
pb.command("mount", "-t", type, "-o", options, remote, mountPoint);
}
LOG.info(pb.command().toString());
pb.redirectErrorStream(true);
Process p = pb.start();
StringBuilder output = new StringBuilder();
try (BufferedReader reader = new BufferedReader(new InputStreamReader(p.getInputStream()))) {
String line;
while ((line = reader.readLine()) != null) {
output.append(line).append(System.lineSeparator());
}
}
int rc = p.waitFor();
if (rc != 0) {
throw new RuntimeException("Mount failed, exit=" + rc + ", output: " + output);
}
} catch (IOException | InterruptedException e) {
throw new RuntimeException(e);
}
}
/**
* 卸载挂载点
*
* @param mountPoint 挂载点路径
* @throws IOException 卸载失败
* @throws InterruptedException 进程等待中断
*/
public static void umount(String mountPoint) throws IOException, InterruptedException {
if (!isMounted(mountPoint)) {
return;
}
ProcessBuilder pb = new ProcessBuilder("umount", "-l", mountPoint);
pb.redirectErrorStream(true);
Process p = pb.start();
StringBuilder output = new StringBuilder();
try (BufferedReader reader = new BufferedReader(new InputStreamReader(p.getInputStream()))) {
String line;
while ((line = reader.readLine()) != null) {
output.append(line).append(System.lineSeparator());
}
}
int rc = p.waitFor();
if (rc != 0) {
throw new RuntimeException("Mount failed, exit=" + rc + ", output: " + output);
}
// 清理空目录
try {
Files.deleteIfExists(Paths.get(mountPoint));
} catch (DirectoryNotEmptyException ignore) {
// 目录非空,保留
}
}
/**
* 判断挂载点是否已挂载
*
* @param mountPoint 挂载点路径
* @return true 表示已挂载
* @throws IOException 读取 /proc/mounts 失败
*/
public static boolean isMounted(String mountPoint) throws IOException {
Path procMounts = Paths.get("/proc/mounts");
if (!Files.exists(procMounts)) {
throw new IOException("/proc/mounts not found");
}
String expected = mountPoint.trim();
List<String> lines = Files.readAllLines(procMounts);
return lines.stream()
.map(l -> l.split("\\s+"))
.filter(a -> a.length >= 2)
.anyMatch(a -> a[1].equals(expected));
}
}

View File

@@ -0,0 +1,100 @@
package com.modelengine.edatamate.plugin.writer.nfswriter;
import com.alibaba.datax.common.element.Record;
import com.alibaba.datax.common.exception.CommonErrorCode;
import com.alibaba.datax.common.exception.DataXException;
import com.alibaba.datax.common.plugin.RecordReceiver;
import com.alibaba.datax.common.spi.Writer;
import com.alibaba.datax.common.util.Configuration;
import org.apache.commons.lang3.StringUtils;
import java.io.File;
import java.io.IOException;
import java.util.Arrays;
import java.util.Collections;
import java.util.List;
import java.util.UUID;
public class NfsWriter extends Writer {
public static class Job extends Writer.Job {
private Configuration jobConfig;
private String mountPoint;
@Override
public void init() {
this.jobConfig = super.getPluginJobConf();
}
@Override
public void prepare() {
this.mountPoint = "/dataset/mount/" + UUID.randomUUID();
this.jobConfig.set("mountPoint", this.mountPoint);
new File(this.mountPoint).mkdirs();
MountUtil.mount(this.jobConfig.getString("ip") + ":" + this.jobConfig.getString("path"),
mountPoint, "nfs", StringUtils.EMPTY);
String destPath = this.jobConfig.getString("destPath");
new File(destPath).mkdirs();
}
@Override
public List<Configuration> split(int mandatoryNumber) {
return Collections.singletonList(this.jobConfig);
}
@Override
public void post() {
try {
MountUtil.umount(this.mountPoint);
new File(this.mountPoint).deleteOnExit();
} catch (IOException | InterruptedException e) {
throw new RuntimeException(e);
}
}
@Override
public void destroy() {
}
}
public static class Task extends Writer.Task {
private Configuration jobConfig;
private String mountPoint;
private String destPath;
private List<String> files;
@Override
public void init() {
this.jobConfig = super.getPluginJobConf();
this.destPath = this.jobConfig.getString("destPath");
this.mountPoint = this.jobConfig.getString("mountPoint");
this.files = this.jobConfig.getList("files", Collections.emptyList(), String.class);
}
@Override
public void startWrite(RecordReceiver lineReceiver) {
try {
Record record;
while ((record = lineReceiver.getFromReader()) != null) {
String fileName = record.getColumn(0).asString();
if (StringUtils.isBlank(fileName)) {
continue;
}
if (!files.isEmpty() && !files.contains(fileName)) {
continue;
}
String filePath = this.mountPoint + "/" + fileName;
ShellUtil.runCommand("rsync", Arrays.asList("--no-links", "--chmod=750", "--", filePath,
this.destPath + "/" + fileName));
}
} catch (Exception e) {
throw DataXException.asDataXException(CommonErrorCode.RUNTIME_ERROR, e);
}
}
@Override
public void destroy() {
}
}
}

View File

@@ -0,0 +1,43 @@
package com.modelengine.edatamate.plugin.writer.nfswriter;
import java.io.BufferedReader;
import java.io.InputStreamReader;
import java.util.ArrayList;
import java.util.List;
public class ShellUtil {
/**
* 执行 rsync 命令
*
* @param cmd 命令
* @param extraArgs 额外参数,可为空
* @return 命令完整输出(stdout + stderr)
* @throws Exception 如果 rsync 返回非 0 或发生 IO 异常
*/
public static String runCommand(String cmd, List<String> extraArgs) throws Exception {
List<String> commands = new ArrayList<>();
commands.add(cmd);
if (extraArgs != null && !extraArgs.isEmpty()) {
commands.addAll(extraArgs);
}
ProcessBuilder pb = new ProcessBuilder(commands);
pb.redirectErrorStream(true); // 合并 stdout & stderr
Process p = pb.start();
StringBuilder sb = new StringBuilder();
try (BufferedReader br = new BufferedReader(
new InputStreamReader(p.getInputStream()))) {
String line;
while ((line = br.readLine()) != null) {
sb.append(line).append(System.lineSeparator());
}
}
int exit = p.waitFor();
if (exit != 0) {
throw new RuntimeException("rsync exited with code " + exit + System.lineSeparator() + sb);
}
return sb.toString();
}
}

View File

@@ -0,0 +1,6 @@
{
"name": "nfswriter",
"class": "com.modelengine.edatamate.plugin.writer.nfswriter.NfsWriter",
"description": "write to local",
"developer": "modelengine"
}

View File

@@ -0,0 +1,8 @@
{
"name": "nfswriter",
"parameter": {
"ip": "127.0.0.1",
"path": "/test",
"destPath": ""
}
}