S3

给我写信
GitHub

AWS

参考链接：

• S3 Java API
• S3 Java Examples

目录

1. 用Java API 访问S3

需要用到的依赖是：com.amazonaws:aws-java-sdk:1.9.6
Scala代码：

package cn.gridx.aws.s3
import java.io._
import java.util
import com.amazonaws.regions.{Regions, Region}
import com.amazonaws.services.s3.AmazonS3Client
import com.amazonaws.services.s3.model._
/**
  * Created by tao on 1/26/16.
  */
object Fundamentals extends App {
  val s3: AmazonS3Client = new AmazonS3Client()
  val region_us_west_1: Region = Region.getRegion(Regions.US_WEST_1)
  s3.setRegion(region_us_west_1)
  /**
    * 关于 access key / private key
    *
    * 由于本机存在文件 ~/.aws/credentials,里面的内容为
    *   [default]
    *   aws_access_key_id = YOUR_ACCESS_KEY_ID
    *   aws_secret_access_key = YOUR_SECRET_ACCESS_KEY
    * 因此,本代码运行时可以直接访问S3
    *
    * 如果没有这个文件,则需要在代码中指定
    * 参考[http://docs.ceph.com/docs/master/radosgw/s3/java/]
    * */
  /**
    * 列出全部的buckets
    * */
  println("\n\n列出所有buckets的名字 ...")
  val buckets: util.List[Bucket] = s3.listBuckets
  for (i <- 0 until buckets.size) {
    val bucket = buckets.get(i)
    println(s"Bucket Name: ${bucket.getName}\t, Creation Time: ${bucket.getCreationDate}")
  }
  /**
    * 向名为`gridx-tmp`的bucket中放入1个object
    * */
  println("\n\n向名为`gridx-tmp`的bucket中写入1个object ...")
  def BucketName = "gridx-tmp"
  def MyObjKey   = "my_obj_key"
  def MyFileName = "my_file_name"
  s3.putObject(new PutObjectRequest(BucketName, MyObjKey, createTmpFile(MyFileName)))
  /**
    * 从名为`gridx-tmp`的bucket中下载刚才放入的object
    * */
  println("从名为`gridx-tmp`的bucket中下载刚才放入的object ...")
  val obj: S3Object = s3.getObject(new GetObjectRequest(BucketName, MyObjKey))
  val meta: ObjectMetadata = obj.getObjectMetadata
  println(s"Content-Type : ${meta.getContentType}, " +
    s" Content-Length : ${meta.getContentLength}, " +
    s"Content-Encoding : ${meta.getContentEncoding}" )
  println("内容为:")
  val in = obj.getObjectContent
  displayTextStream(in)
  in.close
  /**
    * 如果某个bucket中有很多的objects,那么listObjects返回的可能只是部分结果,
    * 这需要用`ObjectListing.isTruncated`来判断
    * 如果确实是只返回了部分结果,那么需要用`listNextBatchOfObjects`来继续查询剩下的数据
    *
    * `listObjects`并不是只返回最上层的目录,而是返回所有的objects(如果返回结果没有被截断)
    * */
  println("\n\n列出名为`gridx-jars`的bucket中的全部objects")
  // 名为`gridx-jars`的bucket所在的Region是 US Standard
  s3.setRegion(Region.getRegion(Regions.US_EAST_1))
  var objList: ObjectListing = s3.listObjects("gridx-jars")
  var loop = false
  do {
    val summeryList: util.List[S3ObjectSummary] = objList.getObjectSummaries
    for (i <- 0 until summeryList.size) {
      val summery = summeryList.get(i)
      println(s"  Key : ${summery.getKey},  Size : ${summery.getSize}")
    }
    if (objList.isTruncated) {
      objList = s3.listNextBatchOfObjects(objList)
      loop = true
    }
  } while (loop)
  /**
    * 删除指定bucket中的某个object
    * */
  println("\n\n删除指定bucket中的某个object ...")
  s3.deleteObject("gridx-tmp", MyObjKey)
  // 创建一个临时文件
  def createTmpFile(path: String): File = {
    val file = File.createTempFile(path, "txt")
    file.deleteOnExit
    val writer = new OutputStreamWriter(new FileOutputStream(file))
    writer.write("美国国务卿克里今访华 制裁朝鲜或为头号议题\n")
    writer.write("副军长曝南沙守礁生活:狗上礁三个月就疯了\n")
    writer.write("2016日本小姐大赛决赛在东京举行，庆应大学文学系的二年级学生松野未佳夺得冠军")
    writer.close
    file
  }
  // 展示输入流中的文本内容
  def displayTextStream(in: InputStream): Unit = {
    val reader = new BufferedReader(new InputStreamReader(in))
    var line = reader.readLine
    while (null != line) {
      println(line)
      line = reader.readLine
    }
  }
}