Packages

package aws

Ordering
  1. Alphabetic
Visibility
  1. Public
  2. All

Type Members

  1. class AWSAnonymousCredentials extends Credentials
  2. class AWSBasicCredentials extends Credentials
  3. class AWSClient extends CloudClient
  4. class AWSCredentialsProvider extends Credentials
  5. class AWSGateway extends AutoCloseable with CloudStorage
  6. class AWSProfileCredentials extends Credentials
  7. class AWSTokenCredentials extends Credentials
  8. case class CredentialParams(accessKeyId: String, secretAccessKey: String, sessionToken: String, profile: String, region: String) extends Product with Serializable
  9. trait Credentials extends AnyRef

Value Members

  1. object AWSGateway
  2. object DownloadProgressTracker

    Byte counts for the download currently in flight, readable from outside the JVM.

    Byte counts for the download currently in flight, readable from outside the JVM.

    A Spark driver launched through py4j writes to a stdout that a notebook front end does not surface, so DownloadProgress's own output is invisible there however it is formatted. Python can print to the notebook, but has no view of the transfer -- it is blocked in a py4j call for its whole duration. This is the bridge: the JVM keeps the counters, and the Python side polls them and renders.

    Every download on the driver shares this one set of counters. Two downloads at once (a job that builds several pipelines on threads) therefore mix their numbers in the display, which is cosmetic: each download's own data is unaffected. Accessors are plain public methods returning primitives, which is what py4j can reach.

Ungrouped