Struct Manager

pub struct Manager { /* private fields */ }

Implements AWS S3 manager.

Implementations

impl Manager

pub fn new(shared_config: &AwsSdkConfig) -> Self

pub async fn create_bucket(&self, s3_bucket: &str) -> Result<()>

Creates a S3 bucket.

pub async fn delete_bucket(&self, s3_bucket: &str) -> Result<()>

Deletes a S3 bucket.

pub async fn delete_objects( &self, s3_bucket: Arc<String>, prefix: Option<Arc<String>>, ) -> Result<()>

Deletes objects by “prefix”. If “prefix” is “None”, empties a S3 bucket, deleting all files. ref. https://github.com/awslabs/aws-sdk-rust/blob/main/examples/s3/src/bin/delete-objects.rs

“If a single piece of data must be accessible from more than one task concurrently, then it must be shared using synchronization primitives such as Arc.” ref. https://tokio.rs/tokio/tutorial/spawning

pub async fn list_objects( &self, s3_bucket: Arc<String>, prefix: Option<Arc<String>>, ) -> Result<Vec<Object>>

List objects in the bucket with an optional prefix, in the descending order of “last_modified” timestamps. “bucket_name” implies the suffix “/”, so no need to prefix sub-directory with “/”. Passing “bucket_name” + “directory” is enough!

e.g. “foo-mydatabucket” for bucket_name “mydata/myprefix/” for prefix

pub async fn put_object( &self, file_path: Arc<String>, s3_bucket: Arc<String>, s3_key: Arc<String>, ) -> Result<()>

Writes an object to a S3 bucket using stream.

WARN: use stream! otherwise it can cause OOM – don’t do the following! “fs::read” reads all data onto memory “.body(ByteStream::from(contents))” passes the whole data to an API call

“If a single piece of data must be accessible from more than one task concurrently, then it must be shared using synchronization primitives such as Arc.” ref. https://tokio.rs/tokio/tutorial/spawning

pub async fn get_object( &self, s3_bucket: Arc<String>, s3_key: Arc<String>, file_path: Arc<String>, ) -> Result<()>

Downloads an object from a S3 bucket using stream.

WARN: use stream! otherwise it can cause OOM – don’t do the following! “aws_smithy_http::byte_stream:ByteStream.collect” reads all the data into memory “File.write_all_buf(&mut bytes)” to write bytes

“If a single piece of data must be accessible from more than one task concurrently, then it must be shared using synchronization primitives such as Arc.” ref. https://tokio.rs/tokio/tutorial/spawning

pub async fn compress_seal_put_object( &self, envelope: Arc<Envelope>, file_path: Arc<String>, s3_bucket: Arc<String>, s3_key: Arc<String>, ) -> Result<()>

Compresses the file, encrypts, and uploads to S3.

pub async fn get_object_unseal_decompress( &self, envelope: Arc<Envelope>, s3_bucket: Arc<String>, s3_key: Arc<String>, file_path: Arc<String>, ) -> Result<()>

Reverse of “compress_seal_put_object”.

Trait Implementations

impl Clone for Manager

fn clone(&self) -> Manager

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Manager

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.