README
¶
s3
A very simple interface for reading/writing blobs from S3 using the multipart REST API (which allows for >5GB uploads -- a surprisingly uncommon feature).
Check the docs for more details!
Documentation
¶
Overview ¶
s3 provides a very simple interface for reading/writing blobs from S3 using the multipart REST API (which allows for >5GB uploads).
Index ¶
- type S3
- func (s3 *S3) Get(path string) (io.ReadCloser, http.Header, error)
- func (s3 *S3) Head(path string) (http.Header, error)
- func (s3 *S3) Put(r io.Reader, size int64, path string, md5sum []byte, contentType string) error
- func (s3 *S3) StartMultipart(path string) (*S3Multipart, error)
- func (s3 *S3) Test() error
- type S3Error
- type S3Multipart
- type S3NewEndpointError
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type S3 ¶
type S3 struct {
// contains filtered or unexported fields
}
S3 provides a wrapper around your S3 credentials. It carries no other internal state and can be copied freely.
func (*S3) Get ¶
Get fetches content from S3, returning both a ReadCloser for the data and the HTTP headers returned by S3. You can use the headers to extract the Content-Type that the data was sent with.
func (*S3) Head ¶
Head is similar to Get, but returns only the response headers. The response body is not transferred across the network. This is useful for checking if a file exists remotely, and what headers it was configured with.
func (*S3) Put ¶
Put uploads content to S3. The length of r must be passed as size. md5sum optionally contains the MD5 hash of the content for end-to-end integrity checking; if omitted no checking is done. contentType optionally contains the MIME type to send to S3 as the Content-Type header; when files are later accessed, S3 will echo back this in their response headers.
If the passed size exceeds 3GB, the multipart API is used, otherwise the single-request API is used. It should be noted that the multipart API uploads in 7MB segments and computes checksums of each one -- it does NOT use the passed md5sum, so don't bother with it if you're uploading huge files.
func (*S3) StartMultipart ¶
func (s3 *S3) StartMultipart(path string) (*S3Multipart, error)
StartMultipart initiates a multipart upload.
type S3Multipart ¶
type S3Multipart struct {
// contains filtered or unexported fields
}
S3Multipart tracks the state of a multipart upload, and provides an interface for streaming data to S3 in chunks. All methods on S3Multipart are mutually locked to ensure state doesn't become corrupt.
func (*S3Multipart) Abort ¶
func (mp *S3Multipart) Abort() error
Abort cancels the upload. If an upload is started but not completed, the storage space will be counted against your AWS account (and getting rid of it is difficult), so you should make sure either Abort or Complete is called.
For your convenience, Abort is set as the finalizer for S3Multipart objects as a failsafe, but you shouldn't rely on that.
func (*S3Multipart) AddPart ¶
AddPart uploads the contents of r to S3. The number of bytes that r will read must be passed as size (otherwise the request cannot be signed). Optionally, you can pass the md5sum of the bytes which will be verified on S3's end; if md5sum is nil no end-to-end integrity checking is performed. As per S3's API, size must always exceed 5MB (1024 * 1024 * 5) bytes, except for the last part. This is not enforced locally.
func (*S3Multipart) Complete ¶
func (mp *S3Multipart) Complete(contentType string) error
Complete finalizes the upload, and should be called after all parts have been added.
Source Files