Skip to content

Latest commit

 

History

History
34 lines (26 loc) · 2.58 KB

ChunkedUploadSessionDto.md

File metadata and controls

34 lines (26 loc) · 2.58 KB

ChunkedUploadSessionDto

The reserved chunked upload: where the parts are sent, how much was declared and when the reservation lapses. No content of the file is described here.

Properties

Name Type Description Notes
id str The identifier of the reserved upload, repeated in the path of every call that follows it - the chunk uploads, the finalize and the abort. It is thirty-two hexadecimal characters without separators, and it is the only thing the server checks, so anyone holding it can write into this upload. [optional]
path List[int] The chain of folders leading to the destination, outermost first and the destination itself last, with folders the caller cannot read left out. An answer that reports a stored part carries the destination folder alone instead of the whole chain. [optional]
created datetime The moment the upload was reserved, in UTC. [optional]
expired datetime The moment the reservation lapses and the parts buffered for it are dropped, in UTC. It is a gap rather than a deadline for the whole transfer: every accepted part pushes it twelve hours past that part, so only a long silence loses the upload. [optional]
location str The absolute address of the separate chunk handler that also accepts the parts of this upload, kept for clients written against it. A caller working through this API does not need it and sends the parts to the session operations instead. [optional]
bytes_total int The size in bytes that was declared when the upload was reserved, echoed back. It is what the arriving parts are counted against to decide the file is complete, not the amount received so far. [optional]

Example

from docspace_api_sdk.models.chunked_upload_session_dto import ChunkedUploadSessionDto

# TODO update the JSON string below
json = "{}"
# create an instance of ChunkedUploadSessionDto from a JSON string
chunked_upload_session_dto_instance = ChunkedUploadSessionDto.from_json(json)
# print the JSON string representation of the object
print(ChunkedUploadSessionDto.to_json())

# convert the object into a dict
chunked_upload_session_dto_dict = chunked_upload_session_dto_instance.to_dict()
# create an instance of ChunkedUploadSessionDto from a dict
chunked_upload_session_dto_from_dict = ChunkedUploadSessionDto.from_dict(chunked_upload_session_dto_dict)

[Back to Model list] [Back to API list] [Back to README]