You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
upload() does a single PUT .../content, which the API caps at 100 MiB. Larger files fail with a 404 and the CLI prints "Skipped. Uploads larger than 100MiB are not yet supported."
The classic API supports resumable uploads:
POST /drive/root:/{path}:/createUploadSession
PUT {uploadUrl} Content-Range: bytes 0-10485759/{total}
Chunks must be a multiple of 320 KiB. The session response carries nextExpectedRanges, so an interrupted upload can be resumed rather than restarted.
Notes:
upload() currently gzips the whole body in memory (call() does Zlib.gzipSync) and verifies the SHA1 of the buffer afterwards. A chunked path needs to stream from disk instead, and can verify the SHA1 from the final item response.
upload()does a singlePUT .../content, which the API caps at 100 MiB. Larger files fail with a 404 and the CLI prints "Skipped. Uploads larger than 100MiB are not yet supported."The classic API supports resumable uploads:
Chunks must be a multiple of 320 KiB. The session response carries
nextExpectedRanges, so an interrupted upload can be resumed rather than restarted.Notes:
upload()currently gzips the whole body in memory (call()doesZlib.gzipSync) and verifies the SHA1 of the buffer afterwards. A chunked path needs to stream from disk instead, and can verify the SHA1 from the final item response.@name.conflictBehavioris passed on the session-creation POST, not per chunk. See overwrite existing files #22.README TODO: "Uploads larger than 100MiB are not yet supported (needs range API)".