Import status: running/complete/failed (provider errors verbatim)
Status of an import (import_data): "running" (with bytes_transferred and files_done/files_total when the provider enumerates up front), "complete" ({total_bytes, files, dest_path}), or "failed" with the provider's real error relayed verbatim - an auth failure names the connection so a human can fix it (autor connection get <name> shows what to repair); a missing object is the provider's own 404, not ours. A transfer that breaks mid-object resumes itself with a ranged re-read from the byte that landed rather than restarting the file, so one upstream hiccup no longer costs a multi-hour transfer its progress; bytes_transferred stays monotonic across that. Status checks never wake a stopped node; an import interrupted by node stop or worker loss reports "failed" honestly and is safe to re-run (re-call import_data; completed files are overwritten, never doubled, and a file that failed part-way is kept as <name>.part rather than published truncated).
Request
- Base URL: https://autoresearch.sfcompute.com
- URL: https://autoresearch.sfcompute.com/preview/nodes/{name}/imports/{id}
- Auth: HTTP bearer
Path parameters
Node name (from list_nodes).
Import id from import_data (e.g. "imp-4kd93").
Query parameters
Act in this workspace (its slug). Omitted: the token's bound workspace, then the organization's default. Never ambient - there is no switch-workspace mutation on this door.
Response
The capability's JSON result.
Changes
No recorded changes to this operation across all 1 revision of this API.