# Episodes_ListEpisodes `GET /v1/accounts/{account}/datasets/{dataset}/episodes` Lists the episodes of a dataset in recording order (for a mixture: its pool, source by source). `parent` may be `accounts/a/datasets/-` to read across every dataset of the account (ExploreLake's successor). - Base URL: `https://junction-api.qualiastudios.dev` - Authentication: an API key in the `X-API-Key` header ## Parameters - `account` (path, string, required): The account id. - `dataset` (path, string, required): The dataset id. - `pageSize` (query, integer (int32)): At most this many episodes; 0 means 100, the maximum is 1000. - `pageToken` (query, string): From a previous response's next_page_token. - `filter` (query, string): AIP-160 filter. e.g. `state = UNREVIEWED AND open_issue_codes:"jerky"`, `note_codes:"graph_outcome"`, `methods:"umi"`, `version = "1.0"`, `metrics.duration_seconds < 20`, `viewed = false`, `linked_task_inherited = false`, `draft = true` (takes with a draft, Episode.draft). - `orderBy` (query, string): "index" (default: recording order, source by source in a mixture), "record_time", "metrics. desc", "open_issue_count desc". - `view` (query, string (enum)): EPISODE_VIEW_ANALYSIS adds each episode's Episode.analysis, its channels' progress only: a channel's curves are GetEpisode's. It signs no posters (Episode.posters is empty): a reader of every take's profile shows few posters, and reads those with GetEpisode. One of: `EPISODE_VIEW_UNSPECIFIED`, `EPISODE_VIEW_BASIC`, `EPISODE_VIEW_ANALYSIS`. - `version` (query, string): Read as this version froze it: a tag, `stable`, `v0` or a digest prefix; empty or `head` reads the episode as it is now. The list holds that version's episodes, each with the fields, trim and subtasks of the revision it pinned (task and tags as its snapshot froze them, also for an episode never edited); standing, decision, issues and markers are current. A version the dataset does not have is NOT_FOUND. ## Responses - `200` OK: `application/json` `qualia.v1.ListEpisodesResponse` - `default` Default error response: `application/json` `google.rpc.Status` ## Schemas ```yaml qualia.v1.ListEpisodesResponse: type: object properties: episodes: type: array items: $ref: "#/components/schemas/qualia.v1.Episode" description: One page of episodes. nextPageToken: type: string description: Empty on the last page. totalSize: type: integer description: Episodes matching the filter, across every page. format: int32 description: Response for Episodes.ListEpisodes. qualia.v1.Episode: type: object properties: name: type: string description: accounts/{account}/datasets/{dataset}/episodes/{episode} upload: readOnly: true type: string description: |- accounts/{account}/datasets/{dataset}: the upload it was recorded in. The `upload` filter compares it with a dataset name, `=` and `!=`; a merged dataset's name matches the episodes of every upload it holds, as one comparison. index: readOnly: true type: integer description: Its position in the upload, from 0. format: int32 position: readOnly: true type: integer description: |- Its place in the dataset it is read through, from 0: what it is numbered by there. In a merged dataset, across its uploads, in the order they were merged, each upload's episodes in their own order, so no two share a number; everywhere else it is `index`. Optional for older servers that do not yet report it; clients can distinguish an absent position from the valid value zero. format: int32 parentEpisode: readOnly: true type: string description: The episode it was split from, when it is a child. childCount: readOnly: true type: integer description: Children it was split into. format: int32 parentWindow: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Trim" description: |- Where a split child sits in its parent, in seconds from the start of the parent's recording: what `parent_episode` traces it back to. It is the child's base window, the stretch the split gave it; a later trim of the child narrows the child, not this window. Unset for an episode that was not split from another. parentIndex: readOnly: true type: integer description: The `index` of the episode it was split from, when it is a split child. format: int32 state: readOnly: true enum: - STATE_UNSPECIFIED - UNREVIEWED - KEPT - REJECTED type: string description: Where it stands in this scope. format: enum decision: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Decision" description: The standing ruling in this scope; unset when nobody (and not validation) ruled. openIssueCount: readOnly: true type: integer description: Open blocking issues on it, in this scope. format: int32 noteCount: readOnly: true type: integer description: Notes (non-blocking) on it, in this scope. format: int32 openIssueCodes: readOnly: true type: array items: type: string description: Codes of the open issues, for rail chips without a second call. noteCodes: readOnly: true type: array items: type: string description: |- Codes of the open notes (issues that do not block a keep) that carry one: a check's findings, such as a graph check's, and ingest's non-blocking ones. A person's note has no code and is not listed. viewed: readOnly: true type: boolean description: |- Whether the caller has opened it. Per person: always false for an API key, which acts as itself, not as the person who made it. frameCount: readOnly: true type: string description: Frames in the recording. versionFrameCount: readOnly: true type: string description: |- Frames the read version holds: the trim window it froze, the rows its read plan and a download carry. Set when the read is of a version (the request's `version`, or the tag a branch reads); at head, frame_count and trim describe the episode. fps: readOnly: true type: number description: Frames per second of the data rows. format: double recordTime: readOnly: true type: string description: When it was recorded. format: date-time cameras: readOnly: true type: array items: type: string description: Camera keys it carries. posters: readOnly: true type: array items: $ref: "#/components/schemas/qualia.v1.Episode_Poster" description: First and last frame per camera. task: type: string description: Task label (editable). instruction: type: string description: Language instruction (editable). operator: readOnly: true type: string description: Operator label. rig: readOnly: true type: string description: Recording rig key, as the recording carries it (see Rig.key). sourceRobot: readOnly: true type: string description: "accounts/{account}/robots/{robot}: the fleet robot it was recorded on." method: type: string description: Legacy single-method projection retained for older clients. methods: type: array items: type: string description: |- Canonical collection-method tags (teleop, vr, umi, egocentric, policy-rollout, sim). Multiple tags may describe one episode. scene: type: string description: Scene label (editable). session: readOnly: true type: string description: Recorder session id. tags: type: array items: type: string description: Free tags (editable). outcome: enum: - OUTCOME_UNSPECIFIED - SUCCESS - PARTIAL - FAILURE type: string description: How the attempt went (editable; the operator's own mark by default). format: enum outcomeActor: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Actor" description: Who set the outcome. markers: readOnly: true type: array items: $ref: "#/components/schemas/qualia.v1.Episode_Marker" description: Recorder markers. trim: allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Trim" description: The trim window; unset means untrimmed (editable). subtasks: type: array items: $ref: "#/components/schemas/qualia.v1.Episode_Subtask" description: Subtask spans (editable). metrics: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Metrics" description: Health metrics and judgements. originalMetrics: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Metrics" description: |- The take as recorded, before curation: its measurements (speed, jerk, smoothness, pauses, gripper events, corrections, alignment) over its whole recording, measured once; a split's piece has its take's. duration_seconds is the recording's length. Unset until measured; filter as `original.`. curatedMetrics: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Metrics" description: |- The episode as curated: the same measurements over the window it trains on (after its trim; a piece's own window), measured when the trim or split is applied, or else within minutes of the publish that made it. duration_seconds is after the trim. Unset until measured, and for a revision no upload holds now; filter as `curated.`. revision: readOnly: true type: integer description: Content revision of the episode edit chain (trim, subtasks, tags, ...). format: int32 etag: type: string description: Sent back on UpdateEpisodes; equals the revision. createTime: readOnly: true type: string description: When it was ingested. format: date-time updateTime: readOnly: true type: string description: When its content or standing last changed. format: date-time upstreamState: readOnly: true enum: - STATE_UNSPECIFIED - UNREVIEWED - KEPT - REJECTED type: string description: |- Read through a mixture: where it stands upstream (the least settled of its upload and any mixture in between). Unset in its upload's scope. format: enum analysis: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Analysis" description: How the take moves; set only in the EPISODE_VIEW_ANALYSIS view. originalAnalysis: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Analysis" description: |- How the take moves as recorded, before curation: the same numbers over its whole recording, measured once, with no channels; a split's piece has its take's. Where `analysis` follows a trim or a split, this stays, so the outliers tab sets the episodes as curated against the takes as recorded. Set only in the EPISODE_VIEW_ANALYSIS view, once measured. extentSeconds: readOnly: true type: number description: |- Length of the recording that edits are checked against, in seconds: a subtask or trim may not end after it. It is the stored millisecond length, so it can be shorter than frame_count / fps. Unset on the response an edit replays. format: double linkedTask: type: string description: |- accounts/{account}/tasks/{task}: the Task the take is recorded against, or empty when unlinked. A link, not content: `task` stays the label and language instruction training reads, and linking changes no revision, digest or version. Names a task of the episode's account (NOT_FOUND otherwise). Deleting the task unlinks it. Set with AssignTask, or UpdateEpisodes with the path on its own in one edit. A take with no link of its own reads the Dataset.linked_task of the dataset that holds it (linked_task_inherited): its upload's, or once a merge took the upload, the merged dataset's own, wherever the take is read from. Its own link overrides that, and clearing it (UpdateEpisodes with "") makes the take read the dataset's again. Filters, facets and the VLM check read this value. linkedTaskInherited: readOnly: true type: boolean description: |- Whether linked_task is the dataset's that holds the take (its upload's, or its merged dataset's): the take has no link of its own. False when nothing links it, or when it links a task itself. linkedTaskRevision: readOnly: true type: integer description: |- The revision of linked_task the link carries, the one checks judge the take against: the dataset's when inherited. 0 when unlinked, or when the Task had no published revision when it was linked (its draft is judged). Moved only by RelinkTask on the dataset. format: int32 lease: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Lease" description: |- Who is editing the take now; unset when nobody holds a live lease. ListEpisodes and GetEpisode fill it. draft: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_DraftSummary" description: |- The take's draft (EpisodeDraft), its edit not yet in a version; unset when it has none. One per take, so it reads the same from every scope. ListEpisodes and GetEpisode fill it. cut: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Episode_Cut" description: |- Set on an episode a version published from a draft that cut a take: the take and where in it the episode lies. Unset on a take. description: |- One take. An episode is recorded in exactly one upload, and can be read through any dataset that contains it: the parent in its name is the scope it is seen from. `accounts/a/datasets/{upload}/episodes/{e}` is the take as its upload stands on it; `accounts/a/datasets/{mixture}/episodes/{e}` is the same take as that mixture stands on it. The {episode} id is the same uuid in both. Rulings (Keep, Reject, ...) made through a mixture's name belong to that mixture. Upstream wins: seen from a mixture, an episode its upload (or a mixture in between) does not keep stands as it stands there, read-only; only an upstream KEPT episode is the mixture's to rule on. A merged dataset (Datasets.MergeDatasets) holds its uploads' episodes where they were recorded: `accounts/a/datasets/{merged}/episodes/{e}` is the take as its upload stands on it, the same state, decision, issues, notes, tags and trim, and `upload` still names the upload it came from. A merge is final: the take is the merged dataset's, and everything written to it (a ruling, an issue, UpdateEpisodes, AssignTask, splits, suggestions) is sent through the merged name and lands in the upload that holds it. The upload's own name for it still reads; a write by that name is FAILED_PRECONDITION naming the merged dataset. The account-wide list (`datasets/-`) names such a take under its merged dataset. Served by qds-api. qualia.v1.Episode_Trim: type: object properties: startSeconds: type: number description: First second kept. format: double endSeconds: type: number description: Last second kept. format: double description: The trim window, in seconds from the start of the recording. qualia.v1.Episode_Decision: type: object properties: verdict: readOnly: true enum: - VERDICT_UNSPECIFIED - KEEP - REJECT type: string description: Keep or reject. format: enum actor: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Actor" description: "Who ruled: a person, or VALIDATION for an ingest reject." decideTime: readOnly: true type: string description: When. format: date-time reason: readOnly: true type: string description: Why, in the person's words; for validation, the invalid reasons. dataset: readOnly: true type: string description: |- accounts/{account}/datasets/{dataset}: the dataset the ruling was made in. Equals the upload for an upload ruling; a mixture for its own rejects. locked: readOnly: true type: boolean description: |- True when seen from a mixture and the deciding ruling was made upstream: only that dataset can take it back. description: The standing ruling in this scope, when there is one. qualia.v1.Actor: type: object properties: kind: readOnly: true enum: - KIND_UNSPECIFIED - PERSON - VALIDATION - RULE - MODEL - OPERATOR - SYSTEM type: string description: What kind of actor it is. format: enum user: readOnly: true type: string description: The person's user id when `kind` is PERSON; empty otherwise. displayName: readOnly: true type: string description: The person's display name at the time, for rendering without a lookup. agent: readOnly: true type: string description: |- The rule code, or for a MODEL the Qualia model tier (e.g. "qualia-vlm"; "qualia" for any other model), when not a person. run: readOnly: true type: string description: The run a model or rule finding came from, when it has one. apiKey: readOnly: true type: string description: |- The API key's id when a person acted through one; empty for a session. Never the key itself. attribution: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Attribution" description: |- Signed identity evidence recorded with the change, kept apart from the principal above. Unset on older records and wherever none was signed. description: Who did something. Every decision, issue and reply names one. qualia.v1.Attribution: type: object properties: authentication: readOnly: true enum: - AUTHENTICATION_UNSPECIFIED - SESSION - API_KEY - SERVICE type: string description: How the principal authenticated. format: enum initiator: readOnly: true type: string description: |- The user id of a person verified as starting the work: the person of a session, or the person a trusted service's signed job record names. Empty when nobody was verified. initiatorDisplayName: readOnly: true type: string description: The initiator's display name at the time, never their email. credentialCreator: readOnly: true type: string description: The user id of the key's creator at the time. Not who used the key. credentialOwner: readOnly: true enum: - CREDENTIAL_OWNER_UNSPECIFIED - PERSON - SPACE type: string description: Who owned the key at the time. format: enum credentialOwnerSpace: readOnly: true type: string description: |- The owning space's id (a workspace or the organization) when `credential_owner` is SPACE. description: |- How a change was authenticated and who verifiably started it. Only the trusted mesh signs it. It never authorizes anything, and a key's creator or owner is never the person who used the key. qualia.v1.Episode_Poster: type: object properties: camera: readOnly: true type: string description: Camera key. firstFrameUri: readOnly: true type: string description: |- Signed URI of the midpoint frame of the window ingest recorded. An edit and a split piece keep their take's posters, so after a trim or a split this frame may lie outside the episode's own frames. The field name is retained for compatibility; ingest publishes a midpoint, not a first frame. lastFrameUri: readOnly: true type: string description: |- Signed URI of the last frame of the window ingest recorded; like the midpoint, it is the take's after a trim or a split. description: Midpoint and last-frame posters of one camera. qualia.v1.Episode_Marker: type: object properties: timeSeconds: readOnly: true type: number description: Seconds from the start of the recording. format: double label: readOnly: true type: string description: Its label. role: readOnly: true type: string description: Its role (start, end, success, event). description: A recorder marker, e.g. the operator's success button. qualia.v1.Episode_Subtask: type: object properties: startSeconds: type: number description: Start, in seconds from the start of the recording. format: double endSeconds: type: number description: End, in seconds. format: double label: type: string description: The subtask's one name (also one of the episode's tags). review: enum: - REVIEW_UNSPECIFIED - PROPOSED - ACCEPTED - REJECTED type: string description: Proposed, accepted or rejected. format: enum origin: enum: - ORIGIN_UNSPECIFIED - HUMAN - MODEL - HEURISTIC type: string description: Who proposed it. format: enum description: One subtask span. qualia.v1.Episode_Metrics: type: object properties: durationSeconds: type: number description: Length, seconds (after trim). format: double speedVariance: type: number description: Standard deviation of per-frame speed. format: double speedMean: type: number description: Mean joint speed. format: double sparc: type: number description: Spectral arc length (smoothness; closer to 0 is smoother). format: double ldlj: type: number description: Log dimensionless jerk. format: double psdHfRatio: type: number description: Share of power above the high-frequency cut. format: double idleRatio: type: number description: Share of time the arm is idle. format: double startPoseZ: type: number description: Start pose distance from its robot's median start, in MADs. format: double endPoseZ: type: number description: End pose distance from its robot's median end, in MADs. format: double alignmentLagMs: type: number description: Lag between action and state, milliseconds. format: double correctionCount: type: integer description: Corrections near gripper events. format: int32 gripperEventCount: type: integer description: Gripper open or close events. format: int32 cameraLagMs: type: number description: |- Camera motion against joint speed offset, milliseconds (camera lag): the camera_lags entry the panel opens on (a recorded one, else the trusted estimate with the strongest peak). format: double jerkMean: type: number description: Mean jerk of the joints (the curation fence's jerk_mean). format: double gripperFlickerCount: type: integer description: Gripper open-close flickers. format: int32 jerky: type: boolean description: psd_hf_ratio or tremor above p90 of its robot's population. hasCorrections: type: boolean description: At least one correction. qualityTier: readOnly: true enum: - QUALITY_TIER_UNSPECIFIED - A - B - REJECT type: string description: Composite grade. format: enum cameraLags: readOnly: true type: array items: $ref: "#/components/schemas/qualia.v1.Episode_CameraLag" description: Every camera's offset, by camera key; empty until measured. description: |- Per-episode health metrics and judgements. Unset fields are not computed yet. Each is a filter field and a facet. qualia.v1.Episode_CameraLag: type: object properties: camera: readOnly: true type: string description: Camera key. lagMs: readOnly: true type: number description: |- Milliseconds; positive when the video runs behind the robot's motion, negative when it runs ahead. format: double confidence: readOnly: true type: number description: For an estimate, the correlation peak (-1 to 1); 1 when recorded. format: double method: readOnly: true type: string description: How it was measured, e.g. "motion_xcorr/1" or "sync_event". source: readOnly: true enum: - SOURCE_UNSPECIFIED - ESTIMATED - RECORDED type: string description: Estimated or recorded. format: enum trusted: readOnly: true type: boolean description: |- An estimate whose peak is strong and whose two halves of the episode agree; always true when recorded. Only trusted offsets are judged. description: "One camera's time offset against the robot's motion: what the camera lag panel shows per camera." qualia.v1.Episode_Analysis: type: object properties: version: readOnly: true type: string description: |- The definition the numbers follow; a reader treats another one as not measured yet. unmeasuredReason: readOnly: true type: string description: |- Why the take could not be measured: "no_joints" (no state) or "channel_never_read". Empty when measured. fps: readOnly: true type: number description: Rows per second it was measured at. format: double frames: readOnly: true type: integer description: Rows read. format: int32 speedVariance: readOnly: true type: number description: Variance of the joint speed over every state channel, (units/s)^2. format: double speedMean: readOnly: true type: number description: Mean joint speed over every state channel, units/s. format: double jerkMean: readOnly: true type: number description: Mean norm of the state's third difference, units/s^3. format: double highFrequencyRatio: readOnly: true type: number description: Share of the speed spectrum's power above 3 Hz. format: double idleRatio: readOnly: true type: number description: Share of frames under 5% of the top speed. format: double tremble: readOnly: true type: number description: Mean |v_smooth - v| / (|v_smooth| + |v|) over the arm joints. format: double gripperEventCount: readOnly: true type: integer description: Commanded gripper moves of at least 30% of their travel. format: int32 gripperFlickerCount: readOnly: true type: integer description: The most events of one gripper inside 0.5 s. format: int32 correctionCount: readOnly: true type: integer description: Regrasps within 3 s, plus events the arm retraces its approach round. format: int32 stateChannels: readOnly: true type: array items: type: string description: State channel names, in the order of the poses. startPose: readOnly: true type: array items: type: number format: double description: First state row, every channel. endPose: readOnly: true type: array items: type: number format: double description: Last state row, every channel. channels: readOnly: true type: array items: $ref: "#/components/schemas/qualia.v1.Analysis_Channel" description: Every commanded channel, in recording order. description: |- How the take moves, as the dataset page's outliers and training tabs read it: measured once from its series, within a few minutes of a publish, and again only when its data changes (a re-ingest, or a trim or a split, measured as it is applied); a ruling or a tag measures nothing. The set's fences, embedding and chart averages are laid out over these by the reader. Read in the EPISODE_VIEW_ANALYSIS view of ListEpisodes (each channel's progress only) and of GetEpisode (with the curves); unset when the take is not measured yet, or its profile measured another revision of it than the one the listed scope reads (before a trim, a re-ingest or a split; or through a mixture pinned to an earlier revision, which is not measured: only uploads' current members are). The definitions are dataset-ux-proto's (scripts/extract.py, src/lib/mock): they differ on purpose from the like-named Metrics, which are command-space and feed curation. qualia.v1.Analysis_Channel: type: object properties: name: readOnly: true type: string description: Channel name, as the recording names it. progress: readOnly: true type: array items: type: number format: double description: The value at 50 evenly spaced points of the take, first to last. autocorrelation: readOnly: true type: array items: type: number format: double description: |- Autocorrelation at lags 1..min(100, frames / 2); empty when the channel never moves, and in a list. alignment: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Channel_Alignment" description: Unset when the state has no channel of this name, and in a list. description: One commanded channel's training-chart inputs. qualia.v1.Channel_Alignment: type: object properties: correlation: readOnly: true type: array items: type: number format: double description: |- Correlation at lags -30..30 frames; NaN where a lag has no variance. Empty when the take is under 5 frames. description: Cross-correlation of the channel's step against its measured twin's. qualia.v1.Episode_Lease: type: object properties: holder: readOnly: true type: string description: The holder's user id (as Actor.user). holderName: readOnly: true type: string description: |- The holder's display name at the time they took it; empty when they have none. expireTime: readOnly: true type: string description: When it lapses unless renewed. format: date-time description: |- Someone editing the take now: the lease AcquireEpisodeLease takes. One per take, so it reads the same from every scope. qualia.v1.Episode_DraftSummary: type: object properties: updateActor: readOnly: true allOf: - $ref: "#/components/schemas/qualia.v1.Actor" description: "Who last changed the drawing: a person, or an API key acting for none." updateTime: readOnly: true type: string description: When the drawing was last saved. format: date-time state: readOnly: true enum: - STATE_UNSPECIFIED - OPEN type: string description: "Always OPEN: a draft is never applied." format: enum baseRevision: readOnly: true type: integer description: The revision it was drawn on (Episode.revision then). format: int32 pieceCount: readOnly: true type: integer description: |- The pieces the drawing keeps: 1 for a trim, 2 or more for a split (each becomes an episode of the version published from it), 0 when it keeps nothing. format: int32 keptSeconds: readOnly: true type: number description: The seconds its kept pieces last, together. format: double description: |- The take's draft, its edit not yet in a version, as its row shows it: who last drew it, when, and what it keeps. GetEpisodeDraft reads the drawing. qualia.v1.Episode_Cut: type: object properties: take: readOnly: true type: string description: |- accounts/{account}/datasets/{dataset}/episodes/{episode}: the take it was cut from. takeIndex: readOnly: true type: integer description: The take's index in its upload (Episode.index of the take). format: int32 piece: readOnly: true type: integer description: Its place among the take's pieces, from 1. format: int32 pieceCount: readOnly: true type: integer description: How many pieces the take was cut into. format: int32 startSeconds: readOnly: true type: number description: Where it starts, in seconds of the take's recording. format: double endSeconds: readOnly: true type: number description: Where it ends, in seconds of the take's recording. format: double description: |- Where a cut piece came from: an episode a version published from a draft that split or trimmed a take. It is the take's piece `piece` of `piece_count`, kept from `start_seconds` to `end_seconds` of the take's recording, and reads as the dashboard numbers it ("12.1": the take's index, then the piece's, from 1). google.rpc.Status: oneOf: - $ref: "#/components/schemas/ApiError" - $ref: "#/components/schemas/GatewayError" description: "The body of a failed call: one of the two shapes below." ApiError: title: API error type: object required: - code - message - retryable - details properties: code: type: string description: The gRPC code name in upper case, e.g. `NOT_FOUND`. message: type: string retryable: type: boolean description: Whether the same call may succeed if retried. details: type: array items: type: object description: "Empty, except when a dataset copy or move is refused: then one `google.rpc.ErrorInfo` with the refusal's `reason` code, its `domain` and, in `metadata`, the datasets it names." description: A refusal from the API itself. GatewayError: title: Gateway error type: object required: - detail properties: detail: type: string description: What went wrong. description: A refusal from the gateway before the call reaches the API, for example `401` bad or missing API key, `403` key lacks the scope or environment closed, `413` body too large. ```