import { Journal } from "./journal.cjs";
import { Logger } from "./logger.cjs";
import { MetricsRegistry } from "./metrics.cjs";
import * as http$1 from "node:http";
import * as net from "node:net";

//#region src/types.d.ts
interface Mountable {
  handleRequest(req: http$1.IncomingMessage, res: http$1.ServerResponse, pathname: string): Promise<boolean>;
  handleUpgrade?(socket: net.Socket, head: Buffer, pathname: string): Promise<boolean>;
  health?(): {
    status: string;
    [key: string]: unknown;
  };
  setJournal?(journal: Journal): void;
  setBaseUrl?(url: string): void;
  setRegistry?(registry: MetricsRegistry): void;
}
interface ContentPart {
  type: string;
  text?: string;
  [key: string]: unknown;
}
interface ChatMessage {
  role: "system" | "user" | "assistant" | "tool";
  content: string | ContentPart[] | null;
  name?: string;
  tool_calls?: ToolCallMessage[];
  tool_call_id?: string;
}
interface ToolCallMessage {
  id: string;
  type: "function";
  function: {
    name: string;
    arguments: string;
  };
}
interface ChatCompletionRequest {
  model: string;
  messages: ChatMessage[];
  stream?: boolean;
  stream_options?: {
    include_usage?: boolean;
    [key: string]: unknown;
  };
  temperature?: number;
  max_tokens?: number;
  tools?: ToolDefinition[];
  tool_choice?: string | object;
  response_format?: {
    type: string;
    [key: string]: unknown;
  };
  /** Embedding input text, set by the embeddings handler for fixture matching. */
  embeddingInput?: string;
  /** Endpoint type, set by handlers for fixture endpoint filtering. */
  _endpointType?: string;
  /** Context identifier, set by handlers for fixture context routing. */
  _context?: string;
  [key: string]: unknown;
}
interface ToolDefinition {
  type: "function";
  function: {
    name: string;
    description?: string;
    parameters?: object;
  };
}
interface FixtureMatch {
  userMessage?: string | RegExp;
  /**
   * Substring, regexp, or array of substrings matched against the concatenated
   * text content of every `system` role message in the request. Gates fixture
   * activation on values the host plumbs in via system messages (agent
   * context, persona, dynamic config) instead of the user-typed prompt — so
   * changing context state in the calling app causes stale fixtures to fall
   * through to a real upstream instead of silently returning a baked response
   * that no longer reflects reality.
   *
   * When given an array of strings, ALL substrings must be present (AND
   * semantics). Useful when the gate must combine multiple non-adjacent
   * tokens — e.g., a default name AND a default activity list whose JSON
   * positions vary across requests.
   */
  systemMessage?: string | string[] | RegExp;
  inputText?: string | RegExp;
  toolCallId?: string;
  toolName?: string;
  model?: string | RegExp;
  responseFormat?: string;
  predicate?: (req: ChatCompletionRequest) => boolean;
  /** Which occurrence of this match to respond to (0-indexed). Undefined means match any. */
  sequenceIndex?: number;
  turnIndex?: number;
  hasToolResult?: boolean;
  endpoint?: "chat" | "image" | "speech" | "transcription" | "translation" | "video" | "embedding" | "audio-gen" | "elevenlabs-tts" | "fal-audio" | "fal" | "realtime" | "realtime-transcription" | "realtime-translation";
  context?: string;
}
/**
 * Fields that override auto-generated envelope values in the built response.
 * Scalar fields (finishReason, role) use OpenAI-canonical values — provider
 * handlers translate automatically. For usage, provide field names native to
 * your target provider (OpenAI Chat: prompt_tokens, completion_tokens;
 * Responses API: input_tokens, output_tokens; Anthropic: input_tokens,
 * output_tokens; Gemini: promptTokenCount, candidatesTokenCount).
 *
 * When total_tokens (or provider equivalent) is omitted, it is auto-computed
 * from the component fields.
 *
 * Provider support: OpenAI Chat (all 7), Responses API (5: no role,
 * systemFingerprint), Claude (5: no created, systemFingerprint),
 * Gemini (2: only finishReason, usage).
 */
interface ResponseOverrides {
  id?: string;
  created?: number;
  model?: string;
  usage?: {
    prompt_tokens?: number;
    completion_tokens?: number;
    total_tokens?: number;
    input_tokens?: number;
    output_tokens?: number;
    promptTokenCount?: number;
    candidatesTokenCount?: number;
    totalTokenCount?: number;
  };
  systemFingerprint?: string;
  finishReason?: string;
  role?: string;
}
interface TextResponse extends ResponseOverrides {
  content: string;
  reasoning?: string;
  /**
   * The real cryptographic `signature` captured from a recorded Anthropic
   * thinking turn. When present it is emitted on the replayed thinking block;
   * otherwise replay falls back to aimock's round-trip-safe placeholder.
   * Persisted only alongside a non-empty `reasoning`: a signature captured
   * without plaintext reasoning (e.g. whitespace-only thinking) is intentionally
   * discarded at the persistence layer, since replay attaches signatures only to
   * plaintext thinking blocks.
   */
  reasoningSignature?: string;
  /**
   * Opaque `data` payload(s) of any Anthropic `redacted_thinking` blocks
   * captured from a recorded thinking turn, in stream order. When present they
   * are replayed as faithful `redacted_thinking` content blocks so the encrypted
   * reasoning round-trips; absent for non-Anthropic providers and turns without
   * redacted thinking. Recorded redacted blocks replay as a leading group: the
   * original interleaving of `thinking` and `redacted_thinking` blocks is not
   * preserved (see the CHANGELOG fidelity caveat).
   */
  redactedThinking?: string[];
  webSearches?: string[];
}
interface ToolCall {
  name: string;
  arguments: string;
  id?: string;
}
interface ToolCallResponse extends ResponseOverrides {
  toolCalls: ToolCall[];
  reasoning?: string;
  /** Real Anthropic thinking-block signature; see {@link TextResponse.reasoningSignature}. */
  reasoningSignature?: string;
  /** Anthropic redacted_thinking block data; see {@link TextResponse.redactedThinking}. */
  redactedThinking?: string[];
  webSearches?: string[];
}
interface ContentWithToolCallsResponse extends ResponseOverrides {
  content: string;
  toolCalls: ToolCall[];
  reasoning?: string;
  /** Real Anthropic thinking-block signature; see {@link TextResponse.reasoningSignature}. */
  reasoningSignature?: string;
  /** Anthropic redacted_thinking block data; see {@link TextResponse.redactedThinking}. */
  redactedThinking?: string[];
  webSearches?: string[];
}
interface ErrorResponse {
  error: {
    message: string;
    type?: string;
    param?: string | null;
    code?: string;
  };
  status?: number;
  /** Override the Retry-After header value on 429 responses. Default: 1. */
  retryAfter?: number;
}
interface EmbeddingResponse {
  embedding: number[];
}
interface ImageItem {
  url?: string;
  b64Json?: string;
  revisedPrompt?: string;
}
interface ImageResponse {
  image?: ImageItem;
  images?: ImageItem[];
}
interface AudioResponse {
  audio: string | {
    b64Json: string;
    contentType?: string;
  };
  format?: string;
  /**
   * Companion modalities that can accompany streamed audio. A single Gemini turn
   * may interleave inlineData audio with a functionCall and/or text/thought
   * parts; the recorder preserves them here so the tool call / content / reasoning
   * are not silently discarded when audio is also present.
   */
  toolCalls?: ToolCall[];
  content?: string;
  reasoning?: string;
}
interface TranscriptionResponse {
  transcription: {
    text: string;
    language?: string;
    duration?: number;
    words?: Array<{
      word: string;
      start: number;
      end: number;
    }>;
    segments?: Array<{
      id: number;
      text: string;
      start: number;
      end: number;
    }>;
  };
}
interface VideoResponse {
  video: {
    id: string;
    status: "processing" | "completed" | "failed";
    url?: string;
    /** Failure message surfaced by async video jobs (e.g. OpenRouter `error`). */
    error?: string;
    /** Base64-encoded video bytes served by content-download endpoints. */
    b64?: string;
    /** Generation cost surfaced in usage envelopes (e.g. OpenRouter `usage.cost`). */
    cost?: number;
  };
}
/**
 * Pass-through JSON response. Used by handlers (e.g. fal.ai) that record
 * arbitrary upstream JSON payloads and replay them verbatim, without the
 * provider-specific shape coercion the other response types impose.
 */
interface RawJSONResponse extends ResponseOverrides {
  json: unknown;
  status?: number;
  /**
   * Billed quantity surfaced on the completed fal `queue-result` response via
   * the `x-fal-billable-units` header (emitted alongside `x-fal-request-id` on
   * the completed result). Real fal sets this header on its queue responses;
   * recent versions of `@tanstack/ai-fal` read it to populate a consumer-side
   * billed-units field (e.g. `usage.unitsBilled`). Omit it to preserve the
   * header-less default — present only to let a fixture opt into exercising a
   * consumer's cost/billing accounting path on replay.
   */
  billableUnits?: number;
}
type FixtureResponse = TextResponse | ToolCallResponse | ContentWithToolCallsResponse | ErrorResponse | EmbeddingResponse | ImageResponse | AudioResponse | TranscriptionResponse | VideoResponse | RawJSONResponse;
interface StreamingProfile {
  ttft?: number;
  tps?: number;
  jitter?: number;
}
/**
 * Per-frame arrival timestamps captured during proxy recording.
 * Used during replay to reproduce real-world streaming timing instead of
 * the synthetic model (StreamingProfile / flat latency).
 */
interface RecordedTimings {
  ttftMs: number;
  interChunkDelaysMs: number[];
  totalDurationMs: number;
}
/**
 * Probabilistic chaos injection rates.
 *
 * Rates are evaluated sequentially per request — drop → malformed → disconnect
 * — and the first hit wins. Consequently malformedRate is conditional on drop
 * not firing, and disconnectRate is conditional on neither drop nor malformed
 * firing. A config of `{ dropRate: 0.5, malformedRate: 0.5 }` yields a ~25 %
 * effective malformed rate, not 50 %.
 */
interface ChaosConfig {
  dropRate?: number;
  malformedRate?: number;
  disconnectRate?: number;
}
type ChaosAction = "drop" | "malformed" | "disconnect";
type ResponseFactory = (req: ChatCompletionRequest) => FixtureResponse | Promise<FixtureResponse>;
interface Fixture {
  match: FixtureMatch;
  response: FixtureResponse | ResponseFactory;
  latency?: number;
  chunkSize?: number;
  truncateAfterChunks?: number;
  disconnectAfterMs?: number;
  streamingProfile?: StreamingProfile;
  recordedTimings?: RecordedTimings;
  replaySpeed?: number;
  chaos?: ChaosConfig;
  metadata?: {
    systemHash?: string;
    toolsHash?: string;
  };
}
type FixtureOpts = Omit<Fixture, "match" | "response">;
type EmbeddingFixtureOpts = Pick<FixtureOpts, "latency" | "chaos">;
/**
 * Options for the fal queue/run fixture builders. Adds `billableUnits`, which
 * is emitted as the `x-fal-billable-units` header on the completed
 * `queue-result` response (see {@link RawJSONResponse.billableUnits}).
 */
type FalQueueOpts = FixtureOpts & {
  billableUnits?: number;
};
interface FixtureFileToolCall {
  name: string;
  /** Accepts a JSON object or array for convenience — the loader will JSON.stringify it. */
  arguments: string | Record<string, unknown> | unknown[];
  id?: string;
}
interface FixtureFileToolCallResponse extends ResponseOverrides {
  toolCalls: FixtureFileToolCall[];
  reasoning?: string;
  /** Real Anthropic thinking-block signature; see {@link TextResponse.reasoningSignature}. */
  reasoningSignature?: string;
  /** Anthropic redacted_thinking block data; see {@link TextResponse.redactedThinking}. */
  redactedThinking?: string[];
  webSearches?: string[];
}
interface FixtureFileTextResponse extends ResponseOverrides {
  /** Accepts a JSON object or array (structured output) — the loader will JSON.stringify it. */
  content: string | Record<string, unknown> | unknown[];
  reasoning?: string;
  /** Real Anthropic thinking-block signature; see {@link TextResponse.reasoningSignature}. */
  reasoningSignature?: string;
  /** Anthropic redacted_thinking block data; see {@link TextResponse.redactedThinking}. */
  redactedThinking?: string[];
  webSearches?: string[];
}
interface FixtureFileContentWithToolCallsResponse extends ResponseOverrides {
  /** Accepts a JSON object or array (structured output) — the loader will JSON.stringify it. */
  content: string | Record<string, unknown> | unknown[];
  toolCalls: FixtureFileToolCall[];
  reasoning?: string;
  /** Real Anthropic thinking-block signature; see {@link TextResponse.reasoningSignature}. */
  reasoningSignature?: string;
  /** Anthropic redacted_thinking block data; see {@link TextResponse.redactedThinking}. */
  redactedThinking?: string[];
  webSearches?: string[];
}
type FixtureFileResponse = FixtureFileTextResponse | FixtureFileToolCallResponse | FixtureFileContentWithToolCallsResponse | ErrorResponse | EmbeddingResponse | ImageResponse | AudioResponse | TranscriptionResponse | VideoResponse | RawJSONResponse;
interface FixtureFile {
  fixtures: FixtureFileEntry[];
}
interface FixtureFileEntry {
  match: {
    userMessage?: string;
    /**
     * String (single substring) or array of strings (all must be present).
     * Mirrors the runtime FixtureMatch.systemMessage but without the RegExp
     * form, which JSON cannot express.
     */
    systemMessage?: string | string[];
    inputText?: string;
    toolCallId?: string;
    toolName?: string;
    model?: string;
    responseFormat?: string;
    sequenceIndex?: number;
    turnIndex?: number;
    hasToolResult?: boolean;
    endpoint?: "chat" | "image" | "speech" | "transcription" | "translation" | "video" | "embedding" | "audio-gen" | "elevenlabs-tts" | "fal-audio" | "fal" | "realtime" | "realtime-transcription" | "realtime-translation";
    context?: string;
  };
  response: FixtureFileResponse;
  latency?: number;
  chunkSize?: number;
  truncateAfterChunks?: number;
  disconnectAfterMs?: number;
  streamingProfile?: StreamingProfile;
  recordedTimings?: RecordedTimings;
  replaySpeed?: number;
  chaos?: ChaosConfig;
  metadata?: {
    systemHash?: string;
    toolsHash?: string;
  };
}
interface JournalEntry {
  id: string;
  timestamp: number;
  method: string;
  path: string;
  headers: Record<string, string>;
  body: ChatCompletionRequest | null;
  service?: string;
  response: {
    status: number;
    fixture: Fixture | null;
    /**
     * What was going to serve this request. "fixture" = a fixture matched (or
     * would have, before chaos intervened). "proxy" = no fixture matched and
     * proxy was configured. "internal" = entries where the request was served
     * by aimock's own synthetic logic — neither a matched fixture nor a
     * configured proxy: chaos-path entries (e.g. chaos on the OpenRouter
     * video lifecycle endpoints; in REPLAY mode their normal 200/400/401/404
     * entries omit source, while record-mode 200s on those endpoints carry
     * source:"proxy") AND the OpenRouter video models listing synthesized as
     * the fallback after a FAILED proxy attempt. Absent when the distinction
     * doesn't apply (e.g. 404/503 fallback where nothing was going to serve).
     */
    source?: "fixture" | "proxy" | "internal";
    interrupted?: boolean;
    interruptReason?: string;
    chaosAction?: ChaosAction;
    /** When the X-AIMock-Strict header overrode the server default. */
    strictOverride?: boolean;
  };
}
interface SSEChunk {
  id: string;
  object: "chat.completion.chunk";
  created: number;
  model: string;
  choices: SSEChoice[];
  system_fingerprint?: string;
  usage?: {
    prompt_tokens: number;
    completion_tokens: number;
    total_tokens: number;
  };
}
interface SSEChoice {
  index: number;
  delta: SSEDelta;
  logprobs: null;
  finish_reason: string | null;
}
interface SSEDelta {
  role?: string;
  content?: string | null;
  reasoning_content?: string;
  tool_calls?: SSEToolCallDelta[];
}
interface SSEToolCallDelta {
  index: number;
  id?: string;
  type?: "function";
  function?: {
    name?: string;
    arguments?: string;
  };
}
interface ChatCompletion {
  id: string;
  object: "chat.completion";
  created: number;
  model: string;
  choices: ChatCompletionChoice[];
  usage: {
    prompt_tokens: number;
    completion_tokens: number;
    total_tokens: number;
  };
  system_fingerprint?: string;
}
interface ChatCompletionChoice {
  index: number;
  message: ChatCompletionMessage;
  logprobs: null;
  finish_reason: string;
}
interface ChatCompletionMessage {
  role: string;
  content: string | null;
  refusal: string | null;
  reasoning_content?: string;
  tool_calls?: ToolCallMessage[];
}
type RecordProviderKey = "openai" | "anthropic" | "gemini" | "gemini-interactions" | "vertexai" | "bedrock" | "azure" | "ollama" | "cohere" | "elevenlabs" | "fal" | "openrouter";
interface RecordConfig {
  providers: Partial<Record<RecordProviderKey, string>>;
  fixturePath?: string;
  /** Proxy unmatched requests without saving fixtures or caching in memory. */
  proxyOnly?: boolean;
  /**
   * When true, record the exact model version string returned by the provider
   * (e.g. "gpt-4o-2024-08-06") instead of stripping the date suffix to a
   * canonical alias (e.g. "gpt-4o"). Default: false.
   */
  recordFullModelVersion?: boolean;
  /**
   * fal-specific recording knobs for the queue-walk recorder. During recording
   * the queue handler POSTs submit, polls `status_url` until COMPLETED, then
   * GETs `response_url` for the final job body — saved as the fixture. Tune
   * the poll cadence and timeout here if upstream is unusually slow or fast.
   */
  fal?: FalRecordConfig;
  /**
   * OpenRouter-video-specific recording knobs for the live lifecycle proxy on
   * `/api/v1/videos` (submit/poll proxied 1:1; completed jobs are captured
   * eagerly as fixtures).
   */
  openRouterVideo?: OpenRouterVideoRecordConfig;
  /**
   * Connection idle timeout (ms) on the upstream request socket — fires if the
   * socket is inactive for this duration at any point before the response body
   * begins. Default: 30_000 (30s). Increase for upstreams with slow initial
   * responses (reasoning models, queue-backed providers).
   *
   * Nuance on the OpenRouter video surface: the small-JSON lifecycle fetches
   * (submit, status poll, models listing) apply this value as a TOTAL
   * deadline on the whole fetch — indistinguishable from an idle timeout for
   * envelope-sized bodies. The byte-bearing content fetches (eager capture,
   * proxy-only content relay) gate only the response HEADERS on this value;
   * their body progress is governed by `bodyTimeoutMs` idle semantics.
   *
   * Ceiling note: the fetch-based OpenRouter proxy paths run on Node's
   * built-in undici dispatcher, which carries its own ~300s
   * headers/body-timeout defaults — values above that ceiling do not take
   * effect on those paths without installing a custom undici dispatcher
   * (out of aimock's scope).
   */
  upstreamTimeoutMs?: number;
  /**
   * Idle timeout (ms) on the upstream response body — fires if the upstream
   * goes silent (no bytes) for this long after the response has started.
   * Default: 30_000 (30s). Reasoning models under concurrent load can leave
   * 30s+ gaps between streaming chunks while the model is thinking; lift this
   * to e.g. 180_000 in those setups. The OpenRouter video content fetches use
   * the same idle semantics: the timer re-arms on every chunk, so a steadily
   * downloading multi-minute video never times out — only a silent stall does.
   */
  bodyTimeoutMs?: number;
  /**
   * Maximum number of bytes aimock will accumulate in memory from a single
   * proxied upstream response on the record/proxy path. The full response is
   * still relayed to the client byte-for-byte; this cap only bounds the
   * in-memory buffer used to collapse/journal the response. Once the cap is
   * exceeded aimock stops appending to the buffer, marks the response as
   * truncated, and skips collapse/recording — preventing both unbounded heap
   * growth and the `RangeError: Invalid string length` that a >512MB string
   * would otherwise throw. Default: 64 MiB.
   */
  maxProxyBufferBytes?: number;
  /**
   * Maximum number of SSE/NDJSON/EventStream frames whose per-frame state
   * (`frameTimestamps`, parse buffers) aimock retains for a single proxied
   * response. Frame state is count-indexed, not byte-sized, so a long-lived or
   * never-ending stream accumulates it unbounded even when `maxProxyBufferBytes`
   * is generous. Tripping truncation on EITHER bytes OR frame count bounds both;
   * on the trip the accumulated frame state is freed and collapse/recording is
   * skipped (the full response is still relayed to the client byte-for-byte).
   * Default: 5,000,000 frames.
   */
  maxProxyBufferFrames?: number;
}
interface OpenRouterVideoRecordConfig {
  /**
   * Maximum decoded video size (bytes) embedded as `b64` in a recorded
   * fixture. Captures larger than the cap are persisted WITHOUT `b64` (with a
   * `_warning` in the fixture file) so a multi-hundred-MB video cannot bloat
   * the fixture directory. The cap is also a memory guard: an upstream
   * response that DECLARES an over-cap Content-Length is skipped without
   * downloading, and a response with no declared length is streamed with the
   * byte count enforced during the read — on exceed the download is aborted
   * and nothing oversized is retained in memory. In both cases the
   * same-session job serves the placeholder MP4. Default: 33554432 (32 MB
   * decoded). Set 0 for unlimited. Negative or non-integer values are treated
   * as the default (createServer warns at startup).
   */
  maxContentBytes?: number;
}
interface FalRecordConfig {
  /** Interval between status polls upstream during recording. Default: 1000ms. */
  pollIntervalMs?: number;
  /** Total budget for an upstream queue walk before aborting. Default: 900000ms (15 min) to accommodate video generation. */
  timeoutMs?: number;
}
interface MockServerOptions {
  port?: number;
  host?: string;
  latency?: number;
  chunkSize?: number;
  replaySpeed?: number;
  /** Log verbosity. CLI default is "info"; programmatic default (when omitted) is "silent". */
  logLevel?: "silent" | "warn" | "info" | "debug";
  chaos?: ChaosConfig;
  /** Enable Prometheus-compatible /metrics endpoint. */
  metrics?: boolean;
  /** Strict mode: return 503 instead of 404 when no fixture matches. */
  strict?: boolean;
  /** Record-and-replay: proxy unmatched requests to upstream and save fixtures. */
  record?: RecordConfig;
  /**
   * Maximum number of request/response entries to retain in the in-memory
   * journal. Oldest entries are dropped FIFO when the cap is exceeded.
   * Set to 0 (or omit) for unbounded retention. Negative values are
   * rejected at the CLI parse layer; programmatically they are treated
   * as 0 (unbounded) for back-compat.
   *
   * Default: 1000 (applied by `createServer` when omitted). The CLI passes
   * through its own default. Short-lived test harnesses that want every
   * request recorded can opt in to unbounded retention by passing 0.
   */
  journalMaxEntries?: number;
  /**
   * Maximum number of unique testIds retained in the journal's fixture
   * match-count map. Oldest testIds are dropped FIFO when the cap is
   * exceeded. Set to 0 (or omit) for unbounded retention. Negative values
   * are rejected at the CLI parse layer; programmatically they are treated
   * as 0 (unbounded) for back-compat.
   *
   * Default: 500 (applied by `createServer` when omitted). Without a cap
   * this map can grow over time in long-running servers that see many
   * unique testIds.
   */
  fixtureCountsMaxTestIds?: number;
  /**
   * Normalize requests before matching and recording. Useful for stripping
   * dynamic data (timestamps, UUIDs, session IDs) that would cause fixture
   * mismatches on replay.
   *
   * When set, string matching for `userMessage` and `inputText` uses exact
   * equality (`===`) instead of substring (`includes`) to prevent false
   * positives from shortened keys.
   */
  requestTransform?: (req: ChatCompletionRequest) => ChatCompletionRequest;
  /**
   * Configure fal.ai queue polling progression. By default a job completes
   * on submit (preserves the legacy `COMPLETED`-on-submit shape). Opt into a
   * realistic `IN_QUEUE → IN_PROGRESS → COMPLETED` progression by setting
   * positive poll thresholds — useful for exercising client code that polls
   * `/status` and reacts to intermediate states.
   *
   * Applies to the general fal handler (`x-fal-target-host`-routed); the
   * legacy `/fal/queue/...` audio handler is unaffected.
   */
  falQueue?: FalQueueConfig;
  /**
   * Configure OpenRouter async video job polling progression
   * (`pending → in_progress → completed | failed` on `GET /api/v1/videos/{id}`).
   * Same threshold semantics as `falQueue`. By default (0/0) the job is
   * seeded terminal at submit — content is downloadable with zero polls, and
   * the first status poll merely reports the already-terminal status.
   */
  openRouterVideo?: FalQueueConfig;
}
/**
 * Poll-progression thresholds, documented below in fal queue terms; when used
 * as `openRouterVideo` the states map to `pending` / `in_progress` /
 * `completed | failed`.
 */
interface FalQueueConfig {
  /**
   * Status polls before transitioning `IN_QUEUE → IN_PROGRESS`. Unset and an
   * explicit `0` differ: when BOTH fields are unset the job completes
   * synchronously on submit (no IN_QUEUE / IN_PROGRESS polls emitted), but
   * explicitly setting this field — even to `0` — enables progression when
   * `pollsBeforeCompleted` is unset, with `pollsBeforeCompleted` defaulting
   * to `pollsBeforeInProgress + 1` so the job passes through IN_PROGRESS
   * (an explicit `pollsBeforeCompleted: 0` completes on submit only when
   * `pollsBeforeInProgress` is `0` or unset — otherwise the clamp below
   * lifts it to `pollsBeforeInProgress`).
   */
  pollsBeforeInProgress?: number;
  /**
   * Status polls before transitioning to `COMPLETED`. Default: 0 when
   * `pollsBeforeInProgress` is also unset (no progression), otherwise
   * `pollsBeforeInProgress + 1` so the job spends one poll in IN_PROGRESS.
   * An explicit value lower than `pollsBeforeInProgress` is clamped up so
   * IN_PROGRESS is never skipped. When equal to a nonzero
   * `pollsBeforeInProgress` — or when `pollsBeforeInProgress` is unset or 0
   * and this field is 1 (the IN_PROGRESS branch consumes the first poll) —
   * the job still spends one poll in IN_PROGRESS, so the terminal status
   * lands one poll later than configured.
   */
  pollsBeforeCompleted?: number;
}
interface HandlerDefaults {
  latency: number;
  chunkSize: number;
  replaySpeed: number;
  logger: Logger;
  chaos?: ChaosConfig;
  registry?: MetricsRegistry;
  record?: RecordConfig;
  strict?: boolean;
  requestTransform?: (req: ChatCompletionRequest) => ChatCompletionRequest;
  falQueue?: FalQueueConfig;
  openRouterVideo?: FalQueueConfig;
}
//# sourceMappingURL=types.d.ts.map
//#endregion
export { AudioResponse, ChaosAction, ChaosConfig, ChatCompletion, ChatCompletionChoice, ChatCompletionMessage, ChatCompletionRequest, ChatMessage, ContentPart, ContentWithToolCallsResponse, EmbeddingFixtureOpts, EmbeddingResponse, ErrorResponse, FalQueueOpts, Fixture, FixtureFile, FixtureFileContentWithToolCallsResponse, FixtureFileEntry, FixtureFileResponse, FixtureFileTextResponse, FixtureFileToolCall, FixtureFileToolCallResponse, FixtureMatch, FixtureOpts, FixtureResponse, HandlerDefaults, ImageItem, ImageResponse, JournalEntry, MockServerOptions, Mountable, RawJSONResponse, RecordConfig, RecordProviderKey, RecordedTimings, ResponseFactory, ResponseOverrides, SSEChoice, SSEChunk, SSEDelta, SSEToolCallDelta, StreamingProfile, TextResponse, ToolCall, ToolCallMessage, ToolCallResponse, ToolDefinition, TranscriptionResponse, VideoResponse };
//# sourceMappingURL=types.d.cts.map