pulse-ts-sdk 0.0.65 → 1.0.4

package/dist/cjs/BaseClient.js

@@ -43,8 +43,8 @@ function normalizeClientOptions(options) {
     const headers = (0, headers_js_1.mergeHeaders)({
         "X-Fern-Language": "JavaScript",
         "X-Fern-SDK-Name": "pulse-ts-sdk",
-        "X-Fern-SDK-Version": "0.0.65",
-        "User-Agent": "pulse-ts-sdk/0.0.65",
+        "X-Fern-SDK-Version": "1.0.4",
+        "User-Agent": "pulse-ts-sdk/1.0.4",
         "X-Fern-Runtime": core.RUNTIME.type,
         "X-Fern-Runtime-Version": core.RUNTIME.version,
     }, options === null || options === void 0 ? void 0 : options.headers);

package/dist/cjs/Client.d.ts

@@ -24,10 +24,6 @@ export declare class PulseClient {
      * Set `async: true` to return immediately with a job_id for polling via
      * GET /job/{jobId}. Otherwise processes synchronously.
      *
-     * **Note:** Both sync and async modes return HTTP 200. When `async` is true
-     * the response body contains `{ job_id, status }` instead of the full
-     * extraction result.
-     *
      * @param {Pulse.ExtractRequest} request
      * @param {PulseClient.RequestOptions} requestOptions - Request-specific configuration.
      *
@@ -116,4 +112,31 @@ export declare class PulseClient {
      */
     schema(request?: Pulse.SchemaInput, requestOptions?: PulseClient.RequestOptions): core.HttpResponsePromise<Pulse.SchemaResponse>;
     private __schema;
+    /**
+     * Extract tables from a previously completed extraction. Processes the
+     * extraction's document content and returns structured table data.
+     *
+     * Requires the `tables_endpoint` feature flag to be enabled for your
+     * organization.
+     *
+     * Set `async: true` to return immediately with a `tables_id` for
+     * polling via `GET /job/{tables_id}`.
+     *
+     * @param {Pulse.TablesInput} request
+     * @param {PulseClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
+     * @throws {@link Pulse.BadRequestError}
+     * @throws {@link Pulse.UnauthorizedError}
+     * @throws {@link Pulse.ForbiddenError}
+     * @throws {@link Pulse.NotFoundError}
+     * @throws {@link Pulse.TooManyRequestsError}
+     * @throws {@link Pulse.InternalServerError}
+     *
+     * @example
+     *     await client.tables({
+     *         extraction_id: "extraction_id"
+     *     })
+     */
+    tables(request: Pulse.TablesInput, requestOptions?: PulseClient.RequestOptions): core.HttpResponsePromise<Pulse.TablesResponse>;
+    private __tables;
 }

package/dist/cjs/Client.js

@@ -74,10 +74,6 @@ class PulseClient {
      * Set `async: true` to return immediately with a job_id for polling via
      * GET /job/{jobId}. Otherwise processes synchronously.
      *
-     * **Note:** Both sync and async modes return HTTP 200. When `async` is true
-     * the response body contains `{ job_id, status }` instead of the full
-     * extraction result.
-     *
      * @param {Pulse.ExtractRequest} request
      * @param {PulseClient.RequestOptions} requestOptions - Request-specific configuration.
      *
@@ -102,17 +98,27 @@ class PulseClient {
             if (request.fileUrl != null) {
                 _request.append("fileUrl", request.fileUrl);
             }
+            if (request.pages != null) {
+                _request.append("pages", request.pages);
+            }
+            if (request.figureProcessing != null) {
+                _request.append("figureProcessing", (0, json_js_1.toJson)(request.figureProcessing));
+            }
+            if (request.extensions != null) {
+                _request.append("extensions", (0, json_js_1.toJson)(request.extensions));
+            }
+            if (request.storage != null) {
+                _request.append("storage", (0, json_js_1.toJson)(request.storage));
+            }
+            if (request.async != null) {
+                _request.append("async", request.async.toString());
+            }
             if (request.structuredOutput != null) {
                 _request.append("structuredOutput", (0, json_js_1.toJson)(request.structuredOutput));
             }
             if (request.schema != null) {
                 _request.append("schema", typeof request.schema === "string" ? request.schema : (0, json_js_1.toJson)(request.schema));
             }
-            if (request.experimentalSchema != null) {
-                _request.append("experimentalSchema", typeof request.experimentalSchema === "string"
-                    ? request.experimentalSchema
-                    : (0, json_js_1.toJson)(request.experimentalSchema));
-            }
             if (request.schemaPrompt != null) {
                 _request.append("schemaPrompt", request.schemaPrompt);
             }
@@ -125,9 +131,6 @@ class PulseClient {
             if (request.chunkSize != null) {
                 _request.append("chunkSize", request.chunkSize.toString());
             }
-            if (request.pages != null) {
-                _request.append("pages", request.pages);
-            }
             if (request.extractFigure != null) {
                 _request.append("extractFigure", request.extractFigure.toString());
             }
@@ -140,18 +143,9 @@ class PulseClient {
             if (request.returnHtml != null) {
                 _request.append("returnHtml", request.returnHtml.toString());
             }
-            if (request.effort != null) {
-                _request.append("effort", request.effort.toString());
-            }
             if (request.thinking != null) {
                 _request.append("thinking", request.thinking.toString());
             }
-            if (request.storage != null) {
-                _request.append("storage", (0, json_js_1.toJson)(request.storage));
-            }
-            if (request.async != null) {
-                _request.append("async", request.async.toString());
-            }
             const _maybeEncodedRequest = yield _request.getRequest();
             const _authRequest = yield this._options.authProvider.getAuthRequest();
             const _headers = (0, headers_js_1.mergeHeaders)(_authRequest.headers, (_a = this._options) === null || _a === void 0 ? void 0 : _a.headers, (0, headers_js_1.mergeOnlyDefinedHeaders)(Object.assign({}, _maybeEncodedRequest.headers)), requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers);
@@ -222,17 +216,27 @@ class PulseClient {
             if (request.fileUrl != null) {
                 _request.append("fileUrl", request.fileUrl);
             }
+            if (request.pages != null) {
+                _request.append("pages", request.pages);
+            }
+            if (request.figureProcessing != null) {
+                _request.append("figureProcessing", (0, json_js_1.toJson)(request.figureProcessing));
+            }
+            if (request.extensions != null) {
+                _request.append("extensions", (0, json_js_1.toJson)(request.extensions));
+            }
+            if (request.storage != null) {
+                _request.append("storage", (0, json_js_1.toJson)(request.storage));
+            }
+            if (request.async != null) {
+                _request.append("async", request.async.toString());
+            }
             if (request.structuredOutput != null) {
                 _request.append("structuredOutput", (0, json_js_1.toJson)(request.structuredOutput));
             }
             if (request.schema != null) {
                 _request.append("schema", typeof request.schema === "string" ? request.schema : (0, json_js_1.toJson)(request.schema));
             }
-            if (request.experimentalSchema != null) {
-                _request.append("experimentalSchema", typeof request.experimentalSchema === "string"
-                    ? request.experimentalSchema
-                    : (0, json_js_1.toJson)(request.experimentalSchema));
-            }
             if (request.schemaPrompt != null) {
                 _request.append("schemaPrompt", request.schemaPrompt);
             }
@@ -245,9 +249,6 @@ class PulseClient {
             if (request.chunkSize != null) {
                 _request.append("chunkSize", request.chunkSize.toString());
             }
-            if (request.pages != null) {
-                _request.append("pages", request.pages);
-            }
             if (request.extractFigure != null) {
                 _request.append("extractFigure", request.extractFigure.toString());
             }
@@ -260,18 +261,9 @@ class PulseClient {
             if (request.returnHtml != null) {
                 _request.append("returnHtml", request.returnHtml.toString());
             }
-            if (request.effort != null) {
-                _request.append("effort", request.effort.toString());
-            }
             if (request.thinking != null) {
                 _request.append("thinking", request.thinking.toString());
             }
-            if (request.storage != null) {
-                _request.append("storage", (0, json_js_1.toJson)(request.storage));
-            }
-            if (request.async != null) {
-                _request.append("async", request.async.toString());
-            }
             const _maybeEncodedRequest = yield _request.getRequest();
             const _authRequest = yield this._options.authProvider.getAuthRequest();
             const _headers = (0, headers_js_1.mergeHeaders)(_authRequest.headers, (_a = this._options) === null || _a === void 0 ? void 0 : _a.headers, (0, headers_js_1.mergeOnlyDefinedHeaders)(Object.assign({}, _maybeEncodedRequest.headers)), requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers);
@@ -458,5 +450,80 @@ class PulseClient {
             return (0, handleNonStatusCodeError_js_1.handleNonStatusCodeError)(_response.error, _response.rawResponse, "POST", "/schema");
         });
     }
+    /**
+     * Extract tables from a previously completed extraction. Processes the
+     * extraction's document content and returns structured table data.
+     *
+     * Requires the `tables_endpoint` feature flag to be enabled for your
+     * organization.
+     *
+     * Set `async: true` to return immediately with a `tables_id` for
+     * polling via `GET /job/{tables_id}`.
+     *
+     * @param {Pulse.TablesInput} request
+     * @param {PulseClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
+     * @throws {@link Pulse.BadRequestError}
+     * @throws {@link Pulse.UnauthorizedError}
+     * @throws {@link Pulse.ForbiddenError}
+     * @throws {@link Pulse.NotFoundError}
+     * @throws {@link Pulse.TooManyRequestsError}
+     * @throws {@link Pulse.InternalServerError}
+     *
+     * @example
+     *     await client.tables({
+     *         extraction_id: "extraction_id"
+     *     })
+     */
+    tables(request, requestOptions) {
+        return core.HttpResponsePromise.fromPromise(this.__tables(request, requestOptions));
+    }
+    __tables(request, requestOptions) {
+        return __awaiter(this, void 0, void 0, function* () {
+            var _a, _b, _c, _d, _e, _f, _g, _h, _j;
+            const _authRequest = yield this._options.authProvider.getAuthRequest();
+            const _headers = (0, headers_js_1.mergeHeaders)(_authRequest.headers, (_a = this._options) === null || _a === void 0 ? void 0 : _a.headers, requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.headers);
+            const _response = yield core.fetcher({
+                url: core.url.join((_c = (_b = (yield core.Supplier.get(this._options.baseUrl))) !== null && _b !== void 0 ? _b : (yield core.Supplier.get(this._options.environment))) !== null && _c !== void 0 ? _c : environments.PulseEnvironment.Default, "tables"),
+                method: "POST",
+                headers: _headers,
+                contentType: "application/json",
+                queryParameters: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.queryParams,
+                requestType: "json",
+                body: request,
+                timeoutMs: ((_f = (_d = requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) !== null && _d !== void 0 ? _d : (_e = this._options) === null || _e === void 0 ? void 0 : _e.timeoutInSeconds) !== null && _f !== void 0 ? _f : 60) * 1000,
+                maxRetries: (_g = requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries) !== null && _g !== void 0 ? _g : (_h = this._options) === null || _h === void 0 ? void 0 : _h.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
+                fetchFn: (_j = this._options) === null || _j === void 0 ? void 0 : _j.fetch,
+                logging: this._options.logging,
+            });
+            if (_response.ok) {
+                return { data: _response.body, rawResponse: _response.rawResponse };
+            }
+            if (_response.error.reason === "status-code") {
+                switch (_response.error.statusCode) {
+                    case 400:
+                        throw new Pulse.BadRequestError(_response.error.body, _response.rawResponse);
+                    case 401:
+                        throw new Pulse.UnauthorizedError(_response.error.body, _response.rawResponse);
+                    case 403:
+                        throw new Pulse.ForbiddenError(_response.error.body, _response.rawResponse);
+                    case 404:
+                        throw new Pulse.NotFoundError(_response.error.body, _response.rawResponse);
+                    case 429:
+                        throw new Pulse.TooManyRequestsError(_response.error.body, _response.rawResponse);
+                    case 500:
+                        throw new Pulse.InternalServerError(_response.error.body, _response.rawResponse);
+                    default:
+                        throw new errors.PulseError({
+                            statusCode: _response.error.statusCode,
+                            body: _response.error.body,
+                            rawResponse: _response.rawResponse,
+                        });
+                }
+            }
+            return (0, handleNonStatusCodeError_js_1.handleNonStatusCodeError)(_response.error, _response.rawResponse, "POST", "/tables");
+        });
+    }
 }
 exports.PulseClient = PulseClient;

package/dist/cjs/api/client/requests/ExtractAsyncRequest.d.ts

@@ -8,57 +8,96 @@ export interface ExtractAsyncRequest {
     file?: core.file.Uploadable | undefined;
     /** Public or pre-signed URL that Pulse will download and extract. Required unless file is provided. */
     fileUrl?: string;
+    /** Page range filter supporting segments such as `1-2` or mixed ranges like `1-2,5`. */
+    pages?: string;
+    /** Settings that control how figures in the document are processed. These affect the markdown output directly (e.g. figure descriptions, chart-to-table conversion, image embedding) and do not produce additional output fields in the response. */
+    figureProcessing?: ExtractAsyncRequest.FigureProcessing;
+    /** Settings that enable additional processing passes or alternate output formats. Each enabled extension produces a corresponding output field under `response.extensions.*`. */
+    extensions?: ExtractAsyncRequest.Extensions;
+    /** Options for persisting extraction artifacts. When enabled (default), artifacts are saved to storage and a database record is created. */
+    storage?: ExtractAsyncRequest.Storage;
+    /** If true, returns immediately with a job_id for polling via GET /job/{jobId}. Otherwise processes synchronously. */
+    async?: boolean;
     /** **⚠️ DEPRECATED** — Use the `/schema` endpoint after extraction instead. Pass the `extraction_id` from the extract response to `/schema` with your `schema_config`. This parameter still works for backward compatibility but will be removed in a future version. */
     structuredOutput?: ExtractAsyncRequest.StructuredOutput;
     /** (Deprecated) JSON schema describing structured data to extract. Use structuredOutput instead. Accepts either a JSON object or a stringified JSON representation. */
     schema?: ExtractAsyncRequest.Schema;
-    /** (Deprecated) Experimental schema definition used for feature flagged behaviour. Accepts either a JSON object or a stringified JSON representation. */
-    experimentalSchema?: ExtractAsyncRequest.ExperimentalSchema;
     /** (Deprecated) Natural language prompt for schema-guided extraction. Use structuredOutput.schemaPrompt instead. */
     schemaPrompt?: string;
-    /** (Deprecated) Custom instructions that augment the default extraction behaviour. */
+    /** (Deprecated) Custom instructions that augment the default extraction behaviour. Use `figureProcessing` or `extensions` instead. */
     customPrompt?: string;
-    /** Comma-separated list of chunking strategies to apply (for example `semantic,header,page,recursive`). */
+    /** **⚠️ DEPRECATED** — Use `extensions.chunking.chunkTypes` instead. Comma-separated list of chunking strategies to apply (for example `semantic,header,page,recursive`). Still accepted for backward compatibility. */
     chunking?: string;
-    /** Override for maximum characters per chunk when chunking is enabled. */
+    /** **⚠️ DEPRECATED** — Use `extensions.chunking.chunkSize` instead. Override for maximum characters per chunk when chunking is enabled. */
     chunkSize?: number;
-    /** Page range filter supporting segments such as `1-2` or mixed ranges like `1-2,5`. */
-    pages?: string;
-    /** Toggle to enable figure extraction in results. */
+    /** **⚠️ DEPRECATED** — Toggle to enable figure extraction in results. */
     extractFigure?: boolean;
-    /** Toggle to generate descriptive captions for extracted figures. */
+    /** **⚠️ DEPRECATED** — Use `figureProcessing.description` instead. Toggle to generate descriptive captions for extracted figures. */
     figureDescription?: boolean;
-    /** Embed base64-encoded images inline in figure tags in the output. Increases response size. */
+    /** **⚠️ DEPRECATED** — Use `figureProcessing.showImages` instead. Embed base64-encoded images inline in figure tags in the output. Increases response size. */
     showImages?: boolean;
-    /** Whether to include HTML representation alongside markdown in the response. */
+    /** **⚠️ DEPRECATED** — Use `extensions.altOutputs.returnHtml` instead. Whether to include HTML representation alongside markdown in the response. */
     returnHtml?: boolean;
-    /** Enable extended reasoning mode for higher quality extraction on complex documents. Uses a more powerful model at higher latency. */
-    effort?: boolean;
     /** (Deprecated) Enables expanded rationale output for debugging. */
     thinking?: boolean;
-    /** Options for persisting extraction artifacts. When enabled (default), artifacts are saved to storage and a database record is created. */
-    storage?: ExtractAsyncRequest.Storage;
-    /** If true, returns immediately with a job_id for polling via GET /job/{jobId}. Otherwise processes synchronously. */
-    async?: boolean;
 }
 export declare namespace ExtractAsyncRequest {
     /**
-     * **⚠️ DEPRECATED** — Use the `/schema` endpoint after extraction instead. Pass the `extraction_id` from the extract response to `/schema` with your `schema_config`. This parameter still works for backward compatibility but will be removed in a future version.
+     * Settings that control how figures in the document are processed. These affect the markdown output directly (e.g. figure descriptions, chart-to-table conversion, image embedding) and do not produce additional output fields in the response.
      */
-    interface StructuredOutput {
-        /** JSON schema describing the structured data to extract. */
-        schema?: Record<string, unknown>;
-        /** Natural language prompt with additional extraction instructions. */
-        schemaPrompt?: string;
+    interface FigureProcessing {
+        /** Generate descriptive captions for extracted figures. */
+        description?: boolean;
+        /** Embed base64-encoded images inline in figure tags in the output. Increases response size. */
+        showImages?: boolean;
     }
     /**
-     * (Deprecated) JSON schema describing structured data to extract. Use structuredOutput instead. Accepts either a JSON object or a stringified JSON representation.
+     * Settings that enable additional processing passes or alternate output formats. Each enabled extension produces a corresponding output field under `response.extensions.*`.
      */
-    type Schema = Record<string, unknown> | string;
-    /**
-     * (Deprecated) Experimental schema definition used for feature flagged behaviour. Accepts either a JSON object or a stringified JSON representation.
-     */
-    type ExperimentalSchema = Record<string, unknown> | string;
+    interface Extensions {
+        /** Merge tables that span multiple pages into a single table. */
+        mergeTables?: boolean;
+        /** Link footnote markers to their corresponding footnote text. */
+        footnoteReferences?: boolean;
+        /** Chunking configuration. When provided, the document is split into chunks using the specified strategies. Results appear in `response.extensions.chunking`. */
+        chunking?: Extensions.Chunking;
+        /** Alternate output format options. Each enabled format produces a corresponding field under `response.extensions.altOutputs`. */
+        altOutputs?: Extensions.AltOutputs;
+    }
+    namespace Extensions {
+        /**
+         * Chunking configuration. When provided, the document is split into chunks using the specified strategies. Results appear in `response.extensions.chunking`.
+         */
+        interface Chunking {
+            /** List of chunking strategies to apply (e.g. `["semantic", "header", "page", "recursive"]`). */
+            chunkTypes?: Chunking.ChunkTypes.Item[];
+            /** Maximum characters per chunk. */
+            chunkSize?: number;
+        }
+        namespace Chunking {
+            type ChunkTypes = ChunkTypes.Item[];
+            namespace ChunkTypes {
+                const Item: {
+                    readonly Semantic: "semantic";
+                    readonly Header: "header";
+                    readonly Page: "page";
+                    readonly Recursive: "recursive";
+                };
+                type Item = (typeof Item)[keyof typeof Item];
+            }
+        }
+        /**
+         * Alternate output format options. Each enabled format produces a corresponding field under `response.extensions.altOutputs`.
+         */
+        interface AltOutputs {
+            /** Enable word-level bounding boxes. Runs an additional OCR model to derive bounding boxes for each word. Only applies to PDFs. Results in `response.extensions.altOutputs.wlbb`. */
+            wlbb?: boolean;
+            /** Include an HTML representation of the document. When enabled, `response.markdown` is still present and the HTML is available at `response.extensions.altOutputs.html`. */
+            returnHtml?: boolean;
+            /** Include an XML representation of the document. Results in `response.extensions.altOutputs.xml`. (Work in progress.) */
+            returnXml?: boolean;
+        }
+    }
     /**
      * Options for persisting extraction artifacts. When enabled (default), artifacts are saved to storage and a database record is created.
      */
@@ -67,5 +106,22 @@ export declare namespace ExtractAsyncRequest {
         enabled?: boolean;
         /** Target folder name to save the extraction to. Creates the folder if it doesn't exist. */
         folderName?: string;
+        /** Target folder ID to save the extraction to. Takes precedence over folderName if both are provided. */
+        folderId?: string;
+    }
+    /**
+     * **⚠️ DEPRECATED** — Use the `/schema` endpoint after extraction instead. Pass the `extraction_id` from the extract response to `/schema` with your `schema_config`. This parameter still works for backward compatibility but will be removed in a future version.
+     */
+    interface StructuredOutput {
+        /** JSON schema describing the structured data to extract. */
+        schema?: Record<string, unknown>;
+        /** Natural language prompt with additional extraction instructions. */
+        schemaPrompt?: string;
+        /** Use higher quality model for better results. When true, uses a more capable model at the cost of higher latency. */
+        effort?: boolean;
     }
+    /**
+     * (Deprecated) JSON schema describing structured data to extract. Use structuredOutput instead. Accepts either a JSON object or a stringified JSON representation.
+     */
+    type Schema = Record<string, unknown> | string;
 }

package/dist/cjs/api/client/requests/ExtractAsyncRequest.js

@@ -1,3 +1,22 @@
 "use strict";
 // This file was auto-generated by Fern from our API Definition.
 Object.defineProperty(exports, "__esModule", { value: true });
+exports.ExtractAsyncRequest = void 0;
+var ExtractAsyncRequest;
+(function (ExtractAsyncRequest) {
+    let Extensions;
+    (function (Extensions) {
+        let Chunking;
+        (function (Chunking) {
+            let ChunkTypes;
+            (function (ChunkTypes) {
+                ChunkTypes.Item = {
+                    Semantic: "semantic",
+                    Header: "header",
+                    Page: "page",
+                    Recursive: "recursive",
+                };
+            })(ChunkTypes = Chunking.ChunkTypes || (Chunking.ChunkTypes = {}));
+        })(Chunking = Extensions.Chunking || (Extensions.Chunking = {}));
+    })(Extensions = ExtractAsyncRequest.Extensions || (ExtractAsyncRequest.Extensions = {}));
+})(ExtractAsyncRequest || (exports.ExtractAsyncRequest = ExtractAsyncRequest = {}));

package/dist/cjs/api/client/requests/ExtractRequest.d.ts

@@ -8,57 +8,96 @@ export interface ExtractRequest {
     file?: core.file.Uploadable | undefined;
     /** Public or pre-signed URL that Pulse will download and extract. Required unless file is provided. */
     fileUrl?: string;
+    /** Page range filter supporting segments such as `1-2` or mixed ranges like `1-2,5`. */
+    pages?: string;
+    /** Settings that control how figures in the document are processed. These affect the markdown output directly (e.g. figure descriptions, chart-to-table conversion, image embedding) and do not produce additional output fields in the response. */
+    figureProcessing?: ExtractRequest.FigureProcessing;
+    /** Settings that enable additional processing passes or alternate output formats. Each enabled extension produces a corresponding output field under `response.extensions.*`. */
+    extensions?: ExtractRequest.Extensions;
+    /** Options for persisting extraction artifacts. When enabled (default), artifacts are saved to storage and a database record is created. */
+    storage?: ExtractRequest.Storage;
+    /** If true, returns immediately with a job_id for polling via GET /job/{jobId}. Otherwise processes synchronously. */
+    async?: boolean;
     /** **⚠️ DEPRECATED** — Use the `/schema` endpoint after extraction instead. Pass the `extraction_id` from the extract response to `/schema` with your `schema_config`. This parameter still works for backward compatibility but will be removed in a future version. */
     structuredOutput?: ExtractRequest.StructuredOutput;
     /** (Deprecated) JSON schema describing structured data to extract. Use structuredOutput instead. Accepts either a JSON object or a stringified JSON representation. */
     schema?: ExtractRequest.Schema;
-    /** (Deprecated) Experimental schema definition used for feature flagged behaviour. Accepts either a JSON object or a stringified JSON representation. */
-    experimentalSchema?: ExtractRequest.ExperimentalSchema;
     /** (Deprecated) Natural language prompt for schema-guided extraction. Use structuredOutput.schemaPrompt instead. */
     schemaPrompt?: string;
-    /** (Deprecated) Custom instructions that augment the default extraction behaviour. */
+    /** (Deprecated) Custom instructions that augment the default extraction behaviour. Use `figureProcessing` or `extensions` instead. */
     customPrompt?: string;
-    /** Comma-separated list of chunking strategies to apply (for example `semantic,header,page,recursive`). */
+    /** **⚠️ DEPRECATED** — Use `extensions.chunking.chunkTypes` instead. Comma-separated list of chunking strategies to apply (for example `semantic,header,page,recursive`). Still accepted for backward compatibility. */
     chunking?: string;
-    /** Override for maximum characters per chunk when chunking is enabled. */
+    /** **⚠️ DEPRECATED** — Use `extensions.chunking.chunkSize` instead. Override for maximum characters per chunk when chunking is enabled. */
     chunkSize?: number;
-    /** Page range filter supporting segments such as `1-2` or mixed ranges like `1-2,5`. */
-    pages?: string;
-    /** Toggle to enable figure extraction in results. */
+    /** **⚠️ DEPRECATED** — Toggle to enable figure extraction in results. */
     extractFigure?: boolean;
-    /** Toggle to generate descriptive captions for extracted figures. */
+    /** **⚠️ DEPRECATED** — Use `figureProcessing.description` instead. Toggle to generate descriptive captions for extracted figures. */
     figureDescription?: boolean;
-    /** Embed base64-encoded images inline in figure tags in the output. Increases response size. */
+    /** **⚠️ DEPRECATED** — Use `figureProcessing.showImages` instead. Embed base64-encoded images inline in figure tags in the output. Increases response size. */
     showImages?: boolean;
-    /** Whether to include HTML representation alongside markdown in the response. */
+    /** **⚠️ DEPRECATED** — Use `extensions.altOutputs.returnHtml` instead. Whether to include HTML representation alongside markdown in the response. */
     returnHtml?: boolean;
-    /** Enable extended reasoning mode for higher quality extraction on complex documents. Uses a more powerful model at higher latency. */
-    effort?: boolean;
     /** (Deprecated) Enables expanded rationale output for debugging. */
     thinking?: boolean;
-    /** Options for persisting extraction artifacts. When enabled (default), artifacts are saved to storage and a database record is created. */
-    storage?: ExtractRequest.Storage;
-    /** If true, returns immediately with a job_id for polling via GET /job/{jobId}. Otherwise processes synchronously. */
-    async?: boolean;
 }
 export declare namespace ExtractRequest {
     /**
-     * **⚠️ DEPRECATED** — Use the `/schema` endpoint after extraction instead. Pass the `extraction_id` from the extract response to `/schema` with your `schema_config`. This parameter still works for backward compatibility but will be removed in a future version.
+     * Settings that control how figures in the document are processed. These affect the markdown output directly (e.g. figure descriptions, chart-to-table conversion, image embedding) and do not produce additional output fields in the response.
      */
-    interface StructuredOutput {
-        /** JSON schema describing the structured data to extract. */
-        schema?: Record<string, unknown>;
-        /** Natural language prompt with additional extraction instructions. */
-        schemaPrompt?: string;
+    interface FigureProcessing {
+        /** Generate descriptive captions for extracted figures. */
+        description?: boolean;
+        /** Embed base64-encoded images inline in figure tags in the output. Increases response size. */
+        showImages?: boolean;
     }
     /**
-     * (Deprecated) JSON schema describing structured data to extract. Use structuredOutput instead. Accepts either a JSON object or a stringified JSON representation.
+     * Settings that enable additional processing passes or alternate output formats. Each enabled extension produces a corresponding output field under `response.extensions.*`.
      */
-    type Schema = Record<string, unknown> | string;
-    /**
-     * (Deprecated) Experimental schema definition used for feature flagged behaviour. Accepts either a JSON object or a stringified JSON representation.
-     */
-    type ExperimentalSchema = Record<string, unknown> | string;
+    interface Extensions {
+        /** Merge tables that span multiple pages into a single table. */
+        mergeTables?: boolean;
+        /** Link footnote markers to their corresponding footnote text. */
+        footnoteReferences?: boolean;
+        /** Chunking configuration. When provided, the document is split into chunks using the specified strategies. Results appear in `response.extensions.chunking`. */
+        chunking?: Extensions.Chunking;
+        /** Alternate output format options. Each enabled format produces a corresponding field under `response.extensions.altOutputs`. */
+        altOutputs?: Extensions.AltOutputs;
+    }
+    namespace Extensions {
+        /**
+         * Chunking configuration. When provided, the document is split into chunks using the specified strategies. Results appear in `response.extensions.chunking`.
+         */
+        interface Chunking {
+            /** List of chunking strategies to apply (e.g. `["semantic", "header", "page", "recursive"]`). */
+            chunkTypes?: Chunking.ChunkTypes.Item[];
+            /** Maximum characters per chunk. */
+            chunkSize?: number;
+        }
+        namespace Chunking {
+            type ChunkTypes = ChunkTypes.Item[];
+            namespace ChunkTypes {
+                const Item: {
+                    readonly Semantic: "semantic";
+                    readonly Header: "header";
+                    readonly Page: "page";
+                    readonly Recursive: "recursive";
+                };
+                type Item = (typeof Item)[keyof typeof Item];
+            }
+        }
+        /**
+         * Alternate output format options. Each enabled format produces a corresponding field under `response.extensions.altOutputs`.
+         */
+        interface AltOutputs {
+            /** Enable word-level bounding boxes. Runs an additional OCR model to derive bounding boxes for each word. Only applies to PDFs. Results in `response.extensions.altOutputs.wlbb`. */
+            wlbb?: boolean;
+            /** Include an HTML representation of the document. When enabled, `response.markdown` is still present and the HTML is available at `response.extensions.altOutputs.html`. */
+            returnHtml?: boolean;
+            /** Include an XML representation of the document. Results in `response.extensions.altOutputs.xml`. (Work in progress.) */
+            returnXml?: boolean;
+        }
+    }
     /**
      * Options for persisting extraction artifacts. When enabled (default), artifacts are saved to storage and a database record is created.
      */
@@ -67,5 +106,22 @@ export declare namespace ExtractRequest {
         enabled?: boolean;
         /** Target folder name to save the extraction to. Creates the folder if it doesn't exist. */
         folderName?: string;
+        /** Target folder ID to save the extraction to. Takes precedence over folderName if both are provided. */
+        folderId?: string;
+    }
+    /**
+     * **⚠️ DEPRECATED** — Use the `/schema` endpoint after extraction instead. Pass the `extraction_id` from the extract response to `/schema` with your `schema_config`. This parameter still works for backward compatibility but will be removed in a future version.
+     */
+    interface StructuredOutput {
+        /** JSON schema describing the structured data to extract. */
+        schema?: Record<string, unknown>;
+        /** Natural language prompt with additional extraction instructions. */
+        schemaPrompt?: string;
+        /** Use higher quality model for better results. When true, uses a more capable model at the cost of higher latency. */
+        effort?: boolean;
     }
+    /**
+     * (Deprecated) JSON schema describing structured data to extract. Use structuredOutput instead. Accepts either a JSON object or a stringified JSON representation.
+     */
+    type Schema = Record<string, unknown> | string;
 }

package/dist/cjs/api/client/requests/ExtractRequest.js

@@ -1,3 +1,22 @@
 "use strict";
 // This file was auto-generated by Fern from our API Definition.
 Object.defineProperty(exports, "__esModule", { value: true });
+exports.ExtractRequest = void 0;
+var ExtractRequest;
+(function (ExtractRequest) {
+    let Extensions;
+    (function (Extensions) {
+        let Chunking;
+        (function (Chunking) {
+            let ChunkTypes;
+            (function (ChunkTypes) {
+                ChunkTypes.Item = {
+                    Semantic: "semantic",
+                    Header: "header",
+                    Page: "page",
+                    Recursive: "recursive",
+                };
+            })(ChunkTypes = Chunking.ChunkTypes || (Chunking.ChunkTypes = {}));
+        })(Chunking = Extensions.Chunking || (Extensions.Chunking = {}));
+    })(Extensions = ExtractRequest.Extensions || (ExtractRequest.Extensions = {}));
+})(ExtractRequest || (exports.ExtractRequest = ExtractRequest = {}));

package/dist/cjs/api/client/requests/TablesInput.d.ts

@@ -0,0 +1,15 @@
+import type * as Pulse from "../../index.js";
+/**
+ * @example
+ *     {
+ *         extraction_id: "extraction_id"
+ *     }
+ */
+export interface TablesInput {
+    /** ID of a completed extraction to extract tables from. */
+    extraction_id: string;
+    /** Table extraction configuration. If omitted, defaults are used (`merge: false`, `table_format: "html"`). */
+    tables_config?: Pulse.TablesConfig;
+    /** When true, returns immediately with a job ID. Poll `GET /job/{tables_id}` for the result. */
+    async?: boolean;
+}

package/dist/cjs/api/client/requests/TablesInput.js

@@ -0,0 +1,3 @@
+"use strict";
+// This file was auto-generated by Fern from our API Definition.
+Object.defineProperty(exports, "__esModule", { value: true });