@promptbook/legacy-documents 0.72.0-9 → 0.73.0

package/esm/typings/src/llm-providers/openai/OpenAiExecutionTools.d.ts

@@ -27,7 +27,7 @@ export declare class OpenAiExecutionTools implements LlmExecutionTools {
      *
      * @param options which are relevant are directly passed to the OpenAI client
      */
-    constructor(options?: OpenAiExecutionToolsOptions);
+    constructor(options: OpenAiExecutionToolsOptions);
     get title(): string_title & string_markdown_text;
     get description(): string_markdown;
     getClient(): Promise<OpenAI>;
@@ -45,7 +45,7 @@ export declare class OpenAiExecutionTools implements LlmExecutionTools {
     /**
      * List all available OpenAI models that can be used
      */
-    listModels(): Array<AvailableModel>;
+    listModels(): ReadonlyArray<AvailableModel>;
     /**
      * Calls OpenAI API to use a chat model.
      */

package/esm/typings/src/llm-providers/openai/OpenAiExecutionToolsOptions.d.ts

@@ -1,6 +1,6 @@
 import type { ClientOptions } from 'openai';
 import type { CommonToolsOptions } from '../../execution/CommonToolsOptions';
-import type { string_token } from '../../types/typeAliases';
+import type { string_user_id } from '../../types/typeAliases';
 /**
  * Options for `OpenAiExecutionTools`
  *
@@ -16,5 +16,5 @@ export type OpenAiExecutionToolsOptions = CommonToolsOptions & ClientOptions & {
      *
      * @see https://platform.openai.com/docs/guides/safety-best-practices/end-user-ids
      */
-    user?: string_token;
+    userId: string_user_id | null;
 };

package/esm/typings/src/llm-providers/openai/openai-models.d.ts

@@ -9,7 +9,7 @@ import type { number_usd } from '../../types/typeAliases';
  * @see https://openai.com/api/pricing/
  * @public exported from `@promptbook/openai`
  */
-export declare const OPENAI_MODELS: Array<AvailableModel & {
+export declare const OPENAI_MODELS: ReadonlyArray<AvailableModel & {
     pricing?: {
         readonly prompt: number_usd;
         readonly output: number_usd;

package/esm/typings/src/llm-providers/remote/RemoteLlmExecutionTools.d.ts

@@ -19,9 +19,9 @@ import type { RemoteLlmExecutionToolsOptions } from './interfaces/RemoteLlmExecu
  * @see https://github.com/webgptorg/promptbook#remote-server
  * @public exported from `@promptbook/remote-client`
  */
-export declare class RemoteLlmExecutionTools implements LlmExecutionTools {
-    protected readonly options: RemoteLlmExecutionToolsOptions;
-    constructor(options: RemoteLlmExecutionToolsOptions);
+export declare class RemoteLlmExecutionTools<TCustomOptions = undefined> implements LlmExecutionTools {
+    protected readonly options: RemoteLlmExecutionToolsOptions<TCustomOptions>;
+    constructor(options: RemoteLlmExecutionToolsOptions<TCustomOptions>);
     get title(): string_title & string_markdown_text;
     get description(): string_markdown;
     /**
@@ -31,7 +31,7 @@ export declare class RemoteLlmExecutionTools implements LlmExecutionTools {
     /**
      * List all available models that can be used
      */
-    listModels(): Promise<Array<AvailableModel>>;
+    listModels(): Promise<ReadonlyArray<AvailableModel>>;
     /**
      * Creates a connection to the remote proxy server.
      */

package/esm/typings/src/llm-providers/remote/interfaces/PromptbookServer_ListModels_Request.d.ts

@@ -1,21 +1,30 @@
+import type { string_user_id } from '../../../types/typeAliases';
 import type { LlmToolsConfiguration } from '../../_common/register/LlmToolsConfiguration';
+import type { CollectionRemoteServerClientOptions } from './RemoteServerOptions';
 /**
  * Socket.io progress for remote text generation
  *
  * This is a request from client to server
  */
-export type PromptbookServer_ListModels_Request = PromptbookServer_ListModels_CollectionRequest | PromptbookServer_ListModels_AnonymousRequest;
-export type PromptbookServer_ListModels_CollectionRequest = {
+export type PromptbookServer_ListModels_Request<TCustomOptions> = PromptbookServer_ListModels_CollectionRequest<TCustomOptions> | PromptbookServer_ListModels_AnonymousRequest;
+export type PromptbookServer_ListModels_CollectionRequest<TCustomOptions> = CollectionRemoteServerClientOptions<TCustomOptions> & {
     /**
-     * Collection mode
+     * Application mode
      */
-    isAnonymous: false;
+    readonly isAnonymous: false;
 };
 export type PromptbookServer_ListModels_AnonymousRequest = {
     /**
      * Anonymous mode
      */
-    isAnonymous: true;
+    readonly isAnonymous: true;
+    /**
+     * Identifier of the end user
+     *
+     * Note: this is passed to the certain model providers to identify misuse
+     * Note: In anonymous mode, there is no need to identify yourself, nor does it change the actual configuration of LLM Tools (unlike in application mode)
+     */
+    readonly userId: string_user_id | null;
     /**
      * Configuration for the LLM tools
      */
@@ -24,6 +33,6 @@ export type PromptbookServer_ListModels_AnonymousRequest = {
 /**
  * TODO: [👡] DRY `PromptbookServer_Prompt_Request` and `PromptbookServer_ListModels_Request`
  * TODO: [🧠][🛍] Maybe not `isAnonymous: boolean` BUT `mode: 'ANONYMOUS'|'COLLECTION'`
- * TODO: [🧠][🤺] Pass `userId` in `PromptbookServer_ListModels_Request`
+ * TODO: [🧠][🤺] Maybe allow overriding of `userId` for each prompt - Pass `userId` in `PromptbookServer_ListModels_Request`
  * TODO: [👒] Listing models (and checking configuration) probbably should go through REST API not Socket.io
  */

package/esm/typings/src/llm-providers/remote/interfaces/PromptbookServer_ListModels_Response.d.ts

@@ -4,12 +4,12 @@ import type { AvailableModel } from '../../../execution/AvailableModel';
  *
  * This is sent from server to client when models are listed
  */
-export interface PromptbookServer_ListModels_Response {
+export type PromptbookServer_ListModels_Response = {
     /**
      * Available models that can be used
      */
-    models: Array<AvailableModel>;
-}
+    readonly models: ReadonlyArray<AvailableModel>;
+};
 /**
  * TODO: [👒] Listing models (and checking configuration) probbably should go through REST API not Socket.io
- */ 
+ */

package/esm/typings/src/llm-providers/remote/interfaces/PromptbookServer_Prompt_Request.d.ts

@@ -1,24 +1,18 @@
 import type { Prompt } from '../../../types/Prompt';
 import type { string_user_id } from '../../../types/typeAliases';
 import type { LlmToolsConfiguration } from '../../_common/register/LlmToolsConfiguration';
+import type { CollectionRemoteServerClientOptions } from './RemoteServerOptions';
 /**
  * Socket.io progress for remote text generation
  *
  * This is a request from client to server
  */
-export type PromptbookServer_Prompt_Request = PromptbookServer_Prompt_CollectionRequest | PromptbookServer_Prompt_AnonymousRequest;
-export type PromptbookServer_Prompt_CollectionRequest = {
+export type PromptbookServer_Prompt_Request<TCustomOptions> = PromptbookServer_Prompt_CollectionRequest<TCustomOptions> | PromptbookServer_Prompt_AnonymousRequest;
+export type PromptbookServer_Prompt_CollectionRequest<TCustomOptions> = CollectionRemoteServerClientOptions<TCustomOptions> & {
     /**
-     * Collection mode
+     * Application mode
      */
     readonly isAnonymous: false;
-    /**
-     * Identifier of the end user
-     *
-     * Note: this is passed to the certain model providers to identify misuse
-     * Note: In anonymous mode it is not required to identify
-     */
-    readonly userId: string_user_id;
     /**
      * The Prompt to execute
      */
@@ -33,9 +27,9 @@ export type PromptbookServer_Prompt_AnonymousRequest = {
      * Identifier of the end user
      *
      * Note: this is passed to the certain model providers to identify misuse
-     * Note: In anonymous mode it is not required to identify
+     * Note: In anonymous mode, there is no need to identify yourself, nor does it change the actual configuration of LLM Tools (unlike in application mode)
      */
-    readonly userId?: string_user_id;
+    readonly userId: string_user_id | null;
     /**
      * Configuration for the LLM tools
      */

package/esm/typings/src/llm-providers/remote/interfaces/PromptbookServer_Prompt_Response.d.ts

@@ -4,9 +4,9 @@ import type { PromptResult } from '../../../execution/PromptResult';
  *
  * This is sent from server to client when the generated text is completed
  */
-export interface PromptbookServer_Prompt_Response {
+export type PromptbookServer_Prompt_Response = {
     /**
      * The result of the prompt
      */
-    promptResult: PromptResult;
-}
+    readonly promptResult: PromptResult;
+};

package/esm/typings/src/llm-providers/remote/interfaces/RemoteLlmExecutionToolsOptions.d.ts

@@ -3,12 +3,13 @@ import type { string_base_url } from '../../../types/typeAliases';
 import type { string_uri } from '../../../types/typeAliases';
 import type { string_user_id } from '../../../types/typeAliases';
 import type { LlmToolsConfiguration } from '../../_common/register/LlmToolsConfiguration';
+import type { CollectionRemoteServerClientOptions } from './RemoteServerOptions';
 /**
  * Options for `RemoteLlmExecutionTools`
  *
  * @public exported from `@promptbook/remote-client`
  */
-export type RemoteLlmExecutionToolsOptions = CommonToolsOptions & {
+export type RemoteLlmExecutionToolsOptions<TCustomOptions> = CommonToolsOptions & {
     /**
      * URL of the remote PROMPTBOOK server
      * On this server will be connected to the socket.io server
@@ -29,7 +30,7 @@ export type RemoteLlmExecutionToolsOptions = CommonToolsOptions & {
     /**
      * Use anonymous server with anonymous mode
      */
-    isAnonymous: true;
+    readonly isAnonymous: true;
     /**
      * Configuration for the LLM tools
      */
@@ -37,22 +38,16 @@ export type RemoteLlmExecutionToolsOptions = CommonToolsOptions & {
     /**
      * Identifier of the end user
      *
-     * Note: this is passed to the certain model providers to identify misuse
-     * Note: In anonymous mode it is not required to identify
+     * Note: This is passed to the certain model providers to identify misuse
+     * Note: In anonymous mode, there is no need to identify yourself, nor does it change the actual configuration of LLM Tools (unlike in application mode).
      */
-    readonly userId?: string_user_id;
-} | {
+    readonly userId: string_user_id | null;
+} | ({
     /**
      * Use anonymous server with client identification and fixed collection
      */
-    isAnonymous: false;
-    /**
-     * Identifier of the end user
-     *
-     * Note: this is passed to the certain model providers to identify misuse
-     */
-    readonly userId: string_user_id;
-});
+    readonly isAnonymous: false;
+} & CollectionRemoteServerClientOptions<TCustomOptions>));
 /**
  * TODO: [🧠][🛍] Maybe not `isAnonymous: boolean` BUT `mode: 'ANONYMOUS'|'COLLECTION'`
  * TODO: [🧠][🧜‍♂️] Maybe join remoteUrl and path into single value

package/esm/typings/src/llm-providers/remote/interfaces/RemoteServerOptions.d.ts

@@ -1,6 +1,8 @@
+import type { Promisable } from 'type-fest';
 import type { PipelineCollection } from '../../../collection/PipelineCollection';
 import type { CommonToolsOptions } from '../../../execution/CommonToolsOptions';
 import type { LlmExecutionTools } from '../../../execution/LlmExecutionTools';
+import type { string_app_id } from '../../../types/typeAliases';
 import type { string_uri } from '../../../types/typeAliases';
 import type { string_user_id } from '../../../types/typeAliases';
 /**
@@ -8,7 +10,7 @@ import type { string_user_id } from '../../../types/typeAliases';
  *
  * There are two modes of remote server:
  *
- * 1) **Collection mode** Server will recieve `collection` and execute prompts only from this collection
+ * 1) **Application mode** Server will recieve `collection` and execute prompts only from this collection
  * 2) **Anonymous mode** Server will recieve full `LlmToolsConfiguration` (with api keys) and just acts as a proxy
  *    In anonymous mode, `collection` will be ignored and any prompt will be executed
  *
@@ -17,7 +19,7 @@ import type { string_user_id } from '../../../types/typeAliases';
  * @public exported from `@promptbook/remote-client`
  * @public exported from `@promptbook/remote-server`
  */
-export type RemoteServerOptions = CommonToolsOptions & {
+export type RemoteServerOptions<TCustomOptions> = CommonToolsOptions & {
     /**
      * Port on which the server will listen
      */
@@ -29,18 +31,18 @@ export type RemoteServerOptions = CommonToolsOptions & {
      * @example '/promptbook/socket.io'
      */
     readonly path: string_uri;
-} & (AnonymousRemoteServerOptions | CollectionRemoteServerOptions | (AnonymousRemoteServerOptions & CollectionRemoteServerOptions));
+} & (AnonymousRemoteServerOptions | CollectionRemoteServerOptions<TCustomOptions> | (AnonymousRemoteServerOptions & CollectionRemoteServerOptions<TCustomOptions>));
 export type AnonymousRemoteServerOptions = {
     /**
      * Enable anonymous mode
      */
     readonly isAnonymousModeAllowed: true;
 };
-export type CollectionRemoteServerOptions = {
+export type CollectionRemoteServerOptions<TCustomOptions> = {
     /**
-     * Enable collection mode
+     * Enable application mode
      */
-    readonly isCollectionModeAllowed: true;
+    readonly isApplicationModeAllowed: true;
     /**
      * Promptbook collection to use
      *
@@ -50,9 +52,22 @@ export type CollectionRemoteServerOptions = {
     /**
      * Creates llm execution tools for each client
      */
-    createLlmExecutionTools(userId: string_user_id | undefined): LlmExecutionTools;
+    createLlmExecutionTools(options: CollectionRemoteServerClientOptions<TCustomOptions>): Promisable<LlmExecutionTools>;
+};
+export type CollectionRemoteServerClientOptions<TCustomOptions> = {
+    /**
+     * @@@
+     */
+    readonly appId: string_app_id | null;
+    /**
+     * @@@
+     */
+    readonly userId: string_user_id | null;
+    /**
+     * @@@
+     */
+    readonly customOptions?: TCustomOptions;
 };
 /**
  * TODO: Constrain anonymous mode for specific models / providers
- * TODO: [🧠][🤺] Remove `createLlmExecutionTools`, pass just `llmExecutionTools`
  */

package/esm/typings/src/llm-providers/remote/startRemoteServer.d.ts

@@ -9,7 +9,7 @@ import type { RemoteServerOptions } from './interfaces/RemoteServerOptions';
  * @see https://github.com/webgptorg/promptbook#remote-server
  * @public exported from `@promptbook/remote-server`
  */
-export declare function startRemoteServer(options: RemoteServerOptions): IDestroyable;
+export declare function startRemoteServer<TCustomOptions = undefined>(options: RemoteServerOptions<TCustomOptions>): IDestroyable;
 /**
  * TODO: Maybe use `$asDeeplyFrozenSerializableJson`
  * TODO: [🧠][🛍] Maybe not `isAnonymous: boolean` BUT `mode: 'ANONYMOUS'|'COLLECTION'`

package/esm/typings/src/personas/preparePersona.d.ts

@@ -10,7 +10,7 @@ import type { string_persona_description } from '../types/typeAliases';
  */
 export declare function preparePersona(personaDescription: string_persona_description, tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions): Promise<PersonaPreparedJson['modelRequirements']>;
 /**
- * TODO: [🔃][main] !!!!! If the persona was prepared with different version or different set of models, prepare it once again
+ * TODO: [🔃][main] !! If the persona was prepared with different version or different set of models, prepare it once again
  * TODO: [🏢] !! Check validity of `modelName` in pipeline
  * TODO: [🏢] !! Check validity of `systemMessage` in pipeline
  * TODO: [🏢] !! Check validity of `temperature` in pipeline

package/esm/typings/src/prepare/PrepareAndScrapeOptions.d.ts

@@ -1,3 +1,4 @@
+import type { IntermediateFilesStrategy } from '../types/IntermediateFilesStrategy';
 import type { string_dirname } from '../types/typeAliases';
 /**
  * Options for preparation of the pipeline
@@ -21,37 +22,17 @@ export type PrepareAndScrapeOptions = {
      */
     readonly cacheDirname?: string_dirname;
     /**
-     * If true, the cache is cleaned after the scraping
+     * Strategy for caching the intermediate results for knowledge source
      *
-     *
-     * @default false // <- TODO: !!!!!! Change to `cacheStrategy`/`intermediateFiles`, Put to global config, change to `true` and explicitly set to `false` in all playgrounds
+     * @default DEFAULT_INTERMEDIATE_FILES_STRATEGY
      */
-    readonly isCacheCleaned?: boolean;
+    readonly intermediateFilesStrategy?: IntermediateFilesStrategy;
     /**
      * Maximum number of tasks running in parallel
      *
      * @default MAX_PARALLEL_COUNT
      */
     readonly maxParallelCount?: number;
-    /**
-     * Path to the external programs executables
-     *
-     * TODO: !!!!!! Transform to scrapers and make them Classy
-     */
-    readonly externalProgramsPaths?: {
-        /**
-         * Path to the `pandoc` executable
-         *
-         * @example 'C:/Users/me/AppData/Local/Pandoc/pandoc.exe'
-         */
-        readonly pandocPath?: string;
-        /**
-         * Path to the LibreOffice executable
-         *
-         * @example 'C:/Program Files/LibreOffice/program/swriter.exe'
-         */
-        readonly libreOfficePath?: string;
-    };
     /**
      * If true, the missing software is automatically installed
      */

package/esm/typings/src/prepare/isPipelinePrepared.d.ts

@@ -6,11 +6,11 @@ import type { PipelineJson } from '../types/PipelineJson/PipelineJson';
  */
 export declare function isPipelinePrepared(pipeline: PipelineJson): boolean;
 /**
- * TODO: [🔃][main] !!!!! If the pipeline was prepared with different version or different set of models, prepare it once again
+ * TODO: [🔃][main] !! If the pipeline was prepared with different version or different set of models, prepare it once again
  * TODO: [🐠] Maybe base this on `makeValidator`
  * TODO: [🧊] Pipeline can be partially prepared, this should return true ONLY if fully prepared
  * TODO: [🧿] Maybe do same process with same granularity and subfinctions as `preparePipeline`
  *     - [🏍] ? Is context in each template
- *     - [♨] Are samples prepared
+ *     - [♨] Are examples prepared
  *     - [♨] Are templates prepared
  */

package/esm/typings/src/prepare/prepareTemplates.d.ts

@@ -12,7 +12,7 @@ type PreparedTemplates = {
     /**
      * @@@ Sequence of templates that are chained together to form a pipeline
      */
-    readonly templatesPrepared: Array<TemplateJson>;
+    readonly templatesPrepared: ReadonlyArray<TemplateJson>;
 };
 /**
  * @@@
@@ -24,7 +24,7 @@ export {};
 /**
  * TODO: [🧠] Add context to each template (if missing)
  * TODO: [🧠] What is better name `prepareTemplate` or `prepareTemplateAndParameters`
- * TODO: [♨][main] !!! Prepare index the samples and maybe templates
+ * TODO: [♨][main] !!! Prepare index the examples and maybe templates
  * TODO: Write tests for `preparePipeline`
  * TODO: [🏏] Leverage the batch API and build queues @see https://platform.openai.com/docs/guides/batch
  * TODO: [🧊] In future one preparation can take data from previous preparation and save tokens and time

package/esm/typings/src/scrapers/_common/Scraper.d.ts

@@ -17,7 +17,7 @@ export type Scraper = {
     /**
      * Scrapes the markdown file and returns the knowledge pieces or `null` if it can't scrape it
      */
-    scrape(source: ScraperSourceHandler): Promisable<Array<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+    scrape(source: ScraperSourceHandler): Promisable<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
 };
 /**
  * @@@
@@ -51,16 +51,9 @@ export type ScraperSourceHandler = {
      * Get the content as a utf-8 string
      */
     asText(): Promisable<string>;
-    /**
-     * Get the content as a blob
-     */
-    asBlob(): Promisable<Blob>;
 };
 /**
+ * TODO: [🥽] Add ` asBlob(): Promisable<Blob>;` or asFile
  * TODO: [🐝] @@@ Annotate all
  * TODO: [🔼] Export via types
  */
-/**
- * TODO: !!!!!! Test that this is catched
- * Note: [⚫] Code in this file should never be published in any package
- */

package/esm/typings/src/scrapers/_common/prepareKnowledgePieces.d.ts

@@ -8,7 +8,7 @@ import type { ExecutionTools } from '../../execution/ExecutionTools';
  * @see https://github.com/webgptorg/promptbook/discussions/41
  * @public exported from `@promptbook/core`
  */
-export declare function prepareKnowledgePieces(knowledgeSources: Array<KnowledgeSourceJson>, tools: Pick<ExecutionTools, 'llm' | 'fs' | 'scrapers'>, options: PrepareAndScrapeOptions): Promise<Array<Omit<KnowledgePiecePreparedJson, 'preparationIds'>>>;
+export declare function prepareKnowledgePieces(knowledgeSources: ReadonlyArray<KnowledgeSourceJson>, tools: Pick<ExecutionTools, 'llm' | 'fs' | 'scrapers'>, options: PrepareAndScrapeOptions): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'preparationIds'>>>;
 /**
  * TODO: [🧊] In future one preparation can take data from previous preparation and save tokens and time
  *       Put `knowledgePieces` into `PrepareKnowledgeOptions`

package/esm/typings/src/scrapers/_common/register/$provideScrapersForBrowser.d.ts

@@ -2,11 +2,11 @@ import type { ExecutionTools } from '../../../execution/ExecutionTools';
 import type { PrepareAndScrapeOptions } from '../../../prepare/PrepareAndScrapeOptions';
 import type { Scraper } from '../Scraper';
 /**
- * @@@!!!!!!
+ * @@@
  *
  * 1) @@@
  * 2) @@@
  *
  * @public exported from `@promptbook/browser`
  */
-export declare function $provideScrapersForBrowser(tools: Pick<ExecutionTools, 'llm'>, options?: PrepareAndScrapeOptions): Promise<Array<Scraper>>;
+export declare function $provideScrapersForBrowser(tools: Pick<ExecutionTools, 'llm'>, options?: PrepareAndScrapeOptions): Promise<ReadonlyArray<Scraper>>;

package/esm/typings/src/scrapers/_common/register/$provideScrapersForNode.d.ts

@@ -2,14 +2,14 @@ import type { ExecutionTools } from '../../../execution/ExecutionTools';
 import type { PrepareAndScrapeOptions } from '../../../prepare/PrepareAndScrapeOptions';
 import type { Scraper } from '../Scraper';
 /**
- * !!!!!!
+ * @@@
  *
  * 1) @@@
  * 2) @@@
  *
  * @public exported from `@promptbook/node`
  */
-export declare function $provideScrapersForNode(tools: Pick<ExecutionTools, 'fs' | 'llm'>, options?: PrepareAndScrapeOptions): Promise<Array<Scraper>>;
+export declare function $provideScrapersForNode(tools: Pick<ExecutionTools, 'fs' | 'llm' | 'executables'>, options?: PrepareAndScrapeOptions): Promise<ReadonlyArray<Scraper>>;
 /**
  * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
  */

package/esm/typings/src/scrapers/_common/register/ScraperAndConverterMetadata.d.ts

@@ -1,8 +1,8 @@
+import type { Executables } from '../../../execution/Executables';
 import type { string_mime_type } from '../../../types/typeAliases';
 import type { string_promptbook_documentation_url } from '../../../types/typeAliases';
 import type { string_title } from '../../../types/typeAliases';
 import type { Registered } from '../../../utils/$Register';
-import type { TODO_any } from '../../../utils/organization/TODO_any';
 /**
  * @@@
  *
@@ -33,7 +33,7 @@ export type ScraperAndConverterMetadata = Registered & {
     /**
      * @@@
      */
-    readonly requiredExecutables: TODO_any;
+    readonly requiredExecutables: ReadonlyArray<Capitalize<keyof Executables extends `${infer N}Path` ? N : never>>;
     /**
      * Link to documentation
      */

package/esm/typings/src/scrapers/_common/utils/getScraperIntermediateSource.d.ts

@@ -13,7 +13,7 @@ type GetScraperIntermediateSourceSource = Pick<ScraperSourceHandler, 'filename'
  *
  * @private internal utility of `getScraperIntermediateSource`
  */
-type GetScraperIntermediateSourceHandler = Required<Pick<PrepareAndScrapeOptions, 'rootDirname' | 'cacheDirname' | 'isCacheCleaned' | 'isVerbose'>> & {
+type GetScraperIntermediateSourceHandler = Required<Pick<PrepareAndScrapeOptions, 'rootDirname' | 'cacheDirname' | 'intermediateFilesStrategy' | 'isVerbose'>> & {
     readonly extension: string_file_extension;
 };
 /**
@@ -28,7 +28,7 @@ export {};
 /**
  * Note: Not using `FileCacheStorage` for two reasons:
  * 1) Need to store more than serialized JSONs
- * 2) Need to switch between a `rootDirname` and `cacheDirname` <- TODO: !!!!
+ * 2) Need to switch between a `rootDirname` and `cacheDirname` <- TODO: [😡]
  * TODO: [🐱‍🐉][🧠] Make some smart crop
  * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
  */

package/esm/typings/src/scrapers/_common/utils/makeKnowledgeSourceHandler.d.ts

@@ -6,6 +6,6 @@ import type { ScraperSourceHandler } from '../Scraper';
 /**
  * @@@
  *
- * @private for scraper utilities
+ * @public exported from `@promptbook/core`
  */
 export declare function makeKnowledgeSourceHandler(knowledgeSource: SetOptional<KnowledgeSourceJson, 'name'>, tools: Pick<ExecutionTools, 'fs'>, options?: Pick<PrepareAndScrapeOptions, 'rootDirname' | 'isVerbose'>): Promise<ScraperSourceHandler>;

package/esm/typings/src/scrapers/document/DocumentScraper.d.ts

@@ -23,7 +23,7 @@ export declare class DocumentScraper implements Converter, Scraper {
      * Markdown scraper is used internally
      */
     private readonly markdownScraper;
-    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm'>, options: PrepareAndScrapeOptions);
+    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm' | 'executables'>, options: PrepareAndScrapeOptions);
     /**
      * Convert the `.docx` or `.odt`  to `.md` file and returns intermediate source
      *
@@ -33,7 +33,7 @@ export declare class DocumentScraper implements Converter, Scraper {
     /**
      * Scrapes the docx file and returns the knowledge pieces or `null` if it can't scrape it
      */
-    scrape(source: ScraperSourceHandler): Promise<Array<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+    scrape(source: ScraperSourceHandler): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
 }
 /**
  * TODO: [👣] Converted documents can act as cached items - there is no need to run conversion each time

package/esm/typings/src/scrapers/document/DocumentScraper.test.d.ts

@@ -1,4 +1,4 @@
 export {};
 /**
- * TODO: [📓] Maybe test all file in samples (not just 10-simple.docx)
+ * TODO: [📓] Maybe test all file in examples (not just 10-simple.docx)
  */

package/esm/typings/src/scrapers/document/createDocumentScraper.d.ts

@@ -13,7 +13,7 @@ export declare const createDocumentScraper: ((tools: Pick<ExecutionTools, 'llm'>
     mimeTypes: string[];
     documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
     isAvilableInBrowser: false;
-    requiredExecutables: string[];
+    requiredExecutables: "Pandoc"[];
 }>;
 /**
  * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"

package/esm/typings/src/scrapers/document/register-metadata.d.ts

@@ -11,7 +11,7 @@ export declare const documentScraperMetadata: import("type-fest/source/readonly-
     mimeTypes: string[];
     documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
     isAvilableInBrowser: false;
-    requiredExecutables: string[];
+    requiredExecutables: "Pandoc"[];
 }>;
 /**
  * Registration of known scraper metadata

package/esm/typings/src/scrapers/document-legacy/LegacyDocumentScraper.d.ts

@@ -7,7 +7,7 @@ import type { Scraper } from '../_common/Scraper';
 import type { ScraperSourceHandler } from '../_common/Scraper';
 import type { ScraperIntermediateSource } from '../_common/ScraperIntermediateSource';
 /**
- * Scraper for .docx files
+ * Scraper for old document files (like .doc and .rtf)
  *
  * @see `documentationUrl` for more details
  * @public exported from `@promptbook/legacy-documents`
@@ -23,7 +23,7 @@ export declare class LegacyDocumentScraper implements Converter, Scraper {
      * Document scraper is used internally
      */
     private readonly documentScraper;
-    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm'>, options: PrepareAndScrapeOptions);
+    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm' | 'executables'>, options: PrepareAndScrapeOptions);
     /**
      * Convert the `.doc` or `.rtf`  to `.doc` file and returns intermediate source
      *
@@ -33,7 +33,7 @@ export declare class LegacyDocumentScraper implements Converter, Scraper {
     /**
      * Scrapes the `.doc` or `.rtf` file and returns the knowledge pieces or `null` if it can't scrape it
      */
-    scrape(source: ScraperSourceHandler): Promise<Array<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+    scrape(source: ScraperSourceHandler): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
 }
 /**
  * TODO: [👣] Converted documents can act as cached items - there is no need to run conversion each time

package/esm/typings/src/scrapers/document-legacy/LegacyDocumentScraper.test.d.ts

@@ -1,4 +1,4 @@
 export {};
 /**
- * TODO: [📓] Maybe test all file in samples (not just 10-simple.doc)
+ * TODO: [📓] Maybe test all file in examples (not just 10-simple.doc)
  */

package/esm/typings/src/scrapers/document-legacy/createLegacyDocumentScraper.d.ts

@@ -13,7 +13,7 @@ export declare const createLegacyDocumentScraper: ((tools: Pick<ExecutionTools,
     mimeTypes: string[];
     documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
     isAvilableInBrowser: false;
-    requiredExecutables: string[];
+    requiredExecutables: ("Pandoc" | "LibreOffice")[];
 }>;
 /**
  * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"

package/esm/typings/src/scrapers/document-legacy/register-metadata.d.ts

@@ -11,7 +11,7 @@ export declare const legacyDocumentScraperMetadata: import("type-fest/source/rea
     mimeTypes: string[];
     documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
     isAvilableInBrowser: false;
-    requiredExecutables: string[];
+    requiredExecutables: ("Pandoc" | "LibreOffice")[];
 }>;
 /**
  * Registration of known scraper metadata