@promptbook/javascript 0.88.0-1

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

@@ -0,0 +1,34 @@
+import type { PrepareAndScrapeOptions } from '../../../prepare/PrepareAndScrapeOptions';
+import type { string_file_extension } from '../../../types/typeAliases';
+import type { ScraperSourceHandler } from '../Scraper';
+import type { ScraperIntermediateSource } from '../ScraperIntermediateSource';
+/**
+ * @@@
+ *
+ * @private internal utility of `getScraperIntermediateSource`
+ */
+type GetScraperIntermediateSourceSource = Pick<ScraperSourceHandler, 'filename' | 'url'>;
+/**
+ * @@@
+ *
+ * @private internal utility of `getScraperIntermediateSource`
+ */
+type GetScraperIntermediateSourceHandler = Required<Pick<PrepareAndScrapeOptions, 'rootDirname' | 'cacheDirname' | 'intermediateFilesStrategy' | 'isVerbose'>> & {
+    readonly extension: string_file_extension;
+};
+/**
+ * Create a filename for intermediate cache for scrapers
+ *
+ * Note: It also checks if directory exists and creates it if not
+ *
+ * @private as internal utility for scrapers
+ */
+export declare function getScraperIntermediateSource(source: GetScraperIntermediateSourceSource, options: GetScraperIntermediateSourceHandler): Promise<ScraperIntermediateSource>;
+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: [😡]
+ * 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/getScraperIntermediateSource.test.d.ts

@@ -0,0 +1,4 @@
+export {};
+/**
+ * TODO: [🐱‍🐉][🧠] Make some smart crop
+ */

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

@@ -0,0 +1,11 @@
+import type { SetOptional } from 'type-fest';
+import type { ExecutionTools } from '../../../execution/ExecutionTools';
+import type { KnowledgeSourceJson } from '../../../pipeline/PipelineJson/KnowledgeSourceJson';
+import type { PrepareAndScrapeOptions } from '../../../prepare/PrepareAndScrapeOptions';
+import type { ScraperSourceHandler } from '../Scraper';
+/**
+ * @@@
+ *
+ * @public exported from `@promptbook/core`
+ */
+export declare function makeKnowledgeSourceHandler(knowledgeSource: SetOptional<KnowledgeSourceJson, 'name'>, tools: Pick<ExecutionTools, 'fs' | 'fetch'>, options?: Pick<PrepareAndScrapeOptions, 'rootDirname' | 'isVerbose'>): Promise<ScraperSourceHandler>;

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

@@ -0,0 +1,10 @@
+import type { PromptbookFetch } from '../../../execution/PromptbookFetch';
+/**
+ * The built-in `fetch' function with a lightweight error handling wrapper as default fetch function used in Promptbook scrapers
+ *
+ * @private as default `fetch` function used in Promptbook scrapers
+ */
+export declare const scraperFetch: PromptbookFetch;
+/**
+ * TODO: [🧠] Maybe rename because it is not used only for scrapers but also in `$getCompiledBook`
+ */

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

@@ -0,0 +1,43 @@
+import type { KnowledgePiecePreparedJson } from '../../pipeline/PipelineJson/KnowledgePieceJson';
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import type { Converter } from '../_common/Converter';
+import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
+import type { Scraper } from '../_common/Scraper';
+import type { ScraperSourceHandler } from '../_common/Scraper';
+import type { ScraperIntermediateSource } from '../_common/ScraperIntermediateSource';
+/**
+ * Scraper of .docx and .odt files
+ *
+ * @see `documentationUrl` for more details
+ * @public exported from `@promptbook/documents`
+ */
+export declare class DocumentScraper implements Converter, Scraper {
+    private readonly tools;
+    private readonly options;
+    /**
+     * Metadata of the scraper which includes title, mime types, etc.
+     */
+    get metadata(): ScraperAndConverterMetadata;
+    /**
+     * Markdown scraper is used internally
+     */
+    private readonly markdownScraper;
+    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm' | 'executables'>, options: PrepareAndScrapeOptions);
+    /**
+     * Convert the `.docx` or `.odt`  to `.md` file and returns intermediate source
+     *
+     * Note: `$` is used to indicate that this function is not a pure function - it leaves files on the disk and you are responsible for cleaning them by calling `destroy` method of returned object
+     */
+    $convert(source: ScraperSourceHandler): Promise<ScraperIntermediateSource>;
+    /**
+     * Scrapes the docx file and returns the knowledge pieces or `null` if it can't scrape it
+     */
+    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
+ * TODO: [🪂] Do it in parallel 11:11
+ * Note: No need to aggregate usage here, it is done by intercepting the llmTools
+ * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
+ */

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

@@ -0,0 +1,4 @@
+export {};
+/**
+ * TODO: [📓] Maybe test all file in examples (not just 10-simple.docx)
+ */

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

@@ -0,0 +1,20 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { DocumentScraper } from './DocumentScraper';
+/**
+ * @@@
+ *
+ * @public exported from `@promptbook/documents`
+ */
+export declare const createDocumentScraper: ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => DocumentScraper) & import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvilableInBrowser: false;
+    requiredExecutables: "Pandoc"[];
+}>;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

package/esm/typings/src/scrapers/document/playground/document-scraper-playground.d.ts

@@ -0,0 +1,5 @@
+#!/usr/bin/env ts-node
+export {};
+/**
+ * Note: [⚫] Code in this file should never be published in any package
+ */

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

@@ -0,0 +1,15 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Registration of known scraper
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/documents`
+ * @public exported from `@promptbook/wizzard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _DocumentScraperRegistration: Registration;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

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

@@ -0,0 +1,28 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Metadata of the scraper
+ *
+ * @private within the scraper directory
+ */
+export declare const documentScraperMetadata: import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvilableInBrowser: false;
+    requiredExecutables: "Pandoc"[];
+}>;
+/**
+ * Registration of known scraper metadata
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/core`
+ * @public exported from `@promptbook/wizzard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _DocumentScraperMetadataRegistration: Registration;
+/**
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

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

@@ -0,0 +1,43 @@
+import type { KnowledgePiecePreparedJson } from '../../pipeline/PipelineJson/KnowledgePieceJson';
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import type { Converter } from '../_common/Converter';
+import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
+import type { Scraper } from '../_common/Scraper';
+import type { ScraperSourceHandler } from '../_common/Scraper';
+import type { ScraperIntermediateSource } from '../_common/ScraperIntermediateSource';
+/**
+ * Scraper for old document files (like .doc and .rtf)
+ *
+ * @see `documentationUrl` for more details
+ * @public exported from `@promptbook/legacy-documents`
+ */
+export declare class LegacyDocumentScraper implements Converter, Scraper {
+    private readonly tools;
+    private readonly options;
+    /**
+     * Metadata of the scraper which includes title, mime types, etc.
+     */
+    get metadata(): ScraperAndConverterMetadata;
+    /**
+     * Document scraper is used internally
+     */
+    private readonly documentScraper;
+    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm' | 'executables'>, options: PrepareAndScrapeOptions);
+    /**
+     * Convert the `.doc` or `.rtf`  to `.doc` file and returns intermediate source
+     *
+     * Note: `$` is used to indicate that this function is not a pure function - it leaves files on the disk and you are responsible for cleaning them by calling `destroy` method of returned object
+     */
+    $convert(source: ScraperSourceHandler): Promise<ScraperIntermediateSource>;
+    /**
+     * Scrapes the `.doc` or `.rtf` file and returns the knowledge pieces or `null` if it can't scrape it
+     */
+    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
+ * TODO: [🪂] Do it in parallel 11:11
+ * Note: No need to aggregate usage here, it is done by intercepting the llmTools
+ * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
+ */

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

@@ -0,0 +1,4 @@
+export {};
+/**
+ * TODO: [📓] Maybe test all file in examples (not just 10-simple.doc)
+ */

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

@@ -0,0 +1,20 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { LegacyDocumentScraper } from './LegacyDocumentScraper';
+/**
+ * @@@
+ *
+ * @public exported from `@promptbook/legacy-documents`
+ */
+export declare const createLegacyDocumentScraper: ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => LegacyDocumentScraper) & import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvilableInBrowser: false;
+    requiredExecutables: ("Pandoc" | "LibreOffice")[];
+}>;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

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

@@ -0,0 +1,5 @@
+#!/usr/bin/env ts-node
+export {};
+/**
+ * Note: [⚫] Code in this file should never be published in any package
+ */

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

@@ -0,0 +1,15 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Registration of known scraper
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/legacy-documents`
+ * @public exported from `@promptbook/wizzard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _LegacyDocumentScraperRegistration: Registration;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

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

@@ -0,0 +1,28 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Metadata of the scraper
+ *
+ * @private within the scraper directory
+ */
+export declare const legacyDocumentScraperMetadata: import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvilableInBrowser: false;
+    requiredExecutables: ("Pandoc" | "LibreOffice")[];
+}>;
+/**
+ * Registration of known scraper metadata
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/core`
+ * @public exported from `@promptbook/wizzard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _LegacyDocumentScraperMetadataRegistration: Registration;
+/**
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/markdown/MarkdownScraper.d.ts

@@ -0,0 +1,29 @@
+import type { KnowledgePiecePreparedJson } from '../../pipeline/PipelineJson/KnowledgePieceJson';
+import type { Scraper } from '../_common/Scraper';
+import type { ScraperSourceHandler } from '../_common/Scraper';
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
+/**
+ * Scraper for markdown files
+ *
+ * @see `documentationUrl` for more details
+ * @public exported from `@promptbook/markdown-utils`
+ */
+export declare class MarkdownScraper implements Scraper {
+    private readonly tools;
+    private readonly options;
+    /**
+     * Metadata of the scraper which includes title, mime types, etc.
+     */
+    get metadata(): ScraperAndConverterMetadata;
+    constructor(tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions);
+    /**
+     * Scrapes the markdown file and returns the knowledge pieces or `null` if it can't scrape it
+     */
+    scrape(source: ScraperSourceHandler): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+}
+/**
+ * TODO: [🪂] Do it in parallel 11:11
+ * Note: No need to aggregate usage here, it is done by intercepting the llmTools
+ */

package/esm/typings/src/scrapers/markdown/MarkdownScraper.test.d.ts

@@ -0,0 +1,4 @@
+export {};
+/**
+ * TODO: [📓] Maybe test all file in examples (not just 10-simple.md)
+ */

package/esm/typings/src/scrapers/markdown/createMarkdownScraper.d.ts

@@ -0,0 +1,20 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { MarkdownScraper } from './MarkdownScraper';
+/**
+ * @@@
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ */
+export declare const createMarkdownScraper: ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => MarkdownScraper) & import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvilableInBrowser: true;
+    requiredExecutables: never[];
+}>;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

package/esm/typings/src/scrapers/markdown/playground/markdown-scraper-playground.d.ts

@@ -0,0 +1,5 @@
+#!/usr/bin/env ts-node
+export {};
+/**
+ * Note: [⚫] Code in this file should never be published in any package
+ */

package/esm/typings/src/scrapers/markdown/register-constructor.d.ts

@@ -0,0 +1,15 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Registration of known scraper
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ * @public exported from `@promptbook/wizzard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _MarkdownScraperRegistration: Registration;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

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

@@ -0,0 +1,28 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Metadata of the scraper
+ *
+ * @private within the scraper directory
+ */
+export declare const markdownScraperMetadata: import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvilableInBrowser: true;
+    requiredExecutables: never[];
+}>;
+/**
+ * Registration of known scraper metadata
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/core`
+ * @public exported from `@promptbook/wizzard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _MarkdownScraperMetadataRegistration: Registration;
+/**
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/markitdown/MarkitdownScraper.d.ts

@@ -0,0 +1,50 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { KnowledgePiecePreparedJson } from '../../pipeline/PipelineJson/KnowledgePieceJson';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import type { Converter } from '../_common/Converter';
+import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
+import type { Scraper } from '../_common/Scraper';
+import type { ScraperSourceHandler } from '../_common/Scraper';
+import type { ScraperIntermediateSource } from '../_common/ScraperIntermediateSource';
+/**
+ * Integration of Markitdown by Microsoft into Promptbook
+ *
+ * @see https://github.com/microsoft/markitdown
+ * @see `documentationUrl` for more details
+ * @public exported from `@promptbook/markitdown`
+ * @public exported from `@promptbook/pdf`
+ */
+export declare class MarkitdownScraper implements Converter, Scraper {
+    private readonly tools;
+    private readonly options;
+    /**
+     * Metadata of the scraper which includes title, mime types, etc.
+     */
+    get metadata(): ScraperAndConverterMetadata;
+    /**
+     * Markdown scraper is used internally
+     */
+    private readonly markdownScraper;
+    /**
+     * Markdown scraper is used internally
+     */
+    private readonly markitdown;
+    constructor(tools: Pick<ExecutionTools, 'fs' | 'llm' | 'executables'>, options: PrepareAndScrapeOptions);
+    /**
+     * Convert the documents to `.md` file and returns intermediate source
+     *
+     * Note: `$` is used to indicate that this function is not a pure function - it leaves files on the disk and you are responsible for cleaning them by calling `destroy` method of returned object
+     */
+    $convert(source: ScraperSourceHandler): Promise<ScraperIntermediateSource>;
+    /**
+     * Scrapes the docx file and returns the knowledge pieces or `null` if it can't scrape it
+     */
+    scrape(source: ScraperSourceHandler): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+}
+/**
+ * TODO: [🧠][🌜] Export only from `@promptbook/markitdown` or `@promptbook/pdf` NOT both
+ * TODO: [👣] Converted documents can act as cached items - there is no need to run conversion each time
+ * TODO: [🪂] Do it in parallel
+ * Note: No need to aggregate usage here, it is done by intercepting the llmTools
+ * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
+ */

package/esm/typings/src/scrapers/markitdown/createMarkitdownScraper.d.ts

@@ -0,0 +1,22 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { MarkitdownScraper } from './MarkitdownScraper';
+/**
+ * Constructor of `MarkitdownScraper`
+ *
+ * @public exported from `@promptbook/markitdown`
+ * @public exported from `@promptbook/pdf`
+ */
+export declare const createMarkitdownScraper: ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => MarkitdownScraper) & import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvilableInBrowser: false;
+    requiredExecutables: never[];
+}>;
+/**
+ * TODO: [🧠][🌜] Export only from `@promptbook/markitdown` or `@promptbook/pdf` NOT both
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

package/esm/typings/src/scrapers/markitdown/playground/markitdown-scraper-playground.d.ts

@@ -0,0 +1,5 @@
+#!/usr/bin/env ts-node
+export {};
+/**
+ * Note: [⚫] Code in this file should never be published in any package
+ */

package/esm/typings/src/scrapers/markitdown/register-constructor.d.ts

@@ -0,0 +1,17 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Registration of known scraper
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/markitdown`
+ * @public exported from `@promptbook/pdf`
+ * @public exported from `@promptbook/wizzard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _MarkitdownScraperRegistration: Registration;
+/**
+ * TODO: [🧠][🌜] Export only from `@promptbook/markitdown` or `@promptbook/pdf` NOT both
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

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

@@ -0,0 +1,28 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Metadata of the scraper
+ *
+ * @private within the scraper directory
+ */
+export declare const markitdownScraperMetadata: import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvilableInBrowser: false;
+    requiredExecutables: never[];
+}>;
+/**
+ * Registration of known scraper metadata
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/core`
+ * @public exported from `@promptbook/wizzard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _MarkitdownScraperMetadataRegistration: Registration;
+/**
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */

package/esm/typings/src/scrapers/pdf/PdfScraper.d.ts

@@ -0,0 +1,41 @@
+import type { KnowledgePiecePreparedJson } from '../../pipeline/PipelineJson/KnowledgePieceJson';
+import type { Scraper } from '../_common/Scraper';
+import type { ScraperSourceHandler } from '../_common/Scraper';
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import type { Converter } from '../_common/Converter';
+import type { ScraperAndConverterMetadata } from '../_common/register/ScraperAndConverterMetadata';
+import type { ScraperIntermediateSource } from '../_common/ScraperIntermediateSource';
+/**
+ * Scraper for .pdf files
+ *
+ * @see `documentationUrl` for more details
+ * @public exported from `@promptbook/pdf`
+ */
+export declare class PdfScraper implements Converter, Scraper {
+    private readonly tools;
+    private readonly options;
+    /**
+     * Metadata of the scraper which includes title, mime types, etc.
+     */
+    get metadata(): ScraperAndConverterMetadata;
+    /**
+     * Markdown scraper is used internally
+     */
+    private readonly markdownScraper;
+    constructor(tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions);
+    /**
+     * Converts the `.pdf` file to `.md` file and returns intermediate source
+     */
+    $convert(source: ScraperSourceHandler): Promise<ScraperIntermediateSource>;
+    /**
+     * Scrapes the `.pdf` file and returns the knowledge pieces or `null` if it can't scrape it
+     */
+    scrape(source: ScraperSourceHandler): Promise<ReadonlyArray<Omit<KnowledgePiecePreparedJson, 'sources' | 'preparationIds'>> | null>;
+}
+/**
+ * TODO: [👣] Converted pdf documents can act as cached items - there is no need to run conversion each time
+ * TODO: [🪂] Do it in parallel 11:11
+ * Note: No need to aggregate usage here, it is done by intercepting the llmTools
+ * Note: [🟢] Code in this file should never be never released in packages that could be imported into browser environment
+ */

package/esm/typings/src/scrapers/pdf/createPdfScraper.d.ts

@@ -0,0 +1,20 @@
+import type { ExecutionTools } from '../../execution/ExecutionTools';
+import type { PrepareAndScrapeOptions } from '../../prepare/PrepareAndScrapeOptions';
+import { PdfScraper } from './PdfScraper';
+/**
+ * @@@
+ *
+ * @public exported from `@promptbook/pdf`
+ */
+export declare const createPdfScraper: ((tools: Pick<ExecutionTools, 'llm'>, options: PrepareAndScrapeOptions) => PdfScraper) & import("type-fest/source/readonly-deep").ReadonlyObjectDeep<{
+    title: string;
+    packageName: string;
+    className: string;
+    mimeTypes: string[];
+    documentationUrl: "https://github.com/webgptorg/promptbook/discussions/@@";
+    isAvilableInBrowser: false;
+    requiredExecutables: never[];
+}>;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ */

package/esm/typings/src/scrapers/pdf/playground/pdf-scraper-playground.d.ts

@@ -0,0 +1,5 @@
+#!/usr/bin/env ts-node
+export {};
+/**
+ * Note: [⚫] Code in this file should never be published in any package
+ */

package/esm/typings/src/scrapers/pdf/register-constructor.d.ts

@@ -0,0 +1,15 @@
+import type { Registration } from '../../utils/$Register';
+/**
+ * Registration of known scraper
+ *
+ * Warning: This is not useful for the end user, it is just a side effect of the mechanism that handles all available known scrapers
+ *
+ * @public exported from `@promptbook/pdf`
+ * @public exported from `@promptbook/wizzard`
+ * @public exported from `@promptbook/cli`
+ */
+export declare const _PdfScraperRegistration: Registration;
+/**
+ * TODO: [🎶] Naming "constructor" vs "creator" vs "factory"
+ * Note: [💞] Ignore a discrepancy between file name and entity name
+ */