@promptbook/markdown-utils 0.100.4-0 → 0.101.0-1

package/esm/index.es.js

@@ -1,7 +1,4 @@
 import spaceTrim, { spaceTrim as spaceTrim$1 } from 'spacetrim';
-import parserHtml from 'prettier/parser-html';
-import parserMarkdown from 'prettier/parser-markdown';
-import { format } from 'prettier/standalone';
 import { randomBytes } from 'crypto';
 import { Subject } from 'rxjs';
 import { forTime } from 'waitasecond';
@@ -26,7 +23,7 @@ const BOOK_LANGUAGE_VERSION = '1.0.0';
  * @generated
  * @see https://github.com/webgptorg/promptbook
  */
-const PROMPTBOOK_ENGINE_VERSION = '0.100.4-0';
+const PROMPTBOOK_ENGINE_VERSION = '0.101.0-1';
 /**
  * TODO: string_promptbook_version should be constrained to the all versions of Promptbook engine
  * Note: [💞] Ignore a discrepancy between file name and entity name
@@ -51,6 +48,8 @@ class ParseError extends Error {
 /**
  * Makes first letter of a string uppercase
  *
+ * Note: [🔂] This function is idempotent.
+ *
  * @public exported from `@promptbook/utils`
  */
 function capitalize(word) {
@@ -642,6 +641,7 @@ function isValidFilePath(filename) {
 /**
  * Tests if given string is valid URL.
  *
+ * Note: [🔂] This function is idempotent.
  * Note: Dataurl are considered perfectly valid.
  * Note: There are two similar functions:
  * - `isValidUrl` which tests any URL
@@ -705,8 +705,18 @@ function validatePipelineString(pipelineString) {
  * @private withing the package because of HUGE size of prettier dependency
  */
 function prettifyMarkdown(content) {
+    // In browser/Next.js environments, just return the original content
+    // since prettier parsers are not available and would cause bundling issues
+    if (typeof window !== 'undefined') {
+        return content;
+    }
     try {
-        return format(content, {
+        // Use dynamic require to avoid static imports that cause bundling issues
+        // This will only work in Node.js environments
+        const prettierStandalone = eval('require')('prettier/standalone');
+        const parserMarkdown = eval('require')('prettier/parser-markdown');
+        const parserHtml = eval('require')('prettier/parser-html');
+        return prettierStandalone.format(content, {
             parser: 'markdown',
             plugins: [parserMarkdown, parserHtml],
             // TODO: DRY - make some import or auto-copy of .prettierrc
@@ -1054,6 +1064,7 @@ function checkSerializableAsJson(options) {
 /**
  * Creates a deep clone of the given object
  *
+ * Note: [🔂] This function is idempotent.
  * Note: This method only works for objects that are fully serializable to JSON and do not contain functions, Dates, or special types.
  *
  * @param objectValue The object to clone.
@@ -2680,7 +2691,7 @@ const LLM_PROVIDER_PROFILES = {
 };
 /**
  * TODO: Refactor this - each profile must be alongside the provider definition
- * TODO: [🕛] Unite `AvatarProfileProps`, `ChatParticipant`, `LlmExecutionTools` +  `LlmToolsMetadata`
+ * TODO: [🕛] Unite `AgentBasicInformation`, `ChatParticipant`, `LlmExecutionTools` +  `LlmToolsMetadata`
  * Note: [💞] Ignore a discrepancy between file name and entity name
  */
 
@@ -2989,6 +3000,8 @@ function $getGlobalScope() {
 /**
  * Normalizes a text string to SCREAMING_CASE (all uppercase with underscores).
  *
+ * Note: [🔂] This function is idempotent.
+ *
  * @param text The text string to be converted to SCREAMING_CASE format.
  * @returns The normalized text in SCREAMING_CASE format.
  * @example 'HELLO_WORLD'
@@ -3044,6 +3057,8 @@ function normalizeTo_SCREAMING_CASE(text) {
 /**
  * Normalizes a text string to snake_case format.
  *
+ * Note: [🔂] This function is idempotent.
+ *
  * @param text The text string to be converted to snake_case format.
  * @returns The normalized text in snake_case format.
  * @example 'hello_world'
@@ -3461,6 +3476,8 @@ for (let i = 0; i < defaultDiacriticsRemovalMap.length; i++) {
 /**
  * Removes diacritic marks (accents) from characters in a string.
  *
+ * Note: [🔂] This function is idempotent.
+ *
  * @param input The string containing diacritics to be normalized.
  * @returns The string with diacritics removed or normalized.
  * @public exported from `@promptbook/utils`
@@ -3615,6 +3632,8 @@ function mimeTypeToExtension(value) {
 /**
  * Removes emojis from a string and fix whitespaces
  *
+ * Note: [🔂] This function is idempotent.
+ *
  * @param text with emojis
  * @returns text without emojis
  * @public exported from `@promptbook/utils`
@@ -4889,6 +4908,7 @@ function countCharacters(text) {
 }
 /**
  * TODO: [🥴] Implement counting in formats - like JSON, CSV, XML,...
+ * TODO: [🧠][✌️] Make some Promptbook-native token system
  */
 
 /**
@@ -4923,6 +4943,7 @@ function countLines(text) {
 }
 /**
  * TODO: [🥴] Implement counting in formats - like JSON, CSV, XML,...
+ * TODO: [🧠][✌️] Make some Promptbook-native token system
  */
 
 /**
@@ -4937,6 +4958,7 @@ function countPages(text) {
 }
 /**
  * TODO: [🥴] Implement counting in formats - like JSON, CSV, XML,...
+ * TODO: [🧠][✌️] Make some Promptbook-native token system
  */
 
 /**
@@ -4949,6 +4971,7 @@ function countParagraphs(text) {
 }
 /**
  * TODO: [🥴] Implement counting in formats - like JSON, CSV, XML,...
+ * TODO: [🧠][✌️] Make some Promptbook-native token system
  */
 
 /**
@@ -4969,6 +4992,7 @@ function countSentences(text) {
 }
 /**
  * TODO: [🥴] Implement counting in formats - like JSON, CSV, XML,...
+ * TODO: [🧠][✌️] Make some Promptbook-native token system
  */
 
 /**
@@ -4985,6 +5009,8 @@ function countWords(text) {
 }
 /**
  * TODO: [🥴] Implement counting in formats - like JSON, CSV, XML,...
+ * TODO: [🧠][✌️] Make some Promptbook-native token system
+ * TODO: [✌️] `countWords` should be just `splitWords(...).length`, and all other counters should use this pattern as well
  */
 
 /**
@@ -6609,6 +6635,8 @@ function createMarkdownChart(options) {
  * Function escapeMarkdownBlock will escape markdown block if needed
  * It is useful when you want have block in block
  *
+ * Note: [🔂] This function is idempotent.
+ *
  * @public exported from `@promptbook/markdown-utils`
  */
 function escapeMarkdownBlock(value) {
@@ -6773,6 +6801,114 @@ function flattenMarkdown(markdown) {
  *       NOW we are working just with markdown string and its good enough
  */
 
+/**
+ * Change ellipsis character to three dots `…` -> `...`
+ *
+ * Note: [🔂] This function is idempotent.
+ * Tip: If you want to do the full cleanup, look for `humanizeAiText` exported `@promptbook/markdown-utils`
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ */
+function humanizeAiTextEllipsis(aiText) {
+    return aiText.replace(/…/g, '...');
+}
+/**
+ * Note: [🏂] This function is not tested by itself but together with other cleanup functions with `humanizeAiText`
+ */
+
+/**
+ * Change em-dashes to regular dashes `—` -> `-`
+ *
+ * Note: [🔂] This function is idempotent.
+ * Tip: If you want to do the full cleanup, look for `humanizeAiText` exported `@promptbook/markdown-utils`
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ */
+function humanizeAiTextEmdashed(aiText) {
+    return aiText.replace(/—/g, '-');
+}
+/**
+ * Note: [🏂] This function is not tested by itself but together with other cleanup functions with `humanizeAiText`
+ */
+
+/**
+ * Change smart quotes to regular quotes
+ *
+ * Note: [🔂] This function is idempotent.
+ * Tip: If you want to do the full cleanup, look for `humanizeAiText` exported `@promptbook/markdown-utils`
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ */
+function humanizeAiTextQuotes(aiText) {
+    return aiText
+        .replace(/[“”]/g, '"')
+        .replace(/[‚‘’]/g, "'")
+        .replace(/«/g, '"')
+        .replace(/»/g, '"')
+        .replace(/„/g, '"')
+        .replace(/‹/g, "'")
+        .replace(/›/g, "'");
+}
+/**
+ * Note: [🏂] This function is not tested by itself but together with other cleanup functions with `humanizeAiText`
+ */
+
+/**
+ * Change unprintable hard spaces to regular spaces
+ *
+ * Note: [🔂] This function is idempotent.
+ * Tip: If you want to do the full cleanup, look for `humanizeAiText` exported `@promptbook/markdown-utils`
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ */
+function humanizeAiTextWhitespace(aiText) {
+    return aiText.replace(/\u00A0/g, ' ');
+}
+/**
+ * Note: [🏂] This function is not tested by itself but together with other cleanup functions with `humanizeAiText`
+ */
+
+/**
+ * Function `humanizeAiText` will remove traces of AI text generation artifacts
+ *
+ * Note: [🔂] This function is idempotent.
+ * Tip: If you want more control, look for other functions for example `humanizeAiTextEmdashed` exported `@promptbook/markdown-utils`
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ */
+function humanizeAiText(aiText) {
+    let cleanedText = aiText;
+    cleanedText = humanizeAiTextEllipsis(cleanedText);
+    cleanedText = humanizeAiTextEmdashed(cleanedText);
+    cleanedText = humanizeAiTextQuotes(cleanedText);
+    cleanedText = humanizeAiTextWhitespace(cleanedText);
+    return cleanedText;
+}
+/**
+ * TODO: [🅾️] !!! Use this across the project where AI text is involved
+ */
+
+/**
+ * Function `promptbookifyAiText` will slightly modify the text so we know it was processed by Promptbook
+ *
+ * @public exported from `@promptbook/markdown-utils`
+ */
+function promptbookifyAiText(text) {
+    // Note: Duplicating some spaces
+    const words = text.split(' '); // <- Note: [✌️] Use `splitWords` when available
+    const wordLength = words.length; // <- Note: [✌️] `countWords` should be just `splitWords(...).length`
+    for (const wordIndex of [3, 7, 11, 19].filter((i) => i < wordLength)) {
+        words[wordIndex] = ' ' + words[wordIndex];
+    }
+    const promptbookifiedText = words.join(' ');
+    return promptbookifiedText;
+}
+/**
+ * TODO: !!!!! Make the function idempotent and add "Note: [🔂] This function is idempotent."
+ * TODO: [🅾️]!!! Use this across the project where AI text is involved
+ * TODO: [🧠][✌️] Make some Promptbook-native token system
+ */
+
 /**
  * Removes Markdown formatting tags from a string.
  *
@@ -6790,5 +6926,20 @@ function removeMarkdownFormatting(str) {
     return str;
 }
 
-export { BOOK_LANGUAGE_VERSION, MarkdownScraper, PROMPTBOOK_ENGINE_VERSION, _MarkdownScraperRegistration, addAutoGeneratedSection, createMarkdownChart, createMarkdownScraper, createMarkdownTable, escapeMarkdownBlock, extractAllBlocksFromMarkdown, extractAllListItemsFromMarkdown, extractBlock, extractJsonBlock, extractOneBlockFromMarkdown, flattenMarkdown, parseMarkdownSection, removeMarkdownComments, removeMarkdownFormatting, splitMarkdownIntoSections };
+/**
+ * Removes Markdown link tags from a string.
+ *
+ * @param {string} str - The string to remove Markdown tags from.
+ * @returns {string} The input string with all Markdown tags removed.
+ * @public exported from `@promptbook/markdown-utils`
+ */
+function removeMarkdownLinks(str) {
+    // Use regular expressions to remove Markdown tags from the string
+    return str.replace(/\[([^\]]+)\]\(([^)]+)\)/g, '$1');
+}
+/**
+ * @see https://chat.openai.com/chat/bb7c3a5b-fe9c-4ccc-9057-f47e0fd66489
+ */
+
+export { BOOK_LANGUAGE_VERSION, MarkdownScraper, PROMPTBOOK_ENGINE_VERSION, _MarkdownScraperRegistration, addAutoGeneratedSection, createMarkdownChart, createMarkdownScraper, createMarkdownTable, escapeMarkdownBlock, extractAllBlocksFromMarkdown, extractAllListItemsFromMarkdown, extractBlock, extractJsonBlock, extractOneBlockFromMarkdown, flattenMarkdown, humanizeAiText, humanizeAiTextEllipsis, humanizeAiTextEmdashed, humanizeAiTextQuotes, humanizeAiTextWhitespace, parseMarkdownSection, promptbookifyAiText, removeMarkdownComments, removeMarkdownFormatting, removeMarkdownLinks, splitMarkdownIntoSections };
 //# sourceMappingURL=index.es.js.map