mdream 0.6.1 → 0.7.1

package/dist/_chunks/{src-Eo8j0-9L.mjs → src-Dbe3WLUq.mjs}

@@ -1,6 +1,6 @@
 import { BLOCKQUOTE_SPACING, DEFAULT_BLOCK_SPACING, ELEMENT_NODE, HTML_ENTITIES, LIST_ITEM_SPACING, MARKDOWN_CODE_BLOCK, MARKDOWN_EMPHASIS, MARKDOWN_HORIZONTAL_RULE, MARKDOWN_INLINE_CODE, MARKDOWN_STRIKETHROUGH, MARKDOWN_STRONG, MAX_TAG_ID, NO_SPACING, NodeEventEnter, NodeEventExit, TABLE_ROW_SPACING, TAG_A, TAG_ABBR, TAG_ADDRESS, TAG_AREA, TAG_ASIDE, TAG_AUDIO, TAG_B, TAG_BASE, TAG_BDO, TAG_BLOCKQUOTE, TAG_BODY, TAG_BR, TAG_BUTTON, TAG_CANVAS, TAG_CENTER, TAG_CITE, TAG_CODE, TAG_COL, TAG_DD, TAG_DEL, TAG_DETAILS, TAG_DFN, TAG_DIALOG, TAG_DIV, TAG_DL, TAG_DT, TAG_EM, TAG_EMBED, TAG_FIELDSET, TAG_FOOTER, TAG_FORM, TAG_H1, TAG_H2, TAG_H3, TAG_H4, TAG_H5, TAG_H6, TAG_HEAD, TAG_HR, TAG_I, TAG_IFRAME, TAG_IMG, TAG_INPUT, TAG_INS, TAG_KBD, TAG_KEYGEN, TAG_LABEL, TAG_LEGEND, TAG_LI, TAG_LINK, TAG_MAP, TAG_MARK, TAG_META, TAG_METER, TAG_NAV, TAG_NOFRAMES, TAG_NOSCRIPT, TAG_OL, TAG_OPTION, TAG_P, TAG_PARAM, TAG_PLAINTEXT, TAG_PRE, TAG_PROGRESS, TAG_Q, TAG_RP, TAG_RT, TAG_RUBY, TAG_SAMP, TAG_SCRIPT, TAG_SELECT, TAG_SMALL, TAG_SOURCE, TAG_SPAN, TAG_STRONG, TAG_STYLE, TAG_SUB, TAG_SUMMARY, TAG_SUP, TAG_SVG, TAG_TABLE, TAG_TBODY, TAG_TD, TAG_TEMPLATE, TAG_TEXTAREA, TAG_TFOOT, TAG_TH, TAG_THEAD, TAG_TIME, TAG_TITLE, TAG_TR, TAG_TRACK, TAG_U, TAG_UL, TAG_VAR, TAG_VIDEO, TAG_WBR, TAG_XMP, TEXT_NODE, TagIdMap, assembleBufferedContent, collectNodeContent, extractionPlugin } from "./extraction-D28Kr1J3.mjs";
 import { readFile } from "node:fs/promises";
-import { basename, dirname, relative } from "node:path";
+import { basename, dirname, relative, sep } from "pathe";
 import { glob } from "tinyglobby";
 
 //#region src/tags.ts
@@ -1431,6 +1431,7 @@ function extractMetadata(html, url) {
 */
 function pathToUrl(filePath, baseDir) {
 	let url = relative(baseDir, filePath);
+	url = url.split(sep).join("/");
 	if (url.endsWith(".html")) url = url.slice(0, -5);
 	if (url.endsWith("/index")) url = url.slice(0, -6);
 	if (url === "index") return "/";
@@ -1514,7 +1515,8 @@ function generateLlmsFullTxtContent(files, options) {
 function generateMarkdownFilesContent(files) {
 	const markdownFiles = [];
 	for (const file of files) {
-		const mdPath = file.url === "/" ? "md/index.md" : `md${file.url}.md`;
+		const urlPath = file.url === "/" ? "index" : file.url.replace(/^\//, "").replace(/\/$/, "");
+		const mdPath = `md/${urlPath}.md`;
 		markdownFiles.push({
 			path: mdPath,
 			content: file.content

package/dist/cli.mjs

@@ -1,9 +1,9 @@
 import "./_chunks/extraction-D28Kr1J3.mjs";
-import { generateLlmsTxtArtifacts, streamHtmlToMarkdown } from "./_chunks/src-Eo8j0-9L.mjs";
+import { generateLlmsTxtArtifacts, streamHtmlToMarkdown } from "./_chunks/src-Dbe3WLUq.mjs";
 import "./_chunks/plugins-DXY-fo9h.mjs";
 import { withMinimalPreset } from "./_chunks/minimal-CCnrG7a1.mjs";
 import { mkdir, writeFile } from "node:fs/promises";
-import { dirname, join } from "node:path";
+import { dirname, join, resolve } from "pathe";
 import { readFileSync } from "node:fs";
 import { Readable } from "node:stream";
 import { fileURLToPath } from "node:url";
@@ -24,6 +24,7 @@ async function generateLlms(patterns, options) {
 			"llms-full.txt",
 			"markdown"
 		];
+		const outputDir = resolve(options.output);
 		const result = await generateLlmsTxtArtifacts({
 			patterns,
 			siteName: options.siteName,
@@ -32,18 +33,19 @@ async function generateLlms(patterns, options) {
 			generateFull: artifacts.includes("llms-full.txt"),
 			generateMarkdown: artifacts.includes("markdown")
 		});
-		await mkdir(options.output, { recursive: true });
-		const llmsPath = join(options.output, "llms.txt");
+		await mkdir(outputDir, { recursive: true });
+		const llmsPath = join(outputDir, "llms.txt");
 		await writeFile(llmsPath, result.llmsTxt, "utf-8");
 		if (artifacts.includes("llms-full.txt") && result.llmsFullTxt) {
-			const fullPath = join(options.output, "llms-full.txt");
+			const fullPath = join(outputDir, "llms-full.txt");
 			await writeFile(fullPath, result.llmsFullTxt, "utf-8");
 		}
 		if (artifacts.includes("markdown") && result.markdownFiles) for (const mdFile of result.markdownFiles) {
-			const fullPath = join(options.output, mdFile.path);
+			const fullPath = join(outputDir, mdFile.path);
 			await mkdir(dirname(fullPath), { recursive: true });
 			await writeFile(fullPath, mdFile.content, "utf-8");
 		}
+		console.log(`✅ Generated llms.txt artifacts in: ${outputDir}`);
 	} catch (error) {
 		console.error("❌ Error generating llms.txt:", error);
 		process.exit(1);
@@ -57,7 +59,7 @@ const cli = cac();
 cli.command("[options]", "Convert HTML from stdin to Markdown on stdout").option("--origin <url>", "Origin URL for resolving relative image paths").option("--preset <preset>", "Conversion presets: minimal").action(async (_, opts) => {
 	await streamingConvert(opts);
 });
-cli.command("llms <patterns...>", "Generate llms.txt artifacts from HTML files").option("--site-name <name>", "Name of the site for llms.txt header").option("--description <desc>", "Description of the site for llms.txt").option("--origin <url>", "Origin URL for resolving relative paths and generating absolute URLs").option("-o, --output <dir>", "Output directory for generated files", { default: "." }).option("--artifacts <list>", "Comma-separated list of artifacts to generate: llms.txt,llms-full.txt,markdown", { default: "llms.txt,llms-full.txt,markdown" }).action(async (patterns, opts) => {
+cli.command("llms <patterns...>", "Generate llms.txt artifacts from HTML files").option("--site-name <name>", "Name of the site for llms.txt header").option("--description <desc>", "Description of the site for llms.txt").option("--origin <url>", "Origin URL for resolving relative paths and generating absolute URLs").option("-o, --output <dir>", "Output directory for generated files", { default: process.cwd() }).option("--artifacts <list>", "Comma-separated list of artifacts to generate: llms.txt,llms-full.txt,markdown", { default: "llms.txt,llms-full.txt,markdown" }).action(async (patterns, opts) => {
 	await generateLlms(patterns, {
 		patterns,
 		...opts

package/dist/index.mjs

@@ -1,4 +1,4 @@
 import { TagIdMap, createPlugin } from "./_chunks/extraction-D28Kr1J3.mjs";
-import { MarkdownProcessor, generateLlmsTxtArtifacts, htmlToMarkdown, parseHtml, streamHtmlToMarkdown } from "./_chunks/src-Eo8j0-9L.mjs";
+import { MarkdownProcessor, generateLlmsTxtArtifacts, htmlToMarkdown, parseHtml, streamHtmlToMarkdown } from "./_chunks/src-Dbe3WLUq.mjs";
 
 export { MarkdownProcessor, TagIdMap, createPlugin, generateLlmsTxtArtifacts, htmlToMarkdown, parseHtml, streamHtmlToMarkdown };

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "mdream",
   "type": "module",
-  "version": "0.6.1",
+  "version": "0.7.1",
   "description": "Ultra-performant HTML to Markdown Convertor Optimized for LLMs and llm.txt artifacts.",
   "author": {
     "name": "Harlan Wilton",
@@ -54,7 +54,8 @@
   ],
   "dependencies": {
     "cac": "^6.7.14",
-    "tinyglobby": "^0.2.10"
+    "pathe": "^2.0.3",
+    "tinyglobby": "^0.2.14"
   },
   "scripts": {
     "flame": "pnpm build && unbuild bench/bundle && clinic flame -- node bench/bundle/dist/string.mjs 10",