mdream 0.8.0 → 0.8.2

package/dist/_chunks/{src-CrlAO7kH.mjs → src-ChFJhyZI.mjs}

@@ -1492,6 +1492,41 @@ function generateLlmsTxtContent(files, options) {
 	return content;
 }
 /**
+* Parse frontmatter from markdown content
+*/
+function parseFrontmatter(content) {
+	const frontmatterRegex = /^---\n([\s\S]*?)\n---\n([\s\S]*)$/;
+	const match = content.match(frontmatterRegex);
+	if (!match) return {
+		frontmatter: null,
+		body: content
+	};
+	const frontmatterContent = match[1];
+	const body = match[2];
+	const frontmatter = {};
+	const lines = frontmatterContent.split("\n");
+	for (const line of lines) {
+		const colonIndex = line.indexOf(":");
+		if (colonIndex > 0) {
+			const key = line.substring(0, colonIndex).trim();
+			const value = line.substring(colonIndex + 1).trim();
+			frontmatter[key] = value;
+		}
+	}
+	return {
+		frontmatter,
+		body
+	};
+}
+/**
+* Serialize frontmatter object to YAML-like format
+*/
+function serializeFrontmatter(data) {
+	const lines = [];
+	for (const [key, value] of Object.entries(data)) if (value !== void 0 && value !== null) lines.push(`${key}: ${String(value)}`);
+	return lines.join("\n");
+}
+/**
 * Generate llms-full.txt content with complete page content
 */
 function generateLlmsFullTxtContent(files, options) {
@@ -1507,13 +1542,27 @@ function generateLlmsFullTxtContent(files, options) {
 		content += `\n---\n\n`;
 		for (const file of files) {
 			const url = file.url.startsWith("http://") || file.url.startsWith("https://") ? file.url : origin ? origin + file.url : file.url;
-			content += `## ${file.title}\n\n`;
-			content += `**URL:** ${url}\n`;
-			if (file.filePath && options.outputDir) {
-				const relativePath = relative(options.outputDir, file.filePath);
-				content += `**File:** ${relativePath}\n`;
-			} else if (file.filePath) content += `**File:** ${file.filePath}\n`;
-			content += `\n${file.content}\n\n---\n\n`;
+			const { frontmatter, body } = parseFrontmatter(file.content);
+			const metadata = {
+				title: file.title,
+				url
+			};
+			if (file.filePath && options.outputDir) metadata.file = relative(options.outputDir, file.filePath);
+			else if (file.filePath) metadata.file = file.filePath;
+			if (file.metadata) {
+				if (file.metadata.description) metadata.description = file.metadata.description;
+				if (file.metadata.keywords) metadata.keywords = file.metadata.keywords;
+				if (file.metadata.author) metadata.author = file.metadata.author;
+			}
+			const mergedFrontmatter = frontmatter ? {
+				...frontmatter,
+				...metadata
+			} : metadata;
+			const frontmatterString = serializeFrontmatter(mergedFrontmatter);
+			let contentBody = frontmatter ? body : file.content;
+			const titleLine = contentBody.trim().split("\n")[0];
+			if (titleLine === file.title || titleLine === `# ${file.title}`) contentBody = contentBody.trim().split("\n").slice(1).join("\n").trimStart();
+			content += `---\n${frontmatterString}\n---\n\n${contentBody}\n\n---\n\n`;
 		}
 	}
 	return content;

package/dist/cli.mjs

@@ -1,5 +1,5 @@
 import "./_chunks/extraction-D28Kr1J3.mjs";
-import { generateLlmsTxtArtifacts, streamHtmlToMarkdown } from "./_chunks/src-CrlAO7kH.mjs";
+import { generateLlmsTxtArtifacts, streamHtmlToMarkdown } from "./_chunks/src-ChFJhyZI.mjs";
 import "./_chunks/plugins-DXY-fo9h.mjs";
 import { withMinimalPreset } from "./_chunks/minimal-CCnrG7a1.mjs";
 import { mkdir, writeFile } from "node:fs/promises";

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-CrlAO7kH.mjs";
+import { MarkdownProcessor, generateLlmsTxtArtifacts, htmlToMarkdown, parseHtml, streamHtmlToMarkdown } from "./_chunks/src-ChFJhyZI.mjs";
 
 export { MarkdownProcessor, TagIdMap, createPlugin, generateLlmsTxtArtifacts, htmlToMarkdown, parseHtml, streamHtmlToMarkdown };

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "mdream",
   "type": "module",
-  "version": "0.8.0",
+  "version": "0.8.2",
   "description": "Ultra-performant HTML to Markdown Convertor Optimized for LLMs and llm.txt artifacts.",
   "author": {
     "name": "Harlan Wilton",