@uniweb/content-reader 1.0.3 → 1.0.4

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@uniweb/content-reader",
-  "version": "1.0.3",
+  "version": "1.0.4",
   "description": "Markdown to ProseMirror document structure converter",
   "type": "module",
   "main": "src/index.js",
@@ -16,6 +16,7 @@
   "author": "Proximify Inc.",
   "license": "GPL-3.0-or-later",
   "dependencies": {
+    "js-yaml": "^4.1.0",
     "marked": "^11.1.0"
   },
   "devDependencies": {

package/src/parser/block.js

@@ -3,22 +3,23 @@
  */
 
 import { marked } from "marked";
+import yaml from "js-yaml";
 import { parseInline } from "./inline.js";
 import { parseList } from "./lists.js";
 import { parseTable } from "./tables.js";
 
 /**
- * Process code block info string (e.g., "javascript:example.js")
+ * Process code block info string (e.g., "json:tag-name")
  * @param {string} info - Code block info string
- * @returns {Object} Language and filename
+ * @returns {Object} Language and optional tag
  */
 function processCodeInfo(info) {
-    if (!info) return { language: null, filename: null };
+    if (!info) return { language: null, tag: null };
 
     const parts = info.split(":");
     return {
         language: parts[0] || null,
-        filename: parts[1] || null,
+        tag: parts[1] || null,
     };
 }
 
@@ -39,6 +40,37 @@ function cleanCodeText(text) {
         .trim();
 }
 
+/**
+ * Parse code block content based on language
+ * Only parses tagged blocks with json/yaml language
+ * @param {string} text - Raw code block text
+ * @param {string} language - Code block language
+ * @returns {*} Parsed data or null if not parseable
+ */
+function parseCodeBlockData(text, language) {
+    if (!text) return null;
+
+    const lang = (language || "").toLowerCase();
+
+    if (lang === "json") {
+        try {
+            return JSON.parse(text);
+        } catch {
+            return null;
+        }
+    }
+
+    if (lang === "yaml" || lang === "yml") {
+        try {
+            return yaml.load(text);
+        } catch {
+            return null;
+        }
+    }
+
+    return null;
+}
+
 /**
  * Parse a paragraph's content by tokenizing with marked
  * @param {Object} token - Marked token for paragraph
@@ -142,16 +174,32 @@ function parseBlock(token, schema) {
     }
 
     if (token.type === "code") {
-        const { language, filename } = processCodeInfo(token.lang);
+        const { language, tag } = processCodeInfo(token.lang);
+        const rawText = cleanCodeText(token.text);
+
+        // Tagged blocks become dataBlocks (structured data, not code for display)
+        if (tag) {
+            const parsedData = parseCodeBlockData(rawText, language);
+            if (parsedData !== null) {
+                // Successfully parsed - it's a dataBlock
+                return {
+                    type: "dataBlock",
+                    attrs: { tag, data: parsedData },
+                };
+            }
+            // Parsing failed - fall back to codeBlock with language for runtime fallback
+            return {
+                type: "codeBlock",
+                attrs: { language, tag },
+                content: [{ type: "text", text: rawText }],
+            };
+        }
+
+        // Untagged code block - for display with syntax highlighting
         return {
             type: "codeBlock",
-            attrs: { language, filename },
-            content: [
-                {
-                    type: "text",
-                    text: cleanCodeText(token.text),
-                },
-            ],
+            attrs: { language },
+            content: [{ type: "text", text: rawText }],
         };
     }
 

package/tests/code.test.js

@@ -12,7 +12,6 @@ describe("Code Parsing", () => {
           type: "codeBlock",
           attrs: {
             language: "javascript",
-            filename: null,
           },
           content: [
             {
@@ -25,25 +24,19 @@ describe("Code Parsing", () => {
     });
   });
 
-  test("parses code blocks with filenames", () => {
-    const markdown = "```javascript:example.js\nconst x = 1;\n```";
+  test("parses tagged code blocks as dataBlocks", () => {
+    const markdown = "```json:nav-links\n[{\"label\": \"Home\"}]\n```";
     const result = markdownToProseMirror(markdown);
 
     expect(result).toEqual({
       type: "doc",
       content: [
         {
-          type: "codeBlock",
+          type: "dataBlock",  // Structured data, not code for display
           attrs: {
-            language: "javascript",
-            filename: "example.js",
+            tag: "nav-links",
+            data: [{ label: "Home" }],
           },
-          content: [
-            {
-              type: "text",
-              text: "const x = 1;",
-            },
-          ],
         },
       ],
     });
@@ -60,7 +53,6 @@ describe("Code Parsing", () => {
           type: "codeBlock",
           attrs: {
             language: null,
-            filename: null,
           },
           content: [
             {
@@ -107,7 +99,6 @@ describe("Code Parsing", () => {
           type: "codeBlock",
           attrs: {
             language: null,
-            filename: null,
           },
           content: [
             {