fix: Tables does not rendered correctly #7465

[gippole]

Issue

Fixes CON-7465: Pycharm: Tables does not rendered correctly

Changes

• The content of 'codeInline' was not included in the 'text' of the tree passed to remarkTables in remarkTables.tsx, so the content disappeared after rendering.
• This has been fixed by adding the node contents within the paragraph.

Files Changed

remarkTables.tsx - Remark plugin for github-flavor markdown tables

Checklist

• [] I've read the contributing guide
• [] The relevant docs, if any, have been updated or created
• [] The relevant tests, if any, have been updated or created

Screen recording or screenshot

before

after

Tests

Verified extension startup and functionality in VS Code debugging on macOS 15.6.1.

---

I have read the CLA Document and I hereby sign the CLA

---

Summary by cubic

Fixes CON-7465 by restoring correct GFM table rendering in Markdown previews. Inline content (like inline code and emphasis) inside table cells is now preserved.

• Bug Fixes
  • Include all paragraph child nodes (not just text) when building the table parsing buffer.
  • Track node positions to split and keep content before/after tables.
  • Render inline nodes to strings for regex matching to avoid losing formatting.

[github-actions]

All contributors have signed the CLA ✍️ ✅
_{Posted by the CLA Assistant Lite bot.}

[cubic-dev-ai]

1 issue found across 1 file

Prompt for AI agents (all 1 issues)

Understand the root cause of the following 1 issues and fix them.


<file name="gui/src/components/StyledMarkdownPreview/utils/remarkTables.tsx">

<violation number="1" location="gui/src/components/StyledMarkdownPreview/utils/remarkTables.tsx:171">
Clearing the tracked child positions here means subsequent tables in the same paragraph keep their raw markdown text, so a second table renders twice (once as text, once as a table). This regresses multi-table paragraphs.</violation>
</file>

---

Since this is your first cubic review, here's how it works:

• cubic automatically reviews your code and comments on bugs and improvements
• Teach cubic by replying to its comments. cubic learns from your replies and gets better over time
• Ask questions if you need clarification on any suggestion

_{React with 👍 or 👎 to teach cubic. Mention @cubic-dev-ai to give feedback, ask questions, or re-run the review.}

[cubic-dev-ai]

Clearing the tracked child positions here means subsequent tables in the same paragraph keep their raw markdown text, so a second table renders twice (once as text, once as a table). This regresses multi-table paragraphs.

Prompt for AI agents

Address the following comment on gui/src/components/StyledMarkdownPreview/utils/remarkTables.tsx at line 171:

<comment>Clearing the tracked child positions here means subsequent tables in the same paragraph keep their raw markdown text, so a second table renders twice (once as text, once as a table). This regresses multi-table paragraphs.</comment>

<file context>
@@ -104,16 +129,46 @@ export function remarkTables() {
+          newNodes.push(...afterNodes);
+
+          // Mark positions as consumed
+          positions.length = 0;
         } catch (e) {
           console.error(&quot;Failed to parse markdown table after regex match&quot;, e);
</file context>

[gippole]

This is intentional. All elements in positions are consumed at this stage, so the variable won’t be referenced again later. Thanks for pointing it out, though!

[cubic-dev-ai]

Thanks, I've learnt: "In remarkTables, the positions array is fully consumed when constructing before/after nodes around a parsed table; it is not used later, so clearing positions.length = 0 at that point is intentional. In future reviews, verify whether a tracking array/variable is referenced later before flagging its clearing as a regression.".

Edit this learning here.

[gippole]

I have read the CLA Document and I hereby sign the CLA