[WIP] Add article analytics dashboard for writers

[Copilot]

Thanks for asking me to work on this. I will get started on it and keep this PR's description up to date as I form a plan and make progress.

---

This section details on the original issue you should resolve

<issue_title>Article analytics dashboard for writers</issue_title>
<issue_description>Writers want to know their reach. Add a per-article analytics view visible to the author showing views, estimated reads, reactions, and bookmarks over time.

Context: Part of the TechDiary comeback strategy — writers choose platforms where they can see their impact.

Acceptance criteria:

• View count per article
• Reaction and bookmark counts
• Basic time-series chart (views over last 30 days)
• Accessible from the article editor or dashboard</issue_description>

Comments on the Issue (you are @copilot in this section)

@kingRayhan ## Implementation Plan: ClickHouse for Article Analytics

Why ClickHouse?

PostgreSQL (current DB) is not suited for analytics — counting millions of page view events with time-series aggregation is slow and pollutes the main DB. ClickHouse is purpose-built for this:

• Columnar storage — aggregations over millions of rows are instant
• Append-only inserts — perfect for event streams (page views)
• Time-series queries — built-in date functions, fast GROUP BY toDate(timestamp)
• Free tier available — ClickHouse Cloud has a generous free tier, or self-host via Docker
• No impact on the main PostgreSQL DB

---

Architecture

Browser
  └── POST /api/analytics/pageview   (lightweight API route)
        └── ClickHouse: INSERT INTO article_views
              └── (batch insert via @clickhouse/client)

Dashboard page
  └── Server Action: getArticleAnalytics(articleId)
        └── ClickHouse: SELECT toDate(viewed_at), count() GROUP BY 1

---

ClickHouse Schema

CREATE TABLE article_views (
  article_id   UUID,
  viewer_id    Nullable(UUID),   -- NULL for anonymous
  session_id   String,           -- fingerprint / session token hash
  referrer     Nullable(String),
  country_code Nullable(String), -- from IP geo (optional)
  viewed_at    DateTime DEFAULT now()
) ENGINE = MergeTree()
PARTITION BY toYYYYMM(viewed_at)
ORDER BY (article_id, viewed_at);

Use MergeTree — fast inserts, efficient range scans by article_id + viewed_at.

---

New Files

src/backend/persistence/
  clickhouse.client.ts          # ClickHouse client singleton

src/backend/services/
  analytics.actions.ts          # Server actions for querying analytics

src/app/api/analytics/
  pageview/route.ts             # POST endpoint — receives view events

src/components/
  ArticleAnalyticsChart.tsx     # Recharts time-series chart (client component)

src/app/(dashboard-editor)/dashboard/
  analytics/[articleId]/page.tsx

---

Environment Variables to Add

CLICKHOUSE_HOST=
CLICKHOUSE_DATABASE=
CLICKHOUSE_USERNAME=
CLICKHOUSE_PASSWORD=

---

Tracking Logic

Client-side — fire-and-forget fetch on article page load:

// src/app/[username]/[articleHandle]/page.tsx (or a client component)
useEffect(() => {
  fetch('/api/analytics/pageview', {
    method: 'POST',
    body: JSON.stringify({ article_id, session_id: getSessionId() }),
  });
}, []);

API route — inserts into ClickHouse, returns 200 immediately (no await needed by client):

// POST /api/analytics/pageview
export async function POST(req: Request) {
  const { article_id, session_id } = await req.json();
  await clickhouseClient.insert({
    table: 'article_views',
    values: [{ article_id, session_id, viewed_at: new Date() }],
    format: 'JSONEachRow',
  });
  return new Response(null, { status: 200 });
}

Deduplication — count unique session_id per article per day to avoid reload inflation:

SELECT toDate(viewed_at) as date, uniq(session_id) as views
FROM article_views
WHERE article_id = {articleId}
  AND viewed_at >= now() - INTERVAL 30 DAY
GROUP BY date
ORDER BY date

---

Analytics Server Action

// src/backend/services/analytics.actions.ts
export async function getArticleAnalytics(articleId: string) {
  const userId = await authID();
  // verify ownership before returning data

  const rows = await clickhouseClient.query({
    query: `
      SELECT toDate(viewed_at) as date, uniq(session_id) as views
      FROM article_views
      WHERE article_id = {articleId: UUID}
        AND viewed_at >= now() - INTERVAL 30 DAY
      GROUP BY date ORDER BY date
    `,
    query_params: { articleId },
    format: 'JSONEachRow',
  });

  return rows.json();
}

---

Dashboard UI

• Time-series line chart using Recharts (already likely in deps via shadcn)
• Summary cards: total views (all time), views this month, reactions, bookmarks
• Reactions + bookmarks still queried from PostgreSQL via existing repositories
• Date range picker (7d / 30d / 90d / all time)

---

Package

bun add @clickhouse/client

---

Rollout Order

1. Set up ClickHouse Cloud instance + create table
2. Add clickhouse.client.ts
3. Build POST /api/analytics/pageview route
4. Add tracking call to article detail page
5. Build analytics.actions.ts query
6. Build dashboard UI page + chart component
7. Add env vars to production</comment_new>

[coderabbitai]

Important

Review skipped

Bot user detected.

To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 19b92d38-d8eb-4888-b087-94ea10cb934c

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch copilot/add-article-analytics-dashboard

---

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[kingRayhan]

Closing this cause I have done this in #125