<Note>You can use any Redis® client, but we recommend @upstash/redis</Note>

<Note type="info">
  You can use any Redis® client, but we recommend @upstash/redis
</Note>

<Note type="tip">
  You can use any Redis® client, but we recommend @upstash/redis
</Note>

<Note type="caution">
  You can use any Redis® client, but we recommend @upstash/redis
</Note>

<Note type="danger">
  You can use any Redis® client, but we recommend @upstash/redis
</Note>

<Note title="custom">
  You can use any Redis® client, but we recommend @upstash/redis
</Note>
<Note title="custom" danger>
  You can use any Redis® client, but we recommend @upstash/redis
</Note>

```javascript title:"pages/api/hello.ts"
import type { NextRequest } from 'next/server';
import { Redis } from '@upstash/redis';
```

```javascript:pages/api/hello.ts showLineNumbers {5-8,12}
import type { NextRequest } from 'next/server';
import { Redis } from '@upstash/redis'
```

---
slug: qstash-announcement
title: "qStash: Messaging for the Serverless"
author: enes
tags: [announce]
---

Tags group posts under /blog/tag/<tag>. Conventions:

• Always use lowercase kebab-case in frontmatter: aws-lambda, not AWS Lambda or awslambda. Don't wrap tags in quotes.
• Tag matching is case- and whitespace-insensitive (via normalizeTag in src/utils/tags.ts), so Vector, vector, semantic search and semantic-search resolve to the same page — but please write the normalized form so counts and the popular-tag row dedupe correctly.
• Add every topic that applies. An article about Upstash Redis Search should include both redis and search, so it surfaces under /blog/tag/redis and /blog/tag/search. Posts that are about semantic search must carry both search and semantic-search.
• Display labels for multi-word or acronym tags live in TAG_NAMES (src/utils/const.ts). If you introduce a new multi-word tag, add an entry so the pill row renders it nicely (e.g. "feature-flags": "Feature Flags"). Single-word tags fall back to CSS capitalize.

---
title: "A First Look at Upstash Redis Search"
slug: first-look-at-upstash-redis-search
authors: [josh]
tags: [redis, search]
---

Canonical tag names (use these; don't invent variants):

If you ever need to re-run the consolidation (e.g. after importing external posts), scripts/consolidate-blog-tags.mjs is idempotent — it rewrites tags: [...] in every data/blog/*.mdx according to the rules above.

We use snipsync to synchronize snippets from other repos (like redis-examples) to the docs, so they are deduplicated. Snipsync runs automatically when you do yarn build

Use comments to identify code snippets and the locations where they should be merged. In the source repo, wrap the code snippets in comments with a unique snippet identifier like this:

// @@@SNIPSTART hellouniverse
func HelloUniverse() {
        fmt.Println("Hello Universe!")
}
// @@@SNIPEND

In the example above, "hellouniverse" is the unique identifier for the code snippet. Unique identifiers can contain letters, numbers, hyphens, and underscores.

In the target files wrap the location with comments that reference the identifier of the code snippet that will be placed there:

In the example above, the "hellouniverse" code snippet will be spliced between the comments. Any text inside of the placeholders will be replaced by the code snippet when the tool runs.