Strings, GET and SET

Where we left off

Lesson 3 stored one string ("greeting") without pausing on it. This lesson is that pause: the string is Redis's simplest and most-used value type, and SET/GET are the two commands most others in this course build on top of.

SET and GET

r.set("agent:last_response", "The weather in Paris is 18C.")
value = r.get("agent:last_response")

SET key value stores value under key, overwriting whatever was there before. GET key returns it, or None if the key doesn't exist, there's no exception to catch, a missing key is just None, check for it the same way you'd check any other Python value.

Keys are just strings, but a naming convention pays off

Redis has no tables or schemas, every key lives in one flat namespace per database. The agent:last_response style, colon-separated segments acting like a namespace, is a convention, not a Redis feature, but it's the one used throughout this course (and most real Redis codebases): agent:last_response, session:abc123:messages, ratelimit:user42. It makes keys readable and lets later lessons use patterns like session:* to find related keys.

A string can hold more than text

r.set("agent:call_count", 0)
r.incr("agent:call_count")

INCR treats the string as an integer, increments it, and returns the new value, atomically: two callers incrementing the same key at the same time can't lose an update to each other, the way a Python count += 1 on a shared variable could. Lesson 10's rate limiter is built entirely on this one guarantee.

SET isn't only "create"

r.set("agent:last_response", "Updated answer.")

Calling SET on a key that already exists just overwrites it, there's no separate "update" command, the same way Python's dict[key] = value works whether or not key was already there.