Redis Lists and Sets Skill

Lists Overview

Redis Lists are linked lists of string values, perfect for queues and stacks.

List Commands

# Push operations
LPUSH key value [value ...]   # Push to head - O(1) per element
RPUSH key value [value ...]   # Push to tail - O(1) per element

# Pop operations
LPOP key [count]              # Pop from head - O(N)
RPOP key [count]              # Pop from tail - O(N)
BLPOP key [key ...] timeout   # Blocking pop - O(1)
BRPOP key [key ...] timeout   # Blocking pop from tail

# Range operations
LRANGE key start stop         # Get range - O(S+N)
LINDEX key index              # Get by index - O(N)
LLEN key                      # Get length - O(1)

# Manipulation
LMOVE source dest LEFT|RIGHT LEFT|RIGHT
LINSERT key BEFORE|AFTER pivot value
LSET key index value          # Set by index
LTRIM key start stop          # Trim list
LPOS key element              # Find position (Redis 6.0.6+)

Sets Overview

Redis Sets are unordered collections of unique strings.

Set Commands

# Basic operations
SADD key member [member ...]   # Add members - O(N)
SREM key member [member ...]   # Remove members - O(N)
SMEMBERS key                   # Get all members - O(N)
SISMEMBER key member           # Check membership - O(1)
SMISMEMBER key member [member ...]  # Multi-check (Redis 6.2+)
SCARD key                      # Get cardinality - O(1)

# Random operations
SRANDMEMBER key [count]        # Random members
SPOP key [count]               # Pop random members

# Set operations
SINTER key [key ...]           # Intersection - O(N*M)
SUNION key [key ...]           # Union - O(N)
SDIFF key [key ...]            # Difference - O(N)
SINTERSTORE dest key [key ...] # Store intersection
SUNIONSTORE dest key [key ...] # Store union
SDIFFSTORE dest key [key ...]  # Store difference

# Scanning
SSCAN key cursor [MATCH pattern] [COUNT count]

Production Patterns

Pattern 1: Reliable Message Queue

# Producer
RPUSH queue:tasks '{"id":1,"action":"process","retry":0}'

# Consumer with reliability (move to processing)
LMOVE queue:tasks queue:processing LEFT RIGHT

# After processing complete
LREM queue:processing 1 '{"id":1,"action":"process","retry":0}'

# Dead letter queue for failures
RPUSH queue:dlq '{"id":1,"action":"process","error":"timeout"}'

Pattern 2: Priority Queue

# High priority
LPUSH queue:tasks:high '{"priority":"high"}'

# Normal priority
RPUSH queue:tasks:normal '{"priority":"normal"}'

# Consumer checks high first
BLPOP queue:tasks:high queue:tasks:normal 30

Pattern 3: Unique Visitors Tracking

# Track unique visitors per day
SADD visitors:2024-01-15 "user:123"
SADD visitors:2024-01-15 "user:456"

# Count unique
SCARD visitors:2024-01-15

# Weekly unique visitors (union)
SUNIONSTORE visitors:week:3 visitors:2024-01-15 visitors:2024-01-16 visitors:2024-01-17
SCARD visitors:week:3

# Set TTL for automatic cleanup
EXPIRE visitors:2024-01-15 604800  # 7 days

Pattern 4: Tag System

# Add tags to items
SADD item:123:tags "redis" "database" "cache"
SADD item:456:tags "redis" "nosql"

# Find items with specific tag
SADD tag:redis:items "item:123" "item:456"

# Find items with ALL tags (intersection)
SINTER tag:redis:items tag:database:items

# Find items with ANY tag (union)
SUNION tag:redis:items tag:nosql:items

Pattern 5: Social Features

# User follows
SADD user:123:following "user:456" "user:789"
SADD user:456:followers "user:123"

# Mutual friends
SINTER user:123:following user:456:following

# Friend suggestions (friends of friends)
SDIFF user:456:following user:123:following

Command Complexity

Assets

• queue-config.yaml - Queue configuration template
• reliable-queue.lua - Atomic queue operations

Scripts

• queue-consumer.py - Python queue consumer example

References

• LIST_SET_PATTERNS.md - Common patterns guide

Troubleshooting Guide

Common Issues & Solutions

1. WRONGTYPE Error

WRONGTYPE Operation against a key holding the wrong kind of value

Cause: Using list command on set or vice versa

Diagnosis:

TYPE mykey  # Check actual type

Prevention: Use consistent naming conventions (e.g., list:*, set:*)

2. Blocking Operation Timeout

# Returns nil after timeout
BLPOP queue:empty 5

Cause: No elements in list within timeout

Fix:

• Increase timeout for low-traffic queues
• Check if producers are running
• Use non-blocking LPOP with retry logic

3. Large Set Operations Blocking

# Can block for seconds on large sets
SMEMBERS huge:set  # 1M+ members

Fix:

# Use SSCAN instead
SSCAN huge:set 0 COUNT 1000

4. Memory Issues with Long Lists

Cause: Unbounded list growth

Fix:

# Cap list length
LTRIM queue:logs 0 9999  # Keep last 10000

# Or use RPUSH with LTRIM atomically
MULTI
RPUSH queue:logs "entry"
LTRIM queue:logs -10000 -1
EXEC

Debug Checklist

□ Key exists? (EXISTS key)
□ Correct type? (TYPE key)
□ List/Set not empty? (LLEN/SCARD)
□ Blocking timeout appropriate?
□ Memory usage acceptable? (MEMORY USAGE key)
□ Consumer running?
□ Producer running?

Performance Considerations

Error Codes Reference

Test Template

# test_redis_lists_sets.py
import redis
import pytest

@pytest.fixture
def r():
    return redis.Redis(decode_responses=True)

class TestLists:
    def test_queue_pattern(self, r):
        r.delete("test:queue")
        r.rpush("test:queue", "task1", "task2")
        assert r.llen("test:queue") == 2
        assert r.lpop("test:queue") == "task1"
        r.delete("test:queue")

    def test_stack_pattern(self, r):
        r.delete("test:stack")
        r.lpush("test:stack", "a", "b", "c")
        assert r.lpop("test:stack") == "c"  # LIFO
        r.delete("test:stack")

    def test_lrange(self, r):
        r.delete("test:list")
        r.rpush("test:list", *range(10))
        assert len(r.lrange("test:list", 0, 4)) == 5
        r.delete("test:list")

class TestSets:
    def test_unique_members(self, r):
        r.delete("test:set")
        r.sadd("test:set", "a", "b", "a")  # Duplicate ignored
        assert r.scard("test:set") == 2
        r.delete("test:set")

    def test_set_operations(self, r):
        r.delete("test:set1", "test:set2")
        r.sadd("test:set1", "a", "b", "c")
        r.sadd("test:set2", "b", "c", "d")

        assert r.sinter("test:set1", "test:set2") == {"b", "c"}
        assert r.sunion("test:set1", "test:set2") == {"a", "b", "c", "d"}
        assert r.sdiff("test:set1", "test:set2") == {"a"}

        r.delete("test:set1", "test:set2")

    def test_membership(self, r):
        r.delete("test:set")
        r.sadd("test:set", "member1")
        assert r.sismember("test:set", "member1") == 1
        assert r.sismember("test:set", "nonexistent") == 0
        r.delete("test:set")