Python SDK

Installation

pip install altbasedb

Requires Python 3.11 or later. Uses httpx for HTTP and websockets for realtime.

Dependency	Version
httpx	>=0.27, <1
websockets	>=13, <15

Development Dependencies

pip install altbasedb[dev]
# Includes: pytest, pytest-asyncio, respx

Creating a Client

from altbasedb import create_client

client = create_client("http://localhost:4000", "your-api-key")

The client is async-first. All data methods return awaitable results.

Querying Data

Use the fluent query builder to construct queries:

# Select with filters
result = await client.from_("users").select("id, email, name").eq("id", "123").execute()
print(result.data)

# Insert
await client.from_("posts").insert({"title": "Hello", "body": "First post"}).execute()

# Update
await client.from_("posts").update({"title": "Updated"}).eq("id", 42).execute()

# Delete
await client.from_("posts").delete().eq("id", 42).execute()

Filter Methods

The query builder supports the same filter operators as the JS SDK:

.eq("status", "active")
.neq("role", "admin")
.gt("age", 18)
.gte("price", 10)
.lt("stock", 5)
.lte("priority", 3)
.like("name", "%smith%")
.ilike("email", "%@gmail.com")
.in_("id", [1, 2, 3])
.is_("deleted_at", "null")

Modifiers

.order("created_at", ascending=False)
.limit(10)
.range(0, 9)
.single()

Authentication

# Sign up
await client.auth.sign_up("user@example.com", "securepassword")

# Sign in with password
await client.auth.sign_in_with_password("user@example.com", "password")

# Get current user
user = await client.auth.get_user()

# Sign out
await client.auth.sign_out()

Storage

# Upload a file
with open("photo.jpg", "rb") as f:
    await client.storage.from_("avatars").upload("photo.jpg", f.read())

# Download a file
data = await client.storage.from_("avatars").download("photo.jpg")

# Create a signed URL
url = await client.storage.from_("avatars").create_signed_url("photo.jpg", 3600)

# List files
files = await client.storage.from_("avatars").list("user/")

Vector Search

# Similarity search
results = await client.vector("proj_123").search(
    "documents",
    "embedding",
    [0.1, 0.2, 0.3],  # query vector
)

# Ask with citations (RAG)
answer = await client.vector("proj_123").ask(
    "support",
    "How do we rotate API keys?"
)

Full-Text Search

results = await client.search.query("contacts", {
    "query": "jane smith",
    "fields": ["first_name", "last_name", "email"],
})

Functions

result = await client.functions.invoke("refresh-search", body={"table": "posts"})

GraphQL

result = await client.graphql.query("""
    query {
        posts(limit: 10) {
            id
            title
        }
    }
""")

Events

# Publish an event
await client.events.publish("order.created", {"order_id": "123"})

Realtime

import asyncio

# Subscribe to table changes
channel = client.realtime.channel("realtime:public:posts")

@channel.on("INSERT")
async def on_insert(event):
    print("New post:", event)

await channel.subscribe()

# Keep the connection alive
await asyncio.sleep(3600)

Workflows

# Start a workflow run
run = await client.workflows.start("wf_abc123", trigger_data={"email": "user@example.com"})
print(run.id, run.status)

SDK Modules

Module	Description
client	Main AltBaseClient class
auth	Authentication (signup, login, sessions)
query_builder	Fluent query builder
storage	Bucket and object operations
realtime	WebSocket subscriptions
vector	Vector embeddings and RAG
search	Full-text search
functions	Function invocation
graphql	GraphQL queries
events	Event pub/sub
workflows	Workflow automation