belljar 4.2.0

Extremely simple to use callable memoization decorator library.

belljar 🫙

Mid-execution memoization for dynamic state.

Standard caching fails when your function relies on hidden or changing state (like a database cursor or an open file). belljar solves this by letting you build the cache key while the function runs.

uv add belljar

The Difference

Concept	functools.lru_cache	belljar
When is it checked?	Before the function runs.	Mid-execution, exactly when you tell it to.
What defines identity?	Static function arguments.	Function source code + runtime state.
Handling Mutable State	Fails. Returns stale/wrong data.	Succeeds. Hashes current state dynamically.

Usage

You need exactly three primitives: @store to set the boundary, include() to build identity, and check() to short-circuit.

import belljar
import io

# A simulated file. The object stays the same, but its internal cursor moves.
log_file = io.StringIO("chunk1 chunk2 chunk3")

@belljar.store  # Automatically seeds identity using this function's source code
def process_chunk(file_handle):
    # 1. Add the file's exact runtime cursor position to the identity hash
    belljar.include(file_handle.tell())

    # 2. If we've processed from this exact position before, STOP.
    # The function aborts right here and returns the saved result from disk.
    belljar.check()

    # 3. Otherwise, do the heavy processing
    print("Doing heavy work...")
    return file_handle.read(6)

# Call 1: Reads "chunk1", saves to disk. (Takes time)
process_chunk(log_file)

# Call 2: Reads "chunk2", saves to disk. (Takes time)
process_chunk(log_file)

# If we reset the file and run it again:
log_file.seek(0)

# Call 3: belljar.check() detects the cursor is at 0, aborts the function,
# and instantly returns the cached "chunk1".
process_chunk(log_file)

Core Mechanics

• Auto-Invalidation: belljar hashes your actual source code. If you edit the function, the cache invalidates automatically.
• Disk Persistent: Caches are saved to a .jar/ directory by default, surviving script restarts. Pass a path to change it: @store(Path("/tmp/cache")).
• Deep Serialization: Powered by dill (not pickle), meaning it safely handles lambdas, nested classes, and complex closures.
• Async Ready: @store() works on async def functions too, and concurrent calls keep their identities isolated.