llll

Python support for lisp-like linked lists (llll).

Key Features

• Read/write operations in both text (.txt) and native (.llll) formats.
• 1-based indexing following the bach library and bell language conventions.
• Element-wise arithmetic with automatic broadcasting.
• Flexible mapping with depth control.
• Address and key syntax support for data access and assignment.
• Fraction support for exact rational arithmetic.

Installation

pip3 install llll-py

Basic Usage

Creating lllls

from llll import llll

# Atomic values
x = llll(42)      # 42
y = llll("hello") # hello

# Nested lists
z = llll(1, 2, [3, 4], 5)  # 1 2 [ 3 4 ] 5

# From Python objects
data = llll.from_python([1, [2, 3], 4]) # 1 [ 2 3 ] 4

File I/O

# Save to file
l = llll(1, 2, [3, 4])
l.write("data.txt")      # Human-readable format
l.write("data.llll")     # Native llll format

# Load from file
from_txt = llll.read("data.txt")
from_llll = llll.read("data.llll")
print(l == from_txt == from_llll) # True

Indexing

llll uses 1-based indexing by convention:

l = llll(10, 20, [30, 40], 50)

l[1]      # 10
l[-1]     # 50
l[3]      # 30 40
l[3, 2]   # 40 (address-based access)

# Slicing (1-based, inclusive)
l[2:3]    # 20 [ 30 40 ]

Keys

Access and modify lllls via keys:

l = llll(["name", "Alice"], ["age", 33], ["parents", "Bob", "Chloe"])

# Access by key
l["name"]         # "Alice"
l["age"]          # 33
l["parents"]      # "Bob" "Chloe"

# Assignment by key
l['name'] = "John"
l['age'] = 55
l['parents'] = ["Geoffrey", "Ada"]
# [ name John ] [ age 33 ] [parents Geofrrey Ada ]

Arithmetic Operations

Element-wise operations with broadcasting:

l = llll(1, 2, 3)

l + 10              # 11 12 13
l * llll(2, 3, 4)   # 2 6 12
l / 2               # 1/2 1 3/2 (returns Fractions for int division)

# Nested operations
nested = llll([1, 2], [3, 4])
nested + 10      # [ 11 12 ] [ 13 14 ]

Mapping

Apply functions at specific depths:

l = llll(1, [2, 3], [[4, 5], 6])

# Boolean comparison
l.map(lambda x, addr: int(x > 2))
# 0 [ 0 1 ] [ [ 1 1 ] 1 ]

# Apply only at depth 2
l.map(lambda x, addr: x * 10,
      mindepth=2,
      maxdepth=2)
# 1 [ 20 30 ] [ [ 4 5 ] 60 ]

The mapping function receives both the value and its address (tuple of indices).

Utility Methods

l = llll(1, [2.5, 3])

l.depth()         # 2
l.to_python()     # [1, [2.5, 3]]
l.append([4, 5])  # 1 [ 2.5 3 ] [ 4 5 ]
l.extend([4, 5])  # 1 [ 2.5 3 ] 4 5
l.as_float()      # 1.0 [ 2.5 3.0 ] 4.0 5.0
l.as_int()        # 1 [ 2 3 ] 4 5

License

MIT License - Copyright © 2025 Felipe Tovar-Henao

Development setup

Follow these steps if you intend to run and/or build llll-py from source.

1. Install Miniconda.
2. In the terminal, run:

conda env create -f environment.yml
conda activate llll

To build or publish, run:

hatch build
hatch publish

To run unit tests run:

pytest tests