SameCode library

Search, detect, and identify AI-generated code and other copied code.

The AI-Generated Code Search project provides open source tools to find code that may have been generated using LLMs and GPT tools.

In this project, SameCode is a low level Python library that exposes features to:

  1. Break code content in code fragments

  2. Compute fingerprints for approximate matching these fragments

  3. Provide related utilities for hamming distance computation

These features are fundamental building blocks for code fragments and snippets matching approximately.

WARNING: this is under heavy development and not yet a finished project!

Note that using this library alone is not straightforward. Consider looking at the design and reference documentation at https://ai-gen-code-search.readthedocs.io for more details. It is designed to be used in the context of a larger code matching feature with MatchCode and the PurlDB: https://github.com/aboutcode-org/purldb

Installation

SameCode is standalone library that does not provide a UI and command line. To install

From PyPI:

pip install samecode

The preferred development setup is with these commands to create a development environment:

git clone https://github.com/aboutcode-org/ai-gen-code-search
cd ai-gen-code-search
make dev # to configure the environemnt
make test # to run tests
make check # to run code checks

Alternatively, a checkout of the https://github.com/aboutcode-org/ai-gen-code-search repo can also be installed into an environment using pip’s --editable option

git clone https://github.com/aboutcode-org/ai-gen-code-search
cd ai-gen-code-search
python -m venv venv
venv/bin/pip install --editable .

or built into a wheel and dists and then installed:

pip install build
venv/bin/pyproject-build --wheel --sdist
pip install dist/samecode*.whl

Usage

SameCode provides these functions classes:

In the module samecode.chunking, the main functions are:

  • ngrams(iterable, ngram_length) Return an iterable of ngrams of length ngram_length given an iterable of strings. Each ngram is a tuple of ngram_length items. The returned iterable is empty if the input iterable contains less than ngram_length items.

  • select_ngrams(ngrams, with_pos=False) Return an iterable as a subset of a sequence of ngrams using the hailstorm algorithm. If with_pos is True also include the starting position for the ngram in the original sequence.

In the module: samecode.halohash, the main functions and classes are:

  • BitAverageHaloHash(msg=None, size_in_bits=128)

    A bit matrix averaging hash object, with these methods and properties:

    digest_size

    Digest size in bytes.

    b64digest(self)

    Return a base64 “url safe”-encoded string representing this hash.

    hexdigest(self)

    Return the hex-encoded hash value.

    digest(self)

    Return a binary string representing this hash.

    distance(self, other)

    Return the bit Hamming distance between this hash and another hash.

    hash(self)

    Return this hash as a bitarray.

    update(self, msg)

    Append a bytestring or sequence of bytestrings to the hash.

    BitAverageHaloHash.combine(hashes) (class method)

    Return a BitAverageHaloHash by summing and averaging the columns of the BitAverageHaloHashes in hashes together, putting the resulting columns into a new BitAverageHaloHash and returning it

  • bit_to_num(bits)

    Return an int (or long) for a bitarray.

  • bitarray_from_bytes(b)

    Return a bitarray built from a byte string b.

  • byte_hamming_distance(b1, b2)

    Return the Hamming distance between b1 and b2 byte strings

  • common_chunks(h1, h2, chunk_bytes_length=4)

    Compute the number of common chunks of byte length chunk_bytes_length between to hashes h1 and h2 using their digest.

  • common_chunks_from_hexdigest(h1, h2, chunk_bytes_length=4)

    Compute the number of common chunks of byte length chunk_bytes_length between two strings h1 and h2, each representing a BitAverageHaloHash hexdigest value.

  • decode_vector(b64_str)

    Return a bit array from an encoded string representation.

  • hamming_distance(bv1, bv2)

    Return the Hamming distance between bv1 and bv2 bitvectors as the number of equal bits for all positions. (e.g. the count of bits set to one in an XOR between two bit strings.)

    bv1 and bv2 must both be either hash-like Halohash instances (with a hash() function) or bitarray instances (that can be manipulated as-is).

  • slices(s, size)

    Given a sequence s, return a sequence of non-overlapping slices of size. Raise an AssertionError if the sequence length is not a multiple of size.

See also code examples in the test suite under /tests.

Tests

Run the tests with:

pytest -vvs

or with:

make test

License

SPDX-License-Identifier: Apache-2.0

Acknowledgements, Funding, Support and Sponsoring

Europa logo

NGI logo

Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or European Commission. Neither the European Union nor the granting authority can be held responsible for them. Funded within the framework of the NGI Search project under grant agreement No 101069364

This project is also supported and sponsored by:

  • Generous support and contributions from users like you!

  • Microsoft and Microsoft Azure

  • AboutCode ASBL

AboutCode logo