[build-system]
requires = ["hatchling>=1.21"]
build-backend = "hatchling.build"

[project]
name = "composer-replication"
version = "0.1.0"
description = "Open replication framework for Cursor Composer 2.5: GRPO + SDPO + multi-teacher trace-replay DPO with optional DiLoCo outer loop."
readme = "README.md"
license = { file = "LICENSE" }
authors = [
    { name = "Codeseys", email = "bbaladithyab@gmail.com" }
]
keywords = [
    "rl-training",
    "rlvr",
    "grpo",
    "sdpo",
    "simpo",
    "taid",
    "dpo",
    "diloco",
    "decoupled-diloco",
    "agentic",
    "coding-agents",
    "composer-2-5",
    "cursor",
    "trl",
    "verl",
    "prime-rl",
    "openenv",
    "torchft",
    "modal",
    "huggingface-jobs",
]
classifiers = [
    "Development Status :: 3 - Alpha",
    "Intended Audience :: Science/Research",
    "License :: OSI Approved :: MIT License",
    "Programming Language :: Python :: 3.10",
    "Programming Language :: Python :: 3.11",
    "Programming Language :: Python :: 3.12",
    "Topic :: Scientific/Engineering :: Artificial Intelligence",
]
requires-python = ">=3.10"
dependencies = [
    "torch>=2.0",
    "transformers>=4.46",
]

[project.optional-dependencies]
# Real teacher-replay over OpenRouter
replay = [
    "httpx>=0.27",
]
# DiLoCo outer-loop optimizer (single-process)
diloco = [
    "torchft-nightly",
]
# Decoupled DiLoCo over serverless executors (per ADR-005)
# fsspec gives the object-store rendezvous one code path (s3://, gs://, hf://,
# file://); s3fs is the concrete S3 backend (the AWS default per the EKS design);
# boto3 + kubernetes are needed by the AWS leaf adapters (SageMakerExecutor uses
# boto3.create_training_job; EKSExecutor uses the kubernetes BatchV1 client).
serverless = [
    "fsspec>=2024.6",
    "huggingface_hub>=0.27",   # for hf:// fsspec backend + HF Jobs
    "s3fs>=2024.6",            # concrete S3 backend for ObjectStoreAllReduce (AWS default)
    "boto3>=1.34",             # SageMakerExecutor (create_training_job) + S3 IAM
    "kubernetes>=29.0",        # EKSExecutor (indexed k8s Jobs via BatchV1Api)
]
# Amazon EKS / Kubernetes Indexed-Job executor (EKSExecutor, per ADR-005).
# kubernetes is lazy-imported at adapter-init/method time (not at package import).
eks = [
    "kubernetes>=29",
]
# Amazon SageMaker training-job executor (SageMakerExecutor, per ADR-005).
# boto3: the executor uses raw create_training_job. sagemaker (<3): the v2
# Estimator API the GSM8K smoke launcher (examples/gsm8k_grpo/
# run_sagemaker_launch.py) uses — pinned <3 because SDK v3 is an API rewrite
# that dropped sagemaker.estimator.Estimator (F3 §1, verified live 2026-06-09).
aws = [
    "boto3>=1.34",
    "sagemaker>=2.200,<3",
]
# SWE-smith task-synthesis engine (deepread finding V4 buy-vs-build verdict):
# the swesmith toolkit builds env images from arbitrary GitHub repos and
# synthesizes bugs (PR Mirror = this repo's gold-patch-reversion mechanic).
# LIVE synthesis needs Docker on Linux (the toolkit does not support macOS/
# Windows officially); the SwesmithAdapter itself needs nothing beyond core.
swesmith = [
    "swesmith>=0.1",
]
# Replaysim dataset normalization (per ADR-004)
#
# NOTE: data-juicer is intentionally NOT pinned as an extra. The package
# named "data-juicer" does not exist on PyPI (the closest match,
# "py-data-juicer==1.0.0", has broken transitive deps; later py-data-juicer
# releases work but install ~150 transitive packages). Users who want the
# DJNormalizer adapter should install data-juicer from source themselves —
# see docs/TROUBLESHOOTING.md ("monarch / data-juicer install"). The
# replaysim Python module imports data_juicer lazily, so the framework
# package imports cleanly without it; only DJNormalizer use-time fails.
replaysim = [
    "composer-replication[replay]",   # replaysim builds on the replay channel
]
# Production training (TRL GRPOTrainer subclass — Recipe A)
train = [
    "trl>=0.12",
    "peft>=0.13",
    "accelerate>=1.0",
    "datasets>=3.0",
]
# Feature-Deletion synthetic-data generation (ADR-010)
# Inverts OSS SWE substrates into reimplement-to-pass tasks. `datasets` loads
# the substrate instances; `docker` runs tests in the substrate's frozen image.
# Pure-Python core (schema/env/monitor/curriculum/validator/substrate-adapter)
# needs only `datasets`; `docker` is for the real LocalSubprocessSandbox /
# substrate-inversion path.
datagen = [
    "datasets>=3.0",
    "docker>=7.0",
]
# PRIME-RL recipe (Recipe C — per ADR-006)
# NOTE: a `prime-rl` extra used to be advertised here pinning
# `prime-rl>=0.5`. That pin is unsatisfiable: the `prime-rl` PyPI name is
# not registered. Prime Intellect publishes prime-rl from source only
# (https://github.com/PrimeIntellect-ai/prime-rl). The framework's
# composer_replication.recipes.prime_rl adapter handles its absence
# gracefully (the upstream parity test is skip-marked when prime-rl is
# not importable) and the in-file shadow-parity test still verifies the
# loss formula independently. The extra is dropped — see
# docs/TROUBLESHOOTING.md ("prime-rl install") for installation guidance.
# NOTE: a `monarch` extra used to be advertised here pinning
# `monarch>=0.4.1`. That pin is unsatisfiable: PyPI's `monarch` package
# is unrelated to Meta's actor framework and tops out at 0.1.11. The real
# Meta Monarch is published as `torchmonarch-nightly` and ships only as
# nightly wheels with platform constraints. Per ADR-006, full Monarch
# integration is a v0.2+ bet and the `composer_replication.recipes.monarch`
# module is a documentation skeleton (importing it does NOT require
# monarch installed). The extra is dropped — see docs/TROUBLESHOOTING.md
# ("monarch / data-juicer install") for installation guidance.
# Development — the BASE dev set installs on every platform (macOS arm64 incl.).
# NOTE: `diloco` (torchft-nightly) is deliberately NOT in base `dev`: torchft-nightly
# ships Linux-x86_64 wheels only, so including it made `pip install -e '.[dev]'` fail
# outright on Apple Silicon / any non-Linux-x86_64 host. The torchft-dependent tests
# skipif-gate cleanly when it is absent, so the base dev set runs the full suite minus
# the torchft integration tests on any platform.
dev = [
    "pytest>=8.0",
    "ruff>=0.6",
    "composer-replication[replay,train]",
]
# Full development incl. the DiLoCo outer-loop dep (Linux-x86_64 only — torchft-nightly).
# Use on a Linux GPU/CI host to also exercise the torchft integration tests.
dev-full = [
    "composer-replication[dev,diloco,serverless,datagen]",
]

[project.urls]
Homepage = "https://huggingface.co/Codeseys/composer-replication-framework"
Documentation = "https://huggingface.co/Codeseys/composer-replication-framework/blob/main/docs/INTEGRATION_ARCHITECTURE.md"
Repository = "https://huggingface.co/Codeseys/composer-replication-framework"
Issues = "https://huggingface.co/Codeseys/composer-replication-framework/discussions"

[tool.hatch.build.targets.wheel]
packages = ["composer_replication"]

[tool.hatch.build.targets.sdist]
include = [
    "/composer_replication",
    "/README.md",
    "/LICENSE",
    "/CITATION.cff",
    "/CITATION.bib",
]

[tool.ruff]
line-length = 100
target-version = "py310"

[tool.ruff.lint]
select = ["E", "F", "W", "I", "N", "UP", "B"]
ignore = ["E501", "E741"]