Use Ruff instead of Black+isort+Flake8

.pre-commit-config.yaml

@@ -5,21 +5,16 @@
 #   $ pre-commit install
 exclude: '^docs/code-comparisons/'  # skip the code comparisons directory
 repos:
--   repo: https://github.com/ambv/black
-    rev: 24.8.0
-    hooks:
-    - id: black
-      args: [--line-length=100, --exclude=docs/*]
-#- repo: https://github.com/astral-sh/ruff-pre-commit
-#  # Ruff version.
-#  rev: v0.5.6
-#  hooks:
-#    # Run the linter.
-#    - id: ruff
-#      args: [ --fix ]
-#    # Run the formatter.
-#    - id: ruff-format
-#      args: [ --diff ]
+- repo: https://github.com/astral-sh/ruff-pre-commit
+  # Ruff version.
+  rev: v0.5.7
+  hooks:
+    # Run the linter.
+    - id: ruff
+      args: [ --fix ]
+    # Run the formatter.
+    - id: ruff-format
+#      args: [ --diff ]  # Use for previewing changes
 -   repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v4.6.0
     hooks:
@@ -30,16 +25,6 @@ repos:
     -   id: requirements-txt-fixer
     # valid python file
     -   id: check-ast
-# isort python package import sorting
--   repo: https://github.com/pycqa/isort
-    rev: 5.13.2
-    hooks:
-    -   id: isort
-        args: ["--profile", "black",
-               "--line-length=100",
-               "--skip=docs/",
-               "--known-local-folder", "tests",
-               "-p", "hamilton"]
 -   repo: https://github.com/pycqa/flake8
     rev: 7.1.1
     hooks:

README-DOCS.md

@@ -7,7 +7,7 @@ Instructions for managing documentation on read the docs.
 To build locally, you need to run the following -- make sure you're in the root of the repo:
 
 ```bash
-pip install -r requirements-docs.txt
+pip install .[docs]
 ```
 and then one of the following to build and view the documents:
 ```bash

contrib/hamilton/contrib/dagworks/sphinx_doc_chunking/__init__.py

@@ -152,7 +152,6 @@ def collect_chunked_url_text(url_result: Collect[dict]) -> list:
 if __name__ == "__main__":
     # code here for quickly testing the build of the code here.
     import __main__ as sphinx_doc_chunking
-
     from hamilton import driver
     from hamilton.execution import executors
 

contrib/hamilton/contrib/dagworks/text_summarization/__init__.py

@@ -8,11 +8,11 @@
 from hamilton import contrib
 
 with contrib.catch_import_errors(__name__, __file__, logger):
+    import litellm
     import tiktoken
     from pypdf import PdfReader
     from tenacity import retry, stop_after_attempt, wait_random_exponential
     from tqdm import tqdm
-    import litellm
 
 from hamilton.function_modifiers import config
 

contrib/hamilton/contrib/user/skrawcz/air_quality_analysis/__init__.py

@@ -257,7 +257,6 @@ def p_value(t_value: np.ndarray, dof: int) -> float:
 
 if __name__ == "__main__":
     import __main__ as analysis_flow
-
     from hamilton import base, driver
 
     # let's create a dictionary result -- since we want to get a few things from execution for inspection

contrib/hamilton/contrib/user/skrawcz/customize_embeddings/__init__.py

@@ -43,10 +43,10 @@
     import plotly.express as px  # for plots
     import plotly.graph_objs as go  # for plot object type
     import requests
+    import torch  # for matrix optimization
     from sklearn.model_selection import (
         train_test_split,
     )  # for splitting train & test data
-    import torch  # for matrix optimization
     from tenacity import retry, stop_after_attempt, wait_random_exponential
 
 from hamilton.function_modifiers import (
@@ -134,7 +134,8 @@ def snli_dataset(download_path: str = "data") -> pd.DataFrame:
 @config.when(source="snli")
 @check_output(schema=processed_dataset_schema, importance="fail")
 def processed_local_dataset__snli(
-    snli_dataset: pd.DataFrame, num_pairs_to_embed: int = 1000  # 1000 is arbitrary
+    snli_dataset: pd.DataFrame,
+    num_pairs_to_embed: int = 1000,  # 1000 is arbitrary
 ) -> pd.DataFrame:
     """Processes a raw dataset into a dataframe of text pairs to embed; and check that it matches the schema.
 

contrib/hamilton/contrib/user/skrawcz/fine_tuning/__init__.py

@@ -9,10 +9,11 @@
 
 with contrib.catch_import_errors(__name__, __file__, logger):
     import evaluate
-    from datasets.combine import DatasetType
-    from datasets import Dataset, concatenate_datasets
     import numpy as np
     import pandas as pd
+    import torch
+    from datasets import Dataset, concatenate_datasets
+    from datasets.combine import DatasetType
     from peft import (
         LoraConfig,
         PeftConfig,
@@ -22,15 +23,14 @@
         prepare_model_for_int8_training,
     )
     from sklearn.model_selection import train_test_split
-    import torch
     from tqdm import tqdm
     from transformers import (
         AutoModelForSeq2SeqLM,
         AutoTokenizer,
         DataCollatorForSeq2Seq,
+        PreTrainedTokenizerBase,
         Seq2SeqTrainer,
         Seq2SeqTrainingArguments,
-        PreTrainedTokenizerBase,
     )
 
 from hamilton.function_modifiers import extract_fields

contrib/hamilton/contrib/user/zilto/lancedb_vdb/__init__.py

@@ -7,10 +7,10 @@
 from hamilton import contrib
 
 with contrib.catch_import_errors(__name__, __file__, logger):
-    import pyarrow as pa
     import lancedb
     import numpy as np
     import pandas as pd
+    import pyarrow as pa
     from lancedb.pydantic import LanceModel
 
 from hamilton.function_modifiers import tag

contrib/hamilton/contrib/user/zilto/webscraper/__init__.py

@@ -6,9 +6,9 @@
 from hamilton import contrib
 
 with contrib.catch_import_errors(__name__, __file__, logger):
-    from bs4 import BeautifulSoup
     import lxml  # noqa: F401
     import requests
+    from bs4 import BeautifulSoup
     from tenacity import retry, stop_after_attempt, wait_random_exponential
 
 import dataclasses

contrib/hamilton/contrib/user/zilto/xgboost_optuna/__init__.py

@@ -8,10 +8,10 @@
 
 with contrib.catch_import_errors(__name__, __file__, logger):
     import numpy as np
+    import optuna
     import pandas as pd
     import xgboost
-    import optuna
-    from optuna.distributions import IntDistribution, FloatDistribution
+    from optuna.distributions import FloatDistribution, IntDistribution
     from sklearn.metrics import accuracy_score, mean_squared_error
     from sklearn.model_selection import KFold, StratifiedKFold
 

dev_tools/language_server/hamilton_lsp/server.py

@@ -2,7 +2,6 @@
 import re
 from typing import Type
 
-from hamilton_lsp import __version__
 from lsprotocol.types import (
     TEXT_DOCUMENT_COMPLETION,
     TEXT_DOCUMENT_DID_CHANGE,
@@ -28,6 +27,7 @@
 from hamilton import ad_hoc_utils
 from hamilton.graph import FunctionGraph, create_graphviz_graph
 from hamilton.graph_types import HamiltonGraph
+from hamilton_lsp import __version__
 
 
 def _type_to_string(type_: Type):

dev_tools/language_server/tests/conftest.py

@@ -17,6 +17,7 @@
 # limitations under the License.                                           #
 ############################################################################
 import pytest
+
 from hamilton_lsp.server import HamiltonLanguageServer, register_server_features
 
 from .ls_setup import NativeClientServer

developer_setup.md

@@ -26,10 +26,7 @@ Install the project's dependencies in your preferred method for managing python
 python -m venv ./venv
 . ./venv/bin/activate
 
-pip install \
-    -r ./requirements.txt \
-    -r ./requirements-dev.txt \
-    -r ./requirements-test.txt
+pip install .[dev,test]
 ```
 
 Set up `pre-commit`, which will run some lightweight formatting and linting tasks on every commit.

examples/LLM_Workflows/RAG_document_extract_chunk_embed/pipeline.py

@@ -159,7 +159,6 @@ def store(
 
 if __name__ == "__main__":
     import __main__ as doc_pipeline
-
     from hamilton import driver
 
     # create the driver

examples/LLM_Workflows/langchain_comparison/hamilton_anthropic.py

@@ -38,7 +38,6 @@ def joke_response__anthropic(llm_client: anthropic.Anthropic, joke_prompt: str)
 
 if __name__ == "__main__":
     import hamilton_invoke_anthropic
-
     from hamilton import driver
 
     dr = (

examples/LLM_Workflows/langchain_comparison/hamilton_async.py

@@ -28,7 +28,6 @@ async def joke_response(llm_client: openai.AsyncOpenAI, joke_messages: List[dict
     import asyncio
 
     import hamilton_async
-
     from hamilton import base
     from hamilton.experimental import h_async
 

examples/LLM_Workflows/langchain_comparison/hamilton_batch.py

@@ -38,7 +38,6 @@ def joke_responses(joke_response: Collect[str]) -> List[str]:
 
 if __name__ == "__main__":
     import hamilton_batch
-
     from hamilton import driver
 
     dr = (

examples/LLM_Workflows/langchain_comparison/hamilton_completion.py

@@ -38,7 +38,6 @@ def joke_response__chat(llm_client: openai.OpenAI, joke_messages: List[dict]) ->
 
 if __name__ == "__main__":
     import hamilton_completion
-
     from hamilton import driver
 
     dr = (

examples/LLM_Workflows/langchain_comparison/hamilton_fallbacks.py

@@ -1,5 +1,4 @@
 import hamilton_anthropic
-
 from hamilton import driver
 
 anthropic_driver = (

examples/LLM_Workflows/langchain_comparison/hamilton_invoke.py

@@ -26,7 +26,6 @@ def joke_response(llm_client: openai.OpenAI, joke_messages: List[dict]) -> str:
 
 if __name__ == "__main__":
     import hamilton_invoke
-
     from hamilton import driver
 
     dr = driver.Builder().with_modules(hamilton_invoke).build()

examples/LLM_Workflows/langchain_comparison/hamilton_logging.py

@@ -1,6 +1,5 @@
 # run.py
 import hamilton_anthropic
-
 from hamilton import driver, lifecycle
 
 dr = (

examples/LLM_Workflows/langchain_comparison/hamilton_streamed.py

@@ -28,7 +28,6 @@ def joke_response(llm_client: openai.OpenAI, joke_messages: List[dict]) -> Itera
 
 if __name__ == "__main__":
     import hamilton_streaming
-
     from hamilton import driver
 
     dr = driver.Builder().with_modules(hamilton_streaming).build()

examples/LLM_Workflows/modular_llm_stack/qdrant_module.py

@@ -11,7 +11,6 @@ def initialize_vector_db_indices(
     class_name: str,
     embedding_dimension: int,
 ) -> bool:
-
     if client_vector_db.collection_exists(class_name):
         client_vector_db.delete_collection(class_name)
 

examples/LLM_Workflows/scraping_and_chunking/run.py

@@ -13,7 +13,6 @@
 from hamilton.execution import executors
 
 if __name__ == "__main__":
-
     dr = (
         driver.Builder()
         .with_modules(doc_pipeline)

examples/LLM_Workflows/scraping_and_chunking/spark/spark_pipeline.py

@@ -70,7 +70,6 @@ def chunked_url_text(urls_from_sitemap: ps.DataFrame) -> ps.DataFrame:
 
 
 if __name__ == "__main__":
-
     import spark_pipeline
 
     from hamilton import driver

examples/airflow/dags/hamilton/hamilton_how_to_dag.py

@@ -1,4 +1,4 @@
-""" This file shows different usage pattern to integrate Hamilton with Apache Airflow
+"""This file shows different usage pattern to integrate Hamilton with Apache Airflow
 
 For the purpose of this example, we will read and write data from the Airflow
 installation location (${AIRFLOW_HOME}/plugins/data).

examples/async/fastapi_example.py

@@ -5,9 +5,9 @@
 import async_module
 import fastapi
 from aiohttp import client_exceptions
-from hamilton_sdk import adapters
 
 from hamilton import async_driver
+from hamilton_sdk import adapters
 
 logger = logging.getLogger(__name__)
 

examples/aws/glue/processing.py

@@ -4,12 +4,11 @@
 
 # awsglue is installed in the AWS Glue worker environment
 from awsglue.utils import getResolvedOptions
-from hamilton_functions import functions
 
 from hamilton import driver
+from hamilton_functions import functions
 
 if __name__ == "__main__":
-
     args = getResolvedOptions(sys.argv, ["input-table", "output-table"])
 
     df = pd.read_csv(args["input_table"])

examples/aws/lambda/app/lambda_handler.py

@@ -6,7 +6,6 @@
 
 
 def lambda_handler(event, context):
-
     df = pd.DataFrame(**event["body"])
 
     dr = driver.Driver({}, functions)

examples/aws/sagemaker/processing.py

@@ -4,7 +4,6 @@
 from hamilton import driver
 
 if __name__ == "__main__":
-
     df = pd.read_csv("/opt/ml/processing/input/data/input_table.csv")
 
     dr = driver.Driver({}, functions)

examples/dagster/dagster_code/tutorial/resources/__init__.py

@@ -130,6 +130,7 @@ class DataGeneratorResource(ConfigurableResource):
             from dagster import Definitions, asset
             from dagster_data_generator import DataGeneratorResource, DataGeneratorConfig
 
+
             @asset
             def my_table(data_gen: DataGeneratorConfig):
                 return data_gen.get_signups()

examples/dagster/hamilton_code/mock_api.py

@@ -131,6 +131,7 @@ class DataGeneratorResource(ConfigurableResource):
             from dagster import Definitions, asset
             from dagster_data_generator import DataGeneratorResource, DataGeneratorConfig
 
+
             @asset
             def my_table(data_gen: DataGeneratorConfig):
                 return data_gen.get_signups()

examples/dbt/python_transforms/feature_transforms.py

@@ -8,8 +8,10 @@
 import pandas as pd
 
 # from sklearn.preprocessing import OneHotEncoder
-from sklearn import impute  # import KNNImputer
-from sklearn import preprocessing
+from sklearn import (
+    impute,  # import KNNImputer
+    preprocessing,
+)
 
 from hamilton.function_modifiers import check_output, config
 

examples/due_date_probabilities/probability_estimation.py

@@ -98,9 +98,7 @@ def raw_data() -> str:
 43 weeks, 3 days	99.8%	0.1%	0.1%
 43 weeks, 4 days	99.9%	0.1%	< 0.1%
 43 weeks, 5 days	99.9%	< 0.1%	< 0.1%
-43 weeks, 6 days	> 99.9%	< 0.1%""".replace(
-            "weeks\t", "weeks, 0 days "
-        )
+43 weeks, 6 days	> 99.9%	< 0.1%""".replace("weeks\t", "weeks, 0 days ")
         .replace("\t", " ")
         .split("\n")
     )

examples/feature_engineering/feature_engineering_multiple_contexts/scenario_1/fastapi_server.py

@@ -1,4 +1,4 @@
-""""
+""" "
 This is a simple example of a FastAPI server that uses Hamilton on the request
 path to transform the data into features, and then uses a fake model to make
 a prediction.