Switch code formatter from Black to Ruff (#3317)

Replace Black with Ruff for code formatting across the project:
- Update Makefile format target to use `ruff format .`
- Replace Black GitHub Actions with `ruff format --check .` in CI
- Add ruff>=0.9.0 as a dev dependency in setup.py
- Reformat all Python files with ruff defaults

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

Author: MaxGhenis

.github/workflows/pr.yml

@@ -14,10 +14,10 @@ jobs:
         uses: actions/checkout@v4
       - name: Setup Python
         uses: actions/setup-python@v5
-      - name: Format with Black
-        uses: psf/black@stable
-        with:
-          options: ". -l 79 --check"
+      - name: Install ruff
+        run: pip install ruff>=0.9.0
+      - name: Format check with ruff
+        run: ruff format --check .
   check-version:
     name: Check version
     runs-on: ubuntu-latest

.github/workflows/push.yml

@@ -20,10 +20,12 @@ jobs:
     steps:
       - name: Checkout repo
         uses: actions/checkout@v4
-      - name: Check formatting
-        uses: "lgeiger/black-action@master"
-        with:
-          args: ". -l 79 --check"
+      - name: Setup Python
+        uses: actions/setup-python@v5
+      - name: Install ruff
+        run: pip install ruff>=0.9.0
+      - name: Format check with ruff
+        run: ruff format --check .
   ensure-model-version-aligns-with-sim-api:
     name: Ensure model version aligns with simulation API
     runs-on: ubuntu-latest

Makefile

@@ -16,7 +16,7 @@ debug-test:
 	MAX_HOUSEHOLDS=1000 FLASK_DEBUG=1 pytest -vv --durations=0 tests
 
 format:
-	black . -l 79
+	ruff format .
 
 deploy:
 	python gcp/export.py

changelog_entry.yaml

@@ -0,0 +1,4 @@
+- bump: minor
+  changes:
+    changed:
+      - Switched code formatter from Black to Ruff.

dashboard/app.py

@@ -19,12 +19,8 @@
 
 st.subheader("Look up a policy")
 
-policy_id = int(
-    st.text_input("Enter a policy ID", "1", key="policy_lookup_text")
-)
-country_id = st.text_input(
-    "Enter a country ID", "uk", key="policy_lookup_country"
-)
+policy_id = int(st.text_input("Enter a policy ID", "1", key="policy_lookup_text"))
+country_id = st.text_input("Enter a country ID", "uk", key="policy_lookup_country")
 if st.button("Look up policy", key="policy_lookup"):
     try:
         results = database.query(
@@ -40,9 +36,7 @@
 
 policy_id = int(st.text_input("Enter a policy ID", "1"))
 country_id = st.text_input("Enter a country ID", "uk")
-new_label = st.text_input(
-    "Enter a new label", "New label", key="policy_label_text"
-)
+new_label = st.text_input("Enter a new label", "New label", key="policy_label_text")
 if st.button("Set policy label", key="policy_label"):
     try:
         database.set_policy_label(policy_id, country_id, new_label)
@@ -54,12 +48,8 @@
 
 st.subheader("Delete a policy")
 
-policy_id = int(
-    st.text_input("Enter a policy ID", "1", key="policy_delete_text")
-)
-country_id = st.text_input(
-    "Enter a country ID", "uk", key="policy_delete_country"
-)
+policy_id = int(st.text_input("Enter a policy ID", "1", key="policy_delete_text"))
+country_id = st.text_input("Enter a country ID", "uk", key="policy_delete_country")
 if st.button("Delete policy", key="policy_delete"):
     try:
         database.delete_policy(policy_id, country_id)

dashboard/experiments/gpt4_api_user.py

@@ -13,11 +13,9 @@
 
 @st.cache_data
 def get_embeddings_df():
-    embeddings_df = pd.read_csv(
-        "parameter_embeddings.csv.gz", compression="gzip"
-    )
-    embeddings_df.parameter_embedding = (
-        embeddings_df.parameter_embedding.apply(lambda x: eval(x))
+    embeddings_df = pd.read_csv("parameter_embeddings.csv.gz", compression="gzip")
+    embeddings_df.parameter_embedding = embeddings_df.parameter_embedding.apply(
+        lambda x: eval(x)
     )
     return embeddings_df
 
@@ -55,11 +53,7 @@ def embed(prompt, engine="text-embedding-ada-002"):
     lambda x: cosine_similarity(x, embedding)
 )
 
-top5 = (
-    embeddings_df.sort_values("similarities", ascending=False)
-    .head(5)["json"]
-    .values
-)
+top5 = embeddings_df.sort_values("similarities", ascending=False).head(5)["json"].values
 # display in streamlit
 
 

gcp/bump_country_package.py

@@ -16,9 +16,7 @@ def main():
         help="Country package to bump",
         choices=["policyengine-uk", "policyengine-us", "policyengine-canada"],
     )
-    parser.add_argument(
-        "--version", type=str, required=True, help="Version to bump to"
-    )
+    parser.add_argument("--version", type=str, required=True, help="Version to bump to")
     args = parser.parse_args()
     country = args.country
     version = args.version
@@ -44,9 +42,9 @@ def bump_country_package(country, version):
     with open(setup_py_path, "w") as f:
         f.write(setup_py)
 
-    country_package_full_name = country.replace(
-        "policyengine", "PolicyEngine"
-    ).replace("-", " ")
+    country_package_full_name = country.replace("policyengine", "PolicyEngine").replace(
+        "-", " "
+    )
     country_id = country.replace("policyengine-", "")
     country_package_full_name = country_package_full_name.replace(
         country_id, country_id.upper()

gcp/export.py

@@ -32,13 +32,9 @@
         dockerfile = dockerfile.replace(
             ".github_microdata_token", GITHUB_MICRODATA_TOKEN
         )
-        dockerfile = dockerfile.replace(
-            ".anthropic_api_key", ANTHROPIC_API_KEY
-        )
+        dockerfile = dockerfile.replace(".anthropic_api_key", ANTHROPIC_API_KEY)
         dockerfile = dockerfile.replace(".openai_api_key", OPENAI_API_KEY)
-        dockerfile = dockerfile.replace(
-            ".hugging_face_token", HUGGING_FACE_TOKEN
-        )
+        dockerfile = dockerfile.replace(".hugging_face_token", HUGGING_FACE_TOKEN)
 
     with open(dockerfile_location, "w") as f:
         f.write(dockerfile)

policyengine_api/ai_prompts/simulation_analysis_prompt.py

@@ -95,18 +95,12 @@ def generate_simulation_analysis_prompt(params: InboundParameters) -> str:
     )
 
     impact_budget: str = json.dumps(parameters.impact["budget"])
-    impact_intra_decile: dict[str, Any] = json.dumps(
-        parameters.impact["intra_decile"]
-    )
+    impact_intra_decile: dict[str, Any] = json.dumps(parameters.impact["intra_decile"])
     impact_decile: str = json.dumps(parameters.impact["decile"])
     impact_inequality: str = json.dumps(parameters.impact["inequality"])
     impact_poverty: str = json.dumps(parameters.impact["poverty"]["poverty"])
-    impact_deep_poverty: str = json.dumps(
-        parameters.impact["poverty"]["deep_poverty"]
-    )
-    impact_poverty_by_gender: str = json.dumps(
-        parameters.impact["poverty_by_gender"]
-    )
+    impact_deep_poverty: str = json.dumps(parameters.impact["poverty"]["deep_poverty"])
+    impact_poverty_by_gender: str = json.dumps(parameters.impact["poverty_by_gender"])
 
     all_parameters: AllParameters = AllParameters.model_validate(
         {

policyengine_api/api.py

@@ -132,9 +132,7 @@ def log_timing(message):
 
 app.route("/<country_id>/calculate-full", methods=["POST"])(
     cache.cached(make_cache_key=make_cache_key)(
-        lambda *args, **kwargs: get_calculate(
-            *args, **kwargs, add_missing=True
-        )
+        lambda *args, **kwargs: get_calculate(*args, **kwargs, add_missing=True)
     )
 )
 log_timing("Calculate-full endpoint registered")
@@ -153,9 +151,7 @@ def log_timing(message):
 app.route("/<country_id>/user-policy", methods=["PUT"])(update_user_policy)
 log_timing("User policy update endpoint registered")
 
-app.route("/<country_id>/user-policy/<user_id>", methods=["GET"])(
-    get_user_policy
-)
+app.route("/<country_id>/user-policy/<user_id>", methods=["GET"])(get_user_policy)
 log_timing("User policy get endpoint registered")
 
 app.register_blueprint(user_profile_bp)
@@ -177,19 +173,15 @@ def log_timing(message):
 
 @app.route("/liveness-check", methods=["GET"])
 def liveness_check():
-    return flask.Response(
-        "OK", status=200, headers={"Content-Type": "text/plain"}
-    )
+    return flask.Response("OK", status=200, headers={"Content-Type": "text/plain"})
 
 
 log_timing("Liveness check endpoint registered")
 
 
 @app.route("/readiness-check", methods=["GET"])
 def readiness_check():
-    return flask.Response(
-        "OK", status=200, headers={"Content-Type": "text/plain"}
-    )
+    return flask.Response("OK", status=200, headers={"Content-Type": "text/plain"})
 
 
 log_timing("Readiness check endpoint registered")