ci: Use ruff in pre-commit to further limit code complexity

.pre-commit-config.yaml

@@ -16,6 +16,11 @@ repos:
     - id: trailing-whitespace  # trims trailing whitespace
       args: [--markdown-linebreak-ext=md]
 
+- repo: https://github.com/astral-sh/ruff-pre-commit
+  rev: v0.0.289
+  hooks:
+  - id: ruff
+
 - repo: https://github.com/psf/black
   rev: 23.9.1
   hooks:

haystack/mmh3.py

@@ -22,8 +22,10 @@ def xencode(x):
 del _sys
 
 
-def hash128(key, seed=0x0, x64arch=True):
+def hash128(key, seed=0x0, x64arch=True):  # noqa: C901,PLR0915
     """Implements 128bit murmur3 hash."""
+    # This function has a very high McCabe cyclomatic complexity score of 44
+    # (recommended is 10) and contains 212 statements (recommended is 50).
 
     def hash128_x64(key, seed):
         """Implements 128bit murmur3 hash for x64."""
@@ -153,8 +155,9 @@ def fmix(k):
 
         return h2 << 64 | h1
 
-    def hash128_x86(key, seed):
+    def hash128_x86(key, seed):  # noqa: PLR0915
         """Implements 128bit murmur3 hash for x86."""
+        # This function contains 125 statements (recommended is 50).
 
         def fmix(h):
             h ^= h >> 16

haystack/modeling/data_handler/processor.py

@@ -474,7 +474,7 @@ def dataset_from_dicts(
         max_answers = (
             self.max_answers
             if self.max_answers is not None
-            else max(max(len(basket.raw["answers"]) for basket in baskets), 1)
+            else max(*(len(basket.raw["answers"]) for basket in baskets), 1)
         )
 
         # Convert answers from string to token space, skip this step for inference

haystack/pipelines/base.py

@@ -607,7 +607,7 @@ def run(  # type: ignore
 
         return node_output
 
-    def run_batch(  # type: ignore
+    def run_batch(  # noqa: C901,PLR0912 type: ignore
         self,
         queries: Optional[List[str]] = None,
         file_paths: Optional[List[str]] = None,

pyproject.toml

@@ -332,9 +332,11 @@ disable = [
 [tool.pylint.'DESIGN']
 max-args = 37  # Default is 5
 max-attributes = 27  # Default is 7
-max-branches = 33  # Default is 12
-max-locals = 44  # Default is 15
-max-statements = 205  # Default is 50
+max-branches = 34  # Default is 12
+max-locals = 45  # Default is 15
+max-module-lines = 2468  # Default is 1000
+max-nested-blocks = 7  # Default is 5
+max-statements = 206  # Default is 50
 [tool.pylint.'SIMILARITIES']
 min-similarity-lines=6
 
@@ -370,6 +372,31 @@ plugins = [
   "pydantic.mypy",
 ]
 
+[tool.ruff]
+ignore = [
+  "PLR1714",  # repeated-equality-comparison
+  "PLR5501",  # collapsible-else-if
+  "PLW0603",  # global-statement
+  "PLW1510",  # subprocess-run-without-check
+  "PLW2901",  # redefined-loop-name
+  "W605",     # invalid-escape-sequence
+]
+select = [
+  "C90",  # McCabe cyclomatic complexity
+  "PL",   # Pylint
+  "W",    # Pycodestyle warnings
+]
+
+[tool.ruff.mccabe]
+max-complexity = 28
+
+[tool.ruff.pylint]
+allow-magic-value-types = ["float", "int", "str"]
+max-args = 38  # Default is 5
+max-branches = 32  # Default is 12
+max-returns = 9  # Default is 6
+max-statements = 105  # Default is 50
+
 [tool.coverage.run]
 omit = [
     "haystack/testing/*",

test/mocks/pinecone.py

@@ -148,7 +148,7 @@ def fetch(self, ids: List[str], namespace: str = ""):
                 }
         return response
 
-    def _filter(
+    def _filter(  # noqa: C901,PLR0912
         self,
         metadata: dict,
         filters: Optional[Union[FilterType, List[Optional[FilterType]]]],
@@ -158,6 +158,8 @@ def _filter(
         """
         Mock filtering function
         """
+        # This function has a very high McCabe cyclomatic complexity score of 38
+        # (recommended is 10) and contains 55 branches (recommended is 12).
         bools = []
         if type(filters) is list:
             list_bools = []