make mypy happy

Author: KRRT7

codeflash/api/aiservice.py

@@ -414,9 +414,9 @@ def optimize_code_refinement(self, request: list[AIServiceRefinerRequest]) -> li
             List of refined optimization candidates
 
         """
-        payload = []
+        payload: list[dict[str, Any]] = []
         for opt in request:
-            item = {
+            item: dict[str, Any] = {
                 "optimization_id": opt.optimization_id,
                 "original_source_code": opt.original_source_code,
                 "read_only_dependency_code": opt.read_only_dependency_code,

codeflash/benchmarking/plugin/plugin.py

@@ -236,7 +236,7 @@ class Benchmark:  # noqa: D106
         def __init__(self, request: pytest.FixtureRequest) -> None:
             self.request = request
 
-        def __call__(self, func, *args, **kwargs):  # type: ignore  # noqa: ANN001, ANN002, ANN003, ANN204, PGH003
+        def __call__(self, func, *args, **kwargs):  # noqa: ANN001, ANN002, ANN003, ANN204
             """Handle both direct function calls and decorator usage."""
             if args or kwargs:
                 # Used as benchmark(func, *args, **kwargs)

codeflash/code_utils/env_utils.py

@@ -155,7 +155,8 @@ def get_cached_gh_event_data() -> dict[str, Any]:
     if not event_path:
         return {}
     with open(event_path, encoding="utf-8") as f:  # noqa: PTH123
-        return json.load(f)  # type: ignore  # noqa
+        result: dict[str, Any] = json.load(f)
+        return result
 
 
 def is_repo_a_fork() -> bool:

codeflash/code_utils/tabulate.py

@@ -649,7 +649,7 @@ def tabulate(
     headersalign=None,
     rowalign=None,
     maxheadercolwidths=None,
-):
+) -> str:
     if tabular_data is None:
         tabular_data = []
 

codeflash/context/code_context_extractor.py

@@ -39,6 +39,7 @@
     from libcst import CSTNode
 
     from codeflash.context.unused_definition_remover import UsageInfo
+    from codeflash.languages.base import HelperFunction
 
 
 def build_testgen_context(
@@ -265,7 +266,7 @@ def get_code_optimization_context_for_language(
         target_relative_path = function_to_optimize.file_path
 
     # Group helpers by file path
-    helpers_by_file: dict[Path, list] = defaultdict(list)
+    helpers_by_file: dict[Path, list[HelperFunction]] = defaultdict(list)
     helper_function_sources = []
 
     for helper in code_context.helper_functions:

codeflash/result/create_pr.py

@@ -59,19 +59,19 @@ def existing_tests_source_for(
     # Include both behavior and benchmarking paths since test results might come from either
     instrumented_to_original: dict[Path, Path] = {}
     if test_files_registry:
-        for tf in test_files_registry.test_files:
-            if tf.original_file_path:
-                if tf.instrumented_behavior_file_path:
-                    instrumented_to_original[tf.instrumented_behavior_file_path.resolve()] = (
-                        tf.original_file_path.resolve()
+        for registry_tf in test_files_registry.test_files:
+            if registry_tf.original_file_path:
+                if registry_tf.instrumented_behavior_file_path:
+                    instrumented_to_original[registry_tf.instrumented_behavior_file_path.resolve()] = (
+                        registry_tf.original_file_path.resolve()
                     )
                     logger.debug(
-                        f"[PR-DEBUG] Mapping (behavior): {tf.instrumented_behavior_file_path.name} -> {tf.original_file_path.name}"
+                        f"[PR-DEBUG] Mapping (behavior): {registry_tf.instrumented_behavior_file_path.name} -> {registry_tf.original_file_path.name}"
                     )
-                if tf.benchmarking_file_path:
-                    instrumented_to_original[tf.benchmarking_file_path.resolve()] = tf.original_file_path.resolve()
+                if registry_tf.benchmarking_file_path:
+                    instrumented_to_original[registry_tf.benchmarking_file_path.resolve()] = registry_tf.original_file_path.resolve()
                     logger.debug(
-                        f"[PR-DEBUG] Mapping (perf): {tf.benchmarking_file_path.name} -> {tf.original_file_path.name}"
+                        f"[PR-DEBUG] Mapping (perf): {registry_tf.benchmarking_file_path.name} -> {registry_tf.original_file_path.name}"
                     )
 
     # Resolve all paths to absolute for consistent comparison
@@ -249,19 +249,19 @@ def existing_tests_source_for(
                             f"{perf_gain}%✅",
                         ]
                     )
-    output_existing += tabulate(  # type: ignore[no-untyped-call]
+    output_existing += tabulate(
         headers=headers, tabular_data=rows_existing, tablefmt="pipe", colglobalalign=None, preserve_whitespace=True
     )
     output_existing += "\n"
     if len(rows_existing) == 0:
         output_existing = ""
-    output_concolic += tabulate(  # type: ignore[no-untyped-call]
+    output_concolic += tabulate(
         headers=headers, tabular_data=rows_concolic, tablefmt="pipe", colglobalalign=None, preserve_whitespace=True
     )
     output_concolic += "\n"
     if len(rows_concolic) == 0:
         output_concolic = ""
-    output_replay += tabulate(  # type: ignore[no-untyped-call]
+    output_replay += tabulate(
         headers=headers, tabular_data=rows_replay, tablefmt="pipe", colglobalalign=None, preserve_whitespace=True
     )
     output_replay += "\n"

codeflash/result/critic.py

@@ -12,6 +12,7 @@
     MIN_THROUGHPUT_IMPROVEMENT_THRESHOLD,
 )
 from codeflash.models import models
+from codeflash.models.test_type import TestType
 
 if TYPE_CHECKING:
     from codeflash.models.models import ConcurrencyMetrics, CoverageData, OptimizedCandidateResult, OriginalCodeBaseline
@@ -200,7 +201,7 @@ def quantity_of_tests_critic(candidate_result: OptimizedCandidateResult | Origin
     if pass_count >= MIN_TESTCASE_PASSED_THRESHOLD:
         return True
     # If one or more tests passed, check if least one of them was a successful REPLAY_TEST
-    return bool(pass_count >= 1 and report[models.TestType.REPLAY_TEST]["passed"] >= 1)  # type: ignore  # noqa: PGH003
+    return bool(pass_count >= 1 and report[TestType.REPLAY_TEST]["passed"] >= 1)
 
 
 def coverage_critic(original_code_coverage: CoverageData | None) -> bool:

codeflash/verification/comparator.py

@@ -280,7 +280,7 @@ def comparator(orig: Any, new: Any, superset_obj=False) -> bool:  # noqa: ANN001
             return comparator(dict(orig), dict(new), superset_obj)
 
         if HAS_NUMPY:
-            import numpy as np  # type: ignore  # noqa: PGH003
+            import numpy as np
 
             if isinstance(orig, (np.datetime64, np.timedelta64)):
                 # Handle NaT (Not a Time) - numpy's equivalent of NaN for datetime
@@ -343,7 +343,7 @@ def comparator(orig: Any, new: Any, superset_obj=False) -> bool:  # noqa: ANN001
             return (orig != new).nnz == 0
 
         if HAS_PANDAS:
-            import pandas  # type: ignore  # noqa: ICN001, PGH003
+            import pandas  # noqa: ICN001
 
             if isinstance(
                 orig, (pandas.DataFrame, pandas.Series, pandas.Index, pandas.Categorical, pandas.arrays.SparseArray)