Исправлены замечания ревью Workers

scimesh/worker/runners.py

@@ -23,12 +23,21 @@ class SciMeshRunner:
         command = [sys.executable, "-m", "scimesh.cli", task.workload, str(input_path)]
         params = task.parameters
         if task.workload == "similarity-search":
-            query_id = self._string(params, "query_id")
+            self._reject_unknown(params, {"query_id", "query_smiles", "top_k", "threshold", "threshold_direction", "max_rows", "progress_every"})
+            query_id, query_smiles = params.get("query_id"), params.get("query_smiles")
+            if (query_id is None) == (query_smiles is None):
+                raise ValueError("exactly one of query_id or query_smiles is required")
             top_k = self._positive_int(params, "top_k", default=20)
-            command += ["--query-id", query_id, "--top-k", str(top_k)]
+            command += ["--query-id", self._string(params, "query_id")] if query_id is not None else ["--query-smiles", self._string(params, "query_smiles")]
+            command += ["--top-k", str(top_k)]
+            self._append_common_options(command, params)
         elif task.workload == "similarity-graph":
+            self._reject_unknown(params, {"threshold", "threshold_direction", "block_size", "max_rows", "progress_every"})
             threshold = self._number(params, "threshold")
             command += ["--threshold", str(threshold)]
+            self._append_common_options(command, params, include_threshold=False)
+            if "block_size" in params:
+                command += ["--block-size", str(self._positive_int(params, "block_size", default=1_000))]
         else:
             raise ValueError(f"unsupported workload: {task.workload}")
         command += ["--output", str(output_path)]
@@ -38,6 +47,25 @@ class SciMeshRunner:
         processed_rows = max(sum(1 for _ in output_path.open(encoding="utf-8")) - 1, 0)
         return RunResult((ProducedArtifact(output_path, "text/csv"),), {"processed_rows": processed_rows})
 
+    def _append_common_options(self, command: list[str], params: dict[str, object], *, include_threshold: bool = True) -> None:
+        if include_threshold and "threshold" in params:
+            command += ["--threshold", str(self._number(params, "threshold"))]
+        if "threshold_direction" in params:
+            direction = params["threshold_direction"]
+            if direction not in ("greater", "less"):
+                raise ValueError("threshold_direction must be 'greater' or 'less'")
+            command += ["--threshold-direction", str(direction)]
+        if "max_rows" in params:
+            command += ["--max-rows", str(self._positive_int(params, "max_rows", default=1))]
+        if "progress_every" in params:
+            command += ["--progress-every", str(self._nonnegative_int(params, "progress_every"))]
+
+    @staticmethod
+    def _reject_unknown(params: dict[str, object], allowed: set[str]) -> None:
+        unknown = set(params) - allowed
+        if unknown:
+            raise ValueError(f"unsupported parameters: {', '.join(sorted(unknown))}")
+
     @staticmethod
     def _string(params: dict[str, object], name: str) -> str:
         value = params.get(name)
@@ -52,6 +80,13 @@ class SciMeshRunner:
             raise ValueError(f"{name} must be a positive integer")
         return value
 
+    @staticmethod
+    def _nonnegative_int(params: dict[str, object], name: str) -> int:
+        value = params[name]
+        if isinstance(value, bool) or not isinstance(value, int) or value < 0:
+            raise ValueError(f"{name} must be a non-negative integer")
+        return value
+
     @staticmethod
     def _number(params: dict[str, object], name: str) -> float:
         value = params.get(name)