chore: import upstream snapshot with attribution

2026-07-13 13:24:13 +08:00
commit 1037506f2e
6050 changed files with 1731598 additions and 0 deletions
@@ -0,0 +1,87 @@
+import os.path
+
+from datasets import load_dataset
+from glob import glob
+from tqdm import tqdm
+
+
+class CodeContestReader:
+    def __call__(self, file_path):
+        if os.path.exists(file_path):
+            data = load_dataset("parquet", data_files=file_path)["train"]
+        else:
+            data = load_dataset("parquet", data_files=list(glob(file_path)))["train"]
+        print(len(data))
+        outputs = []
+        for item in tqdm(data):
+            description = item["description"]
+            title = item["name"]
+            if item["time_limit"]:
+                time_limits = item["time_limit"]["seconds"]
+            else:
+                time_limits = "N/A"
+            if item["memory_limit_bytes"]:
+                mem_limits = item["memory_limit_bytes"]
+            else:
+                mem_limits = "N/A"
+            pos_solution = item["solutions"]["solution"]
+            neg_solution = item["incorrect_solutions"]["solution"]
+
+            outputs.append({
+                "name": title,
+                "description": description,
+                "time_limit": time_limits,
+                "memory_limit": mem_limits,
+                "pos_solution": pos_solution,
+                "neg_solution": neg_solution,
+                "id": f"{title}",
+            })
+
+        return outputs
+
+
+class CodeContestFlatReader:
+    def __call__(self, file_path):
+        if os.path.exists(file_path):
+            data = load_dataset("parquet", data_files=file_path)["train"]
+        else:
+            data = load_dataset("parquet", data_files=list(glob(file_path)))["train"]
+        print(len(data))
+        outputs = []
+        for item in tqdm(data):
+            description = item["description"]
+            title = item["name"]
+            if item["time_limit"]:
+                time_limits = item["time_limit"]["seconds"]
+            else:
+                time_limits = "N/A"
+            if item["memory_limit_bytes"]:
+                mem_limits = item["memory_limit_bytes"]
+            else:
+                mem_limits = "N/A"
+            pos_solution = item["solutions"]["solution"]
+            neg_solution = item["incorrect_solutions"]["solution"]
+
+            for i, solution in enumerate(pos_solution):
+                outputs.append({
+                    "name": title,
+                    "description": description,
+                    "time_limit": time_limits,
+                    "memory_limit": mem_limits,
+                    "solution": solution,
+                    "id": f"{title}_{i}",
+                    "value": 1,
+                })
+
+            for i, solution in enumerate(neg_solution):
+                outputs.append({
+                    "name": title,
+                    "description": description,
+                    "time_limit": time_limits,
+                    "memory_limit": mem_limits,
+                    "solution": solution,
+                    "id": f"{title}_{i}",
+                    "value": 0,
+                })
+
+        return outputs