auto detect patient id type

app/main.py

@@ -48,6 +48,7 @@ def index() -> str:
             <div>
               <label for="data_type">\u60a3\u8005\u7f16\u53f7\u7c7b\u578b</label>
               <select id="data_type" name="data_type">
+                <option value="auto">\u81ea\u52a8\u8bc6\u522b</option>
                 <option value="pat_no">\u60a3\u8005\u53f7 pat_no</option>
                 <option value="zhuyuanhao">\u4f4f\u9662\u53f7 zhuyuanhao</option>
               </select>
@@ -78,7 +79,7 @@ def index() -> str:
 async def process(
     file: UploadFile = File(...),
     mode: str = Form("auto"),
-    data_type: str = Form("pat_no"),
+    data_type: str = Form("auto"),
     result_name: str = Form("Result"),
     preview_rows: int = Form(20),
     include_basic_sheets: str | None = Form(None),

app/processor.py

@@ -1,3 +1,4 @@
+import csv
 import os
 import shutil
 import subprocess
@@ -56,7 +57,7 @@ def run_processing(
 ) -> ProcessingResult:
     if mode not in {"auto", "v1", "v2"}:
         raise ProcessingError("处理模式不正确。")
-    if data_type not in {"pat_no", "zhuyuanhao"}:
+    if data_type not in {"auto", "pat_no", "zhuyuanhao"}:
         raise ProcessingError("患者编号类型不正确。")
 
     extract_dir = job_dir / "input"
@@ -67,6 +68,7 @@ def run_processing(
     _safe_extract(zip_path, extract_dir)
     data_dir = _find_data_root(extract_dir)
     selected_mode = _detect_mode(data_dir) if mode == "auto" else mode
+    selected_data_type = _detect_data_type(data_dir, selected_mode) if data_type == "auto" else data_type
 
     clean_name = _clean_result_name(result_name)
     if selected_mode == "v1":
@@ -78,7 +80,7 @@ def run_processing(
             str(result_path),
             str(show_not_match),
             str(show_all_infos),
-            data_type,
+            selected_data_type,
         ]
     elif selected_mode == "v2":
         cmd = [
@@ -93,7 +95,7 @@ def run_processing(
             "--show_all_infos",
             str(show_all_infos),
             "--data_type",
-            data_type,
+            selected_data_type,
         ]
     else:
         raise ProcessingError("无法识别数据目录结构，请手动选择 V1 或 V2。")
@@ -115,7 +117,7 @@ def run_processing(
     )
 
     (job_dir / "process.log").write_text(
-        "mode=" + selected_mode + "\n\n" + completed.stdout,
+        "mode=" + selected_mode + "\n" + "data_type=" + selected_data_type + "\n\n" + completed.stdout,
         encoding="utf-8",
     )
     if completed.returncode != 0:
@@ -238,6 +240,60 @@ def _detect_mode(data_dir: Path) -> str:
     raise ProcessingError("无法自动识别 V1/V2 数据结构。")
 
 
+def _detect_data_type(data_dir: Path, selected_mode: str) -> str:
+    raw_ids = _read_patient_ids(data_dir / "Patients_info.csv")
+    if not raw_ids:
+        return "pat_no"
+
+    raw_id_set = set(raw_ids)
+    padded_ids = {_pad_patient_id(value) for value in raw_ids}
+
+    if selected_mode == "v1":
+        evidence_names = {
+            path.stem
+            for path in (data_dir / "Tests_List").glob("*.csv")
+            if path.is_file()
+        }
+    elif selected_mode == "v2":
+        evidence_names = {path.name for path in data_dir.iterdir() if path.is_dir()}
+    else:
+        evidence_names = set()
+
+    if not evidence_names:
+        return "pat_no"
+
+    raw_score = len(evidence_names & raw_id_set)
+    padded_score = len(evidence_names & padded_ids)
+    if raw_score > padded_score:
+        return "zhuyuanhao"
+    return "pat_no"
+
+
+def _read_patient_ids(patients_info_path: Path) -> list[str]:
+    if not patients_info_path.exists():
+        return []
+    try:
+        return _read_patient_ids_with_encoding(patients_info_path, "utf-8-sig")
+    except UnicodeDecodeError:
+        return _read_patient_ids_with_encoding(patients_info_path, "gb18030")
+
+
+def _read_patient_ids_with_encoding(patients_info_path: Path, encoding: str) -> list[str]:
+    with patients_info_path.open("r", encoding=encoding, newline="") as file:
+        return [
+            str(row.get("pat_no", "")).strip()
+            for row in csv.DictReader(file)
+            if str(row.get("pat_no", "")).strip()
+        ]
+
+
+def _pad_patient_id(value: str) -> str:
+    try:
+        return f"{int(value):010}"
+    except ValueError:
+        return value
+
+
 def _clean_result_name(result_name: str) -> str:
     name = (result_name or "Result").strip()
     if name.lower().endswith(".xlsx"):