fix: expose per-file cover identities in import scans

backend/tests/test_import_and_model_errors.py

@@ -24,6 +24,8 @@ def test_scan_selected_student_folder_prefills_cover_identity(tmp_path, monkeypa
     assert project["status"] == "needs_confirmation"
     assert len(project["files"]) == 5
     assert project["missing"] == []
+    assert project["folder_identity"] == {"student_no": "", "name": "", "topic": ""}
+    assert all(file["detected_identity"]["student_no"] == "9000000001" for file in project["files"])
     assert scan("test", root=tmp_path)["projects"] == result["projects"]
 
 
@@ -39,6 +41,26 @@ def test_scan_conflicting_covers_need_confirmation(tmp_path, monkeypatch):
     assert project["student_no"] == ""
     assert project["warnings"]
     assert project["status"] == "needs_confirmation"
+    assert "学号" in project["warnings"][0]
+    assert "student_no" not in project["warnings"][0]
+    assert {file["detected_identity"]["student_no"] for file in project["files"]} == {"9000000001", "9000000002"}
+
+
+def test_scan_unreadable_cover_has_no_stale_identity(tmp_path, monkeypatch):
+    from wisegrade import storage
+    folder = tmp_path / "9000000001_学生甲_机器人系统"
+    folder.mkdir()
+    (folder / "阶段报告1.pdf").write_bytes(b"fixture")
+    (folder / "阶段报告2.pdf").write_bytes(b"broken")
+    def read(path):
+        if path.name.endswith("2.pdf"):
+            raise ValueError("损坏PDF")
+        return "学号：9000000001\n姓名：学生甲\n题目：机器人系统"
+    monkeypatch.setattr(storage, "first_text", read)
+    project = scan(".", root=tmp_path)["projects"][0]
+    assert project["folder_identity"]["student_no"] == "9000000001"
+    assert project["files"][0]["detected_identity"]["student_no"] == "9000000001"
+    assert project["files"][1]["detected_identity"] == {}
 
 
 def test_scan_parent_skips_empty_folders(tmp_path, monkeypatch):

backend/wisegrade/storage.py

@@ -11,6 +11,7 @@ from .config import get_settings
 CATEGORIES = ("stage1", "stage2", "stage3", "final_report", "presentation")
 SUPPORTED = {".pdf", ".docx", ".pptx"}
 NAMES = {"stage1": "阶段报告1", "stage2": "阶段报告2", "stage3": "阶段报告3", "final_report": "最终报告", "presentation": "汇报PPT"}
+IDENTITY_LABELS = {"student_no": "学号", "name": "姓名", "topic": "选题"}
 
 
 def secure_path(relative: str, *, root: Path | None = None, must_exist=True) -> Path:
@@ -114,6 +115,7 @@ def scan(relative: str, *, root: Path | None = None):
         if item.is_dir() and not item.is_symlink() and has_materials(item)]
     for folder in folders:
         info = parse_student_folder(folder.name)
+        folder_identity = {key: info[key] for key in IDENTITY_LABELS}
         files = []
         extracted = {}
         identity_values = {}
@@ -124,6 +126,7 @@ def scan(relative: str, *, root: Path | None = None):
             if candidate.suffix.lower() not in SUPPORTED:
                 continue
             error = ""
+            identity = {}
             try:
                 text = first_text(candidate)
                 category, status = classify(candidate, text)
@@ -137,6 +140,7 @@ def scan(relative: str, *, root: Path | None = None):
             files.append({"relative_path": candidate.relative_to(base).as_posix(), "category": category, "status": status,
                           "size": stat.st_size, "mtime": stat.st_mtime, "sha256": sha256(candidate),
                           "mime_type": mimetypes.guess_type(candidate.name)[0] or "application/octet-stream", "error": error,
+                          "detected_identity": identity,
                           "first_text": text[:500] if not error else ""})
         for category in CATEGORIES:
             matches = [file for file in files if file["category"] == category]
@@ -145,17 +149,18 @@ def scan(relative: str, *, root: Path | None = None):
                     file["status"] = "duplicate"
         for key, values in identity_values.items():
             if len(values) > 1:
-                warnings.append(f"不同材料封面的{key}不一致，请核对后填写")
+                warnings.append(f"不同材料封面的{IDENTITY_LABELS[key]}不一致，请核对后填写")
             else:
                 extracted[key] = next(iter(values))
         for key, value in extracted.items():
             if info.get(key) and value != info[key]:
-                warnings.append(f"封面{key}“{value}”与目录名“{info[key]}”不一致")
+                warnings.append(f"封面{IDENTITY_LABELS[key]}“{value}”与目录名“{info[key]}”不一致")
             elif not info.get(key):
                 info[key] = value
         if warnings:
             info["status"] = "needs_confirmation"
         info.update({"relative_dir": folder.relative_to(base).as_posix(), "files": files,
+                     "folder_identity": folder_identity,
                      "detected_identity": extracted, "warnings": warnings,
                      "missing": [category for category in CATEGORIES if not any(file["category"] == category for file in files)]})
         projects.append(info)