fix: 完善数据预处理与 JSON 上传链路

backend/app/api/v1/endpoints/data_process.py

@@ -8,12 +8,14 @@ import os
 import re
 import socket
 import time
+from collections.abc import Iterator, Mapping
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from contextlib import contextmanager
+from copy import deepcopy
 from dataclasses import asdict
 from pathlib import Path
 from threading import BoundedSemaphore, Lock
-from typing import Any, Iterator, Literal
+from typing import Any, Literal
 from urllib.parse import quote, urlsplit
 
 import httpx
@@ -45,9 +47,10 @@ from app.modules.data_process.algorithms import (
     is_near_duplicate,
     near_duplicate_fingerprint,
     parse_text_content,
-    preprocess_structured_records,
+    preprocess_structured_records_with_lineage,
     remove_document_noise,
     score_quality,
+    structured_json_dumps,
 )
 from app.modules.data_process.document_chunking import (
     DocumentChunk,
@@ -75,9 +78,11 @@ from app.modules.data_process.store import (
     NotFoundError,
     get_data_process_store,
     new_id,
+    repeat_task_id,
 )
 from app.schemas.data_process import (
     DataProcessRegenerateRequest,
+    DataProcessRepeatRequest,
     DataProcessStatus,
     DataProcessTaskCreate,
     DataProcessTaskUpdate,
@@ -261,7 +266,14 @@ def _parse_stored_source(source: dict[str, Any]) -> ParsedText:
     content = str(source.get("content") or "")
     file_format = str(source.get("file_format") or "").lower()
     if file_format == "xlsx":
-        # XLSX 上传阶段已安全解析为 JSONL 后入库。
+        raw_content = source.get("raw_content")
+        if isinstance(raw_content, bytes):
+            return parse_text_content(
+                raw_content,
+                filename=str(source.get("name") or "source.xlsx"),
+                file_format="xlsx",
+            )
+        # 兼容原始对象已缺失的历史文件：退化为上传阶段生成的 JSONL。
         return parse_text_content(content, file_format="jsonl")
     if file_format in {"pdf", "docx", "pptx"}:
         # 文档上传阶段已抽取文本，预览阶段只需要对正文切片。
@@ -381,11 +393,12 @@ def _build_preview_items(
     seen_near_duplicate_bands: dict[tuple[int, int], list[str]] = {}
     items: list[dict[str, Any]] = []
 
-    def append_item(item: dict[str, Any]) -> None:
+    def append_item(item: dict[str, Any], *, dedup_content: str) -> None:
         content = str(item.get("edited_content") or "").strip()
         if should_clean_invalid and not content:
             return
-        content_hash = hashlib.sha256(content.encode("utf-8")).hexdigest()
+        # 去重必须基于脱敏前内容，否则不同原文可能在替换 PII 后被错误合并。
+        content_hash = hashlib.sha256(dedup_content.strip().encode("utf-8")).hexdigest()
         if should_deduplicate and content_hash in seen_content_hashes:
             return
         seen_content_hashes.add(content_hash)
@@ -447,6 +460,7 @@ def _build_preview_items(
                         continue
                     for key in band_keys:
                         seen_near_duplicate_bands.setdefault(key, []).append(content)
+                dedup_content = content
                 pii_counts: dict[str, int] = {}
                 if should_desensitize:
                     content, pii_counts = desensitize_pii(content)
@@ -476,7 +490,8 @@ def _build_preview_items(
                             else "original"
                         ),
                         "quality_score": quality,
-                    }
+                    },
+                    dedup_content=dedup_content,
                 )
             continue
 
@@ -488,48 +503,76 @@ def _build_preview_items(
             "filter_anomaly",
         }
         source_records = list(parsed.records)
-        processed_records = preprocess_structured_records(
+        processed_records = preprocess_structured_records_with_lineage(
             source_records,
             structured_options,
         )
-        if not processed_records and parsed.text and not source_records:
-            processed_records = [{"value": parsed.text}]
-        same_cardinality = len(processed_records) == len(source_records)
-        for index, record in enumerate(processed_records):
-            original_record = source_records[index] if same_cardinality else record
-            original_content = json.dumps(
-                original_record,
-                ensure_ascii=False,
-                separators=(",", ":"),
+        for processed in processed_records:
+            source_index = processed.source_index
+            record = processed.record
+            original_record = (
+                source_records[source_index]
+                if source_index < len(source_records)
+                else record
             )
+            source_locator = (
+                deepcopy(parsed.record_locators[source_index])
+                if source_index < len(parsed.record_locators)
+                else None
+            )
+            original_content = structured_json_dumps(original_record)
             pii_counts: dict[str, int] = {}
             edited_record = record
+            dedup_content = (
+                canonical_record_json(record)
+                if "normalize_format" in preprocess_options
+                else structured_json_dumps(record)
+            )
             if should_desensitize:
                 edited_record, pii_counts = desensitize_structured_record(record)
             content = (
                 canonical_record_json(edited_record)
                 if "normalize_format" in preprocess_options
-                else json.dumps(
-                    edited_record,
-                    ensure_ascii=False,
-                    separators=(",", ":"),
-                )
+                else structured_json_dumps(edited_record)
             )
             quality = _preview_quality(content, config)
             quality["pii_replacements"] = pii_counts
+            if source_locator is not None:
+                quality["source_locator"] = source_locator
+            source_start = (
+                source_locator.get("source_start")
+                if source_locator is not None
+                else None
+            )
+            source_end = (
+                source_locator.get("source_end")
+                if source_locator is not None
+                else None
+            )
+            source_start_line = (
+                source_locator.get("start_line")
+                if source_locator is not None
+                else None
+            )
+            source_end_line = (
+                source_locator.get("end_line")
+                if source_locator is not None
+                else None
+            )
             append_item(
                 {
                     "source_file_id": source["id"],
                     "original_content": original_content,
                     "edited_content": content,
-                    "source_start": None,
-                    "source_end": None,
-                    "source_start_line": None,
-                    "source_end_line": None,
+                    "source_start": source_start,
+                    "source_end": source_end,
+                    "source_start_line": source_start_line,
+                    "source_end_line": source_end_line,
                     "token_count": estimate_token_count(content),
                     "status": "modified" if content != original_content else "original",
                     "quality_score": quality,
-                }
+                },
+                dedup_content=dedup_content,
             )
     return items
 
@@ -850,6 +893,135 @@ def prepare_regeneration(
         )
 
 
+def _repeat_file_copies(
+    store: DataProcessStore,
+    storage: LocalDataProcessStorage,
+    source_task_id: str,
+    request_id: str,
+) -> tuple[dict[str, dict[str, str]], list[StagedSourceObject]]:
+    """为新任务创建独立的源文件引用，避免删除任一任务时互相影响。"""
+
+    target_task_id = repeat_task_id(source_task_id, request_id)
+    copies: dict[str, dict[str, str]] = {}
+    staged: list[StagedSourceObject] = []
+    batch_id = storage.new_batch_id()
+    for summary in store.list_source_files(source_task_id):
+        old_file_id = str(summary["id"])
+        source = store.get_source_file(source_task_id, old_file_id, include_content=True)
+        new_file_id = new_id("dpsf")
+        old_reference = str(source.get("storage_object_id") or "")
+        if old_reference.startswith("local://data-process/"):
+            staged_object = storage.stage_copy(
+                batch_id=batch_id,
+                source_reference=old_reference,
+                expected_source_task_id=source_task_id,
+                expected_source_file_id=old_file_id,
+                task_id=target_task_id,
+                source_file_id=new_file_id,
+                version=1,
+                name=str(source["name"]),
+            )
+            staged.append(staged_object)
+            new_reference = staged_object.reference
+        elif old_reference.startswith("db://data-process/") or not old_reference:
+            new_reference = f"db://data-process/{target_task_id}/{new_file_id}/v1"
+        else:
+            raise ValueError("源任务包含不受支持的文件存储引用")
+        copies[old_file_id] = {
+            "id": new_file_id,
+            "storage_object_id": new_reference,
+        }
+    return copies, staged
+
+
+def _remove_repeated_storage_objects(
+    storage: LocalDataProcessStorage,
+    task_id: str,
+    staged: list[StagedSourceObject],
+    copies: dict[str, dict[str, str]],
+) -> None:
+    source_file_ids = {
+        str(copy["storage_object_id"]): str(copy["id"])
+        for copy in copies.values()
+    }
+    for item in staged:
+        try:
+            storage.delete(
+                item.reference,
+                expected_task_id=task_id,
+                expected_source_file_id=source_file_ids[item.reference],
+            )
+        except Exception:
+            logger.exception(
+                "failed to roll back repeated data process source object task_id=%s",
+                task_id,
+            )
+
+
+@router.post("/{task_id}/repeat", status_code=202)
+def repeat_generation(
+    task_id: str,
+    payload: DataProcessRepeatRequest,
+    background_tasks: BackgroundTasks,
+    store: DataProcessStore = Depends(get_data_process_store),
+    storage: LocalDataProcessStorage = Depends(get_data_process_storage),
+) -> dict[str, Any]:
+    """按原任务快照创建独立任务，并立即在后台开始新一批生成。"""
+
+    with api_errors():
+        repeated = store.find_repeated_task(task_id, payload.request_id)
+        staged: list[StagedSourceObject] = []
+        target_task_id = repeat_task_id(task_id, payload.request_id)
+        if repeated is None:
+            copies, staged = _repeat_file_copies(
+                store,
+                storage,
+                task_id,
+                payload.request_id,
+            )
+            storage.publish(staged)
+            try:
+                repeated = store.repeat_task(
+                    task_id,
+                    expected_updated_at=payload.expected_updated_at,
+                    request_id=payload.request_id,
+                    file_copies=copies,
+                )
+            except Exception:
+                _remove_repeated_storage_objects(
+                    storage,
+                    target_task_id,
+                    staged,
+                    copies,
+                )
+                raise
+            if not repeated["created"]:
+                _remove_repeated_storage_objects(
+                    storage,
+                    target_task_id,
+                    staged,
+                    copies,
+                )
+
+        repeated_task = repeated["task"]
+        if repeated_task.get("status") == "pending":
+            try:
+                started = store.start_generation(target_task_id, replace_existing=True)
+                background_tasks.add_task(
+                    _run_generation,
+                    store,
+                    target_task_id,
+                    str(started["generation_run_id"]),
+                )
+            except ConflictError:
+                latest = store.get_task(target_task_id)
+                if latest.get("status") != "running":
+                    raise
+        repeated["task"] = store.get_task(target_task_id)
+        repeated["progress"] = store.progress(target_task_id)
+        return ok(repeated, "已按原配置创建新任务并开始后台生成")
+
+
 @router.delete("/{task_id}")
 def delete_task(
     task_id: str,
@@ -919,7 +1091,7 @@ async def upload_source_files(
                         f"{suffix} is not supported for {process_type} data processing",
                     )
                 parsed = parse_text_content(raw, filename=name)
-                if not parsed.text:
+                if not parsed.text.strip():
                     raise fail(400, f"source file is empty: {name}")
                 batch_size += len(raw)
                 if batch_size > MAX_SOURCE_BATCH_BYTES:
@@ -934,7 +1106,11 @@ async def upload_source_files(
                     content=raw,
                 )
                 staged.append(staged_object)
-                record_count = len(parsed.records) or (1 if parsed.text else 0)
+                record_count = (
+                    len(parsed.records)
+                    if process_type == "structured"
+                    else (1 if parsed.text else 0)
+                )
                 prepared.append(
                     {
                         "id": source_file_id,
@@ -1362,15 +1538,28 @@ def _prepare_preview_items(
         _value(config, "chunk_method", "chunkMethod", "layout_hybrid")
     )
     is_unstructured = task.get("process_type") == "unstructured"
-    if is_unstructured and (
+    needs_unstructured_raw = is_unstructured and (
         chunk_method == "layout_hybrid"
         or preprocess_options & {"clean_invalid", "clean_invalid_content"}
-    ):
+    )
+    has_structured_xlsx = not is_unstructured and any(
+        str(source.get("file_format") or "").lower() == "xlsx"
+        for source in sources
+    )
+    if needs_unstructured_raw or has_structured_xlsx:
         for index, source in enumerate(sources):
-            if (
-                chunk_method != "layout_hybrid"
-                and str(source.get("file_format") or "").lower() != "pdf"
-            ):
+            source_format = str(source.get("file_format") or "").lower()
+            needs_structured_xlsx = not is_unstructured and source_format == "xlsx"
+            needs_layout_raw = is_unstructured and chunk_method == "layout_hybrid"
+            needs_pdf_noise = (
+                is_unstructured
+                and not needs_layout_raw
+                and source_format == "pdf"
+                and bool(
+                    preprocess_options & {"clean_invalid", "clean_invalid_content"}
+                )
+            )
+            if not (needs_structured_xlsx or needs_layout_raw or needs_pdf_noise):
                 continue
             storage_object_id = str(source.get("storage_object_id") or "")
             actual_size = storage.file_size(
@@ -1379,7 +1568,7 @@ def _prepare_preview_items(
                 expected_source_file_id=str(source["id"]),
             )
             if actual_size is None:
-                if chunk_method == "layout_hybrid":
+                if needs_layout_raw:
                     raise InvalidStateError(
                         "版面结构混合切分无法读取原始文件，请重新上传后再处理"
                     )
@@ -1396,12 +1585,10 @@ def _prepare_preview_items(
                 )
             )
             enriched = dict(source)
-            if chunk_method == "layout_hybrid":
+            if needs_structured_xlsx or needs_layout_raw:
                 enriched["raw_content"] = raw
                 sources[index] = enriched
                 continue
-            if str(source.get("file_format") or "").lower() != "pdf":
-                continue
             pages = extract_pdf_page_texts(raw)
             extracted_text = "\n\n".join(page.text for page in pages if page.text)
             if extracted_text != str(source.get("content") or ""):
@@ -1413,7 +1600,7 @@ def _prepare_preview_items(
             enriched["document_noise_spans"] = detect_pdf_document_noise(pages)
             sources[index] = enriched
     items = _build_preview_items(task, sources)
-    if not items and source_file_ids is None:
+    if not items and source_file_ids is None and is_unstructured:
         raise InvalidStateError("source files did not produce preview items")
     return items
 
@@ -1435,6 +1622,7 @@ def _run_preview(
         len(source_file_ids),
     )
     try:
+        is_unstructured = store.get_task(task_id).get("process_type") == "unstructured"
         if not store.mark_preview_running(task_id, preview_run_id):
             logger.info(
                 "data process preview skipped inactive run task_id=%s preview_run_id=%s",
@@ -1461,7 +1649,7 @@ def _run_preview(
                 storage,
                 [source_file_id],
             )
-            if not items:
+            if not items and is_unstructured:
                 raise InvalidStateError(
                     f"source file did not produce preview items: {source_file_id}"
                 )
@@ -1657,8 +1845,13 @@ def update_preview_item(
 ) -> dict[str, Any]:
     with api_errors():
         task = store.get_task(task_id)
+        existing = store.get_preview_item(task_id, preview_id)
         update = payload.model_dump(exclude_unset=True, mode="json")
-        update["quality_score"] = _preview_quality(payload.edited_content, task.get("config") or {})
+        quality = _preview_quality(payload.edited_content, task.get("config") or {})
+        source_locator = (existing.get("quality_score") or {}).get("source_locator")
+        if isinstance(source_locator, Mapping):
+            quality["source_locator"] = deepcopy(dict(source_locator))
+        update["quality_score"] = quality
         item = store.update_preview_item(
             task_id,
             preview_id,

backend/app/modules/data_process/storage.py

@@ -137,6 +137,67 @@ class LocalDataProcessStorage:
         self._issued_staged_objects[temporary_path] = staged
         return staged
 
+    def stage_copy(
+        self,
+        *,
+        batch_id: str,
+        source_reference: str,
+        expected_source_task_id: str,
+        expected_source_file_id: str,
+        task_id: str,
+        source_file_id: str,
+        version: int,
+        name: str,
+    ) -> StagedSourceObject:
+        """为不可变源对象创建独立目录项，不把大文件重新读入内存。"""
+
+        batch_id = _safe_component(batch_id, "batch id")
+        task_id = _safe_component(task_id, "task id")
+        source_file_id = _safe_component(source_file_id, "source file id")
+        if isinstance(version, bool) or not isinstance(version, int) or version < 1:
+            raise DataProcessStorageError("invalid source file version")
+        basename = _safe_basename(name)
+        source_relative = self._relative_from_reference(source_reference)
+        if source_relative is None:
+            raise DataProcessStorageError("original source object is not available")
+        self._assert_expected_owner(
+            source_relative,
+            expected_task_id=expected_source_task_id,
+            expected_source_file_id=expected_source_file_id,
+        )
+        descriptor, source_info = self._open_read_descriptor(source_relative)
+        os.close(descriptor)
+
+        batch_directory = self._ensure_directory(self._root / ".staging" / batch_id)
+        temporary_path = batch_directory / f"{source_file_id}-{uuid.uuid4().hex}.tmp"
+        source_path = self._path_for_relative(source_relative)
+        try:
+            os.link(source_path, temporary_path, follow_symlinks=False)
+            copy_info = temporary_path.lstat()
+            if (
+                not stat.S_ISREG(copy_info.st_mode)
+                or source_info.st_dev != copy_info.st_dev
+                or source_info.st_ino != copy_info.st_ino
+            ):
+                raise DataProcessStorageError("source storage object changed while copying")
+        except Exception:
+            temporary_path.unlink(missing_ok=True)
+            raise
+
+        relative_path = PurePosixPath(
+            task_id,
+            source_file_id,
+            f"v{version}",
+            basename,
+        )
+        reference = (
+            "local://data-process/"
+            f"{task_id}/{source_file_id}/v{version}/{quote(basename, safe='')}"
+        )
+        staged = StagedSourceObject(reference, temporary_path, relative_path)
+        self._issued_staged_objects[temporary_path] = staged
+        return staged
+
     def publish(self, objects: Iterable[StagedSourceObject]) -> None:
         staged = list(objects)
         published: list[StagedSourceObject] = []

backend/app/modules/data_process/store.py

@@ -45,6 +45,13 @@ _UNSTRUCTURED_PREVIEW_DEFAULTS: dict[str, Any] = {
     "preserve_lists": True,
 }
 _REGENERATION_MARKER_KEY = "_regeneration_prepared"
+_REPEAT_SOURCE_TASK_KEY = "_repeat_source_task_id"
+_REPEAT_REQUEST_KEY = "_repeat_request_id"
+_INTERNAL_CONFIG_KEYS = {
+    _REGENERATION_MARKER_KEY,
+    _REPEAT_SOURCE_TASK_KEY,
+    _REPEAT_REQUEST_KEY,
+}
 
 
 class DataProcessStoreError(RuntimeError):
@@ -71,6 +78,13 @@ def new_id(prefix: str) -> str:
     return f"{prefix}_{uuid.uuid4().hex[:20]}"
 
 
+def repeat_task_id(source_task_id: str, request_id: str) -> str:
+    """按源任务和请求幂等键生成稳定的新任务 ID。"""
+
+    digest = hashlib.sha256(f"{source_task_id}:{request_id}".encode()).hexdigest()
+    return f"dpt_{digest[:20]}"
+
+
 def json_dumps(value: Any) -> str:
     return json.dumps(value, ensure_ascii=False, separators=(",", ":"))
 
@@ -183,17 +197,25 @@ def _is_regeneration_prepared(task: dict[str, Any]) -> bool:
     return _regeneration_marker(task) is not None
 
 
+def _business_config(config: dict[str, Any] | None) -> dict[str, Any]:
+    """过滤只供服务端维护的工作流标记。"""
+
+    return {
+        key: value
+        for key, value in (config or {}).items()
+        if key not in _INTERNAL_CONFIG_KEYS
+    }
+
+
 def _public_task(item: dict[str, Any] | None) -> dict[str, Any] | None:
-    """从 API 任务快照中移除服务端内部重新生成标记。"""
+    """从 API 任务快照中移除服务端内部工作流标记。"""
 
     if item is None:
         return None
     public = dict(item)
     config = public.get("config")
-    if isinstance(config, dict) and _REGENERATION_MARKER_KEY in config:
-        public["config"] = {
-            key: value for key, value in config.items() if key != _REGENERATION_MARKER_KEY
-        }
+    if isinstance(config, dict):
+        public["config"] = _business_config(config)
     return public
 
 
@@ -353,13 +375,7 @@ class DataProcessStore:
                         payload.get("description") or "",
                         payload["process_type"],
                         payload.get("source_dataset_id"),
-                        json_dumps(
-                            {
-                                key: value
-                                for key, value in (payload.get("config") or {}).items()
-                                if key != _REGENERATION_MARKER_KEY
-                            }
-                        ),
+                        json_dumps(_business_config(payload.get("config"))),
                         payload.get("tenant_id"),
                         payload.get("project_id"),
                         payload.get("owner_id"),
@@ -373,6 +389,268 @@ class DataProcessStore:
             raise ConflictError("data process task name already exists") from exc
         return _public_task(_decode_row(row)) or {}
 
+    @staticmethod
+    def _repeat_response(
+        conn: psycopg.Connection[dict[str, Any]],
+        row: dict[str, Any],
+        *,
+        source_task_id: str,
+        created: bool,
+    ) -> dict[str, Any]:
+        task_id = str(row["id"])
+        counts = conn.execute(
+            """
+            SELECT
+              (SELECT COUNT(*) FROM data_process_source_files
+               WHERE task_id=%s AND deleted_at IS NULL) AS source_file_count,
+              (SELECT COUNT(*) FROM data_process_preview_items
+               WHERE task_id=%s) AS preview_count
+            """,
+            (task_id, task_id),
+        ).fetchone() or {}
+        task = _public_task(_decode_row(row)) or {}
+        task["source_file_count"] = int(counts.get("source_file_count") or 0)
+        task["preview_count"] = int(counts.get("preview_count") or 0)
+        return {
+            "task": task,
+            "source_task_id": source_task_id,
+            "created": created,
+            "copied_source_file_count": task["source_file_count"],
+            "copied_preview_count": task["preview_count"],
+        }
+
+    def find_repeated_task(
+        self,
+        source_task_id: str,
+        request_id: str,
+    ) -> dict[str, Any] | None:
+        """查找同一幂等请求已创建的新任务。"""
+
+        task_id = repeat_task_id(source_task_id, request_id)
+        with self.connect() as conn:
+            row = conn.execute(
+                "SELECT * FROM data_process_tasks WHERE id=%s",
+                (task_id,),
+            ).fetchone()
+            if row is None:
+                return None
+            decoded = _decode_row(row) or {}
+            config = decoded.get("config") or {}
+            if (
+                config.get(_REPEAT_SOURCE_TASK_KEY) != source_task_id
+                or config.get(_REPEAT_REQUEST_KEY) != request_id
+            ):
+                raise ConflictError("再次生成请求与现有任务冲突")
+            if decoded.get("deleted_at"):
+                raise ConflictError("此次再次生成创建的任务已被删除，请重新发起")
+            return self._repeat_response(
+                conn,
+                row,
+                source_task_id=source_task_id,
+                created=False,
+            )
+
+    def repeat_task(
+        self,
+        source_task_id: str,
+        *,
+        expected_updated_at: str,
+        request_id: str,
+        file_copies: dict[str, dict[str, str]],
+    ) -> dict[str, Any]:
+        """复制已确认任务的配置、源文件和预览，结果与发布数据保持独立。"""
+
+        task_id = repeat_task_id(source_task_id, request_id)
+        now = utcnow()
+        try:
+            with self.connect() as conn:
+                existing = conn.execute(
+                    "SELECT * FROM data_process_tasks WHERE id=%s FOR UPDATE",
+                    (task_id,),
+                ).fetchone()
+                if existing is not None:
+                    decoded = _decode_row(existing) or {}
+                    config = decoded.get("config") or {}
+                    if (
+                        config.get(_REPEAT_SOURCE_TASK_KEY) != source_task_id
+                        or config.get(_REPEAT_REQUEST_KEY) != request_id
+                    ):
+                        raise ConflictError("再次生成请求与现有任务冲突")
+                    if decoded.get("deleted_at"):
+                        raise ConflictError("此次再次生成创建的任务已被删除，请重新发起")
+                    return self._repeat_response(
+                        conn,
+                        existing,
+                        source_task_id=source_task_id,
+                        created=False,
+                    )
+
+                source_task = self._task_in_connection(
+                    conn,
+                    source_task_id,
+                    for_update=True,
+                )
+                if (
+                    source_task.get("status") != "completed"
+                    or source_task.get("results_confirmed") is False
+                ):
+                    raise InvalidStateError("只有已完成并确认结果的任务可以再次生成")
+                if source_task.get("preview_status") in ACTIVE_PREVIEW_STATUSES:
+                    raise ConflictError("源任务仍在处理切分，暂时不能再次生成")
+                if expected_updated_at != _serialize_value(source_task.get("updated_at")):
+                    raise ConflictError("源任务已被其他操作修改，请刷新后重试")
+
+                source_files = conn.execute(
+                    """
+                    SELECT * FROM data_process_source_files
+                    WHERE task_id=%s AND deleted_at IS NULL
+                    ORDER BY created_at, id
+                    """,
+                    (source_task_id,),
+                ).fetchall()
+                source_file_ids = {str(row["id"]) for row in source_files}
+                if source_file_ids != set(file_copies):
+                    raise ConflictError("源文件快照已变化，请刷新后重试")
+                previews = conn.execute(
+                    """
+                    SELECT * FROM data_process_preview_items
+                    WHERE task_id=%s
+                    ORDER BY source_file_id NULLS LAST, source_start NULLS LAST,
+                             created_at, id
+                    """,
+                    (source_task_id,),
+                ).fetchall()
+                if not previews:
+                    raise InvalidStateError("源任务没有可用于再次生成的切分结果")
+
+                suffix = f"（再次生成-{task_id[-6:]}）"
+                base_name = str(source_task.get("name") or "数据处理任务")
+                repeated_name = f"{base_name[: max(1, 150 - len(suffix))]}{suffix}"
+                repeated_config = _business_config(source_task.get("config") or {})
+                repeated_config[_REPEAT_SOURCE_TASK_KEY] = source_task_id
+                repeated_config[_REPEAT_REQUEST_KEY] = request_id
+                input_count = sum(int(row.get("record_count") or 0) for row in source_files)
+                task_row = conn.execute(
+                    """
+                    INSERT INTO data_process_tasks
+                    (id, name, description, status, process_type, source_dataset_id,
+                     output_dataset_id, config, progress, input_count, output_count,
+                     filtered_count, duplicate_count, error_count, failure_reason,
+                     generation_run_id, results_confirmed, workflow_step,
+                     preview_status, preview_progress, preview_run_id,
+                     preview_failure_reason, preview_total_files,
+                     preview_completed_files, tenant_id, project_id, owner_id,
+                     approval_status, created_by, updated_by, created_at, updated_at)
+                    VALUES
+                    (%s, %s, %s, 'pending', %s, %s, NULL, %s, 20, %s, 0,
+                     0, 0, 0, NULL, NULL, FALSE, 'preview', 'completed', 100,
+                     NULL, NULL, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s)
+                    RETURNING *
+                    """,
+                    (
+                        task_id,
+                        repeated_name,
+                        source_task.get("description") or "",
+                        source_task["process_type"],
+                        source_task.get("source_dataset_id"),
+                        json_dumps(repeated_config),
+                        input_count,
+                        len(source_files),
+                        len(source_files),
+                        source_task.get("tenant_id"),
+                        source_task.get("project_id"),
+                        source_task.get("owner_id"),
+                        source_task.get("approval_status") or "not_required",
+                        source_task.get("created_by"),
+                        source_task.get("created_by"),
+                        now,
+                        now,
+                    ),
+                ).fetchone()
+
+                file_id_map: dict[str, str] = {}
+                for source in source_files:
+                    old_file_id = str(source["id"])
+                    copy = file_copies[old_file_id]
+                    new_file_id = str(copy["id"])
+                    storage_object_id, metadata = _source_storage_descriptor(
+                        {
+                            "storage_object_id": copy["storage_object_id"],
+                            "metadata": _json_value(source.get("metadata"), {}),
+                        },
+                        task_id,
+                        new_file_id,
+                    )
+                    file_id_map[old_file_id] = new_file_id
+                    conn.execute(
+                        """
+                        INSERT INTO data_process_source_files
+                        (id, task_id, storage_object_id, name, size_bytes, record_count,
+                         file_format, checksum_sha256, version_no, content,
+                         content_preview, metadata, tenant_id, project_id, created_by,
+                         created_at, updated_at)
+                        VALUES (%s, %s, %s, %s, %s, %s, %s, %s, 1, %s, %s, %s,
+                                %s, %s, %s, %s, %s)
+                        """,
+                        (
+                            new_file_id,
+                            task_id,
+                            storage_object_id,
+                            source["name"],
+                            source.get("size_bytes") or 0,
+                            source.get("record_count") or 0,
+                            source.get("file_format"),
+                            source["checksum_sha256"],
+                            source.get("content") or "",
+                            source.get("content_preview"),
+                            json_dumps(metadata),
+                            source_task.get("tenant_id"),
+                            source_task.get("project_id"),
+                            source.get("created_by") or source_task.get("created_by"),
+                            now,
+                            now,
+                        ),
+                    )
+
+                for preview in previews:
+                    old_source_file_id = preview.get("source_file_id")
+                    conn.execute(
+                        """
+                        INSERT INTO data_process_preview_items
+                        (id, task_id, source_file_id, original_content, edited_content,
+                         source_start, source_end, source_start_line, source_end_line,
+                         token_count, status, quality_score, created_at, updated_at)
+                        VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s,
+                                %s, %s)
+                        """,
+                        (
+                            new_id("dpp"),
+                            task_id,
+                            file_id_map.get(str(old_source_file_id))
+                            if old_source_file_id
+                            else None,
+                            preview.get("original_content") or "",
+                            preview.get("edited_content") or "",
+                            preview.get("source_start"),
+                            preview.get("source_end"),
+                            preview.get("source_start_line"),
+                            preview.get("source_end_line"),
+                            max(0, int(preview.get("token_count") or 0)),
+                            preview.get("status") or "original",
+                            json_dumps(_json_value(preview.get("quality_score"), {})),
+                            now,
+                            now,
+                        ),
+                    )
+                return self._repeat_response(
+                    conn,
+                    task_row or {},
+                    source_task_id=source_task_id,
+                    created=True,
+                )
+        except psycopg.errors.UniqueViolation as exc:
+            raise ConflictError("再次生成任务名称或请求发生冲突，请重试") from exc
+
     def get_task(self, task_id: str, *, for_update: bool = False) -> dict[str, Any]:
         lock = " FOR UPDATE" if for_update else ""
         with self.connect() as conn:
@@ -654,14 +932,11 @@ class DataProcessStore:
                         "process type and source dataset cannot change during regeneration"
                     )
                 if payload.get("config") is not None:
-                    next_config = {
-                        key: value
-                        for key, value in payload["config"].items()
-                        if key != _REGENERATION_MARKER_KEY
-                    }
-                    current_marker = _regeneration_marker(task)
-                    if current_marker:
-                        next_config[_REGENERATION_MARKER_KEY] = current_marker
+                    next_config = _business_config(payload["config"])
+                    current_config = dict(task.get("config") or {})
+                    for key in _INTERNAL_CONFIG_KEYS:
+                        if key in current_config:
+                            next_config[key] = current_config[key]
                     values["config"] = json_dumps(next_config)
                 invalidates_results = (
                     ("config" in payload and payload.get("config") != task.get("config"))
@@ -753,8 +1028,10 @@ class DataProcessStore:
                     raise InvalidStateError("process_type cannot be changed during regeneration")
 
                 current_config = dict(task.get("config") or {})
-                next_config = dict(payload.get("config") or {})
-                next_config.pop(_REGENERATION_MARKER_KEY, None)
+                next_config = _business_config(payload.get("config"))
+                for key in (_REPEAT_SOURCE_TASK_KEY, _REPEAT_REQUEST_KEY):
+                    if key in current_config:
+                        next_config[key] = current_config[key]
                 preview_invalidated = _preview_config_changed(
                     process_type,
                     current_config,

backend/app/schemas/data_process.py

@@ -220,6 +220,19 @@ class DataProcessRegenerateRequest(BaseModel):
         return self
 
 
+class DataProcessRepeatRequest(BaseModel):
+    """按已确认任务的完整快照创建一批独立的新生成结果。"""
+
+    model_config = ConfigDict(extra="forbid")
+
+    expected_updated_at: str = Field(min_length=1)
+    request_id: str = Field(
+        min_length=8,
+        max_length=80,
+        pattern=r"^[A-Za-z0-9_-]+$",
+    )
+
+
 class PreviewBuildRequest(BaseModel):
     model_config = ConfigDict(extra="forbid")