Exclude PM room types from shared daily report processing

arr_web/arr_data_review.py

@@ -277,16 +277,19 @@ class DataFieldReviews:
         data["status"] = "collected_with_gaps" if unresolved else "collected"
         return data
 
-    def _issues(self, directory, data):
+    def _normalized(self, directory, data):
         # Use the exact frozen processor's normalizer and field validation; the
         # acquisition layer never maintains a second whitelist or numeric rule.
         with tempfile.NamedTemporaryFile(suffix=".json", dir=directory) as stream:
             stream.write(json_bytes(data))
             stream.flush()
-            business_date, records = self.rules.read_data_source(Path(stream.name))
+            return self.rules.read_data_source(Path(stream.name))
+
+    def _issues(self, directory, data):
+        business_date, records = self._normalized(directory, data)
         issues = {}
         for row, record in zip(data["records"], records):
-            if self.rules.is_cancelled_record(record):
+            if self.rules.is_cancelled_record(record) or self.rules.is_pm_record(record):
                 continue
             rate = record["_NORMALIZED_RATE_CODE"]
             fields = {}
@@ -312,13 +315,24 @@ class DataFieldReviews:
         value = observation.get("value")
         return value if observation.get("state") == "available" and isinstance(value, (str, int)) else ""
 
+    def _excluded_sequences(self, directory, data):
+        cancelled, pm = set(), set()
+        _date, records = self._normalized(directory, data)
+        for row, record in zip(data["records"], records):
+            sequence = row["source_sequence"]
+            if self.rules.is_cancelled_record(record):
+                cancelled.add(sequence)
+            elif self.rules.is_pm_record(record):
+                pm.add(sequence)
+        return cancelled, pm
+
     def _public(self, directory, meta, original, state):
         data = self._derive(directory, original, state)
         issues = self._issues(directory, data)
-        cancelled_sequences = {row["source_sequence"] for row in data["records"]
-            if self.rules.is_cancelled_record({"_RESERVATION_STATUS": row.get("reservation_status", "")})}
+        cancelled_sequences, pm_sequences = self._excluded_sequences(directory, data)
         keys = set(issues) | {key for key in state["decisions"]
-            if int(key.split(":")[0]) not in cancelled_sequences}
+            if int(key.split(":")[0]) not in cancelled_sequences
+            and (int(key.split(":")[0]) not in pm_sequences or key.endswith(":ROOM_CATEGORY_LABEL"))}
         items = []
         for item_id in sorted(keys, key=lambda key: (int(key.split(":")[0]), key.split(":")[1])):
             sequence, field = item_id.split(":")
@@ -335,6 +349,7 @@ class DataFieldReviews:
                 "revision": state["revision"], "status": state["status"], "items": items,
                 "pending_count": len(issues), "total_count": len(items),
                 "excluded_cancelled_count": len(cancelled_sequences),
+                "excluded_pm_count": len(pm_sequences),
                 "can_finalize": not issues and state["status"] == "editing"}
 
     def prepare(self, request_id, payload, manifest_sha256, report_date):
@@ -523,9 +538,11 @@ class DataFieldReviews:
             actor = self._actor(actor)
             data = self._derive(directory, original, state)
             sequence = int(item_id.split(":")[0]) if isinstance(item_id, str) and re.fullmatch(r"[1-9][0-9]*:[A-Z_]+", item_id) else 0
-            if 1 <= sequence <= len(data["records"]) and self.rules.is_cancelled_record(
-                    {"_RESERVATION_STATUS": data["records"][sequence - 1].get("reservation_status", "")}):
+            cancelled_sequences, pm_sequences = self._excluded_sequences(directory, data)
+            if sequence in cancelled_sequences:
                 raise _error(message="已取消预订不参与日报，无需完善字段")
+            if sequence in pm_sequences and not (item_id.endswith(":ROOM_CATEGORY_LABEL") and item_id in state["decisions"]):
+                raise _error(message="PM 房型不参与日报，无需完善字段")
             if item_id not in self._issues(directory, data) and item_id not in state["decisions"]:
                 raise _error(message="只能完善当前任务列出的异常字段")
             sequence, field = item_id.split(":")

arr_web/local_ohip.py

@@ -215,7 +215,7 @@ def open_portal(root, port):
     load_key(credential_file)
     access = AccessState(root)
     with job_lock(root):
-        database = ReplayDatabase(root, schema_version=20)
+        database = ReplayDatabase(root, schema_version=21)
         try:
             database.start()
             def factory(*, root, snapshot, **dependencies):

arr_web/local_replay_database.py

@@ -14,8 +14,8 @@ from integrations.ohip.capture_job import atomic_json, private_directory
 
 
 class ReplayDatabase:
-    def __init__(self, root: Path, *, schema_version: int = 20):
-        if schema_version not in (18, 19, 20):
+    def __init__(self, root: Path, *, schema_version: int = 21):
+        if schema_version not in (18, 19, 20, 21):
             raise ValueError("unsupported_local_schema")
         self.schema_version = schema_version
         self.root = root

arr_web/processing_runtime.py

@@ -80,6 +80,7 @@ def compose_programmatic_processing(
             connect=connect,
         )
         repository.assert_cancelled_scope_schema()
+        repository.assert_pm_scope_schema()
         policy = load_processor_policy(project_root)
         ingestion = IngestionService(DeliveryValidator(object_store, policy), repository)
         processor = LocalDailyProcessor(policy)

arr_web/static/app.js

@@ -698,6 +698,9 @@
     if (Number.isInteger(review?.excluded_cancelled_count) && review.excluded_cancelled_count > 0) {
       $("#arr-data-review-progress").textContent += ` · ${I18N.t("data_review.excluded_cancelled", { count: review.excluded_cancelled_count })}`;
     }
+    if (Number.isInteger(review?.excluded_pm_count) && review.excluded_pm_count > 0) {
+      $("#arr-data-review-progress").textContent += ` · ${I18N.t("data_review.excluded_pm", { count: review.excluded_pm_count })}`;
+    }
     $("#arr-data-review-refresh").textContent = I18N.t("data_review.refresh");
     $("#arr-data-review-refresh").disabled = busy;
     const finalize = $("#arr-data-review-finalize");

arr_web/static/i18n.js

@@ -172,6 +172,7 @@
     "data_review.loading": ["正在读取待完善字段…", "Loading fields for review…", "กำลังโหลดข้อมูลที่ต้องตรวจสอบ…"],
     "data_review.no_items": ["没有待完善字段。", "No fields need review.", "ไม่มีข้อมูลที่ต้องตรวจสอบ"],
     "data_review.excluded_cancelled": ["已排除 {count} 笔取消预订，原始数据保留", "{count} cancelled reservations excluded; original data retained", "ไม่รวมการจองที่ยกเลิก {count} รายการ โดยเก็บข้อมูลต้นฉบับไว้"],
+    "data_review.excluded_pm": ["已排除 {count} 笔 PM 预订，原始数据保留", "{count} PM reservations excluded; original data retained", "ไม่รวมการจองประเภทห้อง PM {count} รายการ โดยเก็บข้อมูลต้นฉบับไว้"],
     "data_review.region": ["待完善报表字段", "Report fields to review", "ข้อมูลรายงานที่ต้องตรวจสอบ"],
     "data_review.disconnected": ["字段读取中断，请刷新字段后继续。", "Field connection lost. Refresh the fields to continue.", "การเชื่อมต่อข้อมูลขัดข้อง โปรดรีเฟรชข้อมูลเพื่อดำเนินการต่อ"],
     "data_review.saving": ["正在保存并确认字段…", "Saving and confirming field…", "กำลังบันทึกและยืนยันข้อมูล…"],