feat: add date-based OHIP data reporting alongside XML

integrations/ohip/arr_xml_contract.py

@@ -0,0 +1,92 @@
+"""Explicit row values for ARR processing XML, not an OHIP mapping policy.
+
+Callers must already have resolved every value, note/trace selection and row
+order from independently accepted source rules. No fields have defaults.
+"""
+from __future__ import annotations
+
+from dataclasses import dataclass
+from datetime import date
+import re
+
+from .collect_arr_source import require
+
+VERSION = "arr-processing-xml/v1"
+# Match the existing frozen handoff/upload boundary, not the larger diagnostic
+# native-XML comparator limit. A serialization result must fit that entrypoint.
+MAX_BYTES = 25 * 1024 * 1024
+MAX_ROWS = 10000
+MAX_ELEMENTS = 500000
+
+# Declarative correspondence only; shared by serializer and independent parser.
+SCALARS = (
+    ("block_code", "BLOCK_CODE"), ("adults", "ADULTS"), ("children", "CF_CHILDREN"),
+    ("company_name", "COMPANY_NAME"), ("confirmation_no", "CONFIRMATION_NO"),
+    ("room_no", "DISP_ROOM_NO"), ("effective_rate", "EFFECTIVE_RATE_AMOUNT"),
+    ("full_name", "FULL_NAME"), ("no_of_rooms", "NO_OF_ROOMS"), ("products", "PRODUCTS"),
+    ("rate_code", "RATE_CODE"), ("room_category_label", "ROOM_CATEGORY_LABEL"),
+    ("arrival", "TRUNC_BEGIN"), ("departure", "TRUNC_END"),
+)
+
+
+@dataclass(frozen=True)
+class SourceRow:
+    reservation_id: str
+    block_code: str
+    adults: str
+    children: str
+    company_name: str
+    confirmation_no: str
+    room_no: str
+    effective_rate: str
+    full_name: str
+    no_of_rooms: str
+    products: str
+    rate_code: str
+    room_category_label: str
+    arrival: str
+    departure: str
+    notes: tuple[str, ...]
+    traces: tuple[str, ...]
+
+
+@dataclass(frozen=True)
+class SourceDocument:
+    hotel_id: str
+    report_date: date
+    rows: tuple[SourceRow, ...]
+
+
+def _text(value):
+    require(type(value) is str, "arr_xml_explicit_text_required")
+    require(all(char in "\t\n\r" or 0x20 <= ord(char) <= 0xD7FF
+                or 0xE000 <= ord(char) <= 0xFFFD or 0x10000 <= ord(char) <= 0x10FFFF
+                for char in value), "arr_xml_invalid_character")
+    return len(value.encode("utf-8"))
+
+
+def validate_document(document: SourceDocument) -> None:
+    """Validate transport shape only; existing processor owns business rules.
+
+An explicit blank is retained even on a required whitelist field. A missing
+value/None is rejected. This preserves invalid and non-whitelist source rows so
+serialization cannot silently remove records to make processing succeed.
+"""
+    require(type(document) is SourceDocument, "arr_xml_document_required")
+    require(type(document.hotel_id) is str and bool(re.fullmatch(r"[A-Za-z0-9_-]{1,64}", document.hotel_id)),
+            "arr_xml_hotel_required")
+    require(type(document.report_date) is date, "arr_xml_explicit_date_required")
+    require(type(document.rows) is tuple and 1 <= len(document.rows) <= MAX_ROWS, "arr_xml_record_count")
+    size, elements = 0, 6
+    for row in document.rows:
+        require(type(row) is SourceRow, "arr_xml_row_required")
+        size += _text(row.reservation_id)
+        require(bool(row.reservation_id.strip()), "arr_xml_reservation_identity_required")
+        for attribute, _ in SCALARS:
+            size += _text(getattr(row, attribute))
+        for collection in (row.notes, row.traces):
+            require(type(collection) is tuple, "arr_xml_explicit_collection_required")
+            for value in collection:
+                size += _text(value)
+        elements += 19 + 2 * (len(row.notes) + len(row.traces))
+        require(size <= MAX_BYTES and elements <= MAX_ELEMENTS, "arr_xml_input_budget_exceeded")