optimization

app/pipeline/mapping.py

@@ -153,11 +153,20 @@ def build_fixed_row(
         "Уникальный номер операции": str(operation_index),
     }
     merged = _merge_fields(operation_fields, participant_fields)
+    merged_suffix_index = _build_suffix_index(merged)
+    participant_suffix_index = _build_suffix_index(participant_fields)
     for rule in REPORT_MAPPING_RULES:
         if row.get(rule.column_name):
             continue
-        source_payload = participant_fields if rule.source == "participant" else merged
+        if rule.source == "participant":
-        row[rule.column_name] = _pick_value(source_payload, rule.candidate_keys)
+            source_payload = participant_fields
+            suffix_index = participant_suffix_index
+        else:
+            source_payload = merged
+            suffix_index = merged_suffix_index
+        row[rule.column_name] = _pick_value(
+            source_payload, suffix_index, rule.candidate_keys
+        )
     for column in FIXED_REPORT_COLUMNS:
         row.setdefault(column, "")
     return row
@@ -173,17 +182,27 @@ def _merge_fields(
     return merged
 
 
-def _pick_value(payload: dict[str, str], candidates: tuple[str, ...]) -> str:
+def _build_suffix_index(payload: dict[str, str]) -> dict[str, str]:
+    index: dict[str, str] = {}
+    for key, value in payload.items():
+        if not value:
+            continue
+        short = key.rsplit(".", maxsplit=1)[-1]
+        index.setdefault(short, value)
+    return index
+
+
+def _pick_value(
+    payload: dict[str, str],
+    suffix_index: dict[str, str],
+    candidates: tuple[str, ...],
+) -> str:
     for key in candidates:
         if key in payload and payload[key]:
             return payload[key]
-        short = key.split(".")[-1]
+        short = key.rsplit(".", maxsplit=1)[-1]
         if short in payload and payload[short]:
             return payload[short]
-        dotted_suffix = f".{short}"
+        if short in suffix_index:
-        for payload_key, payload_value in payload.items():
+            return suffix_index[short]
-            if not payload_value:
-                continue
-            if payload_key.endswith(dotted_suffix):
-                return payload_value
     return ""

app/pipeline/service.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import tempfile
+import time
 from dataclasses import dataclass, field
 from datetime import datetime
 from pathlib import Path
@@ -39,6 +40,7 @@ class BatchResult:
     protocol_path: str
     log_path: str
     messages: list[str] = field(default_factory=list)
+    duration_seconds: float = 0.0
 
 
 def run_batch(
@@ -48,6 +50,7 @@ def run_batch(
     samba_password: str,
     launcher: str,
 ) -> BatchResult:
+    started_at = time.perf_counter()
     messages: list[str] = []
     if not samba_user:
         return BatchResult(
@@ -195,6 +198,8 @@ def run_batch(
             status=status, processed=processed, partial=partial, errors=errors
         )
     )
+    duration_seconds = round(time.perf_counter() - started_at, 2)
+    messages.append(f"Время выполнения: {_format_duration(duration_seconds)}.")
     messages.extend(log_lines[-5:])
     return BatchResult(
         status=status,
@@ -205,6 +210,7 @@ def run_batch(
         protocol_path=protocol_path,
         log_path=log_path,
         messages=messages,
+        duration_seconds=duration_seconds,
     )
 
 
@@ -362,3 +368,11 @@ def _build_result_message(
             f"успешно={processed}, частично={partial}, с ошибками={errors}."
         )
     return f"Ошибка обработки: успешно={processed}, частично={partial}, с ошибками={errors}."
+
+
+def _format_duration(duration_seconds: float) -> str:
+    total_seconds = max(0, int(round(duration_seconds)))
+    minutes, seconds = divmod(total_seconds, 60)
+    if minutes:
+        return f"{minutes} мин {seconds} сек"
+    return f"{seconds} сек"

tests/unit/test_mapping.py

@@ -121,3 +121,31 @@ def test_extract_tag_candidate_keys_supports_comma_separator() -> None:
         "КППЮЛ, ИдентификацияФЛ, ПризнакОргФормаИНБОЮЛ"
     )
     assert candidates == ("КППЮЛ", "ИдентификацияФЛ", "ПризнакОргФормаИНБОЮЛ")
+
+
+def test_build_fixed_row_uses_first_suffix_value_from_index() -> None:
+    row = build_fixed_row(
+        file_name="f.xml",
+        record_id="R1",
+        operation_index=1,
+        operation_fields={},
+        participant_fields={
+            "УчастникФЛИП.Первый.ИННФЛИП": "111111111111",
+            "УчастникФЛИП.Второй.ИННФЛИП": "222222222222",
+        },
+    )
+    assert row[_column_with("ИНН ")] == "111111111111"
+
+
+def test_build_fixed_row_skips_empty_suffix_values_in_index() -> None:
+    row = build_fixed_row(
+        file_name="f.xml",
+        record_id="R1",
+        operation_index=1,
+        operation_fields={},
+        participant_fields={
+            "УчастникФЛИП.Первый.ИННФЛИП": "",
+            "УчастникФЛИП.Второй.ИННФЛИП": "222222222222",
+        },
+    )
+    assert row[_column_with("ИНН ")] == "222222222222"