fix: separate news from publications and add employee refresh

app/services/crawler.py

@@ -80,6 +80,48 @@ def run_crawl(db: Session, settings: Settings) -> CrawlRun:
     return run
 
 
+def refresh_employee(db: Session, employee: Employee, settings: Settings) -> CrawlRun:
+    run = CrawlRun(source_url=employee.canonical_url, status="running", found_count=1)
+    db.add(run)
+    db.commit()
+    db.refresh(run)
+
+    try:
+        with requests.Session() as session:
+            parsed = parse_person_profile(
+                session,
+                employee.canonical_url,
+                HEADERS,
+                settings.request_timeout,
+                settings.parser_use_playwright,
+            )
+        if not parsed:
+            raise ValueError("Профиль не удалось распарсить.")
+        if _parsed_profile_key(parsed) != employee.profile_key:
+            raise ValueError("Распарсенный профиль не совпадает с обновляемым сотрудником.")
+
+        _upsert_employee(db, run, parsed)
+        run.parsed_count = 1
+        run.status = "completed"
+    except Exception as exc:
+        run.status = "failed"
+        run.error_count = 1
+        run.message = str(exc)
+        db.add(
+            CrawlError(
+                crawl_run_id=run.id,
+                profile_url=employee.canonical_url,
+                error_type=type(exc).__name__,
+                message=str(exc),
+            )
+        )
+    finally:
+        run.finished_at = datetime.now(timezone.utc)
+        db.commit()
+        db.refresh(run)
+    return run
+
+
 def _ensure_source(db: Session, source_url: str) -> ParserSource:
     source = db.scalar(select(ParserSource).where(ParserSource.source_url == source_url))
     if source:
@@ -91,10 +133,14 @@ def _ensure_source(db: Session, source_url: str) -> ParserSource:
     return source
 
 
+def _parsed_profile_key(parsed: dict) -> str:
+    return f"{parsed.get('profile_type')}:{parsed.get('profile_id')}"
+
+
 def _upsert_employee(db: Session, run: CrawlRun, parsed: dict) -> Employee:
     html = parsed.pop("_html", None)
     checksum = _checksum(parsed)
-    key = f"{parsed.get('profile_type')}:{parsed.get('profile_id')}"
+    key = _parsed_profile_key(parsed)
     employee = db.scalar(select(Employee).where(Employee.profile_key == key))
     now = datetime.now(timezone.utc)
     if not employee: