Add v2 registry uploads and source CSV exports

src/apps/parsers/api_v2_urls.py

@@ -0,0 +1,39 @@
+"""API v2 routes for parser source exports."""
+
+from apps.parsers.models import ParserLoadLog
+from apps.parsers.source_registry import PARSER_SOURCES, ParserSourceDescriptor
+from apps.parsers.views import SourceResultCsvDownloadView
+from django.urls import path
+
+app_name = "parser_sources"
+
+
+def _download_source_descriptors():
+    """Return unique source result routes that support CSV download."""
+    seen_routes = set()
+    for descriptor in PARSER_SOURCES.values():
+        if not descriptor.api_route or descriptor.api_route in seen_routes:
+            continue
+        seen_routes.add(descriptor.api_route)
+        if descriptor.source == ParserLoadLog.Source.FNS_REPORTS:
+            continue
+        yield descriptor
+
+
+def _download_view(descriptor: ParserSourceDescriptor):
+    class SourceDownloadView(SourceResultCsvDownloadView):
+        source_key = descriptor.key
+
+    return SourceDownloadView.as_view()
+
+
+urlpatterns = []
+for source_descriptor in _download_source_descriptors():
+    route_name = source_descriptor.api_route.replace("/", "-")
+    urlpatterns.append(
+        path(
+            f"{source_descriptor.api_route}/download/",
+            _download_view(source_descriptor),
+            name=f"{route_name}-download",
+        )
+    )

src/apps/parsers/serializers.py

@@ -520,6 +520,7 @@ class ParserSourceSerializer(serializers.Serializer):
     parser_strategy = serializers.CharField()
     source_notes = serializers.CharField(allow_blank=True)
     supports_file_upload = serializers.BooleanField()
+    api_route = serializers.CharField(allow_blank=True)
     result_list_url = serializers.CharField()
     result_detail_url = serializers.CharField()
     upload_url = serializers.CharField(allow_blank=True)

src/apps/parsers/views.py

@@ -2295,6 +2295,79 @@ class SourceResultListView(APIView):
         )
 
 
+SOURCE_CSV_HEADERS = [
+    "id",
+    "load_batch",
+    "source",
+    "external_id",
+    "inn",
+    "ogrn",
+    "organisation_name",
+    "title",
+    "record_date",
+    "amount",
+    "status",
+    "url",
+    "payload",
+    "created_at",
+    "updated_at",
+]
+
+
+def _csv_cell(value) -> str:
+    if value is None:
+        return ""
+    if isinstance(value, dict):
+        return json.dumps(value, ensure_ascii=False, sort_keys=True)
+    if hasattr(value, "isoformat"):
+        return value.isoformat()
+    return str(value)
+
+
+def _source_csv_filename(descriptor) -> str:
+    route_name = descriptor.api_route.replace("/", "-") or descriptor.key
+    return f"{get_valid_filename(route_name)}.csv"
+
+
+class SourceResultCsvDownloadView(APIView):
+    """CSV download результата конкретного источника API v2."""
+
+    permission_classes = [IsAuthenticated]
+    source_key = ""
+
+    def get(self, request: Request, source_key: str | None = None):
+        resolved_source_key = source_key or self.source_key
+        query_serializer = ParserResultQuerySerializer(data=request.query_params)
+        query_serializer.is_valid(raise_exception=True)
+        params = query_serializer.validated_data
+        params["include_payload"] = True
+
+        descriptor, queryset = _filter_result_queryset(resolved_source_key, params)
+        if descriptor is None:
+            return _source_not_found_response(resolved_source_key)
+        if descriptor.source == ParserLoadLog.Source.FNS_REPORTS:
+            return _source_not_found_response(resolved_source_key)
+
+        response = HttpResponse(content_type="text/csv; charset=utf-8")
+        response[
+            "Content-Disposition"
+        ] = f'attachment; filename="{_source_csv_filename(descriptor)}"'
+
+        writer = csv.writer(response)
+        writer.writerow(SOURCE_CSV_HEADERS)
+        for record in queryset.iterator(chunk_size=1000):
+            row = _result_record_to_dict(
+                descriptor.source,
+                record,
+                include_payload=True,
+            )
+            writer.writerow(
+                [_csv_cell(row.get(header)) for header in SOURCE_CSV_HEADERS]
+            )
+
+        return response
+
+
 class SourceResultDetailView(APIView):
     """GET одной записи результата источника."""