From 27d66d846bb62e05a5bf0298be7fd3c7554ae1f6 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Petr=20Ja=C5=A1ek?= <jasekpetr@gmail.com>
Date: Fri, 27 Sep 2024 11:26:49 +0200
Subject: [PATCH] fix encoding in csv files on windows (#1098)

excel needs BOM to load it properly

CPCN-761
---
 newsroom/agenda/formatters/csv_formatter.py | 25 ++++++++++++++-------
 tests/core/test_csv_formatter.py            |  2 +-
 2 files changed, 18 insertions(+), 9 deletions(-)

diff --git a/newsroom/agenda/formatters/csv_formatter.py b/newsroom/agenda/formatters/csv_formatter.py
index f71a7b141..0188556a3 100644
--- a/newsroom/agenda/formatters/csv_formatter.py
+++ b/newsroom/agenda/formatters/csv_formatter.py
@@ -1,13 +1,14 @@
-from newsroom.formatter import BaseFormatter
 import csv
 import io
 import arrow
-from werkzeug.utils import secure_filename
-from newsroom.utils import parse_dates
+
 from datetime import datetime
 from typing import List, Dict, Any, Union, Tuple, Optional
-from newsroom.agenda.utils import get_filtered_subject
 from flask import current_app as app
+from werkzeug.utils import secure_filename
+from newsroom.utils import parse_dates
+from newsroom.formatter import BaseFormatter
+from newsroom.agenda.utils import get_filtered_subject
 
 
 class CSVFormatter(BaseFormatter):
@@ -39,7 +40,7 @@ def serialize_to_csv(self, items: List[Dict[str, Any]]) -> bytes:
             csv_writer.writerow(item)
 
         csv_string.seek(0)  # Reset the buffer position
-        return csv_string.getvalue().encode("utf-8")
+        return csv_string.getvalue().encode("utf-8-sig")
 
     def format_event(self, item: Dict[str, Any]) -> Dict[str, Any]:
         subj_schemas = app.config.get("AGENDA_CSV_SUBJECT_SCHEMES", [])
@@ -96,9 +97,7 @@ def format_location(self, item: Dict[str, Any], field: str) -> str:
         return ""
 
     def format_list(self, item: Dict[str, Any], key: str, language: Optional[str] = None) -> str:
-        values = [
-            v.get("translations", {}).get("name", {}).get(language) or v.get("name", "") for v in item.get(key, [])
-        ]
+        values = [get_translated_name(v, language) for v in item.get(key, [])]
         return ",".join(list(filter(bool, values)))
 
     def format_contact_info(self, item: Dict[str, Any]) -> str:
@@ -129,3 +128,13 @@ def format_coverage(self, item: Dict[str, Any], field: str) -> str:
             for coverage in coverages:
                 value.append(coverage.get(field, ""))
         return ",".join(value)
+
+
+def get_translated_name(value: Dict[str, Any], language: Optional[str] = None) -> str:
+    """
+    Get translation for the given language
+    """
+    try:
+        return value["translations"]["name"][language]
+    except (KeyError, TypeError):
+        return value.get("name", "")
diff --git a/tests/core/test_csv_formatter.py b/tests/core/test_csv_formatter.py
index f73505fb2..422897119 100644
--- a/tests/core/test_csv_formatter.py
+++ b/tests/core/test_csv_formatter.py
@@ -50,7 +50,7 @@
 
 def read_csv(data):
     csv_data = formatter.format_item(data, item_type="agenda")
-    csv_string = csv_data.decode("utf-8")
+    csv_string = csv_data.decode("utf-8-sig")
     csv_lines = csv_string.split("\n")
     csv_reader = csv.reader(csv_lines)
     header = next(csv_reader)