From ce48ec82b766686fdc69b9c722b49d1a29065634 Mon Sep 17 00:00:00 2001
From: Benjamin Gutzmann <gutzemann@gmail.com>
Date: Thu, 21 Mar 2024 22:07:00 +0100
Subject: [PATCH] Fix parsing of DWD Observation stations where name contains a
 comma

---
 CHANGELOG.rst                                 |  2 +
 .../dwd/observation/test_api_stations.py      | 43 +++++++++++++++++++
 .../provider/dwd/observation/metaindex.py     | 20 ++++-----
 3 files changed, 55 insertions(+), 10 deletions(-)

diff --git a/CHANGELOG.rst b/CHANGELOG.rst
index 1da6c28d5..e6139ae5f 100644
--- a/CHANGELOG.rst
+++ b/CHANGELOG.rst
@@ -4,6 +4,8 @@ Changelog
 Development
 ***********
 
+- Fix parsing of DWD Observation stations where name contains a comma
+
 0.78.0 (09.03.2024)
 *******************
 
diff --git a/tests/provider/dwd/observation/test_api_stations.py b/tests/provider/dwd/observation/test_api_stations.py
index 7f2ab68e2..2d11a7bfd 100644
--- a/tests/provider/dwd/observation/test_api_stations.py
+++ b/tests/provider/dwd/observation/test_api_stations.py
@@ -120,3 +120,46 @@ def test_dwd_observations_stations_minute_1(default_settings):
         },
     )
     assert_frame_equal(given_df, expected_df)
+
+
+@pytest.mark.remote
+def test_dwd_observations_stations_name_with_comma():
+    request = DwdObservationRequest(
+        parameter="kl",
+        resolution="monthly",
+        period="recent",
+    )
+    stations = request.all()
+    stations = stations.df.filter(pl.col("station_id").is_in(["00314", "03164", "06272"]))
+    assert stations.to_dicts() == [
+        {
+            "station_id": "00314",
+            "start_date": dt.datetime(1881, 1, 1, 0, 0, tzinfo=ZoneInfo(key="UTC")),
+            "end_date": dt.datetime(2024, 2, 29, 0, 0, tzinfo=ZoneInfo(key="UTC")),
+            "latitude": 51.1604,
+            "longitude": 14.5042,
+            "height": 234.0,
+            "name": "Kubschütz, Kr. Bautzen",
+            "state": "Sachsen",
+        },
+        {
+            "station_id": "03164",
+            "start_date": dt.datetime(1881, 1, 1, 0, 0, tzinfo=ZoneInfo(key="UTC")),
+            "end_date": dt.datetime(2024, 2, 29, 0, 0, tzinfo=ZoneInfo(key="UTC")),
+            "latitude": 50.8492,
+            "longitude": 8.7745,
+            "height": 187.0,
+            "name": "Cölbe, Kr. Marburg-Biedenkopf",
+            "state": "Hessen",
+        },
+        {
+            "station_id": "06272",
+            "start_date": dt.datetime(2004, 10, 1, 0, 0, tzinfo=ZoneInfo(key="UTC")),
+            "end_date": dt.datetime(2024, 2, 29, 0, 0, tzinfo=ZoneInfo(key="UTC")),
+            "latitude": 50.8426,
+            "longitude": 10.2518,
+            "height": 284.0,
+            "name": "Salzungen, Bad-Gräfen-Nitzendorf",
+            "state": "Thüringen",
+        },
+    ]
diff --git a/wetterdienst/provider/dwd/observation/metaindex.py b/wetterdienst/provider/dwd/observation/metaindex.py
index d27162edd..41cc59601 100644
--- a/wetterdienst/provider/dwd/observation/metaindex.py
+++ b/wetterdienst/provider/dwd/observation/metaindex.py
@@ -182,17 +182,17 @@ def _read_meta_df(file: BytesIO) -> pl.LazyFrame:
     if first.startswith("SP"):
         # Skip first line if it contains a header
         lines = lines[1:]
-    content = BytesIO(b"\n".join(lines))
-    df = pl.read_csv(source=content, encoding="latin-1", has_header=False, truncate_ragged_lines=True)
+    lines = [line.decode("latin-1") for line in lines]
+    df = pl.DataFrame(lines)
     column_specs = (
-        (0, 5),
-        (6, 14),
-        (15, 24),
-        (23, 38),
-        (38, 50),
-        (50, 60),
-        (60, 102),
-        (102, 200),
+        (0, 4),
+        (6, 13),
+        (15, 22),
+        (24, 37),
+        (39, 49),
+        (51, 59),
+        (61, 140),
+        (141, 200),
     )
     df = read_fwf_from_df(df, column_specs)
     return df.rename(mapping=lambda col: DWD_COLUMN_NAMES_MAPPING.get(col, col)).lazy()