Remove raise IndexError on dates in wrong sequence, remove test for d…

…ates in wrong sequence. Add testcases, doctests
statisticsnorway · Jan 30, 2024 · a9b7f11 · a9b7f11
1 parent 3f962ad
commit a9b7f11
Show file tree

Hide file tree

Showing 2 changed files with 30 additions and 48 deletions.
diff --git a/src/datadoc/backend/dapla_dataset_path_info.py b/src/datadoc/backend/dapla_dataset_path_info.py
@@ -192,6 +192,10 @@ def categorize_period_string(period: str) -> IsoDateFormat | SsbDateFormat:
     >>> date_format.name
     SSB_HALF_YEAR
 
+    >>> date_format = categorize_period_string('1876H5') # Not valid SSB date format, number is out of range
+    >>> date_format.name
+    SSB_HALF_YEAR
+
     >>> categorize_period_string('unknown format')
     Traceback (most recent call last):
      ...
@@ -247,6 +251,10 @@ def convert_ssb_period(
     >>> ssb_half_year_period_end
     189812
 
+    >>> convert_ssb_period("2018Q5","start",SSB_QUARTERLY)
+    Traceback (most recent call last):
+    KeyError: 'Q5'
+
     """
     return period_string[:4] + date_format.time_frame[period_string[-2:]][period_type]
 
@@ -265,9 +273,6 @@ def __init__(self, dataset_path: str) -> None:
         with contextlib.suppress(IndexError):
             self.second_period_string = _period_strings[1]
 
-        if self.contains_data_until < self.contains_data_from:
-            raise IndexError
-
     @staticmethod
     def _extract_period_strings(dataset_name_sections: list[str]) -> list[str]:
         """Extract period strings from dataset name sections.
@@ -285,6 +290,10 @@ def _extract_period_strings(dataset_name_sections: list[str]) -> list[str]:
 
         >>> DaplaDatasetPathInfo._extract_period_strings(['p1990Q1', 'kommune', 'v1'])
         ['1990Q1']
+
+        >>> DaplaDatasetPathInfo._extract_period_strings(['varehandel','v1']) # No date will return empty string
+        []
+
         """
         date_format_regex = re.compile(
             r"^p\d{4}(?:-\d{2}-\d{2}|-\d{2}|-{0,1}[QTHWB]\d{1,2})?$",

diff --git a/tests/backend/test_dapla_dataset_path_info.py b/tests/backend/test_dapla_dataset_path_info.py
@@ -81,6 +81,21 @@ class DatasetPathTestCase:
         expected_contains_data_from=datetime.date(2022, 1, 1),
         expected_contains_data_until=datetime.date(2022, 2, 28),
     ),
+    DatasetPathTestCase(
+        path="ufo_observasjoner_p2019_p1920_v1.parquet",  # Will pass, but wrong sequence dates
+        expected_contains_data_from=datetime.date(2019, 1, 1),
+        expected_contains_data_until=datetime.date(1920, 12, 31),
+    ),
+    DatasetPathTestCase(
+        path="varehandel_p2018H2_p2018H1_v1.parquet",  # Will pass, but wrong sequence dates
+        expected_contains_data_from=datetime.date(2018, 7, 1),
+        expected_contains_data_until=datetime.date(2018, 6, 30),
+    ),
+    DatasetPathTestCase(
+        path="varehandel_p2018Q1_p2018H2_v1.parquet",  # Will pass, but combination of two different SSB dates
+        expected_contains_data_from=datetime.date(2018, 1, 1),
+        expected_contains_data_until=datetime.date(2018, 12, 31),
+    ),
 ]
 
 
@@ -121,58 +136,16 @@ def test_extract_period_info_date_until(
     assert dataset_path.contains_data_until == expected_contains_data_until
 
 
-@pytest.mark.parametrize(
-    "data",
-    [
-        (
-            "varehandel_p2018Q5_p2018Q4_v1.parquet",
-            "Period format p2018Q5 is not supported",
-        ),
-        (
-            "varehandel_p2018Q1_p2018H2_v1.parquet",
-            "Period format p2018H2 is not supported",
-        ),
-    ],
-)
-def test_extract_period_info_failures(data: tuple):
-    DaplaDatasetPathInfo(data[0])
-    with pytest.raises(NotImplementedError):
-        raise NotImplementedError(data[1])
-
-
+# Nonsens names, no dates path names will trigger IndexError
 @pytest.mark.parametrize(
     "data",
     [
         "nonsen.data",
         "nonsens2.parquet",
+        "nonsens_v1.parquet",
+        "varehandel_v1.parquet",
     ],
 )
 def test_extract_period_info_failures_index_error(data: str):
     with pytest.raises(IndexError):
         DaplaDatasetPathInfo(data)
-
-
-@pytest.mark.parametrize(
-    "data",
-    [
-        (
-            "varehandel_p2018H2_p2018H1_v1.parquet",
-            "Periods are not in sequence",
-        ),
-        (
-            "varehandel_p2018T3_p2018T1_v1.parquet",
-            "Periods are not in sequence",
-        ),
-        (
-            "varehandel_p2018Q4_p2018Q1_v1.parquet",
-            "Periods are not in sequence",
-        ),
-        (
-            "varehandel_p2018B6_p2018B1_v1.parquet",
-            "Periods are not in sequence",
-        ),
-    ],
-)
-def test_extract_period_info_in_sequence(data: tuple):
-    with pytest.raises(IndexError):
-        DaplaDatasetPathInfo(data[0])