diff --git a/personal_data/parse_util.py b/personal_data/parse_util.py
index 0e0bf76..9a89557 100644
--- a/personal_data/parse_util.py
+++ b/personal_data/parse_util.py
@@ -21,9 +21,9 @@ FORMAT_DATE_HEADER = '%a, %d %b %Y %H:%M:%S GMT'
 
 
 def parse_duration(text: str) -> datetime.timedelta:
-    (num, unit) = text.split(' ')
-    num = int(num)
-    unit = DATETIME_UNITS[unit]
+    (num_str, unit_str) = text.split(' ')
+    num = int(num_str)
+    unit = DATETIME_UNITS[unit_str]
     return unit * num
 
 
@@ -57,12 +57,16 @@ def parse_time(text: str) -> datetime.datetime:
     if time is None and (m := try_parse(text, '%d %b @ %I:%M%p')):
         time = m.replace(year=NOW.year)
 
-    assert time is not None, 'Could not parse format'
+    if time is None:
+        msg = 'Unknown format: ' + text
+        raise RuntimeError(msg)
 
     if time.tzinfo is None:
         time = time.replace(tzinfo=LOCAL_TIMEZONE)
 
-    assert time.tzinfo is not None, time
+    if time.tzinfo is None:
+        msg = 'Could not parse timezone: ' + text
+        raise RuntimeError(msg)
     return time
 
 
@@ -74,4 +78,5 @@ def parse_date(text: str) -> datetime.date:
         return dt.date()
     if dt := try_parse(text, '%B %d, %Y'):
         return dt.date()
-    assert False, text
+    msg = 'Unknown format: ' + text
+    raise RuntimeError(msg)
diff --git a/personal_data/util.py b/personal_data/util.py
index b8eb018..e43289e 100644
--- a/personal_data/util.py
+++ b/personal_data/util.py
@@ -3,9 +3,8 @@ import csv
 import datetime
 import io
 import logging
-import typing
 import urllib.parse
-from collections.abc import Iterable, Mapping, Sequence
+from collections.abc import Iterable, Mapping
 from pathlib import Path
 from typing import Any
 
@@ -20,13 +19,14 @@ def csv_safe_value(v: Any) -> str:
     if isinstance(v, urllib.parse.ParseResult):
         return v.geturl()
     if isinstance(v, datetime.datetime):
-        assert v.tzinfo is not None, v
+        if v.tzinfo is None:
+            raise RuntimeError(v)
     return str(v)
 
 
 def equals_without_fields(
-    a: Mapping[str, object],
-    b: Mapping[str, object],
+    a: Mapping[str, Any],
+    b: Mapping[str, Any],
     fields: Iterable[str] = frozenset(),
 ) -> bool:
     a = dict(a)
@@ -39,14 +39,13 @@ def equals_without_fields(
 
 
 def deduplicate_by_ignoring_certain_fields(
-    dicts: list[dict],
+    dicts: list[frozendict[str, Any]],
     deduplicate_ignore_columns: Iterable[str],
-) -> list[dict]:
+) -> list[frozendict[str, Any]]:
     """Removes duplicates that occur when ignoring certain columns.
 
     Output order is stable.
     """
-
     to_remove = set()
     for idx1, first in enumerate(dicts):
         for idx2, second in enumerate(dicts[idx1 + 1 :], idx1 + 1):
@@ -55,19 +54,21 @@ def deduplicate_by_ignoring_certain_fields(
             del idx2, second
         del idx1, first
 
-    to_remove = sorted(to_remove)
-    while to_remove:
-        del dicts[to_remove.pop()]
+    to_remove_ls = sorted(to_remove)
+    del to_remove
+    while to_remove_ls:
+        del dicts[to_remove_ls.pop()]
 
     return dicts
 
 
 def deduplicate_dicts(
-    dicts: Sequence[dict[str, typing.Any] | frozendict[str, typing.Any]],
+    dicts: list[frozendict[str, Any]],
     deduplicate_mode: data.DeduplicateMode,
     deduplicate_ignore_columns: list[str],
-) -> tuple[Sequence[dict[str, typing.Any]], list[str]]:
-    assert isinstance(deduplicate_ignore_columns, list), deduplicate_ignore_columns
+) -> tuple[list[frozendict[str, Any]], list[str]]:
+    if not isinstance(deduplicate_ignore_columns, list):
+        raise TypeError(deduplicate_ignore_columns)
 
     fieldnames = []
     for d in dicts:
@@ -78,7 +79,7 @@ def deduplicate_dicts(
         del d
 
     if deduplicate_mode == data.DeduplicateMode.ONLY_LATEST:
-        while len(dicts) >= 2 and equals_without_fields(
+        while len(dicts) > 1 and equals_without_fields(
             dicts[-1],
             dicts[-2],
             deduplicate_ignore_columns,
@@ -90,13 +91,13 @@ def deduplicate_dicts(
             deduplicate_ignore_columns,
         )
     elif deduplicate_mode != data.DeduplicateMode.NONE:
-        dicts = set(dicts)
+        dicts = list(set(dicts))
 
     dicts = sorted(dicts, key=lambda d: tuple(str(d.get(fn, '')) for fn in fieldnames))
     return dicts, fieldnames
 
 
-def normalize_dict(d: dict[str, typing.Any]) -> frozendict[str, typing.Any]:
+def normalize_dict(d: dict[str, Any] | frozendict[str, Any]) -> frozendict[str, Any]:
     return frozendict(
         {
             k: csv_import.csv_str_to_value(str(v))
@@ -108,20 +109,23 @@ def normalize_dict(d: dict[str, typing.Any]) -> frozendict[str, typing.Any]:
 
 def extend_csv_file(
     csv_file: Path,
-    new_dicts: list[dict[str, typing.Any]],
+    new_dicts: list[dict[str, Any] | frozendict[str, Any]],
     deduplicate_mode: data.DeduplicateMode,
     deduplicate_ignore_columns: list[str],
 ) -> dict:
-    assert isinstance(deduplicate_ignore_columns, list), deduplicate_ignore_columns
+    if not isinstance(deduplicate_ignore_columns, list):
+        raise TypeError(deduplicate_ignore_columns)
 
     try:
-        dicts = csv_import.load_csv_file(csv_file)
-    except (FileNotFoundError, _csv.Error) as e:
+        original_dicts = csv_import.load_csv_file(csv_file)
+    except (FileNotFoundError, _csv.Error):
         logger.info('Creating file: %s', csv_file)
-        dicts = []
+        original_dicts = []
 
-    original_num_dicts = len(dicts)
-    dicts += [normalize_dict(d) for d in new_dicts]
+    original_num_dicts = len(original_dicts)
+    dicts = [normalize_dict(d) for d in original_dicts] + [
+        normalize_dict(d) for d in new_dicts
+    ]
     del new_dicts
 
     dicts, fieldnames = deduplicate_dicts(
diff --git a/test/test_deduplicate.py b/test/test_deduplicate.py
index 10250f9..4ca4506 100644
--- a/test/test_deduplicate.py
+++ b/test/test_deduplicate.py
@@ -39,7 +39,7 @@ def test_all_fields():
     ]
 
 
-def test_all_fields():
+def test_all_fields_for_duplicated_list():
     ls, fields = deduplicate_dicts(LIST + LIST, DeduplicateMode.BY_ALL_COLUMNS, ['t'])
     assert fields == ['a', 'b', 't']
     print(ls)
diff --git a/test/test_parsing.py b/test/test_parsing.py
index e9b08ad..747f9b7 100644
--- a/test/test_parsing.py
+++ b/test/test_parsing.py
@@ -54,6 +54,6 @@ PARSE_MAPPINGS = [
 ]
 
 
-@pytest.mark.parametrize('text,parsed', PARSE_MAPPINGS)
-def test_csv_str_to_value(text, parsed):
+@pytest.mark.parametrize(('text', 'parsed'), PARSE_MAPPINGS)
+def test_csv_str_to_value(text: str, parsed: object):
     assert csv_str_to_value(text) == parsed, text
diff --git a/test/test_psnprofiles.py b/test/test_psnprofiles.py
index bc90da3..c818cae 100644
--- a/test/test_psnprofiles.py
+++ b/test/test_psnprofiles.py
@@ -10,6 +10,6 @@ URLS_AND_IDS = [
 ]
 
 
-@pytest.mark.parametrize('id, url', URLS_AND_IDS)
-def test_game_psnprofiles_id_from_url(id, url):
-    assert psnprofiles.game_psnprofiles_id_from_url(url) == id
+@pytest.mark.parametrize(('psnprofiles_id', 'url'), URLS_AND_IDS)
+def test_game_psnprofiles_id_from_url(psnprofiles_id: int, url: str):
+    assert psnprofiles.game_psnprofiles_id_from_url(url) == psnprofiles_id
diff --git a/test/test_util.py b/test/test_util.py
index d083a8a..7d51335 100644
--- a/test/test_util.py
+++ b/test/test_util.py
@@ -2,5 +2,4 @@ import personal_data
 
 
 def test_version():
-    assert personal_data._version.__version__ is not None
     assert personal_data.__version__ is not None