Removing mutable default args

This is a common pitfall in Python that causes unexpected behavior.

https://docs.python-guide.org/writing/gotchas/#mutable-default-arguments

.gitignore

@@ -13,6 +13,10 @@ tabbycat/static/jsi18n/
 tabbycat/static/vue/
 tabbycat/static/fonts/
 
+# IDE files
+.idea/
+.vscode/
+
 # Translations
 .tx/
 

tabbycat/adjfeedback/tests/test_progress.py

@@ -61,13 +61,15 @@ def _dt(self, debate, t):
     def _da(self, debate, a):
         return DebateAdjudicator.objects.get(debate=debate, adjudicator=self._adj(a))
 
-    def _create_debate(self, teams, adjs, votes, trainees=[], venue=None):
+    def _create_debate(self, teams, adjs, votes, trainees=None, venue=None):
         """Enters a debate into the database, using the teams and adjudicators specified.
         `votes` should be a string (or iterable of characters) indicating "a" for affirmative or
             "n" for negative, e.g. "ann" if the chair was rolled in a decision for the negative.
         The method will give the winning team all 76s and the losing team all 74s.
         The first adjudicator is the chair; the rest are panellists."""
 
+        if trainees is None:
+            trainees = []
         if venue is None:
             venue = Venue.objects.first()
         debate = Debate.objects.create(round=self.rd, venue=venue)
@@ -134,7 +136,10 @@ def _create_feedback(self, source, target, confirmed=True, ignored=False):
     # From team
     # ==========================================================================
 
-    def assertExpectedFromTeamTracker(self, debate, t, expected, fulfilled, count, submissions, targets, tracker_kwargs={}): # noqa
+    def assertExpectedFromTeamTracker(self, debate, t, expected, fulfilled, count, submissions, targets, tracker_kwargs=None): # noqa
+        if tracker_kwargs is None:
+            tracker_kwargs = {}
+
         tracker = FeedbackExpectedSubmissionFromTeamTracker(self._dt(debate, t), **tracker_kwargs)
         self.assertIs(tracker.expected, expected)
         self.assertIs(tracker.fulfilled, fulfilled)

tabbycat/breakqual/liveness.py

@@ -26,7 +26,9 @@ def get_coefficient(m, k):
     return half_row + half_row[-2::-1]
 
 
-def liveness_twoteam(is_general, current_round, break_size, total_teams, total_rounds, team_scores=[]):
+def liveness_twoteam(is_general, current_round, break_size, total_teams, total_rounds, team_scores=None):
+    if team_scores is None:
+        team_scores = []
 
     if total_teams < break_size or (not is_general and len(team_scores) <= break_size):
         return 0, -1  # special case, everyone is safe
@@ -64,7 +66,9 @@ def liveness_twoteam(is_general, current_round, break_size, total_teams, total_r
     return safe, dead
 
 
-def liveness_bp(is_general, current_round, break_size, total_teams, total_rounds, team_scores=[]):
+def liveness_bp(is_general, current_round, break_size, total_teams, total_rounds, team_scores=None):
+    if team_scores is None:
+        team_scores = []
 
     if total_teams < break_size or (not is_general and len(team_scores) <= break_size):
         return -1, -1  # special case, everyone is safe

tabbycat/draw/generator/pairing.py

@@ -37,10 +37,15 @@ class BasePairing:
     This is a base class for functionality common to both two-team pairings and
     BP pairings."""
 
-    def __init__(self, teams, bracket, room_rank, flags=[], team_flags={}):
+    def __init__(self, teams, bracket, room_rank, flags=None, team_flags=None):
         """'teams' must be a list of two teams, or four teams if it's for BP.
         'bracket' and 'room_rank' are both integers.
         'flags' is a list of strings."""
+        if flags is None:
+            flags = []
+        if team_flags is None:
+            team_flags = {}
+
         self.teams = list(teams)
         self.bracket = bracket
         self.room_rank = room_rank
@@ -95,7 +100,7 @@ class Pairing(BasePairing):
 
     sides = [DebateSide.AFF, DebateSide.NEG]
 
-    def __init__(self, teams, bracket, room_rank, num_sides=2, flags=[], team_flags={}):
+    def __init__(self, teams, bracket, room_rank, num_sides=2, flags=None, team_flags=None):
         super().__init__(teams, bracket, room_rank, flags, team_flags)
         assert len(self.teams) == 2, "There must be two teams in a Pairing"
 
@@ -147,7 +152,7 @@ class ResultPairing(Pairing):
     This class is the data structure expected by DrawGenerator classes, when
     taking information about the results of the previous round."""
 
-    def __init__(self, teams, bracket, room_rank, flags=[], team_flags={}, winner=None):
+    def __init__(self, teams, bracket, room_rank, flags=None, team_flags=None, winner=None):
         super().__init__(teams, bracket, room_rank, flags, team_flags)
         self.set_winner(winner)
 
@@ -176,7 +181,7 @@ def winner(self):
 class PolyPairing(BasePairing):
     """Pairing class for British Parliamentary and Public Speaking."""
 
-    def __init__(self, teams, bracket, room_rank, num_sides=4, flags=[], team_flags={}):
+    def __init__(self, teams, bracket, room_rank, num_sides=4, flags=None, team_flags=None):
         super().__init__(teams, bracket, room_rank, flags, team_flags)
 
     def __repr__(self):
@@ -191,7 +196,7 @@ class BPEliminationResultPairing(PolyPairing):
 
     sides = [DebateSide.OG, DebateSide.OO, DebateSide.CG, DebateSide.CO]
 
-    def __init__(self, teams, bracket, room_rank, flags=[], team_flags={}, advancing=[]):
+    def __init__(self, teams, bracket, room_rank, flags=None, team_flags=None, advancing=None):
         super().__init__(teams, bracket, room_rank, 4, flags, team_flags)
         self.set_advancing(advancing)
 

tabbycat/draw/tables.py

@@ -65,7 +65,10 @@ class PublicDrawTableBuilder(BaseDrawTableBuilder):
     def get_sides(self, debates: List['Debate']) -> int:
         return max([dt.side for debate in debates for dt in debate.debateteams], default=self.tournament.pref('teams_in_debate') - 1) + 1
 
-    def add_debate_team_columns(self, debates, highlight=[]):
+    def add_debate_team_columns(self, debates, highlight=None):
+        if highlight is None:
+            highlight = []
+
         all_sides_confirmed = all(debate.sides_confirmed for debate in debates)  # should already be fetched
 
         for side in range(self.get_sides(debates)):

tabbycat/draw/views.py

@@ -90,7 +90,10 @@ def get_page_subtitle(self):
         else:
             return ""
 
-    def populate_table(self, debates, table, highlight=[]):
+    def populate_table(self, debates, table, highlight=None):
+        if highlight is None:
+            highlight = []
+
         table.add_debate_venue_columns(debates)
         table.add_debate_team_columns(debates, highlight)
         table.add_debate_adjudicators_column(debates, show_splits=False)
@@ -232,7 +235,10 @@ def get_page_subtitle(self):
     def get_page_emoji(self):
         return None
 
-    def populate_table(self, debates, table, highlight=[]):
+    def populate_table(self, debates, table, highlight=None):
+        if highlight is None:
+            highlight = []
+
         table.add_round_column(d.round for d in debates)
         super().populate_table(debates, table, highlight=highlight)
 

tabbycat/importer/importers/base.py

@@ -25,10 +25,13 @@ def convert_bool(value):
         raise ValueError('Invalid boolean value: %s' % (value,))
 
 
-def make_interpreter(DELETE=[], **kwargs):  # noqa: N803
+def make_interpreter(DELETE=None, **kwargs):  # noqa: N803
     """Convenience function for building an interpreter. The default interpreter
     (i.e. the one returned if no arguments are passed to this function) just
     removes blank values."""
+    if DELETE is None:
+        DELETE = []
+
     def interpreter(lineno, line):
         # remove blank and unwanted values
         line = {

tabbycat/options/forms.py

@@ -48,7 +48,10 @@ def get_pref(name, section=section):
         return self.cleaned_data
 
 
-def tournament_preference_form_builder(instance, preferences=[], **kwargs):
+def tournament_preference_form_builder(instance, preferences=None, **kwargs):
+    if preferences is None:
+        preferences = []
+
     if kwargs.get('section') in [str(s) for s in global_preferences_registry.sections()]:
         # Check for global preferences
         return preference_form_builder(GlobalPreferenceForm, preferences, **kwargs)

tabbycat/results/result.py

@@ -664,11 +664,11 @@ class DebateResultWithScoresMixin:
     uses_declared_winners = False
     uses_speakers = True
 
-    def __init__(self, ballotsub, load=True, criteria=[], **kwargs):
+    def __init__(self, ballotsub, load=True, criteria=None, **kwargs):
         super().__init__(ballotsub, load=False, **kwargs)
 
         self.positions = self.tournament.positions
-        self.criteria = criteria or []
+        self.criteria = [] if criteria is None else criteria
 
         if load:
             if self.ballotsub.id is None:

tabbycat/tournaments/models.py

@@ -441,7 +441,7 @@ def num_available_adjudicators_not_allocated(self):
     # Draw retrieval methods
     # --------------------------------------------------------------------------
 
-    def debate_set_with_prefetches(self, filter_args=[], filter_kwargs={}, ordering=(F('venue__name').asc(nulls_last=True),),
+    def debate_set_with_prefetches(self, filter_args=None, filter_kwargs=None, ordering=(F('venue__name').asc(nulls_last=True),),
             teams=True, adjudicators=True, speakers=True, wins=False,
             results=False, venues=True, institutions=False, check_ins=False, iron=False):
         """Returns the debate set, with aff_team and neg_team populated.
@@ -451,6 +451,10 @@ def debate_set_with_prefetches(self, filter_args=[], filter_kwargs={}, ordering=
         from draw.models import DebateTeam
         from participants.models import Speaker
         from results.prefetch import populate_confirmed_ballots, populate_wins, populate_checkins
+        if filter_args is None:
+            filter_args = []
+        if filter_kwargs is None:
+            filter_kwargs = {}
 
         debates = self.debate_set.filter(*filter_args, **filter_kwargs)
         if results:

tabbycat/utils/admin.py

@@ -27,8 +27,12 @@ def log_addition(self, request, object, message):
     def log_change(self, request, object, message):
         return super().log_change(request, object, self.add_ip_to_message(request, message))
 
-    def log_deletion(self, request, object, object_repr, message=[]):
+    def log_deletion(self, request, object, object_repr, message=None):
         from django.contrib.admin.models import DELETION, LogEntry
+
+        if message is None:
+            message = []
+
         return LogEntry.objects.log_action(
             user_id=request.user.pk,
             content_type_id=get_content_type_for_model(object).pk,

tabbycat/utils/tables.py

@@ -803,12 +803,14 @@ def add_ranking_columns(self, standings):
             } for ranking in standing.iterrankings()])
         self.add_columns(headers, data)
 
-    def add_metric_columns(self, standings, integer_score_columns=[]):
+    def add_metric_columns(self, standings, integer_score_columns=None):
         """`integer_score_columns`, if given, indicates which metrics to cast to
         an int if the metric's value is an integer. For example, if the
         tournament preferences are such that the total speaker score should
         always be an integer, a list containing the string 'total' or
         'speaks_sum' should be passed in via this argument."""
+        if integer_score_columns is None:
+            integer_score_columns = []
 
         headers = self._standings_headers(standings.metrics_info())
         data = []