Dropped Courses

evap/evaluation/migrations/0148_evaluation_allow_drop_out_and_more.py

@@ -0,0 +1,44 @@
+# Generated by Django 5.1.1 on 2024-10-14 20:03
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("evaluation", "0147_unusable_password_default"),
+    ]
+
+    operations = [
+        migrations.AddField(
+            model_name="evaluation",
+            name="allow_drop_out",
+            field=models.BooleanField(default=True, verbose_name="allow students to drop out"),
+        ),
+        migrations.AddField(
+            model_name="questionnaire",
+            name="is_active_dropout_questionnaire",
+            field=models.BooleanField(
+                blank=True, default=None, null=True, unique=True, verbose_name="questionnaire is selected as active"
+            ),
+        ),
+        migrations.AlterField(
+            model_name="questionnaire",
+            name="type",
+            field=models.IntegerField(
+                choices=[
+                    (10, "Top questionnaire"),
+                    (20, "Contributor questionnaire"),
+                    (30, "Bottom questionnaire"),
+                    (40, "Dropout questionnaire"),
+                ],
+                default=10,
+                verbose_name="type",
+            ),
+        ),
+        migrations.AddField(
+            model_name="evaluation",
+            name="dropout_count",
+            field=models.IntegerField(default=0, verbose_name="dropout count"),
+        ),
+    ]

evap/evaluation/models.py

@@ -164,6 +164,12 @@ def general_questionnaires(self):
     def contributor_questionnaires(self):
         return super().get_queryset().filter(type=Questionnaire.Type.CONTRIBUTOR)
 
+    def dropout_questionnaires(self):
+        return super().get_queryset().filter(type=Questionnaire.Type.DROPOUT)
+
+    def active_dropout_questionnaire(self):
+        return super().get_queryset().filter(type=Questionnaire.Type.DROPOUT, is_active_dropout_questionnaire=True)
+
 
 class Questionnaire(models.Model):
     """A named collection of questions."""
@@ -172,6 +178,11 @@ class Type(models.IntegerChoices):
         TOP = 10, _("Top questionnaire")
         CONTRIBUTOR = 20, _("Contributor questionnaire")
         BOTTOM = 30, _("Bottom questionnaire")
+        DROPOUT = 40, _("Dropout questionnaire")
+
+    # TODO@Felix: switch logic to add the Dropout Questionnaire to general_contribution, if allow_dropout is true
+    # maybe even allow_dropout <=> exists questionnaire in general_contribution with questionnaire.type=Dropout
+    # TODO@Felix: (?) allow selecting Dropout-Questionnaire when creating evaluation
 
     type = models.IntegerField(choices=Type.choices, verbose_name=_("type"), default=Type.TOP)
 
@@ -193,6 +204,12 @@ class Type(models.IntegerChoices):
 
     order = models.IntegerField(verbose_name=_("ordering index"), default=0)
 
+    # (unique=True, blank=True, null=True) allows having multiple non-active but only one active questionnaire
+    # TODO@Felix: hidden True?
+    is_active_dropout_questionnaire = models.BooleanField(
+        default=None, unique=True, blank=True, null=True, verbose_name=_("questionnaire is selected as active")
+    )
+
     class Visibility(models.IntegerChoices):
         HIDDEN = 0, _("Don't show")
         MANAGERS = 1, _("Managers only")
@@ -224,6 +241,15 @@ def __lt__(self, other):
     def __gt__(self, other):
         return (self.type, self.order, self.pk) > (other.type, other.order, other.pk)
 
+    @transaction.atomic()
+    def set_active_dropout(self):
+        if not self.is_dropout_questionnaire:
+            raise ValueError("Can only set DROPOUT-type Questionnaires as active dropout questionnaire.")
+
+        Questionnaire.objects.active_dropout_questionnaire().update(is_active_dropout_questionnaire=None)
+        self.is_active_dropout_questionnaire = True
+        self.save()
+
     @property
     def is_above_contributors(self):
         return self.type == self.Type.TOP
@@ -233,7 +259,20 @@ def is_below_contributors(self):
         return self.type == self.Type.BOTTOM
 
     @property
-    def can_be_edited_by_manager(self):
+    def is_dropout_questionnaire(self):
+        return self.type == self.Type.DROPOUT
+
+    @property
+    def can_be_edited_by_manager(
+        self,
+    ):
+        if self.is_dropout_questionnaire:
+            assert set(Answer.__subclasses__()) == {TextAnswer, RatingAnswerCounter}
+            return (
+                not TextAnswer.objects.filter(question__questionnaire=self).exists()
+                and not RatingAnswerCounter.objects.filter(question__questionnaire=self).exists()
+            )
+
         if is_prefetched(self, "contributions"):
             if all(is_prefetched(contribution, "evaluation") for contribution in self.contributions.all()):
                 return all(
@@ -244,6 +283,9 @@ def can_be_edited_by_manager(self):
 
     @property
     def can_be_deleted_by_manager(self):
+        if self.is_dropout_questionnaire:
+            return self.can_be_edited_by_manager
+
         return not self.contributions.exists()
 
     @property
@@ -455,6 +497,9 @@ class State(models.IntegerChoices):
     )
     _voter_count = models.IntegerField(verbose_name=_("voter count"), blank=True, null=True, default=None)
 
+    # TODO@Felix: UI for dropout count
+    dropout_count = models.IntegerField(verbose_name=_("dropout count"), default=0)
+
     # when the evaluation takes place
     vote_start_datetime = models.DateTimeField(verbose_name=_("start of evaluation"))
     # Usually the property vote_end_datetime should be used instead of this field
@@ -463,6 +508,8 @@ class State(models.IntegerChoices):
     # Disable to prevent editors from changing evaluation data
     allow_editors_to_edit = models.BooleanField(verbose_name=_("allow editors to edit"), default=True)
 
+    allow_drop_out = models.BooleanField(verbose_name=_("allow students to drop out"), default=True)
+
     evaluation_evaluated = Signal()
 
     # whether to wait for grade uploading before publishing results

evap/evaluation/tests/test_models.py

@@ -1,6 +1,7 @@
 from datetime import date, datetime, timedelta
 from unittest.mock import Mock, call, patch
 
+import django.db.utils
 from django.contrib.auth.models import Group
 from django.core import mail
 from django.core.cache import caches
@@ -1139,3 +1140,30 @@ class QuestionnaireTests(TestCase):
     def test_locked_contributor_questionnaire(self):
         questionnaire = baker.prepare(Questionnaire, is_locked=True, type=Questionnaire.Type.CONTRIBUTOR)
         self.assertRaises(ValidationError, questionnaire.clean)
+
+    def test_two_active_dropout_questionnaires_not_allowed(self):
+        q1 = baker.prepare(Questionnaire, type=Questionnaire.Type.DROPOUT)
+        q2 = baker.prepare(Questionnaire, type=Questionnaire.Type.DROPOUT)
+
+        q1.is_active_dropout_questionnaire = True
+        q1.save()
+
+        with self.assertRaises(django.db.utils.IntegrityError):
+            q2.is_active_dropout_questionnaire = True
+            q2.save()
+
+    def test_set_active_removes_active_for_others(self):
+        q1 = baker.make(Questionnaire, type=Questionnaire.Type.DROPOUT)
+        q2 = baker.make(Questionnaire, type=Questionnaire.Type.DROPOUT)
+        q3 = baker.make(Questionnaire, type=Questionnaire.Type.DROPOUT)
+
+        self.assertEqual(Questionnaire.objects.dropout_questionnaires().count(), 3)
+        self.assertFalse(Questionnaire.objects.active_dropout_questionnaire().exists())
+
+        q1.set_active_dropout()
+        self.assertTrue(q1.is_active_dropout_questionnaire)
+
+        self.assertEqual(Questionnaire.objects.active_dropout_questionnaire().first(), q1)
+
+        q2.set_active_dropout()
+        q3.set_active_dropout()

evap/evaluation/tests/test_views.py

@@ -5,7 +5,16 @@
 from django.urls import reverse
 from model_bakery import baker
 
-from evap.evaluation.models import Evaluation, Question, QuestionType, Semester, UserProfile
+from evap.evaluation.models import (
+    NO_ANSWER,
+    Contribution,
+    Evaluation,
+    Question,
+    Questionnaire,
+    QuestionType,
+    Semester,
+    UserProfile,
+)
 from evap.evaluation.tests.tools import (
     WebTest,
     WebTestWith200Check,
@@ -14,6 +23,7 @@
     store_ts_test_asset,
 )
 from evap.staff.tests.utils import WebTestStaffMode
+from evap.student.tools import answer_field_id
 
 
 class RenderJsTranslationCatalog(WebTest):
@@ -284,3 +294,93 @@ def test_reset_to_new(self) -> None:
             self.reset_from_x_to_new(s, success_expected=True)
         for s in invalid_start_states:
             self.reset_from_x_to_new(s, success_expected=False)
+
+
+class TestDropoutQuestionnaire(WebTest):
+    @classmethod
+    def setUpTestData(cls) -> None:
+        cls.user = baker.make(UserProfile, email="[email protected]")
+        cls.user2 = baker.make(UserProfile, email="[email protected]")
+        cls.evaluation = baker.make(
+            Evaluation, state=Evaluation.State.IN_EVALUATION, participants=[cls.user, cls.user2]
+        )
+
+        cls.q1 = baker.make(
+            Question,
+            type=QuestionType.POSITIVE_LIKERT,
+        )
+        cls.evaluation.general_contribution.questionnaires.add(cls.q1.questionnaire)
+
+        cls.q2 = baker.make(
+            Question,
+            type=QuestionType.NEGATIVE_YES_NO,
+        )
+        cls.contribution = baker.make(
+            Contribution,
+            contributor=baker.make(UserProfile, email="[email protected]"),
+            questionnaires=[cls.q2.questionnaire],
+            evaluation=cls.evaluation,
+        )
+
+    def assert_no_answer_set(self, form):
+        self.assertEqual(
+            form.fields[answer_field_id(self.evaluation.general_contribution, self.q1.questionnaire, self.q1)][0].value,
+            str(NO_ANSWER),
+            f"Rating questions in the general contribution should be set to NO_ANSWER (eg. {NO_ANSWER})",
+        )
+        self.assertEqual(
+            form.fields[answer_field_id(self.contribution, self.q2.questionnaire, self.q2)][0].value,
+            str(NO_ANSWER),
+            f"Rating questions in contributor questionnaires should be set to NO_ANSWER (eg. {NO_ANSWER})",
+        )
+
+    def test_no_dropout_questionnaire_chosen_does_still_work(self):
+        self.assertFalse(Questionnaire.objects.dropout_questionnaires().exists())
+        response = self.app.get(url=reverse("student:drop", args=[self.evaluation.id]), user=self.user, status=200)
+
+        self.assert_no_answer_set(response.forms["student-vote-form"])
+
+    def test_chosing_dropout_sets_to_no_answer(self):
+        dropout_questionnaire = baker.make(Questionnaire, type=Questionnaire.Type.DROPOUT)
+        dropout_question = baker.make(Question, type=QuestionType.TEXT, questionnaire=dropout_questionnaire)
+
+        dropout_questionnaire.set_active_dropout()
+        response = self.app.get(url=reverse("student:drop", args=[self.evaluation.id]), user=self.user, status=200)
+        form = response.forms["student-vote-form"]
+
+        self.assertIn(
+            answer_field_id(self.evaluation.general_contribution, dropout_questionnaire, dropout_question),
+            form.fields.keys(),
+            "The dropout Questionnaire should be shown",
+        )
+        self.assert_no_answer_set(form)
+
+    def test_allow_dropout_is_respected(self):
+        _ = self.app.get(url=reverse("student:vote", args=[self.evaluation.id]), user=self.user, status=200)
+        _ = self.app.get(url=reverse("student:drop", args=[self.evaluation.id]), user=self.user, status=200)
+
+        no_dropout_evaluation = baker.make(
+            Evaluation, state=Evaluation.State.IN_EVALUATION, participants=[self.user], allow_drop_out=False
+        )
+
+        _ = self.app.get(url=reverse("student:vote", args=[no_dropout_evaluation.id]), user=self.user, status=200)
+        _ = self.app.get(url=reverse("student:drop", args=[no_dropout_evaluation.id]), user=self.user, status=400)
+
+    def test_dropping_out_increments_dropout_counter(self):
+        self.assertEqual(self.evaluation.dropout_count, 0, "dropout_count should be initially zero")
+
+        form = self.app.get(url=reverse("student:drop", args=[self.evaluation.id]), user=self.user, status=200).forms[
+            "student-vote-form"
+        ]
+        form.submit()
+        self.evaluation = Evaluation.objects.get(pk=self.evaluation.pk)
+
+        self.assertEqual(self.evaluation.dropout_count, 1, "dropout count should increment with dropout")
+
+        form = self.app.get(url=reverse("student:vote", args=[self.evaluation.id]), user=self.user2, status=200).forms[
+            "student-vote-form"
+        ]
+        form.submit()
+        self.evaluation = Evaluation.objects.get(pk=self.evaluation.pk)
+
+        self.assertEqual(self.evaluation.dropout_count, 1, "dropout_count should not change on normal vote")

evap/results/templates/results_evaluation_detail.html

@@ -227,6 +227,17 @@ <h3>{{ evaluation.full_name }} ({{ evaluation.course.semester.name }})</h3>
         </div>
     {% endif %}
 
+    <!-- TODO@Felix: WIP Dropout view -->
+    <div class="card card-outline-primary">
+        <div class="card-header d-flex flex-row justify-content-between">
+            <span>{% translate 'Dropout' %}</span>
+            <span class="bg-secondary-outline">Dropout count: {{ evaluation.dropout_count }}</span>
+        </div>
+        <div class="card-body">
+            {% include 'results_evaluation_detail_questionnaires.html' with questionnaire_result=dropout_results %}
+        </div>
+    </div>
+
     {# Leave some space for big tooltips #}
     <div class="py-5 py-md-0"></div>
 {% endblock %}

evap/results/tests/test_tools.py

@@ -435,6 +435,25 @@ def test_calculate_average_course_distribution(self):
         self.assertEqual(distribution[3], 0)
         self.assertEqual(distribution[4], 0)
 
+    def test_dropout_questionnaires_are_not_included(self):
+        general_questionnaire = baker.make(Questionnaire, type=Questionnaire.Type.TOP)
+        general_question = baker.make(Question, questionnaire=general_questionnaire, type=QuestionType.GRADE)
+
+        dropout_questionnaire = baker.make(Questionnaire, type=Questionnaire.Type.DROPOUT)
+        dropout_question = baker.make(Question, questionnaire=dropout_questionnaire, type=QuestionType.GRADE)
+
+        contribution = baker.make(
+            Contribution, evaluation=self.evaluation, questionnaires=[general_questionnaire, dropout_questionnaire]
+        )
+
+        make_rating_answer_counters(general_question, contribution, [10, 10, 0, 0, 0])
+        make_rating_answer_counters(dropout_question, contribution, [0, 0, 0, 0, 10])
+
+        cache_results(self.evaluation)
+
+        calculated_grade = distribution_to_grade(calculate_average_distribution(self.evaluation))
+        self.assertAlmostEqual(calculated_grade, 1.5)
+
 
 class TestTextAnswerVisibilityInfo(TestCase):
     @classmethod

evap/results/tools.py

@@ -194,8 +194,10 @@ def cache_results(evaluation, *, refetch_related_objects=True):
 def get_results(evaluation: Evaluation) -> EvaluationResult:
     assert evaluation.state in STATES_WITH_RESULTS_CACHING | {Evaluation.State.IN_EVALUATION}
 
-    if evaluation.state == Evaluation.State.IN_EVALUATION:
-        return _get_results_impl(evaluation)
+    # TODO@remove commenting out
+    # if evaluation.state == Evaluation.State.IN_EVALUATION:
+    #     return _get_results_impl(evaluation)
+    return _get_results_impl(evaluation)
 
     cache_key = get_results_cache_key(evaluation)
     result = caches["results"].get(cache_key)
@@ -392,7 +394,10 @@ def calculate_average_distribution(evaluation):
     grouped_results = defaultdict(list)
     for contribution_result in get_results(evaluation).contribution_results:
         for questionnaire_result in contribution_result.questionnaire_results:
-            grouped_results[contribution_result.contributor].extend(questionnaire_result.question_results)
+            if (
+                questionnaire_result.questionnaire.type != Questionnaire.Type.DROPOUT
+            ):  # dropout questionnaires are not counted
+                grouped_results[contribution_result.contributor].extend(questionnaire_result.question_results)
 
     evaluation_results = grouped_results.pop(None, [])
 

evap/settings.py

@@ -110,6 +110,9 @@
 # Questionnaires automatically added to exam evaluations
 EXAM_QUESTIONNAIRE_IDS: list[int] = []
 
+# Questionnaire that is added on top when marking an evaluation as "dropped out"
+DROPOUT_QUESTIONNAIRE_ID = 1
+
 ### Installation specific settings
 
 # People who get emails on errors.