def test_mcq_basic(self):
"""
Basic question template. Really the minimal possible example.
NOTE: it would be much cleaner to have
{{ a+b |quick_eval }}
instead of
{% quick_eval "a+b" %}
unfortunately the former (a Django template filter) evaluates
"a" and "b" directly and does not pass the context dictionary.
The latter (a Django template tag) is far more powerful and
gives us additional flexibility at render time.
"""
some_text = """
[[type]]
MCQ
[[question]]
If a={{a}}, b={{b}}. What is a*b?
--
& {{a}}{{b}}
&1
^{% quick_eval "a*b" 5 %}
& {% quick_eval "a+b" 2 %}
[[variables]]
a: [2, 5, 0.5, float]
b: [5, 9, 1, int]
"""
qtemplate = views.create_question_template(some_text, user=user)
qt = QTemplate.objects.get(id=qtemplate.id)
_, html_a, var_dict, _ = render(qt)
true_answer = var_dict['a'][1] * var_dict['b'][1]
self.assertEqual(html_a, '<p>The solution is: "%s"</p>' % true_answer)
def test_mcq_more_details(self):
"""
More complete template
"""
some_text = """
[[type]]
MCQ
[[attribs]]
Name: Multiplication warm-up
Contributor: Kevin Dunn
Difficulty: 2
Tags: multiplication, math
Grade: 3
Feedback: False
[[question]]
If a=1, b=2. What is a*b?
--
& 12
& 1
^2
& 4"""
qtemplate = views.create_question_template(some_text, user=user)
q = QTemplate.objects.get(id=qtemplate.id)
self.assertEqual(q.difficulty, 2)
self.assertEqual(q.max_grade, 3)
self.assertEqual(q.enable_feedback, False)
self.assertEqual(q.t_solution, u'The solution is: "2"')
t_grading = json.loads(q.t_grading)
vals = t_grading.values()
vals.sort()
self.assertEqual(vals, [[u'key', u'2'], [u'lure', u'1'],
[u'lure', u'12'], [u'lure', u'4']])
def test_tf_basic(self):
"""
Basic true/false template. Really the minimal possible example.
"""
some_text = """
[[type]]
TF
[[question]]
The sun is hot.
--
& False
^True
"""
qtemplate = views.create_question_template(some_text, user=user)
qt = QTemplate.objects.get(id=qtemplate.id)
html_q, _, _, _ = render(qt)
key, value = views.get_type(qt.t_grading, 'key').next()
self.assertTrue(key.startswith('True'))
start = html_q.find(value)
self.assertEqual(html_q[start + 7:start + 11], 'True')
key, value = views.get_type(qt.t_grading, 'lure').next()
self.assertTrue(key.startswith('False'))
start = html_q.find(value)
self.assertEqual(html_q[start + 7:start + 12], 'False')
def test_tf_final_correct(self):
"""
Lures and an CORRECT option that must be shown as the last option.
"""
some_text = """
[[type]]
TF
[[question]]
The sun is ....
--
& Cold
& Luke warm
& Warm
%^ None of the above.
"""
qtemplate = views.create_question_template(some_text, user=user)
qt = QTemplate.objects.get(id=qtemplate.id)
html_q, _, _, _ = render(qt)
key, value = views.get_type(qt.t_grading, 'final-key').next()
self.assertTrue(key.startswith('None of the above.'))
start = html_q.find(value)
self.assertEqual(html_q[start + 7:start + 11], 'None')
self.assertEqual(qt.t_solution, ('The solution is: "None of the '
'above."'))
def test_long_answer_entry(self):
"""
More complete template
"""
some_text = """
[[type]]
long
[[attribs]]
Contributor: Kevin Dunn
Difficulty: 3
Grade: 5
[[question]]
Your statistical pre-requisite course was taken a few years ago;
tell us your thoughts about that course.
[[grading]]
Full grade for any reasonable answer. Our aim is to collect feedback.
[[solution]]
There is no solution for this question.
"""
qtemplate = views.create_question_template(some_text, user=user)
q = QTemplate.objects.get(id=qtemplate.id)
self.assertEqual(q.difficulty, 3)
self.assertEqual(q.max_grade, 5)
self.assertEqual(q.t_solution, ('There is no solution for this '
'question.'))
def test_mcq_bad_specified(self):
"""
Two correct options for an MCQ.
"""
some_text = """
[[type]]
MCQ
[[question]]
The sun is ....
--
^ Cold
& Luke warm
^ Hot
% None of the above.
"""
with self.assertRaises(views.ParseError):
views.create_question_template(some_text)
def test_peer_evaluation(self):
"""
Template for peer evaluation.
"""
some_text = """
[[type]]
Peer-eval
[[question]]
* Name one aspect of {{person}}'s work that you really appreciated this week: {[[person_slug_positive]]}
* Provide constructive feedback on {{person}}'s work that will help him/her improve. {[[person_slug_feedback]]}
* Please rank {{person}}'s contribution to the assignment: --ranking--
Note: your evaluation for each person will be a number from 0 to 8, with 6 being typical, satisfactory performance.
+ 0 = No show = Made no contribution. I could not recognize this person in a lineup.
+ 2 = Hitchhiker = Made minimal contribution. The group could have received the same grade without this member.
+ 4 = Marginal = Made limited contribution, only when required to do so. Took no initiative, was not always prepared and missed meetings.
+ 5 = Ordinary = Performed some tasks acceptably but was not reliable or consistent.
+ 6 = Fully satisfactory = Made good contributions to work and group organization and morale. This is the average performance for a student in the course.
+ 7 = Very good = Consistently showed initiative and preparation beyond expectations. High quality of work.
+ 8 = Excellent = Lead the group by example and personality. Prepared excellent technical work and assisted others to excel.
[[attribs]]
Name: Peer feedback for Assignment 1 (on personal finance)
Contributor: Kevin Dunn
Difficulty: 1
Grade: 1
Feedback: False
"""
group = Group.objects.create(name='TestA1')
user_2 = User.objects.create(username='******',
first_name='Test',
last_name='user2')
user_2 = UserProfile.objects.create(role='Grader',
group=group,
user=user_2)
user_3 = User.objects.create(username='******',
first_name='Test',
last_name='user3')
user_3 = UserProfile.objects.create(role='Grader',
group=group,
user=user_3)
user.group = group
user.save()
qtemplate = views.create_question_template(some_text, user=user)
qt = QTemplate.objects.get(id=qtemplate.id)
options = {}
options['peers'] = user.get_peers()
html_q, html_a, var_dict, _ = render(qt, options)
def test_dont_allow_no_answers_MCQ(self):
"""
"""
some_text = """
[[type]]
multi
[[question]]
The following plot is from a measurement system at a company showing actual data PLOT HERE
Some of the things that can be noticed in the plot are:
--
&
^
"""
qtemplate = views.create_question_template(some_text)
def test_mcq_basic_parse(self):
"""
Basic question template. Really the minimal possible example.
"""
some_text = """
[[type]]
MCQ
[[question]]
If a=1, b=2. What is a*b?
--
& 12
&1
^2
& 4
"""
qtemplate = views.create_question_template(some_text, user=user)
q = QTemplate.objects.get(id=qtemplate.id)
self.assertEqual(q.difficulty, 1)
self.assertEqual(q.q_type, 'mcq')
self.assertEqual(q.name, 'If a=1, b=2. What is a*b?')
def test_image_location(self):
"""
Tests that images are placed in the correct location
"""
some_text = """
[[type]]
TF
[[question]]
The image here contains oscillations
--
& False
^ True
"""
qtemplate = views.create_question_template(some_text, user=user)
qt = QTemplate.objects.get(id=qtemplate.id)
html_q, _, _, _ = render(qt)
idx = html_q.find('image_file_name.jpg')
self.assertEqual(html_q[idx - 3:idx],
'/0/') # this is the subdir stored in
def test_variables_with_choices(self):
some_text = """
[[type]]
long
[[question]]
Plot a time series plot using rows {{row_start}} to
{% quick_eval "row_start+1000" %} for the `{{variable_name}}` variable.
Save the plot as a JPEG or PNG file and upload it here {[:upload:]}
[[variables]]
row_start:[1, 2000, 100, int]
variable_name: {'choices': ['Opt1', 'Opt2', 'Opt3']}
[[solution]]
Some solution text would go here.
[[grading]]
"""
qtemplate = views.create_question_template(some_text, user=user)
qt = QTemplate.objects.get(id=qtemplate.id)
html_q, _, _, _ = render(qt)
start = html_q.find('for the <code>')
self.assertTrue(html_q[start + 14:start + 18] in ('Opt1', 'Opt2',
'Opt3'))
def test_question_missing_variable(self):
some_text = """
[[type]]
multi
[[question]]
Our process produces defective products at a rate of 1 in {{n_total}}. If we randomly take a sample of {{n_sample}} items from the production line,then ....
--
^{% quick_eval "((n_total-1)/n_total)**n_sample" %}
& {% quick_eval "((n_total-1)/n_total)" %}
& be greater than or equal to ({{n_total}}-1)/{{n_total}})
& is equal to 1/{{n_total}}
[[variables]]
n_total: [6, 10, 1, int]
n: [4,6,1,int]
[[Solution]]
The pass rate for this system is ({{n_total}}-1)/{{n_total}}), so
"""
# The ``n_sample`` variable is not specified
qtemplate = views.create_question_template(some_text, user=user)
qt = QTemplate.objects.get(id=qtemplate.id)
with self.assertRaises(NameError):
render(qt)
def test_short_answer_question(self):
"""
Template test for a short answer question
"""
some_text = """
[[type]]
short
[[question]]
Plots with both category and value axes are known as {[ans1]} plots, while a
plot with the 5-number summary of a univariate series are called {[2]} plots.
[[grading]]
ans1:bar
ans1:BAR
2:box
[[solution]]
Goes here
[[attribs]]
Contributor: Kevin Dunn
Difficulty: 1
Tags: data visualization
Grade: 1
"""
qtemplate = views.create_question_template(some_text, user=user)
qt = QTemplate.objects.get(id=qtemplate.id)
self.assertEqual(qt.t_solution, u'Goes here')
t_grading = json.loads(qt.t_grading)
vals = t_grading.values()
keys = t_grading.keys()
self.assertEqual(keys, ['ans1', '2']) # <--- keys are strings
self.assertEqual(vals, [['bar', 'BAR'], ['box']])
html_q, _, _, _ = render(qt)
self.assertEqual(html_q[0:74],
('<p>Plots with both category '
'and value axes are known as <input type="text"'))