exams.py - Endpoints relating to assignments¶

Imports¶

These are listed in the order prescribed by PEP 8.

Standard library¶

import json
import logging
import datetime
from collections import OrderedDict
import traceback
import logging
 

Third-party imports¶

from psycopg2 import IntegrityError
import pandas as pd
import altair as alt
 

Local application imports¶

 
logger = logging.getLogger(settings.logger)
logger.setLevel(settings.log_level)


@auth.requires(
    lambda: verifyInstructorStatus(auth.user.course_name, auth.user),
    requires_login=True,
)
def one_exam_competency():

Get data about one exam using a very cool query.

 
    assignment_name = request.vars.assignment

    if not assignment_name:
        session.flash = "You must provide an assignment ID"
        return redirect(redirect(URL("admin", "admin")))

    course_id = auth.user.course_id

    logger.debug(f"COMP REPORT for {assignment_name} , {course_id}")
    query = f"""
    SELECT course, assignments.name,  question_grades.sid, div_id, score, selected_questions.points, selected_id, competency.competency, is_primary
    FROM assignments JOIN assignment_questions ON assignment_id = assignments.id
        JOIN questions ON question_id = questions.id
        JOIN question_grades ON questions.name = div_id
        JOIN selected_questions ON selector_id = div_id AND question_grades.sid = selected_questions.sid
        JOIN competency ON selected_id = question_name
    WHERE assignments.name='{assignment_name}' AND course={course_id};
    """

    clx = pd.read_sql_query(
        query, settings.database_uri.replace("postgres://", "postgresql://")
    )
    clx["pct"] = clx.score / clx.points
    clx["correct"] = clx.pct.map(lambda x: 1 if x > 0.9 else 0)

    logging.debug(clx)
    glx = (
        clx[clx.is_primary == "T"]
        .groupby(["competency", "sid"])
        .agg(num_correct=("correct", "sum"), num_answers=("correct", "count"))
    )
    glx = glx.reset_index()
    glx["pct_correct"] = glx.num_correct / glx.num_answers
    c = (
        alt.Chart(glx, title="Primary Competencies")
        .mark_rect()
        .encode(
            x="sid:O",
            y="competency",
            color=alt.Color("pct_correct", scale=alt.Scale(scheme="blues")),
            tooltip="num_answers",
        )
    )
    pcdata = c.to_json()

Supporting

    glx = (
        clx[clx.is_primary == "F"]
        .groupby(["competency", "sid"])
        .agg(num_correct=("correct", "sum"), num_answers=("correct", "count"))
    )
    glx = glx.reset_index()
    glx["pct_correct"] = glx.num_correct / glx.num_answers
    c = (
        alt.Chart(glx, title="Supplemental Competencies")
        .mark_rect()
        .encode(
            x="sid:O",
            y="competency",
            color=alt.Color("pct_correct", scale=alt.Scale(scheme="reds")),
            tooltip="num_answers",
        )
    )

    scdata = c.to_json()

    return dict(
        course_id=auth.user.course_name,
        course=get_course_row(db.courses.ALL),
        pcdata=pcdata,
        scdata=scdata,
    )