Phase 1: scaffold, data models, selection logic, tests

Flask application skeleton with SQLAlchemy models for courses (including
per-course settings), students, enrollments, opt-outs, and calls; the
weighted and cycle selection logic ported from the manual coldcall
scripts; alembic migrations; and a pytest suite covering selection
behavior and the model query helpers.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

.gitignore

@@ -0,0 +1,7 @@
+.venv/
+__pycache__/
+*.pyc
+.pytest_cache/
+instance/
+*.egg-info/
+.claude/

README.md

@@ -0,0 +1,65 @@
+# coldcall-lti
+
+A Canvas external tool (LTI 1.3) for managing cold calls in case-based
+classes: it selects students to call using weighted randomness, records
+what happened with each call, lets students report planned absences, and
+reports participation data back to both instructor and students. It
+replaces a manual workflow built on exported rosters, Google Forms, and
+local scripts.
+
+## How it works
+
+The tool is a Flask application that Canvas launches over LTI 1.3. A
+single URL serves everyone: Canvas identifies the person and course on
+each launch, so instructors get the call-list and reporting views while
+students get the absence form and their own history. The roster comes
+from Canvas through the Names and Role Provisioning Service, which means
+adds and drops are picked up automatically rather than reconciled by
+hand.
+
+Selection uses the same weighting as the manual system it replaces: each
+answered call divides a student's weight by the course's weight factor
+(default 2), so students who have answered more questions become
+progressively less likely to be called. Each course can instead use
+"cycle" mode, which shuffles the roster and calls everyone exactly once.
+These, along with whether students can see their own assessments, are
+per-course settings.
+
+## Layout
+
+- `coldcall_lti/models.py` — SQLAlchemy models: courses (with their
+  settings), students, enrollments, opt-outs, and calls, plus the query
+  helpers that feed selection.
+- `coldcall_lti/selection.py` — the weighted and cycle selection logic.
+  Kept free of database and web dependencies so it can be tested and
+  reasoned about on its own.
+- `coldcall_lti/__init__.py` — the Flask application factory.
+- `migrations/` — alembic migrations. The schema avoids
+  database-specific types so the same migrations run on SQLite (the
+  default) and MariaDB/MySQL; switching is a matter of changing
+  `COLDCALL_DATABASE_URL`.
+- `tests/` — pytest suite covering selection behavior and the model
+  helpers.
+
+## Setup
+
+Development uses a virtualenv that shares the system's Debian-packaged
+libraries (Flask, SQLAlchemy, alembic, pytest) and adds the one
+PyPI-only dependency, the maintained `pylti1p3next` fork of PyLTI1p3:
+
+```
+python3 -m venv --system-site-packages .venv
+.venv/bin/pip install pylti1p3next
+.venv/bin/pip install -e .
+```
+
+Create the database and run the tests:
+
+```
+.venv/bin/python -m alembic upgrade head
+.venv/bin/python -m pytest tests/
+```
+
+Configuration is by environment variable: `COLDCALL_DATABASE_URL` (any
+SQLAlchemy URL; defaults to an SQLite file under `instance/`) and
+`COLDCALL_SECRET_KEY` for Flask sessions.

alembic.ini

@@ -0,0 +1,116 @@
+# A generic, single database configuration.
+
+[alembic]
+# path to migration scripts
+# Use forward slashes (/) also on windows to provide an os agnostic path
+script_location = migrations
+
+# template used to generate migration file names; The default value is %%(rev)s_%%(slug)s
+# Uncomment the line below if you want the files to be prepended with date and time
+# see https://alembic.sqlalchemy.org/en/latest/tutorial.html#editing-the-ini-file
+# for all available tokens
+# file_template = %%(year)d_%%(month).2d_%%(day).2d_%%(hour).2d%%(minute).2d-%%(rev)s_%%(slug)s
+
+# sys.path path, will be prepended to sys.path if present.
+# defaults to the current working directory.
+prepend_sys_path = .
+
+# timezone to use when rendering the date within the migration file
+# as well as the filename.
+# If specified, requires the python>=3.9 or backports.zoneinfo library.
+# Any required deps can installed by adding `alembic[tz]` to the pip requirements
+# string value is passed to ZoneInfo()
+# leave blank for localtime
+# timezone =
+
+# max length of characters to apply to the "slug" field
+# truncate_slug_length = 40
+
+# set to 'true' to run the environment during
+# the 'revision' command, regardless of autogenerate
+# revision_environment = false
+
+# set to 'true' to allow .pyc and .pyo files without
+# a source .py file to be detected as revisions in the
+# versions/ directory
+# sourceless = false
+
+# version location specification; This defaults
+# to migrations/versions.  When using multiple version
+# directories, initial revisions must be specified with --version-path.
+# The path separator used here should be the separator specified by "version_path_separator" below.
+# version_locations = %(here)s/bar:%(here)s/bat:migrations/versions
+
+# version path separator; As mentioned above, this is the character used to split
+# version_locations. The default within new alembic.ini files is "os", which uses os.pathsep.
+# If this key is omitted entirely, it falls back to the legacy behavior of splitting on spaces and/or commas.
+# Valid values for version_path_separator are:
+#
+# version_path_separator = :
+# version_path_separator = ;
+# version_path_separator = space
+version_path_separator = os  # Use os.pathsep. Default configuration used for new projects.
+
+# set to 'true' to search source files recursively
+# in each "version_locations" directory
+# new in Alembic version 1.10
+# recursive_version_locations = false
+
+# the output encoding used when revision files
+# are written from script.py.mako
+# output_encoding = utf-8
+
+sqlalchemy.url = sqlite:///instance/coldcall.sqlite3
+
+
+[post_write_hooks]
+# post_write_hooks defines scripts or Python functions that are run
+# on newly generated revision scripts.  See the documentation for further
+# detail and examples
+
+# format using "black" - use the console_scripts runner, against the "black" entrypoint
+# hooks = black
+# black.type = console_scripts
+# black.entrypoint = black
+# black.options = -l 79 REVISION_SCRIPT_FILENAME
+
+# lint with attempts to fix using "ruff" - use the exec runner, execute a binary
+# hooks = ruff
+# ruff.type = exec
+# ruff.executable = %(here)s/.venv/bin/ruff
+# ruff.options = --fix REVISION_SCRIPT_FILENAME
+
+# Logging configuration
+[loggers]
+keys = root,sqlalchemy,alembic
+
+[handlers]
+keys = console
+
+[formatters]
+keys = generic
+
+[logger_root]
+level = WARN
+handlers = console
+qualname =
+
+[logger_sqlalchemy]
+level = WARN
+handlers =
+qualname = sqlalchemy.engine
+
+[logger_alembic]
+level = INFO
+handlers =
+qualname = alembic
+
+[handler_console]
+class = StreamHandler
+args = (sys.stderr,)
+level = NOTSET
+formatter = generic
+
+[formatter_generic]
+format = %(levelname)-5.5s [%(name)s] %(message)s
+datefmt = %H:%M:%S

coldcall_lti/__init__.py

@@ -0,0 +1,22 @@
+import os
+
+from flask import Flask
+
+from .config import Config
+from .db import make_engine, make_session_factory
+
+
+def create_app(config=Config):
+    app = Flask(__name__, instance_relative_config=True)
+    app.config.from_object(config)
+    os.makedirs(app.instance_path, exist_ok=True)
+
+    engine = make_engine(app.config["DATABASE_URL"])
+    app.extensions["db_engine"] = engine
+    app.extensions["db_session_factory"] = make_session_factory(engine)
+
+    @app.get("/healthz")
+    def healthz():
+        return {"status": "ok"}
+
+    return app

coldcall_lti/config.py

@@ -0,0 +1,8 @@
+import os
+
+
+class Config:
+    SECRET_KEY = os.environ.get("COLDCALL_SECRET_KEY", "dev-only-not-secret")
+    DATABASE_URL = os.environ.get(
+        "COLDCALL_DATABASE_URL", "sqlite:///instance/coldcall.sqlite3"
+    )

coldcall_lti/db.py

@@ -0,0 +1,14 @@
+from sqlalchemy import create_engine
+from sqlalchemy.orm import DeclarativeBase, sessionmaker
+
+
+class Base(DeclarativeBase):
+    pass
+
+
+def make_engine(url, echo=False):
+    return create_engine(url, echo=echo)
+
+
+def make_session_factory(engine):
+    return sessionmaker(bind=engine, expire_on_commit=False)

coldcall_lti/models.py

@@ -0,0 +1,150 @@
+import datetime
+
+from sqlalchemy import (
+    Boolean,
+    Date,
+    DateTime,
+    Float,
+    ForeignKey,
+    String,
+    Text,
+    UniqueConstraint,
+    func,
+    select,
+)
+from sqlalchemy.orm import Mapped, mapped_column, relationship
+
+from .db import Base
+
+
+def utcnow():
+    return datetime.datetime.now(datetime.UTC)
+
+
+# Call outcomes. ANSWERED is the only status that counts toward a
+# student's selection weight; SKIPPED calls are treated as if they never
+# happened; MISSING means the student was called but absent without an
+# opt-out. PENDING is used for pre-generated (printable) lists whose
+# outcomes are recorded after class.
+STATUS_ANSWERED = "answered"
+STATUS_MISSING = "missing"
+STATUS_SKIPPED = "skipped"
+STATUS_PENDING = "pending"
+CALL_STATUSES = (STATUS_ANSWERED, STATUS_MISSING, STATUS_SKIPPED, STATUS_PENDING)
+
+# Default assessment vocabulary, carried over from the manual system.
+ASSESSMENTS = ("GOOD", "SATISFACTORY", "POOR", "NO MEANINGFUL ANSWER")
+
+SELECTION_WEIGHTED = "weighted"
+SELECTION_CYCLE = "cycle"
+
+
+class Course(Base):
+    __tablename__ = "courses"
+
+    id: Mapped[int] = mapped_column(primary_key=True)
+    lti_context_id: Mapped[str] = mapped_column(String(255), unique=True)
+    lti_deployment_id: Mapped[str | None] = mapped_column(String(255))
+    title: Mapped[str | None] = mapped_column(String(255))
+
+    # Per-course settings.
+    weight_factor: Mapped[float] = mapped_column(Float, default=2.0)
+    selection_mode: Mapped[str] = mapped_column(
+        String(16), default=SELECTION_WEIGHTED
+    )
+    show_assessments: Mapped[bool] = mapped_column(Boolean, default=True)
+
+    created_at: Mapped[datetime.datetime] = mapped_column(
+        DateTime, default=utcnow
+    )
+
+    enrollments: Mapped[list["Enrollment"]] = relationship(back_populates="course")
+
+
+class Student(Base):
+    __tablename__ = "students"
+
+    id: Mapped[int] = mapped_column(primary_key=True)
+    # The LTI "sub" claim: stable, opaque, unique per Canvas user.
+    canvas_user_id: Mapped[str] = mapped_column(String(255), unique=True)
+    name: Mapped[str | None] = mapped_column(String(255))
+    sortable_name: Mapped[str | None] = mapped_column(String(255))
+    email: Mapped[str | None] = mapped_column(String(255))
+    avatar_url: Mapped[str | None] = mapped_column(String(1024))
+
+    enrollments: Mapped[list["Enrollment"]] = relationship(back_populates="student")
+
+
+class Enrollment(Base):
+    __tablename__ = "enrollments"
+    __table_args__ = (UniqueConstraint("course_id", "student_id"),)
+
+    id: Mapped[int] = mapped_column(primary_key=True)
+    course_id: Mapped[int] = mapped_column(ForeignKey("courses.id"))
+    student_id: Mapped[int] = mapped_column(ForeignKey("students.id"))
+    role: Mapped[str] = mapped_column(String(32), default="student")
+    # Set False when a student disappears from the NRPS roster (dropped).
+    active: Mapped[bool] = mapped_column(Boolean, default=True)
+
+    course: Mapped[Course] = relationship(back_populates="enrollments")
+    student: Mapped[Student] = relationship(back_populates="enrollments")
+
+
+class OptOut(Base):
+    __tablename__ = "optouts"
+    __table_args__ = (UniqueConstraint("course_id", "student_id", "date"),)
+
+    id: Mapped[int] = mapped_column(primary_key=True)
+    course_id: Mapped[int] = mapped_column(ForeignKey("courses.id"))
+    student_id: Mapped[int] = mapped_column(ForeignKey("students.id"))
+    date: Mapped[datetime.date] = mapped_column(Date)
+    created_at: Mapped[datetime.datetime] = mapped_column(
+        DateTime, default=utcnow
+    )
+
+
+class Call(Base):
+    __tablename__ = "calls"
+
+    id: Mapped[int] = mapped_column(primary_key=True)
+    course_id: Mapped[int] = mapped_column(ForeignKey("courses.id"))
+    student_id: Mapped[int] = mapped_column(ForeignKey("students.id"))
+    session_date: Mapped[datetime.date] = mapped_column(Date)
+    status: Mapped[str] = mapped_column(String(16), default=STATUS_PENDING)
+    assessment: Mapped[str | None] = mapped_column(String(32))
+    note: Mapped[str | None] = mapped_column(Text)
+    created_at: Mapped[datetime.datetime] = mapped_column(
+        DateTime, default=utcnow
+    )
+
+    student: Mapped[Student] = relationship()
+
+
+def answered_call_counts(session, course_id):
+    """Number of answered calls per student, the input to selection
+    weighting. Skipped, pending, and missing calls do not count, matching
+    the old system where only answered calls reduced a student's odds."""
+    rows = session.execute(
+        select(Call.student_id, func.count())
+        .where(Call.course_id == course_id, Call.status == STATUS_ANSWERED)
+        .group_by(Call.student_id)
+    ).all()
+    return dict(rows)
+
+
+def students_present(session, course_id, on_date):
+    """Active student-role enrollees minus those opted out for the date."""
+    opted_out = select(OptOut.student_id).where(
+        OptOut.course_id == course_id, OptOut.date == on_date
+    )
+    rows = session.execute(
+        select(Student)
+        .join(Enrollment, Enrollment.student_id == Student.id)
+        .where(
+            Enrollment.course_id == course_id,
+            Enrollment.role == "student",
+            Enrollment.active.is_(True),
+            Student.id.not_in(opted_out),
+        )
+    ).scalars()
+    return list(rows)

coldcall_lti/selection.py

@@ -0,0 +1,42 @@
+"""Cold-call selection logic.
+
+Ported from the manual system's coldcall.py: each answered call a
+student has already received divides their selection weight by the
+course's weight factor, so with the default factor of 2 a student who
+has answered n questions is 2^n times less likely to be picked than one
+who has answered none. A factor of 1 makes selection uniform.
+
+The alternative "cycle" mode shuffles the present students and calls
+each exactly once, the equivalent of the old --shuffle flag.
+"""
+
+import random
+
+
+def compute_weights(student_ids, prev_counts, weight_factor=2.0):
+    return {
+        s: weight_factor ** -prev_counts.get(s, 0) for s in student_ids
+    }
+
+
+def select_student(student_ids, prev_counts, weight_factor=2.0, rng=random):
+    weights = compute_weights(student_ids, prev_counts, weight_factor)
+    return rng.choices(list(weights), weights=list(weights.values()), k=1)[0]
+
+
+def generate_call_list(student_ids, prev_counts, n, weight_factor=2.0, rng=random):
+    """A list of n weighted picks. Counts update as the list is built,
+    so a student picked early is downweighted for later slots, matching
+    the old manual script's behavior of recording each call before
+    drawing the next."""
+    counts = dict(prev_counts)
+    picks = []
+    for _ in range(n):
+        s = select_student(student_ids, counts, weight_factor, rng)
+        counts[s] = counts.get(s, 0) + 1
+        picks.append(s)
+    return picks
+
+
+def generate_cycle_list(student_ids, rng=random):
+    return rng.sample(list(student_ids), len(student_ids))

migrations/README

@@ -0,0 +1 @@
+Generic single-database configuration.

migrations/env.py

@@ -0,0 +1,83 @@
+import os
+from logging.config import fileConfig
+
+from sqlalchemy import engine_from_config
+from sqlalchemy import pool
+
+from alembic import context
+
+from coldcall_lti.db import Base
+from coldcall_lti import models  # noqa: F401  (populates Base.metadata)
+
+# this is the Alembic Config object, which provides
+# access to the values within the .ini file in use.
+config = context.config
+
+# The database URL comes from the same environment variable the app
+# uses, falling back to the value in alembic.ini.
+if os.environ.get("COLDCALL_DATABASE_URL"):
+    config.set_main_option("sqlalchemy.url", os.environ["COLDCALL_DATABASE_URL"])
+
+# Interpret the config file for Python logging.
+# This line sets up loggers basically.
+if config.config_file_name is not None:
+    fileConfig(config.config_file_name)
+
+target_metadata = Base.metadata
+
+# other values from the config, defined by the needs of env.py,
+# can be acquired:
+# my_important_option = config.get_main_option("my_important_option")
+# ... etc.
+
+
+def run_migrations_offline() -> None:
+    """Run migrations in 'offline' mode.
+
+    This configures the context with just a URL
+    and not an Engine, though an Engine is acceptable
+    here as well.  By skipping the Engine creation
+    we don't even need a DBAPI to be available.
+
+    Calls to context.execute() here emit the given string to the
+    script output.
+
+    """
+    url = config.get_main_option("sqlalchemy.url")
+    context.configure(
+        url=url,
+        target_metadata=target_metadata,
+        literal_binds=True,
+        dialect_opts={"paramstyle": "named"},
+    )
+
+    with context.begin_transaction():
+        context.run_migrations()
+
+
+def run_migrations_online() -> None:
+    """Run migrations in 'online' mode.
+
+    In this scenario we need to create an Engine
+    and associate a connection with the context.
+
+    """
+    connectable = engine_from_config(
+        config.get_section(config.config_ini_section, {}),
+        prefix="sqlalchemy.",
+        poolclass=pool.NullPool,
+    )
+
+    with connectable.connect() as connection:
+        context.configure(
+            connection=connection, target_metadata=target_metadata
+        )
+
+        with context.begin_transaction():
+            context.run_migrations()
+
+
+if context.is_offline_mode():
+    run_migrations_offline()
+else:
+    run_migrations_online()

migrations/script.py.mako

@@ -0,0 +1,26 @@
+"""${message}
+
+Revision ID: ${up_revision}
+Revises: ${down_revision | comma,n}
+Create Date: ${create_date}
+
+"""
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+${imports if imports else ""}
+
+# revision identifiers, used by Alembic.
+revision: str = ${repr(up_revision)}
+down_revision: Union[str, None] = ${repr(down_revision)}
+branch_labels: Union[str, Sequence[str], None] = ${repr(branch_labels)}
+depends_on: Union[str, Sequence[str], None] = ${repr(depends_on)}
+
+
+def upgrade() -> None:
+    ${upgrades if upgrades else "pass"}
+
+
+def downgrade() -> None:
+    ${downgrades if downgrades else "pass"}

migrations/versions/ad7e3f5ea2f3_initial_schema.py

@@ -0,0 +1,90 @@
+"""initial schema
+
+Revision ID: ad7e3f5ea2f3
+Revises: 
+Create Date: 2026-07-31 16:23:01.164289
+
+"""
+from typing import Sequence, Union
+
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision: str = 'ad7e3f5ea2f3'
+down_revision: Union[str, None] = None
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table('courses',
+    sa.Column('id', sa.Integer(), nullable=False),
+    sa.Column('lti_context_id', sa.String(length=255), nullable=False),
+    sa.Column('lti_deployment_id', sa.String(length=255), nullable=True),
+    sa.Column('title', sa.String(length=255), nullable=True),
+    sa.Column('weight_factor', sa.Float(), nullable=False),
+    sa.Column('selection_mode', sa.String(length=16), nullable=False),
+    sa.Column('show_assessments', sa.Boolean(), nullable=False),
+    sa.Column('created_at', sa.DateTime(), nullable=False),
+    sa.PrimaryKeyConstraint('id'),
+    sa.UniqueConstraint('lti_context_id')
+    )
+    op.create_table('students',
+    sa.Column('id', sa.Integer(), nullable=False),
+    sa.Column('canvas_user_id', sa.String(length=255), nullable=False),
+    sa.Column('name', sa.String(length=255), nullable=True),
+    sa.Column('sortable_name', sa.String(length=255), nullable=True),
+    sa.Column('email', sa.String(length=255), nullable=True),
+    sa.Column('avatar_url', sa.String(length=1024), nullable=True),
+    sa.PrimaryKeyConstraint('id'),
+    sa.UniqueConstraint('canvas_user_id')
+    )
+    op.create_table('calls',
+    sa.Column('id', sa.Integer(), nullable=False),
+    sa.Column('course_id', sa.Integer(), nullable=False),
+    sa.Column('student_id', sa.Integer(), nullable=False),
+    sa.Column('session_date', sa.Date(), nullable=False),
+    sa.Column('status', sa.String(length=16), nullable=False),
+    sa.Column('assessment', sa.String(length=32), nullable=True),
+    sa.Column('note', sa.Text(), nullable=True),
+    sa.Column('created_at', sa.DateTime(), nullable=False),
+    sa.ForeignKeyConstraint(['course_id'], ['courses.id'], ),
+    sa.ForeignKeyConstraint(['student_id'], ['students.id'], ),
+    sa.PrimaryKeyConstraint('id')
+    )
+    op.create_table('enrollments',
+    sa.Column('id', sa.Integer(), nullable=False),
+    sa.Column('course_id', sa.Integer(), nullable=False),
+    sa.Column('student_id', sa.Integer(), nullable=False),
+    sa.Column('role', sa.String(length=32), nullable=False),
+    sa.Column('active', sa.Boolean(), nullable=False),
+    sa.ForeignKeyConstraint(['course_id'], ['courses.id'], ),
+    sa.ForeignKeyConstraint(['student_id'], ['students.id'], ),
+    sa.PrimaryKeyConstraint('id'),
+    sa.UniqueConstraint('course_id', 'student_id')
+    )
+    op.create_table('optouts',
+    sa.Column('id', sa.Integer(), nullable=False),
+    sa.Column('course_id', sa.Integer(), nullable=False),
+    sa.Column('student_id', sa.Integer(), nullable=False),
+    sa.Column('date', sa.Date(), nullable=False),
+    sa.Column('created_at', sa.DateTime(), nullable=False),
+    sa.ForeignKeyConstraint(['course_id'], ['courses.id'], ),
+    sa.ForeignKeyConstraint(['student_id'], ['students.id'], ),
+    sa.PrimaryKeyConstraint('id'),
+    sa.UniqueConstraint('course_id', 'student_id', 'date')
+    )
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_table('optouts')
+    op.drop_table('enrollments')
+    op.drop_table('calls')
+    op.drop_table('students')
+    op.drop_table('courses')
+    # ### end Alembic commands ###

pyproject.toml

@@ -0,0 +1,24 @@
+[build-system]
+requires = ["setuptools>=64"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "coldcall-lti"
+version = "0.1.0"
+description = "Canvas LTI 1.3 tool for managing in-class cold calls"
+requires-python = ">=3.11"
+# Known-good versions: flask 3.1.1, SQLAlchemy 2.0.40, alembic 1.13.2,
+# pylti1p3next 2.0.2 (development uses Debian system packages for all
+# but pylti1p3next; these floors record what the code was written against).
+dependencies = [
+    "flask>=3.1",
+    "pylti1p3next>=2.0",
+    "SQLAlchemy>=2.0",
+    "alembic>=1.13",
+]
+
+[project.optional-dependencies]
+test = ["pytest>=8"]
+
+[tool.setuptools]
+packages = ["coldcall_lti"]

tests/conftest.py

@@ -0,0 +1,15 @@
+import pytest
+from sqlalchemy import create_engine
+from sqlalchemy.orm import sessionmaker
+
+from coldcall_lti.db import Base
+
+
+@pytest.fixture
+def db_session():
+    engine = create_engine("sqlite://")
+    Base.metadata.create_all(engine)
+    session = sessionmaker(bind=engine)()
+    yield session
+    session.close()
+    engine.dispose()

tests/test_models.py

@@ -0,0 +1,87 @@
+import datetime
+
+from coldcall_lti import models
+
+
+def make_course_with_students(session, n=3):
+    course = models.Course(lti_context_id="ctx-1", title="Test Course")
+    session.add(course)
+    students = []
+    for i in range(n):
+        s = models.Student(canvas_user_id=f"user-{i}", name=f"Student {i}")
+        session.add(s)
+        students.append(s)
+    session.flush()
+    for s in students:
+        session.add(models.Enrollment(course_id=course.id, student_id=s.id))
+    session.flush()
+    return course, students
+
+
+def test_answered_call_counts_ignores_skipped_and_pending(db_session):
+    course, students = make_course_with_students(db_session)
+    day = datetime.date(2026, 10, 1)
+    a, b, c = students
+    for status, student in [
+        (models.STATUS_ANSWERED, a),
+        (models.STATUS_ANSWERED, a),
+        (models.STATUS_SKIPPED, a),
+        (models.STATUS_ANSWERED, b),
+        (models.STATUS_MISSING, b),
+        (models.STATUS_PENDING, c),
+    ]:
+        db_session.add(
+            models.Call(
+                course_id=course.id,
+                student_id=student.id,
+                session_date=day,
+                status=status,
+            )
+        )
+    db_session.flush()
+
+    counts = models.answered_call_counts(db_session, course.id)
+    assert counts == {a.id: 2, b.id: 1}
+
+
+def test_students_present_excludes_optouts_and_inactive(db_session):
+    course, students = make_course_with_students(db_session)
+    day = datetime.date(2026, 10, 1)
+    a, b, c = students
+
+    db_session.add(
+        models.OptOut(course_id=course.id, student_id=a.id, date=day)
+    )
+    enrollment_b = (
+        db_session.query(models.Enrollment)
+        .filter_by(course_id=course.id, student_id=b.id)
+        .one()
+    )
+    enrollment_b.active = False
+    db_session.flush()
+
+    present = models.students_present(db_session, course.id, day)
+    assert [s.id for s in present] == [c.id]
+
+    # The opt-out only applies on its own date.
+    other_day = datetime.date(2026, 10, 2)
+    present = models.students_present(db_session, course.id, other_day)
+    assert {s.id for s in present} == {a.id, c.id}
+
+
+def test_students_present_excludes_instructor_role(db_session):
+    course, students = make_course_with_students(db_session)
+    instructor = models.Student(canvas_user_id="teacher-1", name="Teacher")
+    db_session.add(instructor)
+    db_session.flush()
+    db_session.add(
+        models.Enrollment(
+            course_id=course.id, student_id=instructor.id, role="instructor"
+        )
+    )
+    db_session.flush()
+
+    present = models.students_present(
+        db_session, course.id, datetime.date(2026, 10, 1)
+    )
+    assert instructor.id not in {s.id for s in present}

tests/test_selection.py

@@ -0,0 +1,45 @@
+import random
+from collections import Counter
+
+from coldcall_lti import selection
+
+
+STUDENTS = ["a", "b", "c", "d"]
+
+
+def test_weights_halve_per_answered_call():
+    weights = selection.compute_weights(STUDENTS, {"a": 0, "b": 1, "c": 3}, 2.0)
+    assert weights == {"a": 1.0, "b": 0.5, "c": 0.125, "d": 1.0}
+
+
+def test_weight_factor_one_is_uniform():
+    weights = selection.compute_weights(STUDENTS, {"a": 5, "b": 2}, 1.0)
+    assert set(weights.values()) == {1.0}
+
+
+def test_heavily_called_student_selected_less():
+    rng = random.Random(42)
+    counts = Counter(
+        selection.select_student(STUDENTS, {"a": 4}, 2.0, rng)
+        for _ in range(4000)
+    )
+    # "a" has weight 1/16 against 1 for the others, so it should get
+    # roughly 1/49th of the picks; the others roughly a third each.
+    assert counts["a"] < 250
+    for s in ("b", "c", "d"):
+        assert 1000 < counts[s] < 1700
+
+
+def test_generate_call_list_downweights_within_list():
+    rng = random.Random(7)
+    picks = selection.generate_call_list(STUDENTS, {}, 200, 2.0, rng)
+    assert len(picks) == 200
+    counts = Counter(picks)
+    # Sequential downweighting keeps the distribution close to even.
+    assert all(35 < counts[s] < 65 for s in STUDENTS)
+
+
+def test_generate_cycle_list_covers_everyone_once():
+    rng = random.Random(1)
+    picks = selection.generate_cycle_list(STUDENTS, rng)
+    assert sorted(picks) == sorted(STUDENTS)