Skip to content

:hedgehog: PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all from Slack, web, desktop, or the MCP.

feat(data-warehouse): add 4 missing tables to the Teachable source (#120037)

Tom Owers74e8ce2ddab8
diff --git a/products/warehouse_sources/backend/temporal/data_imports/sources/COVERAGE_GAPS_APPENDIX.md b/products/warehouse_sources/backend/temporal/data_imports/sources/COVERAGE_GAPS_APPENDIX.md
index 75690baa1a73..70f7f5f7967a 100644
--- a/products/warehouse_sources/backend/temporal/data_imports/sources/COVERAGE_GAPS_APPENDIX.md
+++ b/products/warehouse_sources/backend/temporal/data_imports/sources/COVERAGE_GAPS_APPENDIX.md
@@ -12037,10 +12037,10 @@ Today (5): `course_enrollments`, `courses`, `pricing_plans`, `transactions`, `us
Diffed against: <https://docs.teachable.com/reference/listcourses>
-- [ ] `courses/{course_id}/progress` — per-student completion percentage — the headline learning metric, and the only way to measure course engagement (high)
-- [ ] `courses/{course_id}/lectures/{lecture_id}` — lecture lookup table resolving the lecture IDs referenced by progress and quiz data (high)
-- [ ] `courses/{course_id}/lectures/{lecture_id}/quizzes/{quiz_id}/responses` — individual learner answers and scores — the assessment fact table (medium)
-- [ ] `courses/{course_id}/lectures/{lecture_id}/quizzes` — quiz lookup needed to interpret quiz responses (medium)
+- [x] `courses/{course_id}/progress` — per-student completion percentage — the headline learning metric, and the only way to measure course engagement (high)
+- [x] `courses/{course_id}/lectures/{lecture_id}` — lecture lookup table resolving the lecture IDs referenced by progress and quiz data (high)
+- [x] `courses/{course_id}/lectures/{lecture_id}/quizzes/{quiz_id}/responses` — individual learner answers and scores — the assessment fact table (medium)
+- [x] `courses/{course_id}/lectures/{lecture_id}/quizzes` — quiz lookup needed to interpret quiz responses (medium)
Note: Full v1 path list read out of the embedded ReadMe OAS on the reference page: courses, enrollments, lectures, quizzes, quiz responses, progress, pricing_plans, transactions, users, webhooks. Lectures/quizzes/progress are per-course sub-resources that must be fanned out over the courses we already sync.
diff --git a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/canonical_descriptions.py b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/canonical_descriptions.py
index c0f7572a5d96..3231385dfc41 100644
--- a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/canonical_descriptions.py
+++ b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/canonical_descriptions.py
@@ -76,4 +76,59 @@ CANONICAL_DESCRIPTIONS: CanonicalDescriptions = {
"course_id": "Identifier of the course this pricing plan belongs to.",
},
},
+ "course_progress": {
+ "description": "A student's progress through a course, with completion state for each lecture section and lecture.",
+ "docs_url": "https://docs.teachable.com/reference/courseprogress",
+ "columns": {
+ "course_id": "Unique identifier of the course.",
+ "user_id": "Unique identifier of the student.",
+ "id": "Unique identifier of the student's course progress.",
+ "certificate": "The certificate issued to the student for the course: issued_at, page_id and serial_number.",
+ "completed_at": "When the student completed the course, or null if not yet finished.",
+ "enrolled_at": "When the student enrolled in the course, in ISO8601 format.",
+ "lecture_sections": "Progress for each lecture section, with the completion state of each lecture in it.",
+ "percent_complete": "Percentage of the course the student has marked complete.",
+ },
+ },
+ "lectures": {
+ "description": "A lecture in a course curriculum, with its attachments.",
+ "docs_url": "https://docs.teachable.com/reference/showlecture",
+ "columns": {
+ "course_id": "Unique identifier of the course that contains the lecture.",
+ "id": "Unique identifier of the lecture.",
+ "name": "The lecture name.",
+ "is_published": "Whether the lecture is published.",
+ "position": "Position of the lecture within its section (1 is first).",
+ "lecture_section_id": "Unique identifier of the section that contains the lecture.",
+ "attachments": "Content attached to the lecture: text, video, image, PDF, quiz, code and other blocks.",
+ },
+ },
+ "quizzes": {
+ "description": "A quiz in a course lecture, with its questions and answers.",
+ "docs_url": "https://docs.teachable.com/reference/showquiz",
+ "columns": {
+ "course_id": "Unique identifier of the course that contains the lecture.",
+ "lecture_id": "Unique identifier of the lecture that contains the quiz.",
+ "id": "Unique identifier of the quiz attachment.",
+ "name": "The quiz name.",
+ "kind": "The attachment kind. Always 'quiz'.",
+ "position": "Position of the quiz within the lecture (1 is first).",
+ "quiz": "The quiz content: its id and the list of questions, answers and correct answers.",
+ },
+ },
+ "quiz_responses": {
+ "description": "A student's submission of a quiz, with the score.",
+ "docs_url": "https://docs.teachable.com/reference/showquizresponses",
+ "columns": {
+ "course_id": "Unique identifier of the course.",
+ "lecture_id": "Unique identifier of the lecture that contains the quiz.",
+ "quiz_id": "Unique identifier of the quiz attachment.",
+ "graded": "Whether the quiz is graded.",
+ "student_id": "Unique identifier of the student.",
+ "student_name": "The student's name.",
+ "student_email": "The student's email address.",
+ "submitted_at": "When the student submitted the quiz.",
+ "percent_correct": "Percentage of correct answers.",
+ },
+ },
}
diff --git a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/settings.py b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/settings.py
index 2bc6d1639a50..edae05e75db0 100644
--- a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/settings.py
+++ b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/settings.py
@@ -87,6 +87,40 @@ TEACHABLE_ENDPOINTS: dict[str, TeachableEndpointConfig] = {
partition_key="created_at",
# No server-side updated-since filter, so full refresh (the table is small).
),
+ # The four endpoints below sit two or more levels under /v1/courses, so `teachable.py` walks
+ # them with its own iterators. None has a server-side time filter: all are full refresh.
+ "course_progress": TeachableEndpointConfig(
+ name="course_progress",
+ # Needs a `user_id` query param, so this fans out over each course's enrollments.
+ path="/v1/courses/{course_id}/progress",
+ data_selector="course_progress",
+ primary_key=["course_id", "user_id"],
+ partition_key="enrolled_at",
+ ),
+ "lectures": TeachableEndpointConfig(
+ name="lectures",
+ # There is no lecture list endpoint: lecture ids come from the course detail curriculum.
+ path="/v1/courses/{course_id}/lectures/{lecture_id}",
+ data_selector="lecture",
+ primary_key=["course_id", "id"],
+ ),
+ "quizzes": TeachableEndpointConfig(
+ name="quizzes",
+ # The list endpoint returns only `quiz_ids`, so each id is resolved through
+ # /quizzes/{quiz_id} to get the quiz name and questions.
+ path="/v1/courses/{course_id}/lectures/{lecture_id}/quizzes",
+ data_selector="quiz_ids",
+ primary_key=["course_id", "lecture_id", "id"],
+ ),
+ "quiz_responses": TeachableEndpointConfig(
+ name="quiz_responses",
+ path="/v1/courses/{course_id}/lectures/{lecture_id}/quizzes/{quiz_id}/responses",
+ data_selector="responses",
+ # The docs don't say whether a retake replaces the earlier submission, so the
+ # submission time is part of the key.
+ primary_key=["course_id", "lecture_id", "quiz_id", "student_id", "submitted_at"],
+ partition_key="submitted_at",
+ ),
}
ENDPOINTS = tuple(TEACHABLE_ENDPOINTS)
diff --git a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/source.py b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/source.py
index eb2eae19b490..d2e87fe4bbbb 100644
--- a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/source.py
+++ b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/source.py
@@ -116,8 +116,8 @@ class TeachableSource(ResumableSource[TeachableSourceConfig, TeachableResumeConf
category=DataWarehouseSourceCategory.E_COMMERCE,
label="Teachable",
caption=(
- "Import users, courses, enrollments, sales transactions, and pricing plans from "
- "your Teachable school.\n\n"
+ "Import users, courses, enrollments, course progress, lectures, quizzes, quiz responses, "
+ "sales transactions, and pricing plans from your Teachable school.\n\n"
"Create an API key in your Teachable admin under Settings > API. "
"The Teachable API requires a school on the Growth plan or higher."
),
diff --git a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/teachable.py b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/teachable.py
index 635569d335d2..8e218cec64ca 100644
--- a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/teachable.py
+++ b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/teachable.py
@@ -1,9 +1,11 @@
import dataclasses
-from collections.abc import Callable, Iterable
+from collections.abc import Callable, Iterable, Iterator
from typing import Any, Optional, cast
from requests import Request, Response
+from posthog.dataclasses import frozen
+
from products.warehouse_sources.backend.temporal.data_imports.sources.common.datetime_utils import (
coerce_datetime_to_utc,
)
@@ -12,13 +14,19 @@ from products.warehouse_sources.backend.temporal.data_imports.sources.common.res
RESTAPIConfig,
rest_api_resource,
)
+from products.warehouse_sources.backend.temporal.data_imports.sources.common.rest_source.auth import APIKeyAuth
+from products.warehouse_sources.backend.temporal.data_imports.sources.common.rest_source.config_setup import (
+ create_response_hooks,
+)
from products.warehouse_sources.backend.temporal.data_imports.sources.common.rest_source.fanout import (
build_dependent_resource,
)
from products.warehouse_sources.backend.temporal.data_imports.sources.common.rest_source.paginators import (
BasePaginator,
PageNumberPaginator,
+ SinglePagePaginator,
)
+from products.warehouse_sources.backend.temporal.data_imports.sources.common.rest_source.rest_client import RESTClient
from products.warehouse_sources.backend.temporal.data_imports.sources.common.rest_source.typing import (
ClientConfig,
Endpoint,
@@ -28,6 +36,7 @@ from products.warehouse_sources.backend.temporal.data_imports.sources.common.res
from products.warehouse_sources.backend.temporal.data_imports.sources.common.resumable import ResumableSourceManager
from products.warehouse_sources.backend.temporal.data_imports.sources.common.typings import SortMode, SourceResponse
from products.warehouse_sources.backend.temporal.data_imports.sources.teachable.settings import (
+ DEFAULT_PAGE_SIZE,
TEACHABLE_BASE_URL,
TEACHABLE_ENDPOINTS,
TeachableEndpointConfig,
@@ -190,7 +199,7 @@ def get_resource(
incremental_field_name: str | None = None,
) -> EndpointResource:
config = TEACHABLE_ENDPOINTS[endpoint]
- if config.fanout:
+ if config.fanout or endpoint in NESTED_FANOUT_ROWS:
raise ValueError(f"Fan-out endpoint '{endpoint}' must use the fan-out path")
endpoint_config: Endpoint = {
@@ -218,14 +227,227 @@ def get_resource(
}
+SafePoint = Callable[[], None]
+NestedRows = Iterator[list[dict[str, Any]]]
+
+# A course, lecture, quiz or enrollment can be deleted between the parent listing and the
+# child request; skip it rather than fail the whole sync.
+_IGNORE_NOT_FOUND = create_response_hooks([{"status_code": 404, "action": "ignore"}], resource_name="teachable")
+
+
+def _rest_client(api_key: str) -> RESTClient:
+ return RESTClient(
+ base_url=TEACHABLE_BASE_URL,
+ headers={"Accept": "application/json"},
+ auth=APIKeyAuth(api_key=api_key, name="apiKey", location="header"),
+ allowed_hosts=[],
+ allow_redirects=False,
+ )
+
+
+def _get_object(client: RESTClient, path: str, params: Optional[dict[str, Any]] = None) -> Optional[dict[str, Any]]:
+ """GET one JSON object, or None when the API answers 404."""
+ for page in client.paginate(path=path, params=params, paginator=SinglePagePaginator(), hooks=_IGNORE_NOT_FOUND):
+ if page and isinstance(page[0], dict):
+ return page[0]
+ return None
+
+
+def _required(body: dict[str, Any], key: str, path: str) -> Any:
+ if key not in body:
+ raise ValueError(f"Teachable response for {path} has no {key!r} key. The API response shape may have changed.")
+ return body[key]
+
+
+@frozen
+class LectureKey:
+ course_id: int
+ lecture_id: int
+
+
+@frozen
+class QuizKey:
+ course_id: int
+ lecture_id: int
+ quiz_id: int
+
+
+def _iter_course_ids(client: RESTClient, safe_point: SafePoint) -> Iterator[int]:
+ for courses in client.paginate(
+ path="/v1/courses",
+ params={"per": DEFAULT_PAGE_SIZE},
+ paginator=_page_paginator(),
+ data_selector="courses",
+ data_selector_required=True,
+ ):
+ safe_point()
+ for course in courses:
+ if isinstance(course, dict) and isinstance(course.get("id"), int):
+ yield course["id"]
+
+
+def _lecture_ids(client: RESTClient, course_id: int) -> list[int]:
+ path = f"/v1/courses/{course_id}"
+ body = _get_object(client, path)
+ if body is None:
+ return []
+ course = _required(body, "course", path) or {}
+ return [
+ lecture["id"]
+ for section in course.get("lecture_sections") or []
+ if isinstance(section, dict)
+ for lecture in section.get("lectures") or []
+ if isinstance(lecture, dict) and isinstance(lecture.get("id"), int)
+ ]
+
+
+def _quiz_ids(client: RESTClient, course_id: int, lecture_id: int) -> list[int]:
+ path = f"/v1/courses/{course_id}/lectures/{lecture_id}/quizzes"
+ body = _get_object(client, path)
+ if body is None:
+ return []
+ return [quiz_id for quiz_id in _required(body, "quiz_ids", path) or [] if isinstance(quiz_id, int)]
+
+
+def _iter_lecture_keys(client: RESTClient, safe_point: SafePoint) -> Iterator[LectureKey]:
+ for course_id in _iter_course_ids(client, safe_point):
+ lecture_ids = _lecture_ids(client, course_id)
+ safe_point()
+ for lecture_id in lecture_ids:
+ yield LectureKey(course_id=course_id, lecture_id=lecture_id)
+
+
+def _iter_quiz_keys(client: RESTClient, safe_point: SafePoint) -> Iterator[QuizKey]:
+ for lecture in _iter_lecture_keys(client, safe_point):
+ quiz_ids = _quiz_ids(client, lecture.course_id, lecture.lecture_id)
+ safe_point()
+ for quiz_id in quiz_ids:
+ yield QuizKey(course_id=lecture.course_id, lecture_id=lecture.lecture_id, quiz_id=quiz_id)
+
+
+def _lecture_rows(client: RESTClient, safe_point: SafePoint) -> NestedRows:
+ for key in _iter_lecture_keys(client, safe_point):
+ course_id = key.course_id
+ path = f"/v1/courses/{course_id}/lectures/{key.lecture_id}"
+ body = _get_object(client, path)
+ if body is None:
+ safe_point()
+ continue
+ yield [{**_required(body, "lecture", path), "course_id": course_id}]
+
+
+def _quiz_rows(client: RESTClient, safe_point: SafePoint) -> NestedRows:
+ for key in _iter_quiz_keys(client, safe_point):
+ course_id, lecture_id, quiz_id = key.course_id, key.lecture_id, key.quiz_id
+ quiz = _get_object(client, f"/v1/courses/{course_id}/lectures/{lecture_id}/quizzes/{quiz_id}")
+ if quiz is None:
+ safe_point()
+ continue
+ yield [{**quiz, "course_id": course_id, "lecture_id": lecture_id}]
+
+
+def _quiz_response_rows(client: RESTClient, safe_point: SafePoint) -> NestedRows:
+ for key in _iter_quiz_keys(client, safe_point):
+ course_id, lecture_id, quiz_id = key.course_id, key.lecture_id, key.quiz_id
+ path = f"/v1/courses/{course_id}/lectures/{lecture_id}/quizzes/{quiz_id}/responses"
+ body = _get_object(client, path)
+ responses = _required(body, "responses", path) if body is not None else None
+ if body is None or not responses:
+ safe_point()
+ continue
+ yield [
+ {
+ **response,
+ "course_id": course_id,
+ "lecture_id": lecture_id,
+ "quiz_id": quiz_id,
+ "graded": body.get("graded"),
+ }
+ for response in responses
+ if isinstance(response, dict)
+ ]
+
+
+def _course_progress(
+ client: RESTClient, course_id: int, user_id: int, safe_point: SafePoint
+) -> Optional[dict[str, Any]]:
+ """Fetch one student's progress, joining `lecture_sections` across pages.
+
+ The endpoint documents `page`/`per` and pagination meta but not which list they page, so
+ every page is read and its sections appended to the first page's object.
+ """
+ path = f"/v1/courses/{course_id}/progress"
+ progress: Optional[dict[str, Any]] = None
+ page = 1
+ while True:
+ body = _get_object(client, path, params={"user_id": user_id, "page": page, "per": DEFAULT_PAGE_SIZE})
+ if body is None:
+ return progress
+ page_progress = _required(body, "course_progress", path) or {}
+ if progress is None:
+ progress = dict(page_progress)
+ else:
+ progress["lecture_sections"] = [
+ *(progress.get("lecture_sections") or []),
+ *(page_progress.get("lecture_sections") or []),
+ ]
+ meta = body.get("meta")
+ number_of_pages = meta.get("number_of_pages") if isinstance(meta, dict) else None
+ if not isinstance(number_of_pages, int) or page >= number_of_pages or not page_progress.get("lecture_sections"):
+ return progress
+ safe_point()
+ page += 1
+
+
+def _course_progress_rows(client: RESTClient, safe_point: SafePoint) -> NestedRows:
+ for course_id in _iter_course_ids(client, safe_point):
+ seen_user_ids: set[int] = set()
+ for enrollments in client.paginate(
+ path=f"/v1/courses/{course_id}/enrollments",
+ params={"per": DEFAULT_PAGE_SIZE, "sort_direction": "asc"},
+ paginator=_page_paginator(),
+ data_selector="enrollments",
+ data_selector_required=True,
+ hooks=_IGNORE_NOT_FOUND,
+ ):
+ safe_point()
+ user_ids = [
+ enrollment["user_id"]
+ for enrollment in enrollments
+ if isinstance(enrollment, dict) and isinstance(enrollment.get("user_id"), int)
+ ]
+ for user_id in user_ids:
+ if user_id in seen_user_ids:
+ continue
+ seen_user_ids.add(user_id)
+ progress = _course_progress(client, course_id, user_id, safe_point)
+ if progress is None:
+ safe_point()
+ continue
+ yield [{**progress, "course_id": course_id, "user_id": user_id}]
+
+
+# These tables sit two or more levels under /v1/courses, which the dependent-resource helper
+# can't express. Nothing is resumable: a hand-off restarts the walk, and every request is
+# followed by a yield or a safe point.
+NESTED_FANOUT_ROWS: dict[str, Callable[[RESTClient, SafePoint], NestedRows]] = {
+ "course_progress": _course_progress_rows,
+ "lectures": _lecture_rows,
+ "quizzes": _quiz_rows,
+ "quiz_responses": _quiz_response_rows,
+}
+
+
def _make_source_response(
endpoint_config: TeachableEndpointConfig,
items_fn: Callable[[], Iterable[Any]],
sort_mode: SortMode = "asc",
+ supports_resume: bool = True,
) -> SourceResponse:
return SourceResponse(
name=endpoint_config.name,
items=items_fn,
+ supports_resume=supports_resume,
primary_keys=endpoint_config.primary_key,
sort_mode=sort_mode,
partition_count=1,
@@ -248,6 +470,14 @@ def teachable_source(
) -> SourceResponse:
endpoint_config = TEACHABLE_ENDPOINTS[endpoint]
+ nested_rows = NESTED_FANOUT_ROWS.get(endpoint)
+ if nested_rows is not None:
+ client = _rest_client(api_key)
+ safe_point: SafePoint = (
+ resumable_source_manager.safe_point if resumable_source_manager is not None else lambda: None
+ )
+ return _make_source_response(endpoint_config, lambda: nested_rows(client, safe_point), supports_resume=False)
+
if endpoint_config.fanout:
# Dependent resources don't currently support resume in the rest_source framework;
# the manager is intentionally not threaded into this path.
diff --git a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/tests/test_teachable.py b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/tests/test_teachable.py
index 12b3afda515d..c37bb75742cb 100644
--- a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/tests/test_teachable.py
+++ b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/tests/test_teachable.py
@@ -31,6 +31,21 @@ def _driver() -> SourceDriver:
return SourceDriver(TeachableSource(), TeachableSourceConfig(api_key="key"))
+_COURSES_PAGE = ScriptedResponse(json={"courses": [{"id": 11}, {"id": 12}], "meta": {"number_of_pages": 1}})
+_COURSE_11_DETAIL = ScriptedResponse(
+ json={
+ "course": {
+ "id": 11,
+ "lecture_sections": [
+ {"id": 1, "lectures": [{"id": 5, "position": 1}, {"id": 6, "position": 2}]},
+ {"id": 2, "lectures": []},
+ ],
+ }
+ }
+)
+_NOT_FOUND = ScriptedResponse(status=404, json={"error": "not found"})
+
+
def _users_response(meta: dict[str, Any]) -> Mock:
response = Mock()
response.json.return_value = {"users": [], "meta": meta}
@@ -139,9 +154,10 @@ class TestTeachableTransport:
assert config["allowed_hosts"] == []
assert config["allow_redirects"] is False
- def test_get_resource_rejects_fanout_endpoint(self) -> None:
+ @parameterized.expand([("dependent_resource", "course_enrollments"), ("nested_fanout", "lectures")])
+ def test_get_resource_rejects_fanout_endpoint(self, _name, endpoint) -> None:
with pytest.raises(ValueError, match="Fan-out endpoint"):
- get_resource("course_enrollments", should_use_incremental_field=False)
+ get_resource(endpoint, should_use_incremental_field=False)
def test_enrollments_fanout_row_format(self) -> None:
result = _driver().run(
@@ -207,3 +223,192 @@ class TestTeachableTransport:
assert response.sort_mode == expected
assert response.primary_keys == ["id"]
assert response.partition_keys == ["created_at"]
+
+
+class TestTeachableNestedFanout:
+ @parameterized.expand(
+ [
+ (
+ # Course 12 and lecture 6 were deleted after the parent listing, so their 404s are skipped.
+ "lectures",
+ {
+ "/v1/courses": [_COURSES_PAGE],
+ "/v1/courses/11": [_COURSE_11_DETAIL],
+ "/v1/courses/12": [_NOT_FOUND],
+ "/v1/courses/11/lectures/5": [
+ ScriptedResponse(json={"lecture": {"id": 5, "name": "Intro", "lecture_section_id": 1}})
+ ],
+ "/v1/courses/11/lectures/6": [_NOT_FOUND],
+ },
+ [{"id": 5, "name": "Intro", "lecture_section_id": 1, "course_id": 11}],
+ ),
+ (
+ # Lecture 6 has no quizzes; quiz 10 was deleted after the id listing.
+ "quizzes",
+ {
+ "/v1/courses": [_COURSES_PAGE],
+ "/v1/courses/11": [_COURSE_11_DETAIL],
+ "/v1/courses/12": [ScriptedResponse(json={"course": {"id": 12, "lecture_sections": []}})],
+ "/v1/courses/11/lectures/5/quizzes": [ScriptedResponse(json={"quiz_ids": [9, 10]})],
+ "/v1/courses/11/lectures/6/quizzes": [ScriptedResponse(json={"quiz_ids": []})],
+ "/v1/courses/11/lectures/5/quizzes/9": [
+ ScriptedResponse(json={"id": 9, "name": "Check", "kind": "quiz", "quiz": {"id": 3}})
+ ],
+ "/v1/courses/11/lectures/5/quizzes/10": [_NOT_FOUND],
+ },
+ [{"id": 9, "name": "Check", "kind": "quiz", "quiz": {"id": 3}, "course_id": 11, "lecture_id": 5}],
+ ),
+ (
+ # Quiz 10 has no submissions, so it adds no rows.
+ "quiz_responses",
+ {
+ "/v1/courses": [_COURSES_PAGE],
+ "/v1/courses/11": [_COURSE_11_DETAIL],
+ "/v1/courses/12": [_NOT_FOUND],
+ "/v1/courses/11/lectures/5/quizzes": [ScriptedResponse(json={"quiz_ids": [9, 10]})],
+ "/v1/courses/11/lectures/6/quizzes": [ScriptedResponse(json={"quiz_ids": []})],
+ "/v1/courses/11/lectures/5/quizzes/9/responses": [
+ ScriptedResponse(
+ json={
+ "course_id": 11,
+ "graded": True,
+ "responses": [
+ {"student_id": 7, "submitted_at": "2026-01-02T00:00:00Z", "percent_correct": 80},
+ {"student_id": 8, "submitted_at": "2026-01-03T00:00:00Z", "percent_correct": 50},
+ ],
+ }
+ )
+ ],
+ "/v1/courses/11/lectures/5/quizzes/10/responses": [
+ ScriptedResponse(json={"graded": False, "responses": []})
+ ],
+ },
+ [
+ {
+ "student_id": 7,
+ "submitted_at": "2026-01-02T00:00:00Z",
+ "percent_correct": 80,
+ "course_id": 11,
+ "lecture_id": 5,
+ "quiz_id": 9,
+ "graded": True,
+ },
+ {
+ "student_id": 8,
+ "submitted_at": "2026-01-03T00:00:00Z",
+ "percent_correct": 50,
+ "course_id": 11,
+ "lecture_id": 5,
+ "quiz_id": 9,
+ "graded": True,
+ },
+ ],
+ ),
+ (
+ # User 7's progress spans two pages and is joined into one row; user 8 was unenrolled
+ # after the enrollment listing.
+ "course_progress",
+ {
+ "/v1/courses": [_COURSES_PAGE],
+ "/v1/courses/11/enrollments": [
+ ScriptedResponse(
+ json={"enrollments": [{"user_id": 7}, {"user_id": 8}], "meta": {"number_of_pages": 1}}
+ )
+ ],
+ "/v1/courses/12/enrollments": [_NOT_FOUND],
+ "/v1/courses/11/progress": [
+ ScriptedResponse(
+ json={
+ "course_progress": {
+ "id": 70,
+ "enrolled_at": "2026-01-01T00:00:00Z",
+ "percent_complete": 40,
+ "lecture_sections": [{"lecture_section_progress": {"id": 1}}],
+ },
+ "meta": {"page": 1, "number_of_pages": 2},
+ }
+ ),
+ ScriptedResponse(
+ json={
+ "course_progress": {
+ "id": 70,
+ "enrolled_at": "2026-01-01T00:00:00Z",
+ "percent_complete": 40,
+ "lecture_sections": [{"lecture_section_progress": {"id": 2}}],
+ },
+ "meta": {"page": 2, "number_of_pages": 2},
+ }
+ ),
+ _NOT_FOUND,
+ ],
+ },
+ [
+ {
+ "id": 70,
+ "enrolled_at": "2026-01-01T00:00:00Z",
+ "percent_complete": 40,
+ "lecture_sections": [
+ {"lecture_section_progress": {"id": 1}},
+ {"lecture_section_progress": {"id": 2}},
+ ],
+ "course_id": 11,
+ "user_id": 7,
+ }
+ ],
+ ),
+ ]
+ )
+ def test_nested_fanout_walk(self, endpoint, routes, expected_rows) -> None:
+ result = _driver().run(endpoint, route(routes))
+
+ assert result.raised is None
+ assert result.rows == expected_rows
+ # Every scripted answer was used, so no branch of the walk was skipped.
+ assert len(result.requests) == sum(len(answers) for answers in routes.values())
+
+ def test_course_progress_requests_each_enrolled_user_once(self) -> None:
+ progress = ScriptedResponse(
+ json={"course_progress": {"id": 1, "lecture_sections": []}, "meta": {"number_of_pages": 1}}
+ )
+ result = _driver().run(
+ "course_progress",
+ route(
+ {
+ "/v1/courses": [ScriptedResponse(json={"courses": [{"id": 11}], "meta": {"number_of_pages": 1}})],
+ "/v1/courses/11/enrollments": [
+ ScriptedResponse(
+ json={
+ "enrollments": [{"user_id": 7}, {"user_id": 8}, {"user_id": 7}],
+ "meta": {"number_of_pages": 1},
+ }
+ )
+ ],
+ "/v1/courses/11/progress": [progress, progress],
+ }
+ ),
+ )
+
+ assert [row["user_id"] for row in result.rows] == [7, 8]
+ progress_requests = [r for r in result.requests if r.path.endswith("/progress")]
+ assert [r.param("user_id") for r in progress_requests] == ["7", "8"]
+
+ @parameterized.expand(
+ [
+ ("course_detail", "lectures", "/v1/courses/11", {"id": 11}, "'course'"),
+ ("lecture_detail", "lectures", "/v1/courses/11/lectures/5", {"id": 5}, "'lecture'"),
+ ("quiz_ids", "quizzes", "/v1/courses/11/lectures/5/quizzes", {"ids": []}, "'quiz_ids'"),
+ ]
+ )
+ def test_missing_wrapper_key_fails_loud(self, _name, endpoint, broken_path, broken_body, key) -> None:
+ routes = {
+ "/v1/courses": [ScriptedResponse(json={"courses": [{"id": 11}], "meta": {"number_of_pages": 1}})],
+ "/v1/courses/11": [_COURSE_11_DETAIL],
+ "/v1/courses/11/lectures/5": [ScriptedResponse(json={"lecture": {"id": 5}})],
+ "/v1/courses/11/lectures/5/quizzes": [ScriptedResponse(json={"quiz_ids": []})],
+ }
+ routes[broken_path] = [ScriptedResponse(json=broken_body)]
+
+ result = _driver().run(endpoint, route(routes))
+
+ assert isinstance(result.raised, ValueError)
+ assert key in str(result.raised)
diff --git a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/tests/test_teachable_source.py b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/tests/test_teachable_source.py
index 5a9a2b157848..235731b5853e 100644
--- a/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/tests/test_teachable_source.py
+++ b/products/warehouse_sources/backend/temporal/data_imports/sources/teachable/tests/test_teachable_source.py
@@ -40,7 +40,16 @@ class TestTeachableSource:
assert [f["field"] for f in schemas["transactions"].incremental_fields] == ["created_at"]
assert schemas["transactions"].default_incremental_lookback_seconds == TRANSACTIONS_INCREMENTAL_LOOKBACK_SECONDS
- for name in ("users", "courses", "course_enrollments", "pricing_plans"):
+ for name in (
+ "users",
+ "courses",
+ "course_enrollments",
+ "pricing_plans",
+ "course_progress",
+ "lectures",
+ "quizzes",
+ "quiz_responses",
+ ):
assert schemas[name].supports_incremental is False, name
assert schemas[name].incremental_fields == [], name