Hosts a to l declare READS: places, teaser, stated seniority and employment

Ashby, BambooHR, Deel, Eightfold, Gem, Greenhouse, HiBob, HR Manager, Jibe,
Jobylon and Lever move to feed_url(slug, ask) and page(get, slug, cursor, ask),
declare READS with the robots decision, and drop robots_page. Rows carry
places (Ashby secondaryLocations, Lever allLocations, Greenhouse offices in
place of a Hybrid placeholder), Jobylon keeps its summary as teaser and maps
experience and employment, Eightfold reads creationTs as the posting date and
Ashby reads workplaceType. Fixtures <name>-places.* and jobylon-teaser.html
drive the conformance checks; the eleven names leave UNDECLARED.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
WolfandClaude Opus 5.5 committed 2026-09-30 09:11:53 +02:00
1 parent 4c0da58da7
commit f2e652a8da
35 files changed
+1290 -391

No files matched your search

+33 -27
View File
@@ -9,7 +9,6 @@ from datetime import UTC, datetime
from pathlib import Path
from typing import Any
import httpx
import pytest
ROOT = Path(__file__).resolve().parents[1]
@@ -18,7 +17,7 @@ if str(ROOT) not in sys.path:
from app.jobs import discover # noqa: E402
from plugins import ats, readers # noqa: E402
from plugins.readers import bamboohr # noqa: E402
from plugins.readers import Ask, bamboohr # noqa: E402
FIXTURES = ROOT / "tests" / "fixtures" / "ats"
@@ -99,9 +98,10 @@ def test_detail_reads_the_text_the_date_and_the_country() -> None:
found = bamboohr.detail(_get_of({url: _read("bamboohr-detail-37.json")}, asked),
"configit", row)
assert asked == [(url, {"Accept": "application/json"})]
assert set(found) <= set(ats.DETAIL_FIELDS)
assert set(found) <= set(ats.DETAIL_FIELDS) | {"places"}
assert found["posted"] == datetime(2026, 9, 23, tzinfo=UTC)
assert found["location"] == "København Ø, Denmark"
assert found["places"] == [{"text": "København Ø, Denmark", "remote": None}]
assert found["remote"] == "hybrid"
text = found["description"]
assert text.startswith("Our Global Marketing Team is looking for")
@@ -125,38 +125,44 @@ def test_detail_refuses_an_answer_that_is_not_a_role() -> None:
bamboohr.detail(_get_of({url: body}, []), "configit", row)
def test_the_board_and_its_details_are_read_through_the_app(
monkeypatch: pytest.MonkeyPatch) -> None:
monkeypatch.setattr(ats, "PAGE_DELAY", 0)
def test_the_board_and_its_details_are_read_with_the_json_accept_on_the_detail() -> None:
base = "https://configit.bamboohr.com/careers"
seen: list[str] = []
asked: list[tuple[str, Any]] = []
answers: dict[str, bytes | Exception] = {
f"{base}/list": _read("bamboohr.json"),
f"{base}/37/detail": _read("bamboohr-detail-37.json")}
get = _get_of(answers, asked)
rows = bamboohr.parse(get(bamboohr.feed_url("configit", Ask())), "configit")
assert len(rows) == 2
found = bamboohr.detail(get, "configit", rows[1])
assert [url for url, _ in asked] == [f"{base}/list", f"{base}/37/detail"]
assert asked[1][1] == {"Accept": "application/json"}
assert found["location"] == "København Ø, Denmark"
assert found["posted"] == datetime(2026, 9, 23, tzinfo=UTC)
def handler(request: httpx.Request) -> httpx.Response:
seen.append(str(request.url))
if request.url.path == "/careers/list":
return httpx.Response(200, content=_read("bamboohr.json"))
if request.url.path == "/careers/26/detail":
return httpx.Response(404)
assert request.headers["accept"] == "application/json"
return httpx.Response(200, content=_read("bamboohr-detail-37.json"))
with httpx.Client(transport=httpx.MockTransport(handler)) as client:
rows, complete = ats._fetch_rows(client, "bamboohr:configit")
assert complete and len(rows) == 2 and seen == [f"{base}/list"]
filled, done, failed = ats._read_details(client, "bamboohr:configit", rows, complete,
set(), set())
assert seen[1:] == [f"{base}/26/detail", f"{base}/37/detail"]
# A 404 counts as read; the other role is filled from its own page.
assert done == {"bamboohr:configit:26", "bamboohr:configit:37"} and failed == set()
assert filled[0]["description"] == "" and filled[1]["location"] == "København Ø, Denmark"
assert filled[1]["posted"] == datetime(2026, 9, 23, tzinfo=UTC)
def test_the_role_states_one_place_and_the_employment() -> None:
rows = bamboohr.parse(_read("bamboohr.json"), "configit")
assert rows[1]["places"] == [{"text": "København Ø", "remote": None}]
assert rows[1]["employment"] == "full_time"
body = json.loads(_read("bamboohr.json"))
body["result"][0] |= {"employmentStatusLabel": "Seasonal", "location": {"city": None}}
first = bamboohr.parse(json.dumps(body).encode(), "configit")[0]
assert first["places"] == [] and "employment" not in first
def test_it_declares_what_it_reads() -> None:
reads = bamboohr.READS
assert (reads.listing, reads.text, reads.places, reads.fields.posted) == (
"whole", "detail", "one", "trusted")
assert reads.robots.startswith("read: robots.txt allows /careers")
assert not hasattr(bamboohr, "robots_page")
def test_the_slug_the_urls_and_the_allow_list() -> None:
assert bamboohr.slug_ok("configit") and bamboohr.slug_ok("Configit")
assert bamboohr.careers_url("configit") == "https://configit.bamboohr.com/careers"
assert bamboohr.robots_page("configit") == "https://configit.bamboohr.com/careers"
assert bamboohr.feed_url("Configit") == "https://configit.bamboohr.com/careers/list"
assert bamboohr.feed_url("Configit", Ask()) == "https://configit.bamboohr.com/careers/list"
allow = bamboohr.allowed("Configit")
assert allow("configit.bamboohr.com") and allow("CONFIGIT.bamboohr.com")
assert not allow("bamboohr.com") and not allow("other.bamboohr.com")