Files
jobboard/tests/test_ats_oracle.py
T
WolfandClaude Opus 5.5 c497fe965d Keep a description's headings, bullets and paragraphs, and draw them in the reader
The readers flattened a posting's HTML into one line with strip_html, so
sections and bullets ran together. _shared.html_text gives a line per
paragraph, '• ' before a list item and '## ' before a heading; every
reader's description goes through it, sections read per role get their
own headings, and the reader pane draws the lines as h3, ul and p.
Excerpts leave the marks out.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-30 21:40:12 +02:00

296 lines
13 KiB
Python

"""Oracle Recruiting Cloud: the listing's rows from a saved page (Nokia's,
trimmed), paging and completeness against a fake `get`, the role's own resource
for the description (Milestone's, trimmed), the slug, and discovery's patterns.
No network."""
from __future__ import annotations
import copy
import json
import sys
from pathlib import Path
from typing import Any
import httpx
import pytest
ROOT = Path(__file__).resolve().parents[1]
if str(ROOT) not in sys.path:
sys.path.insert(0, str(ROOT))
from app.jobs.discover import boards_in # noqa: E402
from plugins import ats # noqa: E402
from plugins.readers import oracle # noqa: E402
from plugins.readers._shared import is_web_url # noqa: E402
FIXTURES = ROOT / "tests" / "fixtures" / "ats"
HOST = "fa-evmr-saasfaprod1.fa.ocs.oraclecloud.com"
SLUG = f"{HOST}/CX_1"
API = f"https://{HOST}/hcmRestApi/resources/latest"
def _listing() -> dict[str, Any]:
return json.loads((FIXTURES / "oracle.json").read_bytes())
def _rows() -> list[ats.Row]:
return oracle.parse((FIXTURES / "oracle.json").read_bytes(), SLUG)
# ------------------------------------------------------------------- rows
def test_rows_carry_the_listing_fields() -> None:
rows = _rows()
assert [r["source_id"] for r in rows] == ["39905", "37189", "40316", "40975", "31680"]
first = rows[0]
assert first["title"] == "Supply Chain Program Manager"
assert first["url"] == f"https://{HOST}/hcmUI/CandidateExperience/en/sites/CX_1/job/39905"
assert first["posted"] is not None and first["posted"].isoformat() == "2026-09-28T00:00:00+00:00"
assert first["description"] == "" # the text comes with the role's own read
assert first["teaser"].startswith("Nokia is searching for a Supply Chain Program")
assert first["salary_text"] == "" and "salary_min" not in first
assert all(is_web_url(r["url"]) for r in rows)
def test_the_locations_are_the_primary_and_the_secondary_ones() -> None:
rows = _rows()
assert rows[0]["location"] == "Thailand; Malaysia"
assert rows[1]["location"] == "Malaysia; Philippines; Indonesia; Thailand"
assert rows[2]["location"] == "India"
def test_every_place_is_kept_with_its_work_mode() -> None:
rows = _rows()
assert [p["text"] for p in rows[1]["places"]] == [
"Malaysia", "Philippines", "Indonesia", "Thailand"]
assert [p["remote"] for p in rows[0]["places"]] == [False, False] # hybrid
assert [p["remote"] for p in rows[4]["places"]] == [True] # remote
assert [p["remote"] for p in rows[3]["places"]] == [None] # states none
assert all("country" not in r for r in rows) # prep derives it from the places
def test_remote_is_the_workplace_type_code_as_given() -> None:
assert [r["remote"] for r in _rows()] == [
"ORA_HYBRID", "ORA_HYBRID", "ORA_ON_SITE", None, "ORA_REMOTE"]
def test_a_department_is_taken_where_a_tenant_fills_one() -> None:
"""All four boards read left the department fields null."""
assert {r["department"] for r in _rows()} == {""}
data = _listing()
data["items"][0]["requisitionList"][0]["JobFamily"] = " Supply Chain "
data["items"][0]["requisitionList"][1]["Department"] = "Finance"
rows = oracle.parse(json.dumps(data).encode(), SLUG)
assert [r["department"] for r in rows[:3]] == ["Supply Chain", "Finance", ""]
def test_a_role_without_a_title_or_with_an_id_that_is_no_id_is_left_out() -> None:
data = _listing()
roles = data["items"][0]["requisitionList"]
roles[0]["Title"] = " "
roles[1]["Id"] = "1/../../x"
roles[2]["Id"] = ""
assert [r["source_id"] for r in oracle.parse(json.dumps(data).encode(), SLUG)] == \
["40975", "31680"]
def test_a_listing_without_the_expand_has_no_roles() -> None:
"""Without `expand=requisitionList` the answer is the facets only."""
data = _listing()
del data["items"][0]["requisitionList"]
assert oracle.parse(json.dumps(data).encode(), SLUG) == []
# ----------------------------------------------------------------- paging
@pytest.fixture
def no_delay(monkeypatch: pytest.MonkeyPatch) -> None:
monkeypatch.setattr(ats, "PAGE_DELAY", 0)
def _fake(monkeypatch: pytest.MonkeyPatch, total: int | None = None) -> list[str]:
"""`ats._get` answering the listing from the saved roles, `limit` and
`offset` honoured as Oracle does, and a page past the end empty."""
asked: list[str] = []
every = _listing()["items"][0]["requisitionList"]
def fake_get(client, url, host, slug, *, method="GET", payload=None, headers=None,
www_twin=False):
asked.append(url)
limit = int(url.split("limit=")[1].split(",")[0])
offset = int(url.split("offset=")[1])
data = _listing()
data["items"][0]["requisitionList"] = every[offset:offset + limit]
data["items"][0]["TotalJobsCount"] = len(every) if total is None else total
return json.dumps(data).encode()
monkeypatch.setattr(ats, "_get", fake_get)
return asked
def test_the_listing_is_paged_by_limit_and_offset_to_the_stated_total(
monkeypatch: pytest.MonkeyPatch, no_delay: None) -> None:
asked = _fake(monkeypatch)
monkeypatch.setattr(oracle, "PAGE_SIZE", 2)
rows, complete = ats._fetch_rows(object(), f"oracle:{SLUG}") # type: ignore[arg-type]
find = "onlyData=true&expand=requisitionList.secondaryLocations&finder=findReqs;siteNumber=CX_1"
assert asked == [f"{API}/recruitingCEJobRequisitions?{find},limit=2,offset=0",
f"{API}/recruitingCEJobRequisitions?{find},limit=2,offset=2",
f"{API}/recruitingCEJobRequisitions?{find},limit=2,offset=4"]
assert [r["source_id"] for r in rows] == ["39905", "37189", "40316", "40975", "31680"]
assert complete
def test_a_board_that_ends_short_of_its_total_is_incomplete(
monkeypatch: pytest.MonkeyPatch, no_delay: None) -> None:
"""The total says 7 and the page past the fifth role is empty."""
_fake(monkeypatch, total=7)
monkeypatch.setattr(oracle, "PAGE_SIZE", 2)
rows, complete = ats._fetch_rows(object(), f"oracle:{SLUG}") # type: ignore[arg-type]
assert (len(rows), complete) == (5, False)
def test_a_small_board_is_one_page(monkeypatch: pytest.MonkeyPatch, no_delay: None) -> None:
asked = _fake(monkeypatch)
rows, complete = ats._fetch_rows(object(), f"oracle:{SLUG}") # type: ignore[arg-type]
assert len(asked) == 1 and f"limit={oracle.PAGE_SIZE},offset=0" in asked[0]
assert (len(rows), complete) == (5, True)
def test_an_empty_board_is_complete(monkeypatch: pytest.MonkeyPatch, no_delay: None) -> None:
data = _listing()
data["items"][0]["requisitionList"] = []
data["items"][0]["TotalJobsCount"] = 0
monkeypatch.setattr(ats, "_get", lambda *a, **k: json.dumps(data).encode())
assert ats._fetch_rows(object(), f"oracle:{SLUG}") == ([], True) # type: ignore[arg-type]
# ----------------------------------------------------------------- detail
def _detail_get(body: bytes, asked: list[str]) -> Any:
def get(url: str, *, method: str = "GET", payload: Any = None, headers: Any = None) -> bytes:
asked.append(url)
return body
return get
def test_detail_reads_the_description_and_the_posting_time() -> None:
asked: list[str] = []
row = {"source_id": "24594"}
got = oracle.detail(_detail_get((FIXTURES / "oracle-detail.json").read_bytes(), asked),
"fa-ewto-saasfaprod1.fa.ocs.oraclecloud.com/CX_1001", row)
assert asked == [
"https://fa-ewto-saasfaprod1.fa.ocs.oraclecloud.com/hcmRestApi/resources/latest/"
"recruitingCEJobRequisitionDetails?expand=all&onlyData=true"
"&finder=ById;Id=%2224594%22,siteNumber=CX_1001"]
assert got["description"].startswith("Be the key in unlocking business opportunities")
assert "<" not in got["description"] and "&nbsp;" not in got["description"]
assert "&amp;" not in got["description"] and "\xa0" not in got["description"]
assert got["posted"].isoformat() == "2026-09-28T15:09:24+00:00"
assert set(got) <= set(ats.DETAIL_FIELDS)
def test_detail_appends_the_responsibilities_and_qualifications_where_filled() -> None:
data = json.loads((FIXTURES / "oracle-detail.json").read_bytes())
data["items"][0]["ExternalResponsibilitiesStr"] = "<ul><li>Run the desk</li></ul>"
data["items"][0]["ExternalQualificationsStr"] = "<p>Five years &amp; more</p>"
got = oracle.detail(_detail_get(json.dumps(data).encode(), []), SLUG, {"source_id": "24594"})
assert got["description"].endswith("\n## Responsibilities\n\u2022 Run the desk\n## Qualifications\nFive years & more")
def test_detail_of_a_role_no_longer_open_has_nothing_to_read() -> None:
data = copy.deepcopy(json.loads((FIXTURES / "oracle-detail.json").read_bytes()))
data["items"] = []
assert oracle.detail(_detail_get(json.dumps(data).encode(), []), SLUG,
{"source_id": "1"}) == {}
def test_a_role_id_cannot_add_to_the_detail_url() -> None:
asked: list[str] = []
oracle.detail(_detail_get(b'{"items": []}', asked), SLUG, {"source_id": '1"&x=/../'})
assert asked[0].endswith("finder=ById;Id=%221%22%26x%3D%2F..%2F%22,siteNumber=CX_1")
# ----------------------------------------------------------- slug and URLs
@pytest.mark.parametrize("slug", [
"fa-ewto-saasfaprod1.fa.ocs.oraclecloud.com/CX_1001",
"iaings.fa.ocs.oraclecloud.com/SinchCareer",
"ejqi.fa.ocs.oraclecloud.eu/CX_1001",
"a.oraclecloud.com/CX",
])
def test_a_pod_and_a_site_number_are_a_slug(slug: str) -> None:
assert oracle.slug_ok(slug) and ats.is_known_target(f"oracle:{slug}")
@pytest.mark.parametrize("slug", [
"", "CX_1", "/CX_1", f"{HOST}", f"{HOST}/", f"{HOST}/CX_1/", f"{HOST}/CX_1/jobs",
f"{HOST}/a b", f" {HOST}/CX_1", f"{HOST} /CX_1", f"{HOST}/CX_1\n", f"{HOST}/CX_1 ",
f"{HOST}\n/CX_1", f"{HOST}/CX_1\r\n",
f"{HOST}@evil.com/CX_1", f"evil.com@{HOST}/CX_1", f"{HOST}#x/CX_1", f"{HOST}?x=1/CX_1",
f"{HOST}/CX_1?x=1", f"{HOST}/CX_1#x", f"{HOST}/CX%2f1", f"{HOST}/..", f"{HOST}/.",
f"..{HOST}/CX_1", "../CX_1", f"{HOST}:8443/CX_1", f"{HOST}:443/CX_1",
"localhost/CX_1", "127.0.0.1/CX_1", "[::1]/CX_1", "2130706433/CX_1",
"oraclecloud.com/CX_1", ".oraclecloud.com/CX_1", "evil.com/CX_1",
"fa.oraclecloud.com.evil.com/CX_1", "fa.oraclecloud.com.internal/CX_1",
"fa.oraclecloud.org/CX_1", "fa.oraclecloud.com/a/b", "fa.oraclecloud.com\\CX_1",
])
def test_a_hostile_slug_is_refused(slug: str) -> None:
assert not oracle.slug_ok(slug)
assert not ats.is_known_target(f"oracle:{slug}")
def test_only_the_pod_named_by_the_slug_is_allowed() -> None:
allow = oracle.allowed(SLUG)
assert allow(HOST) and allow(HOST.upper())
assert not allow("fa-ewto-saasfaprod1.fa.ocs.oraclecloud.com.evil.com")
assert not allow("jobs.nokia.com") and not allow("oraclecloud.com")
assert not allow("fa-evmr-saasfaprod2.fa.ocs.oraclecloud.com")
assert ats._allowed_host("oracle", SLUG)(HOST)
def test_the_careers_page_is_the_candidate_site() -> None:
url = f"https://{HOST}/hcmUI/CandidateExperience/en/sites/CX_1/jobs"
assert oracle.careers_url(SLUG) == ats.careers_url(f"oracle:{SLUG}") == url
assert oracle.allowed(SLUG)(httpx.URL(url).host)
def test_a_slug_written_with_capitals_names_the_same_pod() -> None:
slug = f"{HOST.upper()}/CX_1"
assert oracle.slug_ok(slug) and oracle.allowed(slug)(HOST)
assert oracle.careers_url(slug).startswith(f"https://{HOST}/")
# -------------------------------------------------------------- discovery
def test_discovery_finds_the_pod_and_the_site_in_a_candidate_site_link() -> None:
page = f'<a href="https://{HOST}/hcmUI/CandidateExperience/en/sites/CX_1/job/39905">Role</a>'
assert [(h, s) for h, s, _ in boards_in(page)] == [("oracle", SLUG)]
def test_discovery_keeps_the_case_of_a_site_name() -> None:
page = "https://iaings.fa.ocs.oraclecloud.com/hcmUI/CandidateExperience/en/sites/SinchCareer/jobs"
assert [(h, s) for h, s, _ in boards_in(page)] == [("oracle", "iaings.fa.ocs.oraclecloud.com/SinchCareer")]
def test_discovery_reads_the_site_off_the_rest_call() -> None:
page = ("fetch('https://ejqi.fa.ocs.oraclecloud.eu/hcmRestApi/resources/latest/"
"recruitingCEJobRequisitions?onlyData=true&finder=findReqs;siteNumber=CX_1001,limit=25')")
assert [(h, s) for h, s, _ in boards_in(page)] == [("oracle", "ejqi.fa.ocs.oraclecloud.eu/CX_1001")]
def test_discovery_names_a_pod_it_has_no_site_for() -> None:
page = f"https://{HOST}/hcmUI/faces/AtkHomePageWelcome"
assert [(h, s) for h, s, _ in boards_in(page)] == [("oracle", "")]
# A company's own domain, which only redirects to the pod, names nothing.
assert boards_in("https://jobs.nokia.com/en/sites/CX_1") == []
def test_every_slug_discovery_finds_is_a_slug() -> None:
pages = [f"https://{HOST}/hcmUI/CandidateExperience/en/sites/CX_1/jobs",
"https://a.fa.us2.oraclecloud.com/hcmUI/CandidateExperience/fr/sites/CX_1001/job/1",
"https://a.oraclecloud.eu/hcmUI/CandidateExperience/en/sites/X?lang=en"]
for page in pages:
found = boards_in(page)
assert found and all(oracle.slug_ok(slug) for _h, slug, _ in found), page