Declare READS for Oracle, Personio, Phenom, Pinpoint, Recruitee, Rippling, SmartRecruiters, SuccessFactors, Teamtailor, Workable and Workday

Every row carries places (Personio additionalOffices, Workable and Recruitee
locations[], every SuccessFactors address, Workday's own read), the list
teaser of Phenom and Oracle is kept in teaser, and the level, employment and
country a host states are mapped. Workday "N Locations" and "Country - City",
and Recruitee "Remote job", are no longer places. Phenom takes dateCreated
as posted. robots_page and ROBOTS_ANY_CASE go; READS.robots records the
decision. Signatures are feed_url(slug, ask) and page(get, slug, cursor, ask),
with an empty Ask as the default so the runner of today still calls them.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
WolfandClaude Opus 5.5 committed 2026-09-30 09:14:27 +02:00
1 parent 4c0da58da7
commit 4316624769
42 files changed
+3126 -199

No files matched your search

+26 -5
View File
@@ -45,12 +45,25 @@ def test_rows_carry_the_listing_fields() -> None:
first = rows[0]
assert first["title"] == "Principal Economy Designer | New Games"
assert first["department"] == "Games Design & Production"
assert first["posted"] is not None and first["posted"].isoformat() == "2026-09-29T00:00:00+00:00"
assert first["description"].startswith("Become part of our team and help shape the economic")
assert first["description"] == "" # the text comes with the role's own page
assert first["teaser"].startswith("Become part of our team and help shape the economic")
assert first["remote"] is None and first["salary_text"] == ""
assert all(is_web_url(r["url"]) for r in rows)
def test_posted_is_the_date_the_role_was_created_not_the_one_the_site_moves() -> None:
"""The second role was created on 15 June and is listed as posted on 28
September; Phenom refreshes `postedDate`, so the creation date is taken, and
`postedDate` only where a site sends no `dateCreated`."""
rows = _rows()
assert rows[0]["posted"].isoformat() == "2026-09-29T09:59:51.253000+00:00"
assert rows[1]["posted"].isoformat() == "2026-06-15T17:02:06.245000+00:00"
data = _answer()
del data["refineSearch"]["data"]["jobs"][1]["dateCreated"]
assert phenom.parse(json.dumps(data).encode(), SLUG)[1]["posted"].isoformat() == \
"2026-09-28T00:00:00+00:00"
def test_the_url_is_the_role_page_on_the_site_not_the_apply_system() -> None:
"""`applyUrl` points at Workday; the role's page is `/job/<jobSeqNo>/<title>`."""
assert _answer()["refineSearch"]["data"]["jobs"][0]["applyUrl"].startswith("https://xboxgaming.")
@@ -66,6 +79,14 @@ def test_the_locations_are_every_place_the_role_names() -> None:
assert rows[1]["location"] == "Stockholm, Stockholm County, Sweden"
def test_every_place_is_kept_once_and_the_first_is_the_roles_own() -> None:
rows = _rows()
assert [p["text"] for p in rows[0]["places"]] == [
"Barcelona, Spain", "London, London, United Kingdom", "Stockholm, Stockholm County, Sweden"]
assert rows[2]["places"][0]["text"] == "London, London, United Kingdom"
assert all(p["remote"] is None for r in rows for p in r["places"])
def test_a_role_with_one_place_only_uses_location() -> None:
data = _answer()
job = data["refineSearch"]["data"]["jobs"][0]
@@ -264,11 +285,11 @@ def test_only_the_site_named_by_the_slug_is_allowed() -> None:
assert ats._allowed_host("phenom", SLUG)("careers.king.com")
def test_the_careers_page_is_the_search_page_and_robots_reads_it() -> None:
def test_the_careers_page_is_the_search_page() -> None:
url = "https://careers.king.com/us/en/search-results"
assert phenom.careers_url(SLUG) == ats.careers_url(f"phenom:{SLUG}") == url
assert phenom.robots_page(SLUG) == url
assert phenom.allowed(SLUG)(httpx.URL(phenom.robots_page(SLUG)).host)
assert phenom.allowed(SLUG)(httpx.URL(url).host)
assert not hasattr(phenom, "robots_page") # decided when built, see READS.robots
assert phenom.careers_url("www.careers.philips.com/global/en") == \
"https://www.careers.philips.com/global/en/search-results"