Readers oracle to workday and the providers hand on the host's words; their word tables move to app/prep/hosts

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
WolfandClaude Opus 5.5 committed 2026-09-30 11:07:07 +02:00
1 parent a66aac95cf
commit 06eb929eaf
48 files changed
+460 -388

No files matched your search

+7 -6
View File
@@ -132,10 +132,10 @@ def test_workable_builds_the_job_url_and_reads_location_and_remote() -> None:
assert [r["source_id"] for r in rows] == ["6125026", "6125041"]
assert rows[0]["url"] == "https://apply.workable.com/huggingface/j/DB4D7C0EC8/"
assert rows[0]["location"] == "France"
assert rows[0]["remote"] is True
assert rows[0]["remote"] == "remote"
assert rows[0]["department"] == "Product"
assert rows[1]["location"] == "Berlin, Germany"
assert rows[1]["remote"] is False
assert rows[1]["remote"] == "hybrid"
assert rows[0]["description"] == "" # the list endpoint carries no text
assert rows[0]["salary_text"] == "" # Workable's list endpoint states no pay at all
@@ -173,8 +173,9 @@ def test_personio_builds_the_job_url_and_reads_cdata_description() -> None:
assert [r["source_id"] for r in rows] == ["6001", "6002", "6003"]
assert rows[0]["url"] == "https://acme.jobs.personio.de/job/6001"
assert rows[0]["description"] == "Grow our pipeline."
assert rows[0]["remote"] is True # "Remote - Germany"
assert rows[1]["remote"] is None # "Munich"
assert rows[0]["places"][0]["remote"] is True # "Remote - Germany"
assert rows[1]["places"][0]["remote"] is None # "Munich"
assert "remote" not in rows[0] # the feed states none
def test_personio_reads_salary_information_when_present() -> None:
@@ -291,9 +292,9 @@ def test_teamtailor_joins_the_distinct_locations_for_country_of() -> None:
assert country_of(rows[0]["location"]) == "SE"
def test_teamtailor_maps_remote_status_and_falls_back_to_role_for_department() -> None:
def test_teamtailor_hands_on_remote_status_and_falls_back_to_role_for_department() -> None:
rows = ats._parse_teamtailor(TEAMTAILOR, "kry")
assert [r["remote"] for r in rows] == ["hybrid", "remote", None] # none says nothing
assert [r["remote"] for r in rows] == ["hybrid", "fully", "none"] # prep maps them
assert rows[1]["title"] == "Läkare" # trimmed
assert rows[1]["department"] == "Software Development"
assert ats.classify(rows[1]["department"], "Sales Lead") == "sales"
+5 -5
View File
@@ -59,19 +59,19 @@ def test_the_locations_are_the_primary_and_the_secondary_ones() -> None:
assert rows[2]["location"] == "India"
def test_every_place_is_kept_with_its_work_mode_and_the_countrys_code() -> None:
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
# One country when the places share one, none over several.
assert [r["country"] for r in rows] == [None, None, "IN", "DE", "US"]
assert all("country" not in r for r in rows) # prep derives it from the places
def test_remote_follows_the_workplace_type_code() -> None:
assert [r["remote"] for r in _rows()] == ["hybrid", "hybrid", "onsite", None, "remote"]
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:
+5 -19
View File
@@ -3,17 +3,14 @@ No network: the fixtures are saved feeds."""
from __future__ import annotations
import logging
import sys
from pathlib import Path
import pytest
ROOT = Path(__file__).resolve().parents[1]
if str(ROOT) not in sys.path:
sys.path.insert(0, str(ROOT))
from app.models import SENIORITIES # noqa: E402
from plugins.readers import Ask, personio # noqa: E402
FIXTURES = ROOT / "tests" / "fixtures" / "ats"
@@ -37,23 +34,12 @@ def test_additional_offices_are_places_of_the_role_each_once() -> None:
assert [p["text"] for p in first["places"]] == ["Remote - Germany", "Berlin", "Hamburg"]
assert [p["remote"] for p in first["places"]] == [True, None, None]
assert first["location"] == "Remote - Germany; Berlin; Hamburg"
assert first["remote"] is True # one of the places is remote
assert "remote" not in first # the feed states none
assert _rows()[1]["places"] == [{"text": "Munich", "remote": None}]
assert _rows()[1]["remote"] is None
def test_the_stated_level_and_employment_are_mapped() -> None:
def test_the_stated_level_and_employment_are_the_hosts_words() -> None:
rows = _rows()
assert [r["seniority"] for r in rows] == ["junior", "exec", "intern"]
# permanent + full-time; permanent + full-time; working student wins over part-time
assert [r["employment"] for r in rows] == ["full_time", "full_time", "internship"]
def test_the_table_speaks_the_apps_vocabulary_and_a_new_word_is_logged(
caplog: pytest.LogCaptureFixture) -> None:
assert {v for v in personio._SENIORITY.values() if v} <= set(SENIORITIES)
with caplog.at_level(logging.DEBUG, logger=personio.__name__):
assert personio._seniority("wizard") is None
assert personio._seniority("experienced") is None # known, names no level
assert personio._seniority("") is None
assert [r.getMessage() for r in caplog.records] == ["personio: unknown seniority 'wizard'"]
assert [r["seniority"] for r in rows] == ["entry-level", "executive", "student"]
# permanent + full-time gives the schedule; a working student is the contract's word
assert [r["employment"] for r in rows] == ["full-time", "full-time", "working_student"]
+1 -1
View File
@@ -226,7 +226,7 @@ def test_detail_falls_back_to_the_page_state_where_there_is_no_job_posting() ->
got = phenom.detail(_get_of("phenom-detail-3.html", []), SLUG, row)
assert got["description"].startswith("Craft: Games Design & Production Job Description:")
assert "<" not in got["description"] and "&amp;" not in got["description"]
assert got["remote"] == "onsite"
assert got["remote"] == "On-site"
assert "posted" not in got and set(got) <= set(ats.DETAIL_FIELDS)
+2 -5
View File
@@ -161,9 +161,6 @@ def test_a_role_has_its_one_place_with_the_work_mode_of_the_role() -> None:
assert rows[0]["location"] == "Bucharest"
def test_employment_is_read_from_the_type_and_an_unknown_one_is_left() -> None:
def test_employment_is_the_hosts_type_as_given() -> None:
rows = pinpoint.parse((FIXTURES / "pinpoint.json").read_bytes(), "xeneta")
assert [r["employment"] for r in rows] == ["full_time", "temporary"]
assert pinpoint._employment("permanent_part_time") == "part_time"
assert pinpoint._employment("internship") == "internship"
assert pinpoint._employment("something_new") is None
assert [r["employment"] for r in rows] == ["permanent_full_time", "fixed_term_full_time"]
+9 -25
View File
@@ -4,17 +4,14 @@ bunq's board, read on 2026-09-30."""
from __future__ import annotations
import logging
import sys
from pathlib import Path
import pytest
ROOT = Path(__file__).resolve().parents[1]
if str(ROOT) not in sys.path:
sys.path.insert(0, str(ROOT))
from app.models import SENIORITIES # noqa: E402
from plugins.readers import Ask, recruitee # noqa: E402
FIXTURES = ROOT / "tests" / "fixtures" / "ats"
@@ -27,19 +24,18 @@ def _rows() -> list[dict]:
def test_the_declaration_says_what_the_feed_gives() -> None:
reads = recruitee.READS
assert (reads.listing, reads.places, reads.text) == ("whole", "many", "listing")
assert reads.fields.posted == "trusted" and reads.fields.country == "stated"
assert reads.fields.posted == "trusted" and reads.fields.country == "none"
assert reads.robots.startswith("api: no robots.txt check needed")
assert recruitee.feed_url("acme", Ask()) == "https://acme.recruitee.com/api/offers/"
def test_every_location_is_a_place_and_the_country_is_stated_when_they_share_one() -> None:
def test_every_location_is_a_place_and_no_country_is_stated() -> None:
first, _remote, berlin = _rows()
assert [p["text"] for p in first["places"]] == [
"Amsterdam, Noord-Holland, Netherlands", "Sofia, Bulgaria"]
assert first["location"] == "Amsterdam, Noord-Holland, Netherlands; Sofia, Bulgaria"
assert first["country"] is None # NL and BG
assert "country" not in first # prep derives it from the places
assert berlin["places"] == [{"text": "Berlin, Germany", "remote": None}]
assert berlin["country"] == "DE"
def test_remote_job_is_not_a_place() -> None:
@@ -48,28 +44,16 @@ def test_remote_job_is_not_a_place() -> None:
assert remote["location"] == "Remote"
offer = {"id": 1, "title": "T", "careers_url": "https://a.example/o/t", "location": "Remote job",
"remote": False}
assert recruitee._places(offer) == ([], set())
assert recruitee._places(offer) == []
def test_an_offer_without_a_locations_list_falls_back_to_its_location() -> None:
offer = {"location": "Lisbon, Portugal", "country_code": "pt"}
assert recruitee._places(offer) == ([{"text": "Lisbon, Portugal", "remote": None}], {"PT"})
assert recruitee._places(offer) == [{"text": "Lisbon, Portugal", "remote": None}]
def test_the_stated_level_and_employment_are_mapped() -> None:
def test_the_stated_level_and_employment_are_the_hosts_codes() -> None:
rows = _rows()
assert [r["seniority"] for r in rows] == [None, "junior", "intern"]
assert [r["employment"] for r in rows] == ["full_time", "part_time", "internship"]
assert recruitee._employment("fulltime_fixed_term") == "temporary"
assert recruitee._employment("freelance") == "contract"
def test_the_table_speaks_the_apps_vocabulary_and_a_new_word_is_logged(
caplog: pytest.LogCaptureFixture) -> None:
assert {v for v in recruitee._SENIORITY.values() if v} <= set(SENIORITIES)
with caplog.at_level(logging.DEBUG, logger=recruitee.__name__):
assert recruitee._seniority("senior_level") is None
assert recruitee._employment("odd_job") is None
assert [r.getMessage() for r in caplog.records] == [
"recruitee: unknown experience_code 'senior_level'",
"recruitee: unknown employment_type_code 'odd_job'"]
assert [r["seniority"] for r in rows] == ["experienced", "entry_level", "student_school"]
assert [r["employment"] for r in rows] == [
"fulltime_permanent", "parttime_permanent", "internship"]
+10 -14
View File
@@ -41,21 +41,18 @@ def test_rows_merge_one_entry_per_place_into_one_role() -> None:
assert first["description"] == "" and first["salary_text"] == "" and first["posted"] is None
def test_remote_is_true_only_when_every_place_is_a_remote_one() -> None:
def test_a_remote_place_is_flagged_and_the_role_states_no_remote() -> None:
by_id = {r["source_id"][:8]: r for r in rippling.parse(BOARD, "rippling")}
assert by_id["46b4af52"]["remote"] is True and by_id["19ae5b34"]["remote"] is True
assert by_id["75ad50c6"]["remote"] is None # an office label says nothing either way
mixed = json.dumps([
{"uuid": "abcdefgh-1", "name": "A", "workLocation": {"label": "Remote (Germany)"}},
{"uuid": "abcdefgh-1", "name": "A", "workLocation": {"label": "Berlin"}}]).encode()
assert rippling.parse(mixed, "x")[0]["remote"] is None
assert [p["remote"] for p in by_id["46b4af52"]["places"]] == [True]
assert [p["remote"] for p in by_id["75ad50c6"]["places"]] == [None] # an office label
assert all("remote" not in r for r in by_id.values()) # prep derives it from the places
def test_prisma_board_reads_two_roles() -> None:
rows = rippling.parse(PRISMA, "prisma-careers")
assert [(r["title"], r["department"], r["location"], r["remote"]) for r in rows] == [
("General Applications: Engineering", "Engineering", "Remote (Germany)", True),
("General Applications: Product", "Product", "Remote (Germany)", True)]
assert [(r["title"], r["department"], r["location"]) for r in rows] == [
("General Applications: Engineering", "Engineering", "Remote (Germany)"),
("General Applications: Product", "Product", "Remote (Germany)")]
def test_an_empty_board_is_no_roles_and_an_error_answer_is_not_a_board() -> None:
@@ -92,7 +89,7 @@ def test_detail_reads_description_places_date_and_a_single_yearly_range() -> Non
assert "<" not in found["description"]
assert found["location"] == "San Francisco, CA; Seattle, WA; New York, NY"
assert [p["text"] for p in found["places"]] == ["San Francisco, CA", "Seattle, WA", "New York, NY"]
assert found["employment"] == "full_time" # employmentType.label SALARIED_FT
assert found["employment"] == "SALARIED_FT" # employmentType.label
assert found["posted"] is not None and found["posted"].isoformat() == \
"2025-04-07T12:29:19.348000-07:00"
assert (found["salary_min"], found["salary_max"], found["currency"]) == (216000, 378000, "USD")
@@ -203,13 +200,12 @@ def test_discovery_finds_the_board_slug() -> None:
("rippling", "prisma-careers"), ("rippling", "Acme"), ("rippling", "beta")]
def test_a_role_at_several_places_keeps_each_once_and_is_remote_only_if_all_are() -> None:
def test_a_role_at_several_places_keeps_each_once() -> None:
rows = {r["source_id"]: r for r in rippling.parse(BOARD, "rippling")}
director = rows[DIRECTOR]
assert [p["text"] for p in director["places"]] == [
"San Francisco, CA", "Seattle, WA", "New York, NY"]
assert director["location"] == "San Francisco, CA; Seattle, WA; New York, NY"
assert director["remote"] is None
remote = next(r for r in rows.values() if r["places"] == [
{"text": "Remote (United States)", "remote": True}])
assert remote["remote"] is True
assert remote["places"][0]["remote"] is True
+5 -23
View File
@@ -4,17 +4,14 @@ No network: the shapes are those of Bosch's board, read on 2026-09-30."""
from __future__ import annotations
import json
import logging
import sys
from pathlib import Path
import pytest
ROOT = Path(__file__).resolve().parents[1]
if str(ROOT) not in sys.path:
sys.path.insert(0, str(ROOT))
from app.models import SENIORITIES # noqa: E402
from plugins.readers import Ask, smartrecruiters # noqa: E402
FIXTURES = ROOT / "tests" / "fixtures" / "ats"
@@ -54,38 +51,23 @@ def test_page_reads_by_offset_and_takes_the_ask() -> None:
def test_the_place_is_one_clean_line_with_its_remote_flag_and_country() -> None:
[row] = _rows(_posting())
assert row["places"] == [{"text": "Hatvan, Hungary", "remote": False}] # ", ," gone
assert row["location"] == "Hatvan, Hungary" and row["country"] == "HU"
assert row["location"] == "Hatvan, Hungary" and row["country"] == "hu"
[remote] = _rows(_posting(location={"remote": True, "fullLocation": "Remote", "country": ""}))
assert remote["places"] == [{"text": "Remote", "remote": True}] and remote["country"] is None
[none] = _rows(_posting(location={}))
assert none["places"] == [] and none["location"] == ""
def test_the_stated_level_and_employment_are_mapped() -> None:
def test_the_stated_level_and_employment_are_the_hosts_words() -> None:
[row] = _rows(_posting(experienceLevel={"id": "internship", "label": "Internship"},
typeOfEmployment={"id": "intern", "label": "Intern"}))
assert (row["seniority"], row["employment"]) == ("intern", "internship")
levels = {"entry_level": "junior", "director": "director", "executive": "exec",
"associate": None, "mid_senior_level": None, "not_applicable": None}
for word, want in levels.items():
[row] = _rows(_posting(experienceLevel={"id": word}))
assert row["seniority"] == want, word
[row] = _rows(_posting(typeOfEmployment={"id": "permanent"}))
assert row["employment"] is None
assert (row["seniority"], row["employment"]) == ("internship", "intern")
[row] = _rows(_posting(experienceLevel={"id": "mid_senior_level"}))
assert row["seniority"] == "mid_senior_level"
[row] = _rows(_posting())
assert row["seniority"] is None and row["employment"] is None
def test_the_table_speaks_the_apps_vocabulary_and_a_new_word_is_logged(
caplog: pytest.LogCaptureFixture) -> None:
assert {v for v in smartrecruiters._SENIORITY.values() if v} <= set(SENIORITIES)
with caplog.at_level(logging.DEBUG, logger=smartrecruiters.__name__):
[row] = _rows(_posting(experienceLevel={"id": "guru"}))
assert row["seniority"] is None
assert [r.getMessage() for r in caplog.records] == [
"smartrecruiters: unknown experienceLevel 'guru'"]
def test_the_saved_page_still_reads() -> None:
data = json.loads((FIXTURES / "smartrecruiters.json").read_bytes())
assert [r["location"] for r in smartrecruiters.rows(data, "Acme")] == ["Berlin, Germany",
+3 -13
View File
@@ -315,7 +315,7 @@ def _detail(name: str, url: str = "https://careers.vestas.com/job/x/1431772533/"
return successfactors.detail(get, "careers.vestas.com", {"url": url}) # type: ignore[arg-type]
def test_detail_reads_the_description_the_date_and_the_country_code_of_an_address() -> None:
def test_detail_reads_the_description_the_date_and_the_places_of_an_address() -> None:
got = _detail("successfactors-job.html")
assert got["description"].startswith("Senior Management | Full-Time | Commercial Are you")
assert "<" not in got["description"]
@@ -324,7 +324,7 @@ def test_detail_reads_the_description_the_date_and_the_country_code_of_an_addres
# The locality and the country's name; the region is cut to "Seou" by the site.
assert got["location"] == "Seoul, South Korea"
assert got["places"] == [{"text": "Seoul, South Korea", "remote": None}]
assert got["country"] == "KR"
assert "country" not in got # prep derives it from the places
def test_detail_keeps_every_address_of_a_role_at_several_places() -> None:
@@ -333,15 +333,6 @@ def test_detail_keeps_every_address_of_a_role_at_several_places() -> None:
assert [p["text"] for p in got["places"]] == [
"Seoul, South Korea", "Busan, South Korea", "Aarhus, Denmark"]
assert got["location"] == "Seoul, South Korea; Busan, South Korea; Aarhus, Denmark"
assert "country" not in got # KR and DK: no single country for the role
def test_detail_states_the_country_when_every_address_is_in_it() -> None:
text = _read("successfactors-places-detail.html").replace(
'content="Aarhus, DK, 8200"', 'content="Daegu, KR, 41200"')
got = successfactors.detail(lambda url, **kw: text.encode(), "careers.vestas.com",
{"url": "https://careers.vestas.com/job/x/1/"})
assert got["country"] == "KR"
def test_the_listing_place_is_the_roles_only_place_until_its_page_is_read() -> None:
@@ -351,11 +342,10 @@ def test_the_listing_place_is_the_roles_only_place_until_its_page_is_read() -> N
for r in rows)
def test_detail_reads_the_country_from_a_street_address() -> None:
def test_detail_reads_the_place_from_a_street_address() -> None:
"""Systematic and most other sites give one `streetAddress`, "City, CC, zip"."""
got = _detail("successfactors-job-street.html")
assert got["location"] == "Bern, Switzerland"
assert got["country"] == "CH"
assert got["posted"] == datetime(2026, 9, 28, 2, tzinfo=UTC)
assert got["description"].startswith("Senior Solution Engineer Defence")
+3 -11
View File
@@ -5,11 +5,9 @@ No network: the fixture has the shape of Hugging Face's board, read on
from __future__ import annotations
import json
import logging
import sys
from pathlib import Path
import pytest
ROOT = Path(__file__).resolve().parents[1]
if str(ROOT) not in sys.path:
@@ -49,21 +47,15 @@ def test_every_location_of_a_role_is_a_place_once() -> None:
assert [p["text"] for p in first["places"]] == ["Berlin, Germany", "Munich, Bavaria, Germany",
"Paris, France"]
assert first["location"] == "Berlin, Germany; Munich, Bavaria, Germany; Paris, France"
assert first["country"] is None # DE and FR
assert "country" not in first # prep derives it from the places
assert [p["remote"] for p in first["places"]] == [False] * 3
assert remote["places"] == [{"text": "France", "remote": True}]
assert remote["country"] == "FR"
# A list-less answer (the older shape) still gives its one place.
assert oslo["places"] == [{"text": "Oslo, Norway", "remote": False}]
assert oslo["country"] == "NO"
def test_the_stated_hours_are_mapped_and_an_unknown_type_is_logged(
caplog: pytest.LogCaptureFixture) -> None:
with caplog.at_level(logging.DEBUG, logger=workable.__name__):
rows = _rows()
assert [r["employment"] for r in rows] == ["full_time", "part_time", None]
assert [r.getMessage() for r in caplog.records] == ["workable: unknown type 'volunteer'"]
def test_the_stated_hours_are_the_hosts_word() -> None:
assert [r["employment"] for r in _rows()] == ["full", "part", "volunteer"]
def test_the_shape_of_a_real_list_answer_is_read() -> None:
+10 -11
View File
@@ -54,8 +54,8 @@ def test_rows_read_the_listing_of_a_tenant_that_sends_remote_type() -> None:
"Associate-Solutions-Architect_R35358")
assert first["posted"] == TODAY
assert first["description"] == "" and first["department"] == "" and first["salary_text"] == ""
# "Fully Flexible" is Zendesk's word for remote.
assert [r["remote"] for r in rows] == ["remote", "onsite", "hybrid", "remote", "hybrid"]
# The host's own words, "Fully Flexible" among them; prep maps them.
assert [r["remote"] for r in rows] == ["Fully Flexible", "In Office", "Hybrid", "Fully Flexible", "Hybrid"]
def test_rows_of_a_tenant_without_remote_type_leave_remote_unknown() -> None:
@@ -80,7 +80,7 @@ def test_rows_of_a_site_called_jobs_state_the_requisition_id_and_no_date_past_30
assert rows[1]["places"] == [{"text": "Remote US TX", "remote": True}]
assert rows[2]["url"].startswith("https://redhat.wd5.myworkdayjobs.com/jobs/job/")
assert rows[3]["posted"] is None # "Posted 30+ Days Ago" says only that it is older
assert [r["remote"] for r in rows] == ["onsite", "remote", "remote", "hybrid"]
assert [r["remote"] for r in rows] == ["Onsite", "Remote", "Remote", "Hybrid"]
@pytest.mark.parametrize(("text", "expected"), [
@@ -275,13 +275,12 @@ def test_detail_reads_the_description_date_remote_and_place_of_the_role() -> Non
SLUG, row)
assert asked == [f"https://{HOST}/wday/cxs/zendesk/zendesk"
"/job/Remote-Mexico-City-Mexico/Associate-Solutions-Architect_R35358"]
assert set(got) == {"description", "location", "places", "posted", "remote", "country",
"employment"}
assert set(got) == {"description", "location", "places", "posted", "remote", "employment"}
assert got["posted"] == datetime(2026, 9, 29, tzinfo=UTC)
assert got["remote"] == "remote"
assert got["remote"] == "Fully Flexible"
assert got["location"] == "Remote, Mexico City, Mexico"
assert got["places"] == [{"text": "Remote, Mexico City, Mexico", "remote": True}]
assert (got["country"], got["employment"]) == ("MX", "full_time")
assert got["employment"] == "Full time" and "country" not in got
assert got["description"].startswith("Job Description") or len(got["description"]) > 200
assert "<" not in got["description"] and "&nbsp;" not in got["description"]
assert " " not in got["description"]
@@ -296,7 +295,7 @@ def test_detail_names_every_location_of_a_role_listed_as_several() -> None:
"/job/Remote-Italy/Telco-Cloud-Consultant_R-059566-2"]
assert got["location"] == "Remote Italy; Remote Spain"
assert [p["text"] for p in got["places"]] == ["Remote Italy", "Remote Spain"]
assert "country" not in got # the code belongs to the main location alone
assert "country" not in got
def _detail(**info: Any) -> bytes:
@@ -337,9 +336,9 @@ def test_a_country_dash_city_place_is_turned_round() -> None:
assert workday._place("12 Locations", None) is None
def test_detail_takes_no_word_for_remote_it_does_not_know() -> None:
def test_detail_hands_on_a_word_for_remote_it_does_not_know() -> None:
row = workday.rows(_load("workday.json"), SLUG)[0]
assert "remote" not in workday.detail(_get_of(_detail(remoteType="No"), []), SLUG, row)
assert workday.detail(_get_of(_detail(remoteType="No"), []), SLUG, row)["remote"] == "No"
def test_detail_refuses_a_row_that_is_not_a_role_of_the_board() -> None:
@@ -383,7 +382,7 @@ def test_detail_is_read_through_the_getter_ats_binds(
monkeypatch.setattr(ats, "_get", fake_get)
row = workday.rows(_load("workday.json"), SLUG)[0]
got = workday.detail(ats._getter(object(), "workday", SLUG), SLUG, row) # type: ignore[arg-type]
assert got["remote"] == "remote"
assert got["remote"] == "Fully Flexible"
assert seen[0][:2] == ("workday", SLUG)
+12
View File
@@ -0,0 +1,12 @@
"""Oracle's word tables for prep: the host's codes, and what they mean."""
from __future__ import annotations
import importlib
oracle = importlib.import_module("app.prep.hosts.oracle")
def test_workplace_type_codes_read() -> None:
assert oracle.REMOTE == {"ora_remote": "remote", "ora_hybrid": "hybrid",
"ora_on_site": "onsite"}
+22
View File
@@ -0,0 +1,22 @@
"""Personio's word tables for prep: the feed's own words, and what they mean."""
from __future__ import annotations
import importlib
personio = importlib.import_module("app.prep.hosts.personio")
def test_seniority_words_read() -> None:
seniority = personio.SENIORITY
assert [seniority[w] for w in ("entry-level", "executive", "student")] == [
"junior", "exec", "intern"]
assert seniority["experienced"] == "" # known, names no level
assert "wizard" not in seniority
def test_employment_words_read_from_the_contract_or_the_schedule() -> None:
employment = personio.EMPLOYMENT
assert employment["working_student"] == "internship" and employment["freelance"] == "contract"
assert employment["permanent"] == ""
assert employment["full-time"] == "full_time" and employment["part-time"] == "part_time"
+12
View File
@@ -0,0 +1,12 @@
"""Phenom's word table for prep: the words a page's `checkRemote` says."""
from __future__ import annotations
import importlib
phenom = importlib.import_module("app.prep.hosts.phenom")
def test_check_remote_words_read() -> None:
assert phenom.REMOTE["on-site"] == "onsite" # King's pages
assert phenom.REMOTE["remote"] == "remote" and phenom.REMOTE["hybrid"] == "hybrid"
+24
View File
@@ -0,0 +1,24 @@
"""Pinpoint's word tables for prep: `workplace_type` and `employment_type`."""
from __future__ import annotations
import importlib
pinpoint = importlib.import_module("app.prep.hosts.pinpoint")
def _match(code: str) -> str | None:
"""The first entry of `MATCH` found in the lowercased code."""
code = code.strip().lower()
return next((value for fragment, value in pinpoint.MATCH if fragment in code), None)
def test_workplace_types_read() -> None:
assert pinpoint.REMOTE == {"onsite": "onsite", "on_site": "onsite", "hybrid": "hybrid",
"remote": "remote"}
def test_employment_types_read_by_their_fragments() -> None:
assert [_match(w) for w in ("permanent_full_time", "fixed_term_full_time",
"permanent_part_time", "internship", "something_new")] == [
"full_time", "temporary", "part_time", "internship", None]
+27
View File
@@ -0,0 +1,27 @@
"""Recruitee's word tables for prep: the host's codes, and what they mean."""
from __future__ import annotations
import importlib
recruitee = importlib.import_module("app.prep.hosts.recruitee")
def _match(word: str) -> str | None:
"""The first entry of `MATCH` found among the words of a code."""
words = set(word.strip().lower().replace("-", "_").split("_"))
return next((value for fragment, value in recruitee.MATCH if fragment in words), None)
def test_experience_codes_read() -> None:
seniority = recruitee.SENIORITY
assert [seniority[w] for w in ("student_school", "entry_level", "mid_level")] == [
"intern", "junior", "mid"]
assert seniority["experienced"] == ""
assert "senior_level" not in seniority
def test_employment_type_codes_read_by_their_words() -> None:
assert [_match(w) for w in ("fulltime_permanent", "parttime_permanent", "internship",
"fulltime_fixed_term", "freelance", "odd_job")] == [
"full_time", "part_time", "internship", "temporary", "contract", None]
+19
View File
@@ -0,0 +1,19 @@
"""Rippling's word table for prep: the words of `employmentType.label`."""
from __future__ import annotations
import importlib
rippling = importlib.import_module("app.prep.hosts.rippling")
def _match(label: str) -> str | None:
"""The first entry of `MATCH` found among the "_"-separated words of a label."""
words = set(label.strip().lower().split("_"))
return next((value for token, value in rippling.MATCH if token in words), None)
def test_employment_labels_read_by_their_words() -> None:
assert [_match(w) for w in ("SALARIED_FT", "HOURLY_PT", "INTERN", "CONTRACTOR",
"SEASONAL", "ODD")] == [
"full_time", "part_time", "internship", "contract", "temporary", None]
+22
View File
@@ -0,0 +1,22 @@
"""SmartRecruiters' word tables for prep: the host's ids, and what they mean."""
from __future__ import annotations
import importlib
smartrecruiters = importlib.import_module("app.prep.hosts.smartrecruiters")
def test_experience_levels_read() -> None:
seniority = smartrecruiters.SENIORITY
assert [seniority[w] for w in ("internship", "entry_level", "director", "executive")] == [
"intern", "junior", "director", "exec"]
# Known, and naming no level.
assert [seniority[w] for w in ("associate", "mid_senior_level", "not_applicable")] == [""] * 3
assert "guru" not in seniority
def test_employment_types_read() -> None:
employment = smartrecruiters.EMPLOYMENT
assert employment["intern"] == "internship" and employment["permanent"] == ""
assert employment["full_time"] == "full_time" and employment["part_time"] == "part_time"
+14
View File
@@ -0,0 +1,14 @@
"""Teamtailor's word table for prep: the feed's `remoteStatus`."""
from __future__ import annotations
import importlib
teamtailor = importlib.import_module("app.prep.hosts.teamtailor")
def test_remote_status_words_read() -> None:
remote = teamtailor.REMOTE
assert [remote[w] for w in ("fully", "hybrid", "onsite")] == ["remote", "hybrid", "onsite"]
# Known, and saying nothing about where the role is worked.
assert remote["none"] == "" and remote["temporary"] == ""
+18
View File
@@ -0,0 +1,18 @@
"""Workable's word tables for prep: the host's own words, lowercased, and what they mean."""
from __future__ import annotations
import importlib
workable = importlib.import_module("app.prep.hosts.workable")
def test_workplace_words_read() -> None:
assert workable.REMOTE == {"remote": "remote", "hybrid": "hybrid", "on_site": "onsite"}
def test_type_words_read() -> None:
assert workable.EMPLOYMENT == {"full": "full_time", "part": "part_time",
"contract": "contract", "temporary": "temporary",
"internship": "internship"}
assert "volunteer" not in workable.EMPLOYMENT # an unknown word is prep's to report
+21
View File
@@ -0,0 +1,21 @@
"""Workday's word tables for prep: the host's own words, lowercased, and what they mean."""
from __future__ import annotations
import importlib
workday = importlib.import_module("app.prep.hosts.workday")
def test_remote_type_words_of_the_tenants_read() -> None:
remote = workday.REMOTE
# Zendesk's "Fully Flexible", Axis' "On-Site", Red Hat's "Onsite".
assert [remote[w] for w in ("remote", "fully flexible", "hybrid", "in office",
"on-site", "onsite")] == [
"remote", "remote", "hybrid", "onsite", "onsite", "onsite"]
assert "no" not in remote # a bare "No" says nothing
def test_time_type_words_read() -> None:
assert workday.EMPLOYMENT == {"full time": "full_time", "part time": "part_time"}
+1 -1
View File
@@ -46,7 +46,7 @@ def test_it_strips_highlights_keeps_the_excerpt_as_teaser_and_drops_predicted_pa
assert "description" not in first
assert first["company"] == "Acme GmbH"
assert first["places"] == [{"text": "Berlin, Deutschland", "remote": None}]
assert first["country"] == "DE" and first["remote"] == "remote"
assert first["country"] == "DE" and "remote" not in first
assert (first["salary_min"], first["salary_max"], first["currency"]) == (70000, 90000, "EUR")
assert (predicted["salary_min"], predicted["currency"]) == (None, None)
+2 -2
View File
@@ -61,12 +61,12 @@ def test_it_reads_title_company_date_and_the_full_text() -> None:
first, _no_town, remote = platsbanken.rows(decoded, "")
assert first["title"] == "Postdoctoral Researcher"
assert first["company"] == "European Spallation Source Eric"
assert first["country"] == "SE" and first["remote"] is None
assert first["country"] == "SE" and first["remote"] == "Arbete på plats"
assert first["url"] == "https://arbetsformedlingen.se/platsbanken/annonser/31457942"
assert first["posted"].date().isoformat() == "2026-09-09"
assert first["description"].startswith("The European Spallation Source")
assert "teaser" not in first
assert (remote["remote"], remote["company"], remote["posted"]) == ("remote", "", None)
assert (remote["remote"], remote["company"], remote["posted"]) == (None, "", None)
def test_it_sends_the_region_the_offset_and_no_region_for_all_of_sweden() -> None:
+32
View File
@@ -2,6 +2,7 @@
from __future__ import annotations
import importlib
import logging
import sys
from pathlib import Path
@@ -24,6 +25,16 @@ from plugins.readers._shared import ( # noqa: E402
#: The module-level names a reader keeps its employment words in.
_TABLES = ("_EMPLOYMENT", "_CONTRACT", "_SCHEDULE")
#: The hosts whose word tables are prep's, in `app/prep/hosts/<name>.py`.
_PREP_HOSTS = ("oracle", "personio", "phenom", "pinpoint", "recruitee", "rippling",
"smartrecruiters", "teamtailor", "workable", "workday")
def _prep_table(name: str, table_name: str):
"""A word table of `app.prep.hosts.<name>`, or None where it has none."""
module = importlib.import_module(f"app.prep.hosts.{name}")
return getattr(module, table_name, None)
def test_every_employment_table_maps_into_the_shared_vocabulary() -> None:
found = 0
@@ -36,9 +47,30 @@ def test_every_employment_table_maps_into_the_shared_vocabulary() -> None:
values = table.values() if isinstance(table, dict) else [v for _k, v in table]
# "" is a known word that names no employment ("permanent").
assert set(values) <= {*EMPLOYMENTS, ""}, (name, table_name)
for name in _PREP_HOSTS:
for table_name in ("EMPLOYMENT", "MATCH"):
table = _prep_table(name, table_name)
if table is None:
continue
values = table.values() if isinstance(table, dict) else [v for _k, v in table]
found += 1
assert set(values) <= {*EMPLOYMENTS, ""}, (name, table_name)
assert found >= 15
def test_the_prep_tables_of_the_hosts_speak_the_apps_vocabulary() -> None:
from app.models import SENIORITIES
for name in _PREP_HOSTS:
seniority = _prep_table(name, "SENIORITY")
if seniority is not None:
assert set(seniority.values()) <= {*SENIORITIES, ""}, name
remote = _prep_table(name, "REMOTE")
if remote is not None:
assert set(remote.values()) <= {"remote", "hybrid", "onsite", ""}, name
for table in (seniority, remote, _prep_table(name, "EMPLOYMENT")):
assert table is None or all(k == k.lower() for k in table), name
def test_the_vocabulary_is_the_documented_one() -> None:
assert EMPLOYMENTS == ("full_time", "part_time", "contract", "temporary", "internship",
"freelance")