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>
165 lines
7.3 KiB
Python
165 lines
7.3 KiB
Python
"""The Deel host module against a saved page (trimmed from Klarna's board,
|
|
2026-09-29), and the fetch around it. No network."""
|
|
|
|
from __future__ import annotations
|
|
|
|
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, readers # noqa: E402
|
|
from plugins.readers import Ask, deel # noqa: E402
|
|
|
|
PAGE = (ROOT / "tests" / "fixtures" / "ats" / "deel.html").read_bytes()
|
|
URL = "https://jobs.deel.com/klarna"
|
|
|
|
|
|
def test_rows_come_from_the_payload_with_descriptions_resolved() -> None:
|
|
rows = deel.parse(PAGE, "klarna")
|
|
assert [r["source_id"][:8] for r in rows] == [
|
|
"21584046", "a66b42b6", "f96bb9d4", "a8be45ac", "a6649d86"]
|
|
first = rows[0]
|
|
assert first["title"] == "Incident Manager"
|
|
assert first["url"] == ("https://jobs.deel.com/klarna/job-details/"
|
|
"21584046-56d5-44c4-8b18-7e15dc659827/overview")
|
|
assert first["location"] == "Stockholm" and first["department"] == "Engineering"
|
|
assert first["remote"] == "ON_SITE"
|
|
assert first["description"].startswith("## Klarna, briefly\nAt Klarna")
|
|
assert "<" not in first["description"] and "$" not in first["description"][:3]
|
|
assert all(r["description"] for r in rows)
|
|
# The day the posting was published, not the day it was drafted (12:27).
|
|
assert first["posted"] is not None and first["posted"].isoformat() == \
|
|
"2026-09-07T12:33:38.484000+00:00"
|
|
|
|
|
|
def test_several_locations_are_joined_and_remote_is_the_hosts_word() -> None:
|
|
rows = {r["source_id"][:8]: r for r in deel.parse(PAGE, "klarna")}
|
|
assert rows["a8be45ac"]["location"] == "Stockholm; London; Milan; Warsaw"
|
|
assert rows["f96bb9d4"]["remote"] == "REMOTE"
|
|
assert rows["f96bb9d4"]["location"] == "Ohio"
|
|
# Every place is kept, in the host's order; the work arrangement is the
|
|
# places' remote flag.
|
|
assert rows["a8be45ac"]["places"] == [
|
|
{"text": name, "remote": False} for name in ("Stockholm", "London", "Milan", "Warsaw")]
|
|
assert rows["f96bb9d4"]["places"] == [{"text": "Ohio", "remote": True}]
|
|
assert rows["f96bb9d4"]["employment"] == "Consultant"
|
|
assert rows["21584046"]["employment"] == "Full-time"
|
|
|
|
|
|
def test_pay_is_only_text_since_the_period_is_not_stated() -> None:
|
|
rows = {r["source_id"][:8]: r for r in deel.parse(PAGE, "klarna")}
|
|
assert rows["21584046"]["salary_text"] == "SEK 775,398 to 1,108,832"
|
|
assert rows["a66b42b6"]["salary_text"] == "GBP from 105,197"
|
|
assert rows["f96bb9d4"]["salary_text"] == "" # a role with no range
|
|
assert all("salary_min" not in r for r in rows.values())
|
|
|
|
|
|
def _page_with(body: dict[str, Any]) -> bytes:
|
|
"""A page whose one payload row holds `body`, and a description row of
|
|
multibyte text whose length is given in bytes, as the stream does."""
|
|
text = "Zo\u00eb\u2019s caf\u00e9 \u2014 100 \u20ac" # 2, 3 and 1 bytes wide
|
|
row = f"5:T{len(text.encode()):x},{text}" + \
|
|
'6:["$","div",null,' + json.dumps(body, ensure_ascii=False, separators=(",", ":")) + "]\n"
|
|
return f'<script>self.__next_f.push([1,{json.dumps(row, ensure_ascii=False)}])</script>'.encode()
|
|
|
|
|
|
def test_a_text_row_is_read_by_its_length_in_bytes() -> None:
|
|
body = {"jobPostings": [{"id": "a", "title": "Baker", "richtextDescription": "$5",
|
|
"job": {}}, {"id": "b", "title": "Cook",
|
|
"richtextDescription": "<p>Inline & done</p>",
|
|
"job": {}}]}
|
|
rows = deel.parse(_page_with(body), "acme")
|
|
assert [r["description"] for r in rows] == ["Zo\u00eb\u2019s caf\u00e9 \u2014 100 \u20ac",
|
|
"Inline & done"]
|
|
assert rows[0]["remote"] is None and rows[0]["location"] == "" and rows[0]["posted"] is None
|
|
|
|
|
|
def test_a_board_without_roles_is_an_empty_board_but_another_page_is_not_a_board() -> None:
|
|
assert deel.parse(_page_with({"jobPostings": []}), "acme") == []
|
|
with pytest.raises(ValueError, match="not a Deel job board"):
|
|
deel.parse(b"<html><title>Job Board Not Found</title></html>", "nobody")
|
|
|
|
|
|
def test_a_posting_without_a_title_or_id_is_skipped() -> None:
|
|
body = {"jobPostings": [{"id": "a", "title": " ", "job": {}}, {"title": "x", "job": {}},
|
|
{"id": "c", "title": "Cook", "job": {}}]}
|
|
assert [r["source_id"] for r in deel.parse(_page_with(body), "acme")] == ["c"]
|
|
|
|
|
|
# ------------------------------------------------------------ the fetch
|
|
|
|
def test_the_board_is_one_request() -> None:
|
|
asked: list[str] = []
|
|
|
|
def get(url: str, **kwargs: Any) -> bytes:
|
|
asked.append(url)
|
|
return PAGE
|
|
|
|
rows = deel.parse(get(deel.feed_url("klarna", Ask())), "klarna")
|
|
assert len(rows) == 5 and asked == [URL]
|
|
|
|
|
|
def test_it_declares_a_whole_listing_read_from_the_page() -> None:
|
|
reads = deel.READS
|
|
assert (reads.listing, reads.text, reads.places, reads.complete) == (
|
|
"whole", "listing", "many", "whole")
|
|
assert reads.fields.pay == "line" and "/api/ is disallowed and not read" in reads.robots
|
|
|
|
|
|
# ------------------------------------------------------- slugs and hosts
|
|
|
|
def test_the_module_is_registered_as_a_feed_host() -> None:
|
|
assert readers.HOSTS["deel"] is deel and ats.HOSTS["deel"] == "Deel"
|
|
assert deel.NAME == "deel" and deel.KEEP_CASE
|
|
assert not hasattr(deel, "page") and not hasattr(deel, "detail")
|
|
|
|
|
|
def test_the_urls_are_the_boards_own() -> None:
|
|
assert deel.careers_url("klarna") == URL
|
|
assert deel.feed_url("klarna", Ask()) == URL
|
|
assert deel.allowed("klarna")("jobs.deel.com") and deel.allowed("klarna")("JOBS.DEEL.COM")
|
|
for host in ("deel.com", "api.deel.com", "jobs.deel.com.evil.example", "evil.example"):
|
|
assert not deel.allowed("klarna")(host)
|
|
|
|
|
|
@pytest.mark.parametrize("slug", ["klarna", "Klarna", "acme-2", "a_b", "9lives", "localhost"])
|
|
def test_a_slug_is_one_path_segment(slug: str) -> None:
|
|
"""`localhost` is only a path here: the host of every URL is fixed."""
|
|
assert ats.is_known_target(f"deel:{slug}")
|
|
|
|
|
|
@pytest.mark.parametrize("slug", [
|
|
"", "a@b", "a#b", "a?b", "..", ".", "a%2fb", "a/b", "a:443", "a b", "acme\n", " acme",
|
|
"127.0.0.1", "a.b", "_next", "-x", "a\\b", "évil",
|
|
])
|
|
def test_a_hostile_slug_is_refused(slug: str) -> None:
|
|
assert not deel.slug_ok(slug)
|
|
assert not ats.is_known_target(f"deel:{slug}")
|
|
|
|
|
|
def test_no_slug_reaches_another_host() -> None:
|
|
"""The host is fixed in every URL the module builds; a slug only ever
|
|
lands in the path."""
|
|
for slug in ("klarna", "Klarna", "a-b_c"):
|
|
for url in (deel.careers_url(slug), deel.feed_url(slug, Ask())):
|
|
assert httpx.URL(url).host == "jobs.deel.com"
|
|
|
|
|
|
def test_discovery_finds_the_slug_in_a_career_site_url() -> None:
|
|
page = ('<a href="https://jobs.deel.com/job-boards/klarna">All open positions</a>'
|
|
'<a href="https://jobs.deel.com/acme/job-details/1/overview">Role</a>'
|
|
'<a href="https://jobs.deel.com/Beta">Beta</a>'
|
|
'<script src="https://jobs.deel.com/_next/static/x.js"></script>'
|
|
'<a href="https://jobs.deel.com/job-boards">List</a>')
|
|
assert [(h, s) for h, s, _ in boards_in(page)] == [
|
|
("deel", "klarna"), ("deel", "acme"), ("deel", "Beta")]
|