Read the text of SmartRecruiters and Workable roles through detail, keeping Workable's shortcode in the row

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
WolfandClaude Opus 5.5 committed 2026-09-30 13:23:58 +02:00
1 parent a19c528e3f
commit 70aa4b3117
6 files changed
+176 -11

No files matched your search

+39 -4
View File
@@ -13,16 +13,26 @@ A fetch term goes as `?q=engineer`, one a request (`Filters(query="one")`):
it matches title and text, and with `country=` it is an AND (survey,
2026-09-30: `q=engineer` answered 1,713 of Bosch's 4,830, with `country=dk`
none). How several words in one `q` combine is untested.
The list has no text. The text is read per posting by `detail`, `GET
/v1/companies/{slug}/postings/{id}` (checked 2026-09-30, Delivery Hero's
744000151349134): `jobAd.sections` holds `companyDescription`,
`jobDescription`, `qualifications` and `additionalInformation`, each with a
`title` and its `text` as HTML. The description joins the last three; the
company blurb is the same on every role of the board and is left out.
"""
from __future__ import annotations
import html
import json
import re
from collections.abc import Callable
from typing import Any
from urllib.parse import quote
from desk.sources import strip_html
from ._contract import Ask, Fields, Filters, Place, Reads
from ._shared import Get, Page, Row, only, parse_dt, path_slug_ok
@@ -35,7 +45,7 @@ PATTERNS = (
#: An unknown identifier answers an empty list, which is `empty="suspect"`.
READS = Reads(
scope="board", listing="paged", page_size=100, text="none", complete="stated",
scope="board", listing="paged", page_size=100, text="detail", complete="stated",
places="one", empty="suspect", filters=Filters(country="one", query="one"),
fields=Fields(posted="trusted", remote="stated", country="stated", seniority="stated",
employment="stated"),
@@ -96,9 +106,9 @@ def rows(data: dict[str, Any], slug: str) -> list[Row]:
"""One page of https://api.smartrecruiters.com/v1/companies/{slug}/postings
-> `{"offset", "limit", "totalFound", "content": [...]}`. At most 100 a
page, whatever `limit` asks for; the next is `?offset=` further on. The
list omits the job description (only the per-posting detail endpoint has
it, and fetching one per row would be hundreds of extra requests per
board), so `description` stays empty here. A posting has one location,
list omits the job description (only the per-posting endpoint has it,
which `detail` reads for the roles worth it), so `description` stays
empty here. A posting has one location,
its `country` a lower-case ISO code, `experienceLevel` and
`typeOfEmployment` (an `id` and a `label` each)."""
out = []
@@ -130,3 +140,28 @@ def rows(data: dict[str, Any], slug: str) -> list[Row]:
"employment": _word(j, "typeOfEmployment"),
})
return out
#: The sections of a posting's ad that make its description, in reading order.
#: `companyDescription` is the employer's blurb, alike on every role.
_SECTIONS = ("jobDescription", "qualifications", "additionalInformation")
def detail(get: Get, slug: str, row: Row) -> dict[str, Any]:
"""`/v1/companies/{slug}/postings/{id}` -> `jobAd.sections`: the HTML of
`jobDescription`, `qualifications` and `additionalInformation`, one
text. An answer without `jobAd.sections` is
not a posting and raises, so the role is asked again on a later pull."""
posting = quote(str(row["source_id"]), safe="")
data = json.loads(get(f"https://{_API}/v1/companies/{slug}/postings/{posting}"))
sections = ((data.get("jobAd") or {}).get("sections")
if isinstance(data, dict) else None)
if not isinstance(sections, dict):
raise ValueError("not a SmartRecruiters posting: no jobAd.sections")
parts = []
for key in _SECTIONS:
text = " ".join(html.unescape(strip_html(str((sections.get(key) or {}).get("text") or "")))
.split())
if text:
parts.append(text)
return {"description": " ".join(parts)}
+38 -3
View File
@@ -3,14 +3,25 @@ total.
Written from one real board, Hugging Face's, on 2026-09-28, when the first
board with roles was seen; jobhunt-hq did not read Workable.
The list carries no text. The text is read per role by `detail`, `GET
/api/v2/accounts/{slug}/jobs/{shortcode}` (checked 2026-09-30, Hugging Face's
DB4D7C0EC8): the list's fields again, and `description`, `requirements` and
`benefits` as HTML. The list gives the numeric `id` and the `shortcode`; the
role's `source_id` is the number, so the row keeps the `shortcode` for this
request. The API answers 429 to a run of requests; the runner spaces the
reads and stops on one.
"""
from __future__ import annotations
import html
import json
from collections.abc import Callable
from typing import Any
from desk.sources import strip_html
from ._contract import Ask, Fields, Place, Reads
from ._shared import Get, Page, Row, add_place, only, parse_dt, path_slug_ok
@@ -23,7 +34,7 @@ PATTERNS = (
)
READS = Reads(
scope="board", listing="paged", page_size=10, text="none", complete="stated",
scope="board", listing="paged", page_size=10, text="detail", complete="stated",
places="many",
fields=Fields(posted="trusted", remote="stated", employment="stated"),
robots="api: no robots.txt check needed (public board API), 2026-09-30",
@@ -82,8 +93,8 @@ def rows(data: dict[str, Any], slug: str) -> list[Row]:
-> `{"total": int, "results": [...], "nextPage": token}`, as Hugging
Face's board answered on 2026-09-28. A result carries no URL and no text:
the job page is `https://apply.workable.com/{slug}/j/{shortcode}/`, and
the description is only on the per-job endpoint, one request per role,
so `description` stays empty here as it does for SmartRecruiters.
the description is only on the per-job endpoint, which `detail` reads, so
`description` stays empty here. The `shortcode` is kept in the row for it.
`location` is the first place and `locations` all of them; `department`
is a list; `remote` is a flag beside `workplace` (remote, hybrid,
on_site); `type` is the hours or the contract (full, part, ...)."""
@@ -101,6 +112,7 @@ def rows(data: dict[str, Any], slug: str) -> list[Row]:
dept = ", ".join(d for d in dept if d)
out.append({
"source_id": str(j.get("id") or shortcode),
"shortcode": shortcode,
"title": title,
"places": places,
"location": "; ".join(p["text"] for p in places),
@@ -113,3 +125,26 @@ def rows(data: dict[str, Any], slug: str) -> list[Row]:
"employment": str(j.get("type") or "").strip() or None,
})
return out
#: The parts of a role's ad that make its description, in reading order.
_PARTS = ("description", "requirements", "benefits")
def detail(get: Get, slug: str, row: Row) -> dict[str, Any]:
"""`/api/v2/accounts/{slug}/jobs/{shortcode}` -> the HTML of
`description`, `requirements` and `benefits`, joined as one text. An
answer with none of the three keys is not a role and raises, so the role
is asked again on a later pull."""
shortcode = str(row.get("shortcode") or "")
if not path_slug_ok(shortcode):
raise ValueError(f"no Workable shortcode in the row: {shortcode!r}")
data = json.loads(get(f"https://{_API}/api/v2/accounts/{slug}/jobs/{shortcode}"))
if not isinstance(data, dict) or not any(k in data for k in _PARTS):
raise ValueError("not a Workable role: no description")
parts = []
for key in _PARTS:
text = " ".join(html.unescape(strip_html(str(data.get(key) or ""))).split())
if text:
parts.append(text)
return {"description": " ".join(parts)}