Read big boards within the agents' reach: countries and remote from the enabled agents, Europe for favourites, filtered at the source by six hosts

app/jobs/reach.py derives each board's reach once per pull and the pull hands it to the runner through the batch payload. SmartRecruiters, Eightfold and SuccessFactors filter one country a request, Phenom a list, Workday and Oracle by host ids from facets. Remote beside countries is a series of its own, since two facets intersect.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
WolfandClaude Opus 5.5 committed 2026-09-30 09:55:18 +02:00
1 parent 3adafd8950
commit 3462a4abe7
17 files changed
+1324 -116

No files matched your search

+52 -3
View File
@@ -8,6 +8,18 @@ number is `CX_1001` or `CX_1` on the tenants read so far, and the site's own
name (`SinchCareer`) is taken in its place. Written from live boards on
2026-09-29 (Milestone Systems, Nokia, Danske Bank, Sinch); jobhunt-hq did not
read Oracle.
Countries are filtered on the server by `selectedLocationsFacet`, the pod's
geography ids (`Filters(country="ids")`), which `facets` reads off
`locationsFacet`; several ids go in one request, joined with `;`, the way the
site's own search sends a multiple choice. That list holds only the pod's ten
biggest countries (fa-evmr's, 2026-09-30), so when it is full a country
missing from it may still have roles: `facets` then maps every other country
to `*`, and a series asking for one is read without a country filter. An id
the pod does not know answers the whole board (review). Remote is
`selectedWorkplaceTypesFacet=ORA_REMOTE` (`remote="one"`); two facets are an
intersection (Finland and remote answered 0 on 2026-09-30), which is why the
runner asks for remote in a series of its own.
"""
from __future__ import annotations
@@ -21,7 +33,8 @@ from urllib.parse import quote
from desk.sources import strip_html
from ._contract import Ask, Fields, Place, Reads
from ._contract import Ask, Fields, Filters, Place, Reads
from ._countries import NAMES, code_of
from ._shared import Get, Page, Row, only, parse_dt
NAME = "oracle"
@@ -41,7 +54,7 @@ PATTERNS = (
READS = Reads(
scope="board", listing="paged", page_size=200, text="detail", teaser=True,
complete="stated", places="many",
complete="stated", places="many", filters=Filters(country="ids", remote="one"),
fields=Fields(posted="trusted", remote="stated", country="stated"),
robots=("read: robots.txt allows /hcmUI/CandidateExperience/en/sites/<site>/jobs "
"(checked 2026-09-29)"),
@@ -103,7 +116,7 @@ def page(get: Get, slug: str, cursor: Any, ask: Ask = Ask()) -> Page:
data = json.loads(get(
f"{_api(slug)}/recruitingCEJobRequisitions?onlyData=true"
f"&expand=requisitionList.secondaryLocations"
f"&finder=findReqs;siteNumber={site},limit={PAGE_SIZE},offset={offset}"))
f"&finder=findReqs;siteNumber={site},{_selected(ask)}limit={PAGE_SIZE},offset={offset}"))
search = (data.get("items") or [{}])[0]
roles = search.get("requisitionList") or []
# `_paged` counts `id`, which Oracle spells `Id`.
@@ -111,6 +124,42 @@ def page(get: Get, slug: str, cursor: Any, ask: Ask = Ask()) -> Page:
return search, results, search.get("TotalJobsCount"), offset + len(roles) if roles else None
#: How many countries `locationsFacet` lists at most; a list this long may
#: leave some out.
FACET_LIMIT = 10
def _selected(ask: Ask) -> str:
"""The finder's facet parameters for `ask`, each ending in a comma: the
location ids (none when one is `*`, a country the pod did not list) and
remote."""
out = ""
ids = [i for i in ask.countries if re.fullmatch(r"\d+", i)]
if ids and "*" not in ask.countries:
out += f"selectedLocationsFacet={'%3B'.join(ids)},"
if ask.remote:
out += "selectedWorkplaceTypesFacet=ORA_REMOTE,"
return out
def facets(get: Get, slug: str) -> dict[str, dict[str, str]]:
"""The pod's geography id of each country `locationsFacet` names, read
with the facets and no roles. When the list is `FACET_LIMIT` long, every
country it leaves out maps to `*`: its roles may be there, and a read
for it goes unfiltered."""
site = _parts(slug)[1]
data = json.loads(get(
f"{_api(slug)}/recruitingCEJobRequisitions?onlyData=true"
f"&finder=findReqs;siteNumber={site},facetsList=LOCATIONS,limit=1,offset=0"))
listed = (data.get("items") or [{}])[0].get("locationsFacet") or []
country = {code: str(entry["Id"]) for entry in listed
if isinstance(entry, dict) and entry.get("Id")
and (code := code_of(str(entry.get("Name") or "")))}
if len(listed) >= FACET_LIMIT:
country = dict.fromkeys(NAMES, "*") | country
return {"country": country}
def parse(body: bytes, slug: str) -> list[Row]:
return rows((json.loads(body).get("items") or [{}])[0], slug)