Seed 186 companies, add roledesk companies, and pull and close every board

seed/companies.json: 179 companies from jobhunt-hq's seed.json (MIT),
reduced to key/name/board/careers_url, plus a hand-added DACH set.
`roledesk companies seed|add|check` import it, add one company from its
board with a live check first, and report dead boards.

roledesk/jobs/pull.py is now a pipeline task: ten boards per pull_batch job
on the http resource, upserting every role the ats plugin yields. Its final
job's settle closes every board this run answered in full and no longer
lists a role from, reading the plugin's own self.state stamps rather than
each batch job's return value, so one bad board in a batch does not cost the
other nine their closure. `roledesk job pull` runs it, and it registers
itself in pipeline.TASKS so <desk-jobs> can start it too.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
This commit is contained in:
WolfandClaude Sonnet 5 committed 2026-09-23 11:27:30 +02:00
1 parent 21ed472a87
commit 0f59452b2a
5 files changed
+1684 -10

No files matched your search

+141 -4
View File
@@ -1,4 +1,4 @@
"""`roledesk serve | job pull | db init`."""
"""`roledesk serve | job pull | companies seed|add|check | db init`."""
from __future__ import annotations
@@ -6,9 +6,24 @@ import argparse
import json
import logging
import sys
from pathlib import Path
from typing import TYPE_CHECKING
from roledesk import config as config_mod
if TYPE_CHECKING:
from roledesk.storage import Store
#: The one plugin this app ships lives outside the `roledesk` package, in
#: `plugins/`, so it is discovered by `desk.sources.registry` the same way a
#: user's own drop-in plugin would be. `companies` and `job pull` both want
#: its plain functions (`is_known_target`, `careers_url`, `fetch_board`) as
#: well as the `Source` class, so the repo root goes on `sys.path` once, here,
#: rather than each caller guessing where `plugins/` is relative to `cwd`.
_ROOT = Path(__file__).resolve().parents[1]
if str(_ROOT) not in sys.path:
sys.path.insert(0, str(_ROOT))
def build_parser() -> argparse.ArgumentParser:
parser = argparse.ArgumentParser(prog="roledesk",
@@ -21,9 +36,19 @@ def build_parser() -> argparse.ArgumentParser:
serve.add_argument("--host", default="127.0.0.1")
serve.add_argument("--port", type=int, default=8790)
job = sub.add_parser("job", help="queue a job and drain the queue once")
job = sub.add_parser("job", help="queue a job or task and drain the queue once")
job.add_argument("kind", choices=["pull"])
companies = sub.add_parser("companies", help="the target company list")
companies_sub = companies.add_subparsers(dest="action", required=True)
seed = companies_sub.add_parser(
"seed", help="import seed/companies.json, subscribing to each board")
seed.add_argument("--path", default="seed/companies.json")
add = companies_sub.add_parser("add", help="add one company from its board")
add.add_argument("board", help="host:slug, e.g. greenhouse:acme")
add.add_argument("name")
companies_sub.add_parser("check", help="report boards that do not answer")
db = sub.add_parser("db", help="database housekeeping")
db.add_argument("action", choices=["init"])
return parser
@@ -40,6 +65,8 @@ def main(argv: list[str] | None = None) -> int:
return serve(args.host, args.port, args.config)
if args.command == "job":
return run_job(args)
if args.command == "companies":
return run_companies(args)
return init_db(args)
except FileNotFoundError as exc:
print(exc, file=sys.stderr)
@@ -47,8 +74,11 @@ def main(argv: list[str] | None = None) -> int:
def run_job(args: argparse.Namespace) -> int:
"""Queue `kind`, then run everything queued until the queue is idle."""
from pipeline import JobQueue, Jobs
"""Queue `kind` — a job, or a task from `pipeline.TASKS` such as `pull` —
then run everything queued until the queue is idle, and print how it
ended. A task's own result is its final job's, which is the one whose
`task_id` equals the run this started."""
from pipeline import FINAL, JobQueue, Jobs, TASKS
from roledesk import jobs
from roledesk.storage import Store
@@ -57,6 +87,15 @@ def run_job(args: argparse.Namespace) -> int:
queue = JobQueue(Jobs(config.database), config, store,
workers=config.workers.model_dump(),
bootstrap=jobs.BOOTSTRAP, bootstrap_arg=config.path)
if args.kind in TASKS:
task_id = queue.start_task(TASKS[args.kind]())
queue.drain()
final = next((j for j in queue.jobs.of_task(task_id) if j.kind == FINAL), None)
if final is not None:
print(json.dumps({"id": final.id, "state": final.state,
"result": final.result, "error": final.error}))
return 0 if final.state == "done" else 1
return 0
queued = queue.enqueue(args.kind)
queue.drain()
if queued is not None:
@@ -68,6 +107,104 @@ def run_job(args: argparse.Namespace) -> int:
return 0
def run_companies(args: argparse.Namespace) -> int:
from roledesk.storage import Store
config = config_mod.load(args.config)
store = Store(config.database)
if args.action == "seed":
print(json.dumps(seed_companies(store, args.path)))
return 0
if args.action == "add":
return add_company(store, args.board, args.name)
return check_companies(store)
def seed_companies(store: Store, path: str) -> dict[str, int]:
"""Import `seed/companies.json`: one `companies` row and one `ats`
subscription per entry, skipping a key that already exists. Not
idempotent on edits to an existing row — re-running it only ever adds
what is missing."""
from plugins.ats import is_known_target
data = json.loads(Path(path).read_text(encoding="utf-8"))
added = skipped = 0
for row in data.get("companies", []):
key, name, board = row["key"], row["name"], row["board"]
if store.company(key) is not None:
skipped += 1
continue
if not is_known_target(board):
print(f"companies seed: {key}: {board!r} is not a known board; skipped",
file=sys.stderr)
skipped += 1
continue
store.add_company(name, key=key, affinity=int(row.get("affinity", 0)),
careers_url=row.get("careers_url", ""))
host, _, _slug = board.partition(":")
try:
store.add_subscription("ats", board, label=name, company=name, board=host)
except ValueError as exc:
# The company row landed; the subscription is the only thing
# missing, and worth saying rather than losing silently.
print(f"companies seed: {key}: {exc}", file=sys.stderr)
added += 1
return {"added": added, "skipped": skipped, "total": len(data.get("companies", []))}
def add_company(store: Store, board: str, name: str) -> int:
"""`roledesk companies add greenhouse:acme "Acme"`: fetch the board first
and refuse if it does not answer, is not a known host, or the company
already exists. Port of seedgen's `-add`."""
from plugins.ats import careers_url as careers_url_of, fetch_board, is_known_target
from roledesk.models import company_key
if not is_known_target(board):
print(f"companies add: {board!r} is not host:slug for a known ATS host",
file=sys.stderr)
return 1
key = company_key(name)
if store.company(key) is not None:
print(f"companies add: {key!r} already exists", file=sys.stderr)
return 1
try:
rows = fetch_board(board)
except Exception as exc:
print(f"companies add: {board} did not answer: {exc}", file=sys.stderr)
return 1
if not rows:
print(f"companies add: {board} answered but lists no roles; nothing added",
file=sys.stderr)
return 1
host, _, _slug = board.partition(":")
store.add_company(name, key=key, careers_url=careers_url_of(board))
try:
store.add_subscription("ats", board, label=name, company=name, board=host)
except ValueError as exc:
print(f"companies add: {exc}", file=sys.stderr)
return 1
print(json.dumps({"key": key, "board": board, "roles": len(rows)}))
return 0
def check_companies(store: Store) -> int:
"""`roledesk companies check`: every `ats` board, fetched once. Prints a
line per board that did not answer and exits 1 if any did not."""
from plugins.ats import fetch_board
subs = store.subscriptions(source="ats")
dead = []
for sub in subs:
try:
fetch_board(sub.target)
except Exception as exc:
dead.append((sub.target, sub.label, str(exc)))
for target, label, reason in dead:
print(f"{target} ({label}): {reason}")
print(f"checked {len(subs)} boards, {len(dead)} dead")
return 1 if dead else 0
def init_db(args: argparse.Namespace) -> int:
"""Create the file, or bring it to the current schema."""
from roledesk.storage import LATEST, Store