Files
transcription/tests/services/test_job_service.py
T
zoltan57andCopilot App 7b9715b3f1 V4.6 Phase 3: worker and provider reliability
Claim jobs atomically [CRIT-01]
- Replace JobService.read_next_queued_job with claim_next_queued_job, which
  selects and transitions QUEUED -> PROCESSING inside one transaction. The old
  read-then-write sequence left a window in which two workers could observe the
  same QUEUED row.
- Add the missing .limit(1). The poll previously ordered the entire queued set
  and discarded all but the first row.
- Drop the eager loads from the hot poll entirely. They were pure waste:
  process_queued_job immediately re-reads the job through read_job with the
  relationships it actually needs.
- Guard the row with with_for_update(skip_locked=True) on PostgreSQL so the
  claim stays correct once more than one worker exists. On SQLite the claim is a
  bounded single-writer transaction.
- Correct the comment at the remaining direct-call claim site, which described
  the hazard rather than the guarantee.

Reuse the provider connection [HIGH-02]
- Build the ServiceBundle once per worker loop instead of once per job, and
  close it at loop shutdown. Every job previously constructed a new
  SourceService, and with it a new provider adapter and a new httpx.AsyncClient,
  paying a full TLS handshake per page and discarding the connection pool.
- process_next_queued_job now accepts an optional caller-owned bundle and only
  closes bundles it created itself.

Uncap the provider timeout [HIGH-03]
- Remove le=20.0 from worker_provider_timeout_seconds. The cap equalled the
  default, so the ceiling could never be raised, and dense-page vision
  transcription routinely needs longer. Default raised to 180s.
- Pass an explicit httpx.Timeout to the OpenRouter AsyncClient. httpx defaults
  every phase to 5 seconds, so the real read budget was 5s regardless of the
  configured value; the outer asyncio.wait_for could never be the binding
  constraint. Connect stays at 10s.

Tighten the provider boundary [MED-03]
- Declare model, current_request_manifest, current_transport_evidence, and
  aclose on the TranscriptionProvider Protocol.
- Delete the per-call inspect.signature(adapter.transcribe).parameters
  reflection and the untyped kwargs dict it fed. The Protocol had declared
  requested_model all along, so the reflection was dead defensive weight on the
  hot path.
- Replace the three getattr probes for aclose and the evidence attributes with
  direct typed access.

Deduplicate bundle construction [MED-06]
- Add ServiceBundle.from_session_factory and ServiceBundle.aclose, replacing the
  duplicated four-service instantiation blocks in app.py and worker.py.
- _recover_stale_processing_jobs now uses the bundle built moments earlier
  instead of constructing a second JobService.

Tests
- Claiming returns the oldest job, marks it PROCESSING, never hands the same job
  out twice, and emits exactly one unadorned SELECT carrying LIMIT and no JOIN.
- The worker loop threads one bundle through consecutive jobs and closes it once
  at shutdown; a caller-owned bundle is left open.
- Settings accepts a timeout above 20 seconds and still rejects zero.
- The OpenRouter client's read, write, and pool timeouts track the configured
  budget rather than the httpx default.

Note: .env in this checkout still pins WORKER_PROVIDER_TIMEOUT_SECONDS=20 and
should be raised to pick up this fix.

Verified: 268 passed, 4 skipped; ruff check clean.

Co-authored-by: Copilot App <[email protected]>
2026-08-17 16:26:31 -05:00

526 lines
19 KiB
Python

from datetime import UTC
from datetime import datetime
from datetime import timedelta
from uuid import uuid4
import pytest
from sqlalchemy import event
from transcription.db.models import Document
from transcription.db.models import Job
from transcription.db.models import JobSource
from transcription.db.models import JobSourceStatus
from transcription.db.models import JobStatus
from transcription.db.models import Source
from transcription.services.documents import DocumentService
from transcription.services.jobs import JobCancelBlockedError
from transcription.services.jobs import JobDeleteBlockedError
from transcription.services.jobs import JobNotFoundError
from transcription.services.jobs import JobResubmitBlockedError
from transcription.services.jobs import JobService
from transcription.services.sources import SourceService
class TestJobService:
@pytest.mark.asyncio
async def test_create_and_read_job(self, job_service: JobService, document_service: DocumentService):
document = Document(id=uuid4(), name="test-bundle")
await document_service.create_document(document=document)
job = Job(document_id=document.id)
await job_service.create_job(job=job)
fetched = await job_service.read_job(job_id=job.id)
assert fetched.id == job.id
assert fetched.document is not None
assert fetched.document.id == document.id
@pytest.mark.asyncio
async def test_update_job_state_updates_status_and_retry(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="test-bundle")
await document_service.create_document(document=document)
job = Job(document_id=document.id)
await job_service.create_job(job=job)
updated = await job_service.update_job_state(
job_id=job.id,
status=JobStatus.PROCESSING,
retry_count_increment=1,
)
assert updated.status == JobStatus.PROCESSING
assert updated.retry_count == 1
@pytest.mark.asyncio
async def test_query_jobs_by_status(self, job_service: JobService, document_service: DocumentService):
document = Document(id=uuid4(), name="query-doc")
await document_service.create_document(document=document)
await job_service.create_job(job=Job(document_id=document.id, status=JobStatus.PROCESSING))
await job_service.create_job(job=Job(document_id=document.id, status=JobStatus.QUEUED))
result = await job_service.query_jobs(status=JobStatus.PROCESSING)
assert len(result) == 1
assert result[0].status == JobStatus.PROCESSING
@pytest.mark.asyncio
async def test_query_jobs_by_source_filename(self, job_service: JobService, document_service: DocumentService):
document = Document(id=uuid4(), name="source-doc")
await document_service.create_document(document=document)
job = Job(document_id=document.id)
await job_service.create_job(job=job)
async with job_service._session_scope() as session:
source = Source(
document_id=document.id,
page_number=1,
upload_name="letter.jpg",
filename="stored-letter.jpg",
file_path="/uploads/stored-letter.jpg",
file_hash="a" * 64,
file_size_bytes=1,
)
session.add(source)
await session.flush()
session.add(
JobSource(
job_id=job.id,
source_id=source.id,
status=JobSourceStatus.PENDING,
)
)
await session.commit()
result = await job_service.query_jobs(filename="stored-letter.jpg")
assert len(result) == 1
assert result[0].id == job.id
@pytest.mark.asyncio
async def test_claim_next_queued_job_claims_oldest_and_marks_processing(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="ordered-doc")
await document_service.create_document(document=document)
created_at = datetime.now(UTC)
first = Job(document_id=document.id, status=JobStatus.QUEUED, date_created=created_at)
second = Job(
document_id=document.id,
status=JobStatus.QUEUED,
date_created=created_at + timedelta(microseconds=1),
)
await job_service.create_job(job=first)
await job_service.create_job(job=second)
claimed = await job_service.claim_next_queued_job()
assert claimed is not None
assert claimed.id == first.id
assert claimed.status == JobStatus.PROCESSING
# The claim is exclusive: the same job is never handed out twice.
next_claim = await job_service.claim_next_queued_job()
assert next_claim is not None
assert next_claim.id == second.id
assert await job_service.claim_next_queued_job() is None
@pytest.mark.asyncio
async def test_claim_next_queued_job_emits_a_bounded_unadorned_query(
self,
job_service: JobService,
):
"""CRIT-01: the hot poll must not select a subgraph or scan the queue."""
statements: list[str] = []
async with job_service._session_scope() as session:
bind = session.get_bind()
def capture(_conn, _cursor, statement, *_rest):
statements.append(statement)
event.listen(bind, "before_cursor_execute", capture)
try:
await job_service.claim_next_queued_job(session=session)
finally:
event.remove(bind, "before_cursor_execute", capture)
selects = [item for item in statements if item.lstrip().upper().startswith("SELECT")]
assert len(selects) == 1, selects
assert "LIMIT" in selects[0].upper()
assert "JOIN" not in selects[0].upper()
@pytest.mark.asyncio
async def test_create_job_persists_provider_and_model(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="provider-doc")
await document_service.create_document(document=document)
job = Job(
document_id=document.id,
provider="openrouter",
model="google/gemini-2.5-flash",
)
await job_service.create_job(job=job)
fetched = await job_service.read_job(job_id=job.id)
assert fetched.provider == "openrouter"
assert fetched.model == "google/gemini-2.5-flash"
@pytest.mark.asyncio
async def test_read_job_resolves_filename_from_linked_source(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="filename-doc")
await document_service.create_document(document=document)
job = Job(document_id=document.id)
await job_service.create_job(job=job)
async with job_service._session_scope() as session:
source = Source(
document_id=document.id,
page_number=1,
upload_name="page_001.png",
filename="stored_page_001.png",
file_path="/uploads/stored_page_001.png",
file_hash="b" * 64,
file_size_bytes=1,
)
session.add(source)
await session.flush()
session.add(
JobSource(
job_id=job.id,
source_id=source.id,
status=JobSourceStatus.PENDING,
)
)
await session.commit()
fetched = await job_service.read_job(job_id=job.id)
assert fetched.filename == "stored_page_001.png"
@pytest.mark.asyncio
async def test_delete_job_with_guardrails_blocks_processing_jobs(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="processing-delete-doc")
await document_service.create_document(document=document)
job = Job(document_id=document.id, status=JobStatus.PROCESSING)
await job_service.create_job(job=job)
with pytest.raises(JobDeleteBlockedError):
await job_service.delete_job_with_guardrails(job_id=job.id)
@pytest.mark.asyncio
async def test_delete_job_with_guardrails_removes_jobsource_links(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="delete-job-doc")
await document_service.create_document(document=document)
job = Job(document_id=document.id, status=JobStatus.QUEUED)
await job_service.create_job(job=job)
async with job_service._session_scope() as session:
source = Source(
document_id=document.id,
page_number=1,
upload_name="delete-job-source.jpg",
filename="stored-delete-job-source.jpg",
file_path="/uploads/stored-delete-job-source.jpg",
file_hash="c" * 64,
file_size_bytes=1,
)
session.add(source)
await session.flush()
session.add(
JobSource(
job_id=job.id,
source_id=source.id,
status=JobSourceStatus.PENDING,
)
)
await session.commit()
await job_service.delete_job_with_guardrails(job_id=job.id)
with pytest.raises(JobNotFoundError):
await job_service.read_job(job_id=job.id)
@pytest.mark.asyncio
async def test_delete_job_and_evidence_removes_attempts_but_preserves_source(
self,
job_service: JobService,
document_service: DocumentService,
):
source_service = SourceService(session_factory=job_service.session_factory)
document = await document_service.create_document(Document(name="evidence-delete-doc"))
job = await job_service.create_job(Job(document_id=document.id, status=JobStatus.FAILED))
source = await source_service.create_source(
Source(
document_id=document.id,
page_number=1,
upload_name="evidence.jpg",
filename="evidence.jpg",
file_path="/uploads/evidence.jpg",
file_hash="d" * 64,
file_size_bytes=1,
)
)
await source_service.create_job_source(JobSource(job_id=job.id, source_id=source.id))
await source_service.update_job_source_transcription(
job_id=job.id,
source_id=source.id,
text=None,
error_detail="fixture failure",
)
await job_service.delete_job_and_evidence(job_id=job.id)
with pytest.raises(JobNotFoundError):
await job_service.read_job(job_id=job.id)
assert await source_service.list_execution_attempts(job_id=job.id) == []
assert (await source_service.read_source(source.id)).id == source.id
@pytest.mark.asyncio
async def test_cancel_job_marks_non_transcribed_sources_failed(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="cancel-job-doc")
await document_service.create_document(document=document)
job = Job(document_id=document.id, status=JobStatus.QUEUED)
await job_service.create_job(job=job)
async with job_service._session_scope() as session:
source_one = Source(
document_id=document.id,
page_number=1,
upload_name="cancel-1.jpg",
filename="stored-cancel-1.jpg",
file_path="/uploads/stored-cancel-1.jpg",
file_hash="d" * 64,
file_size_bytes=1,
)
source_two = Source(
document_id=document.id,
page_number=2,
upload_name="cancel-2.jpg",
filename="stored-cancel-2.jpg",
file_path="/uploads/stored-cancel-2.jpg",
file_hash="e" * 64,
file_size_bytes=1,
)
session.add(source_one)
session.add(source_two)
await session.flush()
session.add(
JobSource(
job_id=job.id,
source_id=source_one.id,
status=JobSourceStatus.TRANSCRIBED,
raw_transcription="done",
)
)
session.add(
JobSource(
job_id=job.id,
source_id=source_two.id,
status=JobSourceStatus.PENDING,
)
)
await session.commit()
cancelled = await job_service.cancel_job(job_id=job.id)
assert cancelled.status == JobStatus.FAILED
refreshed = await job_service.read_job(job_id=job.id)
statuses = {item.status for item in refreshed.job_sources}
assert JobSourceStatus.TRANSCRIBED in statuses
assert JobSourceStatus.FAILED in statuses
pending_entry = next(item for item in refreshed.job_sources if item.status == JobSourceStatus.FAILED)
assert pending_entry.error_detail == "Cancelled by user"
@pytest.mark.asyncio
async def test_resubmit_failed_sources_resets_only_failed(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="resubmit-job-doc")
await document_service.create_document(document=document)
job = Job(document_id=document.id, status=JobStatus.FAILED)
await job_service.create_job(job=job)
async with job_service._session_scope() as session:
source_one = Source(
document_id=document.id,
page_number=1,
upload_name="resubmit-1.jpg",
filename="stored-resubmit-1.jpg",
file_path="/uploads/stored-resubmit-1.jpg",
file_hash="f" * 64,
file_size_bytes=1,
raw_transcription="existing text",
)
source_two = Source(
document_id=document.id,
page_number=2,
upload_name="resubmit-2.jpg",
filename="stored-resubmit-2.jpg",
file_path="/uploads/stored-resubmit-2.jpg",
file_hash="0" * 64,
file_size_bytes=1,
raw_transcription="done text",
)
session.add(source_one)
session.add(source_two)
await session.flush()
session.add(
JobSource(
job_id=job.id,
source_id=source_one.id,
status=JobSourceStatus.FAILED,
raw_transcription=None,
error_detail="prior error",
)
)
session.add(
JobSource(
job_id=job.id,
source_id=source_two.id,
status=JobSourceStatus.TRANSCRIBED,
raw_transcription="done text",
)
)
await session.commit()
count = await job_service.resubmit_failed_sources(job_id=job.id)
assert count == 1
refreshed = await job_service.read_job(job_id=job.id)
assert refreshed.status == JobStatus.QUEUED
failed_entry = next(
item for item in refreshed.job_sources if item.source is not None and item.source.page_number == 1
)
transcribed_entry = next(
item for item in refreshed.job_sources if item.source is not None and item.source.page_number == 2
)
assert failed_entry.status == JobSourceStatus.PENDING
assert failed_entry.error_detail is None
assert failed_entry.source is not None
assert failed_entry.source.raw_transcription == "existing text"
assert transcribed_entry.status == JobSourceStatus.TRANSCRIBED
@pytest.mark.asyncio
async def test_resubmit_failed_sources_blocks_when_only_pending_or_transcribed(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="resubmit-no-failed-doc")
await document_service.create_document(document=document)
job = Job(document_id=document.id, status=JobStatus.FAILED)
await job_service.create_job(job=job)
async with job_service._session_scope() as session:
source_one = Source(
document_id=document.id,
page_number=1,
upload_name="resubmit-pending.jpg",
filename="stored-resubmit-pending.jpg",
file_path="/uploads/stored-resubmit-pending.jpg",
file_hash="1" * 64,
file_size_bytes=1,
)
source_two = Source(
document_id=document.id,
page_number=2,
upload_name="resubmit-done.jpg",
filename="stored-resubmit-done.jpg",
file_path="/uploads/stored-resubmit-done.jpg",
file_hash="2" * 64,
file_size_bytes=1,
raw_transcription="done text",
)
session.add(source_one)
session.add(source_two)
await session.flush()
session.add(
JobSource(
job_id=job.id,
source_id=source_one.id,
status=JobSourceStatus.PENDING,
)
)
session.add(
JobSource(
job_id=job.id,
source_id=source_two.id,
status=JobSourceStatus.TRANSCRIBED,
raw_transcription="done text",
)
)
await session.commit()
with pytest.raises(JobResubmitBlockedError):
await job_service.resubmit_failed_sources(job_id=job.id)
@pytest.mark.asyncio
async def test_resubmit_failed_sources_blocks_when_processing(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="resubmit-blocked-doc")
await document_service.create_document(document=document)
job = Job(document_id=document.id, status=JobStatus.PROCESSING)
await job_service.create_job(job=job)
with pytest.raises(JobResubmitBlockedError):
await job_service.resubmit_failed_sources(job_id=job.id)
@pytest.mark.asyncio
async def test_cancel_job_blocks_transcribed_terminal_jobs(
self,
job_service: JobService,
document_service: DocumentService,
):
document = Document(id=uuid4(), name="cancel-blocked-doc")
await document_service.create_document(document=document)
job = Job(document_id=document.id, status=JobStatus.TRANSCRIBED)
await job_service.create_job(job=job)
with pytest.raises(JobCancelBlockedError):
await job_service.cancel_job(job_id=job.id)