justinzeus/scholarr
1
0
Fork 0
Code Issues 1 Pull requests 18 Projects Releases Packages Wiki Activity Actions

Initial release prep

Browse source
This commit is contained in:
Justin Visser 2026-02-27 10:46:05 +01:00
parent c9b9d892f4
commit 92395b2b4b
100 changed files with 405 additions and 401 deletions
Download patch file Download diff file Expand all files Collapse all files

56
app/services/publications/__init__.py
View file

@ -1,84 +1,84 @@
from app.services.domains.publications.application import (
from app.services.publications.application import (
MODE_ALL as MODE_ALL,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
MODE_LATEST as MODE_LATEST,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
MODE_NEW as MODE_NEW,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
MODE_UNREAD as MODE_UNREAD,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
PublicationListItem as PublicationListItem,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
UnreadPublicationItem as UnreadPublicationItem,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
count_favorite_for_user as count_favorite_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
count_for_user as count_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
count_latest_for_user as count_latest_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
count_pdf_queue_items as count_pdf_queue_items,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
count_unread_for_user as count_unread_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
enqueue_all_missing_pdf_jobs as enqueue_all_missing_pdf_jobs,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
enqueue_retry_pdf_job_for_publication_id as enqueue_retry_pdf_job_for_publication_id,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
get_latest_run_id_for_user as get_latest_run_id_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
get_publication_item_for_user as get_publication_item_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
hydrate_pdf_enrichment_state as hydrate_pdf_enrichment_state,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
list_for_user as list_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
list_pdf_queue_items as list_pdf_queue_items,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
list_pdf_queue_page as list_pdf_queue_page,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
list_unread_for_user as list_unread_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
mark_all_unread_as_read_for_user as mark_all_unread_as_read_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
mark_selected_as_read_for_user as mark_selected_as_read_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
publications_query as publications_query,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
resolve_publication_view_mode as resolve_publication_view_mode,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
retry_pdf_for_user as retry_pdf_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
schedule_missing_pdf_enrichment_for_user as schedule_missing_pdf_enrichment_for_user,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
schedule_retry_pdf_enrichment_for_row as schedule_retry_pdf_enrichment_for_row,
)
from app.services.domains.publications.application import (
from app.services.publications.application import (
set_publication_favorite_for_user as set_publication_favorite_for_user,
)

16
app/services/publications/application.py
View file

@ -1,46 +1,46 @@
from __future__ import annotations
from app.services.domains.publications.counts import (
from app.services.publications.counts import (
count_favorite_for_user,
count_for_user,
count_latest_for_user,
count_unread_for_user,
)
from app.services.domains.publications.enrichment import (
from app.services.publications.enrichment import (
hydrate_pdf_enrichment_state,
schedule_missing_pdf_enrichment_for_user,
schedule_retry_pdf_enrichment_for_row,
)
from app.services.domains.publications.listing import (
from app.services.publications.listing import (
list_for_user,
list_unread_for_user,
retry_pdf_for_user,
)
from app.services.domains.publications.modes import (
from app.services.publications.modes import (
MODE_ALL,
MODE_LATEST,
MODE_NEW,
MODE_UNREAD,
resolve_publication_view_mode,
)
from app.services.domains.publications.pdf_queue import (
from app.services.publications.pdf_queue import (
count_pdf_queue_items,
enqueue_all_missing_pdf_jobs,
enqueue_retry_pdf_job_for_publication_id,
list_pdf_queue_items,
list_pdf_queue_page,
)
from app.services.domains.publications.queries import (
from app.services.publications.queries import (
get_latest_run_id_for_user,
get_publication_item_for_user,
publications_query,
)
from app.services.domains.publications.read_state import (
from app.services.publications.read_state import (
mark_all_unread_as_read_for_user,
mark_selected_as_read_for_user,
set_publication_favorite_for_user,
)
from app.services.domains.publications.types import PublicationListItem, UnreadPublicationItem
from app.services.publications.types import PublicationListItem, UnreadPublicationItem
__all__ = [
"MODE_ALL",

4
app/services/publications/counts.py
View file

@ -6,13 +6,13 @@ from sqlalchemy import distinct, func, select
from sqlalchemy.ext.asyncio import AsyncSession
from app.db.models import Publication, ScholarProfile, ScholarPublication
from app.services.domains.publications.modes import (
from app.services.publications.modes import (
MODE_ALL,
MODE_LATEST,
MODE_UNREAD,
resolve_publication_view_mode,
)
from app.services.domains.publications.queries import get_latest_run_id_for_user
from app.services.publications.queries import get_latest_run_id_for_user
async def count_for_user(

2
app/services/publications/dedup.py
View file

@ -11,7 +11,7 @@ from sqlalchemy.orm import aliased
from app.db.models import Publication, PublicationIdentifier, ScholarPublication
from app.logging_utils import structured_log
from app.services.domains.ingestion.fingerprints import (
from app.services.ingestion.fingerprints import (
canonical_title_text_for_dedup,
canonical_title_tokens_for_dedup,
normalize_title,

4
app/services/publications/enrichment.py
View file

@ -5,12 +5,12 @@ import logging
from sqlalchemy.ext.asyncio import AsyncSession
from app.logging_utils import structured_log
from app.services.domains.publications.pdf_queue import (
from app.services.publications.pdf_queue import (
enqueue_missing_pdf_jobs,
enqueue_retry_pdf_job,
overlay_pdf_job_state,
)
from app.services.domains.publications.types import PublicationListItem
from app.services.publications.types import PublicationListItem
logger = logging.getLogger(__name__)

8
app/services/publications/listing.py
View file

@ -4,20 +4,20 @@ from datetime import datetime
from sqlalchemy.ext.asyncio import AsyncSession
from app.services.domains.publication_identifiers import application as identifier_service
from app.services.domains.publications.modes import (
from app.services.publication_identifiers import application as identifier_service
from app.services.publications.modes import (
MODE_ALL,
MODE_UNREAD,
resolve_publication_view_mode,
)
from app.services.domains.publications.queries import (
from app.services.publications.queries import (
get_latest_run_id_for_user,
get_publication_item_for_user,
publication_list_item_from_row,
publications_query,
unread_item_from_row,
)
from app.services.domains.publications.types import PublicationListItem, UnreadPublicationItem
from app.services.publications.types import PublicationListItem, UnreadPublicationItem
async def list_for_user(

16
app/services/publications/pdf_queue.py
View file

@ -18,13 +18,13 @@ from app.db.models import (
)
from app.db.session import get_session_factory
from app.logging_utils import structured_log
from app.services.domains.publication_identifiers import application as identifier_service
from app.services.domains.publication_identifiers.types import DisplayIdentifier
from app.services.domains.publications.pdf_resolution_pipeline import (
from app.services.publication_identifiers import application as identifier_service
from app.services.publication_identifiers.types import DisplayIdentifier
from app.services.publications.pdf_resolution_pipeline import (
resolve_publication_pdf_outcome_for_row,
)
from app.services.domains.publications.types import PublicationListItem
from app.services.domains.unpaywall.application import (
from app.services.publications.types import PublicationListItem
from app.services.unpaywall.application import (
FAILURE_RESOLUTION_EXCEPTION,
OaResolutionOutcome,
)
@ -450,7 +450,7 @@ async def _resolve_publication_row(
openalex_api_key: str | None = None,
allow_arxiv_lookup: bool = True,
) -> bool:
from app.services.domains.openalex.client import OpenAlexBudgetExhaustedError
from app.services.openalex.client import OpenAlexBudgetExhaustedError
await _mark_attempt_started(publication_id=row.publication_id, user_id=user_id)
try:
@ -493,8 +493,8 @@ async def _run_resolution_task(
request_email: str | None,
rows: list[PublicationListItem],
) -> None:
from app.services.domains.openalex.client import OpenAlexBudgetExhaustedError
from app.services.domains.settings import application as user_settings_service
from app.services.openalex.client import OpenAlexBudgetExhaustedError
from app.services.settings import application as user_settings_service
# Resolve the best available API key: per-user setting → env var fallback.
openalex_api_key: str | None = None

16
app/services/publications/pdf_resolution_pipeline.py
View file

@ -5,11 +5,11 @@ from dataclasses import dataclass
from typing import Any
from app.logging_utils import structured_log
from app.services.domains.arxiv.errors import ArxivRateLimitError
from app.services.domains.arxiv.guards import arxiv_skip_reason_for_item
from app.services.domains.openalex.client import OpenAlexBudgetExhaustedError
from app.services.domains.publications.types import PublicationListItem
from app.services.domains.unpaywall.application import OaResolutionOutcome, resolve_publication_oa_outcomes
from app.services.arxiv.errors import ArxivRateLimitError
from app.services.arxiv.guards import arxiv_skip_reason_for_item
from app.services.openalex.client import OpenAlexBudgetExhaustedError
from app.services.publications.types import PublicationListItem
from app.services.unpaywall.application import OaResolutionOutcome, resolve_publication_oa_outcomes
from app.settings import settings
logger = logging.getLogger(__name__)
@ -59,8 +59,8 @@ async def _openalex_outcome(
request_email: str | None,
openalex_api_key: str | None = None,
) -> OaResolutionOutcome | None:
from app.services.domains.openalex.client import OpenAlexClient
from app.services.domains.openalex.matching import find_best_match
from app.services.openalex.client import OpenAlexClient
from app.services.openalex.matching import find_best_match
if not row.title:
return None
@ -105,7 +105,7 @@ async def _arxiv_outcome(
request_email: str | None,
allow_lookup: bool = True,
) -> OaResolutionOutcome | None:
from app.services.domains.arxiv.application import discover_arxiv_id_for_publication
from app.services.arxiv.application import discover_arxiv_id_for_publication
if not allow_lookup:
structured_log(

4
app/services/publications/queries.py
View file

@ -13,8 +13,8 @@ from app.db.models import (
ScholarProfile,
ScholarPublication,
)
from app.services.domains.publications.modes import MODE_LATEST, MODE_UNREAD
from app.services.domains.publications.types import PublicationListItem, UnreadPublicationItem
from app.services.publications.modes import MODE_LATEST, MODE_UNREAD
from app.services.publications.types import PublicationListItem, UnreadPublicationItem
def _normalized_citation_count(value: object) -> int:

2
app/services/publications/types.py
View file

@ -3,7 +3,7 @@ from __future__ import annotations
from dataclasses import dataclass
from datetime import datetime
from app.services.domains.publication_identifiers.types import DisplayIdentifier
from app.services.publication_identifiers.types import DisplayIdentifier
@dataclass(frozen=True)