Commit e6b53c99 authored by Jan Reimes's avatar Jan Reimes
Browse files

fix: resolve ty check errors — stale cache_dir, date/datetime mismatch, deprecated utcnow 

- Remove stale cache_dir parameter from TDocQueryConfig calls in crawl.py
  and query.py (field was removed in prior refactoring)
- Align _filter_by_meeting_dates() parameter types from datetime to date,
  matching TDocQueryConfig.meeting_start_date/meeting_end_date types
- Replace deprecated datetime.utcnow with timezone-aware _utcnow() helper
  using datetime.now(UTC) in oxyde_models.py (6 occurrences)
parent 17f45448

src/tdoc_crawler/cli/crawl.py

+0 −1
Original line number Diff line number Diff line
@@ -218,7 +218,6 @@ def crawl_tdocs( 
            if checkout:

                checkout_limit = limit_tdocs if limit_tdocs and limit_tdocs > 0 else None

                query_config = TDocQueryConfig(

                    cache_dir=manager.root,

                    working_groups=working_groups,

                    limit=checkout_limit,

                    order=SortOrder.DESC,

src/tdoc_crawler/cli/query.py

+0 −1
Original line number Diff line number Diff line
@@ -116,7 +116,6 @@ def query_tdocs( 
        raise typer.Exit(code=2) from exc



    config = TDocQueryConfig(

        cache_dir=manager.root,

        output_format=format_val,

        tdoc_ids=tdoc_ids,

        working_groups=working_groups,

src/tdoc_crawler/database/oxyde_models.py

+11 −7
Original line number Diff line number Diff line
@@ -7,13 +7,17 @@ These models replace the previous pydantic-sqlite based persistence layer. 
from __future__ import annotations



import json

from datetime import date, datetime

from datetime import UTC, date, datetime

from typing import Any



from oxyde import Field, Model

from pydantic import field_validator





def _utcnow() -> datetime:

    return datetime.now(UTC)





class WorkingGroupRecord(Model):

    """Reference data for working groups."""
@@ -54,7 +58,7 @@ class CrawlLogEntry(Model): 
    items_updated: int = Field(default=0, ge=0)

    errors_count: int = Field(default=0, ge=0)

    status: str = Field(default="RUNNING")

    created_at: datetime = Field(default_factory=datetime.utcnow)

    created_at: datetime = Field(default_factory=_utcnow)



    class Meta:

        """Oxyde table configuration."""
@@ -76,9 +80,9 @@ class MeetingMetadata(Model): 
    files_url: str | None = None

    portal_url: str | None = None

    tdoc_count: int = Field(default=0, ge=0)

    last_synced: datetime = Field(default_factory=datetime.utcnow)

    created_at: datetime = Field(default_factory=datetime.utcnow)

    updated_at: datetime = Field(default_factory=datetime.utcnow)

    last_synced: datetime = Field(default_factory=_utcnow)

    created_at: datetime = Field(default_factory=_utcnow)

    updated_at: datetime = Field(default_factory=_utcnow)



    class Meta:

        """Oxyde table configuration."""
@@ -103,8 +107,8 @@ class TDocMetadata(Model): 
    is_revision_of: str | None = None

    file_size: int | None = None

    date_created: datetime | None = None

    date_retrieved: datetime = Field(default_factory=datetime.utcnow)

    date_updated: datetime = Field(default_factory=datetime.utcnow)

    date_retrieved: datetime = Field(default_factory=_utcnow)

    date_updated: datetime = Field(default_factory=_utcnow)

    validated: bool = Field(default=False)

    validation_failed: bool = Field(default=False)

src/tdoc_crawler/database/tdocs.py

+3 −3
Original line number Diff line number Diff line
@@ -2,7 +2,7 @@ 


import fnmatch

from collections.abc import Callable, Iterable

from datetime import UTC, datetime

from datetime import UTC, date, datetime

from typing import Any



from tdoc_crawler.database.meetings import MeetingDatabase
@@ -243,8 +243,8 @@ class TDocDatabase(MeetingDatabase): 
    def _filter_by_meeting_dates(

        records: list[TDocMetadata],

        meeting_map: dict[int, Any],

        meeting_start_date: datetime | None,

        meeting_end_date: datetime | None,

        meeting_start_date: date | None,

        meeting_end_date: date | None,

    ) -> list[TDocMetadata]:

        """Filter records by start/end date of the related meeting."""

        filtered = records