Commit b959e437 authored by Jan Reimes's avatar Jan Reimes
Browse files

tests: move AI test imports to module level and fix fixtures; tdocs: minor db helper import cleanup

parent 709c0c21

src/tdoc_crawler/tdocs/models.py

+9 −0
Original line number Diff line number Diff line
@@ -298,6 +298,15 @@ class TDocQueryConfig(BaseConfigModel): 
    limit: int | None = Field(None, ge=1, description="Maximum results")

    order: SortOrder = Field(SortOrder.DESC, description="Sort order applied to date_retrieved")



    # TODO: normalization of tdoc name(s) is most likely done already several times in the project? Consider re-using existing utils function?

    @field_validator("tdoc_ids", mode="before")

    @classmethod

    def _normalize_tdoc_ids(cls, value: Iterable[str] | None) -> list[str] | None:

        """Normalize TDoc IDs to uppercase."""

        if value is None:

            return None

        return [str(item).upper() for item in value]



    @classmethod

    def _normalize_working_groups(cls, value: Iterable[str | WorkingGroup] | None) -> list[WorkingGroup] | None:

        """Ensure the working group list is comprised of enum members."""

tests/ai/conftest.py

+1 −2
Original line number Diff line number Diff line
@@ -12,6 +12,7 @@ import pytest 
from typer.testing import CliRunner



from tdoc_crawler.ai.container import AiServiceContainer

from tdoc_crawler.ai.operations import workspaces as workspace_ops

from tdoc_crawler.ai.storage import AiStorage

from tdoc_crawler.http_client import download_to_file
@@ -177,8 +178,6 @@ def test_workspace(ai_storage: AiStorage) -> str: 
    Returns:

        Workspace name ("default")

    """

    from tdoc_crawler.ai.operations import workspaces as workspace_ops



    workspace_name = "default"

    workspace_ops.create_workspace(workspace_name, auto_build=False)

    return workspace_name

tests/ai/test_ai_pipeline.py

+0 −3
Original line number Diff line number Diff line
@@ -140,9 +140,6 @@ class TestProcessTdocApi: 
        """new_only mode should return only non-completed items."""

        # This test verifies the filtering logic in process_all

        # by mocking the storage.get_status call

        from tdoc_crawler.ai.models import PipelineStage, ProcessingStatus

        from tdoc_crawler.ai.storage import AiStorage



        completed = ProcessingStatus(document_id="S4-251003", current_stage=PipelineStage.COMPLETED)

        pending = ProcessingStatus(document_id="S4-260001", current_stage=PipelineStage.PENDING)

tests/ai/test_ai_summarization.py

+1 −2
Original line number Diff line number Diff line
@@ -7,6 +7,7 @@ from unittest.mock import patch 


import pytest



from tdoc_crawler.ai.container import AiServiceContainer

from tdoc_crawler.ai.models import DocumentSummary, LlmConfigError

from tdoc_crawler.ai.operations import summarize

from tdoc_crawler.ai.operations.summarize import _count_words, _should_skip_summary
@@ -81,8 +82,6 @@ class TestSummarization: 


    def test_missing_llm_model_config_raises_error(self, monkeypatch: pytest.MonkeyPatch) -> None:

        """Missing/invalid LLM config should raise LlmConfigError."""

        from tdoc_crawler.ai.container import AiServiceContainer



        monkeypatch.setattr(

            AiServiceContainer,

            "get_instance",