Skip to content

Feature/processing steps slider - #92

Open
suung wants to merge 11 commits into
mainfrom
feature/processing-steps-slider
Open

Feature/processing steps slider#92
suung wants to merge 11 commits into
mainfrom
feature/processing-steps-slider

Conversation

@suung

@suung suung commented Jul 30, 2026

Copy link
Copy Markdown
Collaborator

Acceptance Criteria

  • All steps of the slider are applied and the analyis is only performed to the respective step
  • It can be continued from there in a future step

@cursor

cursor Bot commented Jul 30, 2026

Copy link
Copy Markdown

Bugbot is not enabled for your account, so this pull request was not reviewed.

Enable Bugbot in the Cursor dashboard to get automatic reviews on future PRs.

@suung
suung marked this pull request as draft July 30, 2026 10:49
Comment thread tests/test_all_results_file_configs.py Fixed
Comment thread tests/test_consolidated_chunk_search.py Fixed
Comment thread tests/test_consolidated_chunk_search.py Fixed
suung added 2 commits July 30, 2026 12:51
Honor max_processing_step in analysis, fix slider bypass when
questions are selected, and show chunk-only results after partial runs.
Reset the analyzer singleton per test so cache/chunks do not leak
across cases and flake embed/chunk assertions.
@suung
suung force-pushed the feature/processing-steps-slider branch from 0502459 to acb05e9 Compare July 30, 2026 12:57
@suung suung linked an issue Jul 30, 2026 that may be closed by this pull request
5 tasks
Black/isort/ruff were failing CI test_linting and QG; expand tests so changed-line coverage clears the 90% gate.


def test_normalize_processing_step_via_custom_label(monkeypatch):
import report_analyst.core.analyzer as analyzer_mod
suung added 3 commits August 20, 2026 22:17
Keep pre_retrieved_chunks and max_processing_step kwargs; retain
selected_question_ids_from_editor for analyze button selection.
Main's AppTest mocks rejected the processing-steps kwarg and failed CI.
@suung
suung marked this pull request as ready for review August 25, 2026 22:04

@mounacheikho-cmd mounacheikho-cmd left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The Question Mapping step was running indefinitely on my machine while waiting for embedding or LLM requests. The reason seems to be that there is no timeout or cancellation handling

@suung

suung commented Sep 1, 2026

Copy link
Copy Markdown
Collaborator Author

The Question Mapping step was running indefinitely on my machine while waiting for embedding or LLM requests. The reason seems to be that there is no timeout or cancellation handling

I looked a bit into this. We could set a shorter timeout, but do you know what's the actual problem, why you didn't get a response from openai?

- Fail fast with max_retries=0 and a 60s request timeout
- Re-raise permanent quota errors instead of swallowing them
- Prove stock LlamaIndex retries a bounded 110 times then stops
_PERMANENT_QUOTA_CODES = frozenset({"insufficient_quota", "credit_balance_exhausted"})


def is_permanent_openai_quota_error(exc: BaseException) -> bool:

Copy link
Copy Markdown
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@mounacheikho-cmd does this fix the problem? It retried quota-errors for a long time.

Keep processing-step slider and fail-fast OpenAI clients; add PDF viewer
from main on analyze and All Results.
Main now passes display_pdf_viewer into the consolidated view.
Keep PDF viewer errors from crashing the app and format cache_manager for black 26.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Bug in Processing-step slider Bug: Processing step slider ignored — Analyze always runs full Answer pipeline

2 participants