curo/Pyra
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
52 Commits 4 Branches 0 Tags
833d1445f03b7a54163591d06edb010b91637353
Commit Graph

5 Commits

Author SHA1 Message Date
curo1305 833d1445f0 feat(setup,chat): detect actually-loaded local model via provider-specific API 
For Ollama, /api/tags returns all installed models, not running ones.
Add fetch_loaded_models() using /api/ps for Ollama (and /v1/models for
LM Studio/llama.cpp, which already return only loaded models).

_show_local_model_status() now calls fetch_loaded_models() so the
setup wizard correctly shows only in-memory models for Ollama.

At chat session startup, local providers warn when the configured model
is not currently loaded, or when nothing is loaded at all.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-05-19 14:46:54 +02:00
curo1305 b3851a2715 test: add tests for draft persistence, model status, and model re-entry 
10 new tests covering:
- _save_draft / _load_draft / _delete_draft / _mark_step_done helpers
- draft file permissions (chmod 600)
- _show_local_model_status with zero, one, and multiple loaded models
- _test_connection change_model path (model error → change → retry succeeds)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-05-19 13:43:53 +02:00
curo1305 efc589cc56 test: add tests for _classify_error and _check_local_server retry behaviour 
_classify_error: covers all litellm error types (auth, not-found, rate-limit,
service-unavailable, connection, timeout, bad-request), httpx connect and
timeout errors, and generic fallback — using dynamically constructed fake
exception classes to avoid importing litellm in tests.

_check_local_server: covers success, retry-then-success, abort (SystemExit),
and continue-anyway paths via monkeypatched httpx and questionary.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-05-19 13:27:04 +02:00
curo1305 bafdafea02 test: add unit tests for wizard model-discovery helpers 
Cover _fetch_local_models (LM Studio and Ollama parsing, error paths,
missing base_url), _fetch_lmstudio_available_models (happy path and
errors), and _load_lmstudio_model (success, API failure, exception).
All mocked via monkeypatch/MagicMock — no real HTTP calls.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-05-19 10:53:22 +02:00
curo1305 9735a5559e test: add tests for setup wizard personalization and system prompt builder 
Cover _USE_CASE_PLUGINS mapping, _suggest_plugins side effects, _build_system_base
output for all name/purpose combinations, and GeneralConfig.purpose round-trip.
Also update CLAUDE.md with the testing workflow rule.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
 2026-05-19 10:43:20 +02:00