fegger 081084816b Improve VLM client lifecycle, retries, metrics, and cache keys ...

- Adds abstract aclose() and shared load_prompt() helper to BaseVLMClient.

- OllamaClient now logs latency, tokens, and prompt hash per call.

- Retries cover HTTP 429 and 5xx in addition to network/timeout errors.

- Cache key now includes response_format and max_tokens.

- Warns when JSON schema generation falls back to plain 'json' format.

- Adds FakeVLMClient.aclose() and cache unit tests.

2026-08-21 16:44:14 +02:00

..

__init__.py

Initial scaffold: local invoice OCR pipeline with Ollama, classifier branches, structured extraction, review, and Odoo XML export

2026-08-21 14:04:42 +02:00

conftest.py

Improve VLM client lifecycle, retries, metrics, and cache keys

2026-08-21 16:44:14 +02:00

test_cache.py

Improve VLM client lifecycle, retries, metrics, and cache keys

2026-08-21 16:44:14 +02:00

test_classifier.py

Add uv.lock and bump mypy target to 3.12

2026-08-21 16:44:04 +02:00

test_extractor.py

Initial scaffold: local invoice OCR pipeline with Ollama, classifier branches, structured extraction, review, and Odoo XML export

2026-08-21 14:04:42 +02:00

test_ollama_client.py

Initial scaffold: local invoice OCR pipeline with Ollama, classifier branches, structured extraction, review, and Odoo XML export

2026-08-21 14:04:42 +02:00

test_schemas.py

Add OcrResponse schema and tighten invoice validation

2026-08-21 16:44:09 +02:00

test_xml_builder.py

Initial scaffold: local invoice OCR pipeline with Ollama, classifier branches, structured extraction, review, and Odoo XML export

2026-08-21 14:04:42 +02:00