kshitijk4poor c870589831 fix(custom): tolerate malformed ports in the Ollama URL heuristic ...

urlparse raises ValueError on non-integer / out-of-range ports, and
http://myhost:99999/v1 passes OpenAI-client construction (only httpx
rejects it later), so the crash was reachable from build_kwargs on
every request for such a URL. Wrap the parsed.port check in the same
try/except ValueError guard hermes_cli.models already uses around its
11434 check, and pin it with parametrized tests.

2026-08-29 21:58:00 +05:30

..

test_commandcode_profile.py

fix(providers): honor a custom base_url over models_url in fetch_models

2026-08-18 14:27:36 -07:00

test_copilot_profile.py

…

test_custom_profile.py

fix(custom): tolerate malformed ports in the Ollama URL heuristic

2026-08-29 21:58:00 +05:30

test_deepseek_profile.py

…

test_fireworks_profile.py

…

test_gemini_profile.py

…

test_kimi_profile.py

fix: map Hermes reasoning efforts onto K3's low/high/max vocabulary

2026-08-13 01:05:49 +05:30

test_minimax_profile.py

…

test_nous_profile.py

fix(nous): treat "takes no reasoning parameter" as a definitive no

2026-08-19 23:28:14 -05:00

test_ollama_cloud_profile.py

refactor: single canonical reasoning-effort vocabulary ends the per-vendor clamp drift

2026-08-19 19:29:10 -07:00

test_opencode_go_profile.py

fix: Ox Alpha reasoning effort reaches the wire clamped — shared across zen and free providers

2026-08-21 03:04:06 -07:00

test_upstage_profile.py

…

test_zai_profile.py

fix(zai): GLM-5.3 low/medium reasoning effort reaches the wire instead of clamping to high

2026-08-21 14:58:37 -07:00