WIP: Claude Code progress on #1342

Automated salvage commit — agent session ended (exit 124).
Work in progress, may need continuation.

src/infrastructure/router/config_loader.py

@@ -0,0 +1,123 @@
+"""Config loading helpers for the Cascade LLM Router.
+
+Parses providers.yaml, expands env vars, and checks provider availability.
+"""
+
+from __future__ import annotations
+
+import logging
+
+from infrastructure.router.models import Provider, RouterConfig
+
+logger = logging.getLogger(__name__)
+
+try:
+    import yaml
+except ImportError:
+    yaml = None  # type: ignore
+
+try:
+    import requests
+except ImportError:
+    requests = None  # type: ignore
+
+
+def expand_env_vars(content: str) -> str:
+    """Expand ${VAR} syntax in YAML content.
+
+    Uses os.environ directly (not settings) because this is a generic
+    YAML config loader that must expand arbitrary variable references.
+    """
+    import os
+    import re
+
+    def replace_var(match: "re.Match[str]") -> str:
+        var_name = match.group(1)
+        return os.environ.get(var_name, match.group(0))
+
+    return re.sub(r"\$\{(\w+)\}", replace_var, content)
+
+
+def parse_router_config(data: dict) -> RouterConfig:
+    """Build a RouterConfig from parsed YAML data."""
+    cascade = data.get("cascade", {})
+    cb = cascade.get("circuit_breaker", {})
+    multimodal = data.get("multimodal", {})
+
+    return RouterConfig(
+        timeout_seconds=cascade.get("timeout_seconds", 30),
+        max_retries_per_provider=cascade.get("max_retries_per_provider", 2),
+        retry_delay_seconds=cascade.get("retry_delay_seconds", 1),
+        circuit_breaker_failure_threshold=cb.get("failure_threshold", 5),
+        circuit_breaker_recovery_timeout=cb.get("recovery_timeout", 60),
+        circuit_breaker_half_open_max_calls=cb.get("half_open_max_calls", 2),
+        auto_pull_models=multimodal.get("auto_pull", True),
+        fallback_chains=data.get("fallback_chains", {}),
+    )
+
+
+def load_providers(data: dict) -> list[Provider]:
+    """Load and filter providers from parsed YAML data (unsorted)."""
+    providers: list[Provider] = []
+    for p_data in data.get("providers", []):
+        if not p_data.get("enabled", False):
+            continue
+
+        provider = Provider(
+            name=p_data["name"],
+            type=p_data["type"],
+            enabled=p_data.get("enabled", True),
+            priority=p_data.get("priority", 99),
+            tier=p_data.get("tier"),
+            url=p_data.get("url"),
+            api_key=p_data.get("api_key"),
+            base_url=p_data.get("base_url"),
+            models=p_data.get("models", []),
+        )
+
+        if check_provider_available(provider):
+            providers.append(provider)
+        else:
+            logger.warning("Provider %s not available, skipping", provider.name)
+
+    return providers
+
+
+def check_provider_available(provider: Provider) -> bool:
+    """Check if a provider is actually available."""
+    from config import settings
+
+    if provider.type == "ollama":
+        # Check if Ollama is running
+        if requests is None:
+            # Can't check without requests, assume available
+            return True
+        try:
+            url = provider.url or settings.ollama_url
+            response = requests.get(f"{url}/api/tags", timeout=5)
+            return response.status_code == 200
+        except Exception as exc:
+            logger.debug("Ollama provider check error: %s", exc)
+            return False
+
+    elif provider.type == "vllm_mlx":
+        # Check if local vllm-mlx server is running (OpenAI-compatible)
+        if requests is None:
+            return True
+        try:
+            base_url = provider.base_url or provider.url or "http://localhost:8000"
+            # Strip /v1 suffix — health endpoint is at the root
+            server_root = base_url.rstrip("/")
+            if server_root.endswith("/v1"):
+                server_root = server_root[:-3]
+            response = requests.get(f"{server_root}/health", timeout=5)
+            return response.status_code == 200
+        except Exception as exc:
+            logger.debug("vllm-mlx provider check error: %s", exc)
+            return False
+
+    elif provider.type in ("openai", "anthropic", "grok"):
+        # Check if API key is set
+        return provider.api_key is not None and provider.api_key != ""
+
+    return True