Struct HttpBackendConfig 

pub struct HttpBackendConfig {
    pub base_url: String,
    pub api_key: Option<String>,
    pub forward_auth: bool,
    pub extra_headers: BTreeMap<String, String>,
    pub extra_body: BTreeMap<String, Value>,
    pub omit_body_fields: BTreeSet<String>,
    pub reasoning_effort: Option<String>,
    pub max_retries: u32,
    pub failure_cooldown: Duration,
    pub timeout: Option<Duration>,
}

Shared HTTP configuration for one upstream backend.

Fields

base_url: String

Base URL of the provider API (e.g. https://api.openai.com/v1).

api_key: Option<String>

API key for the provider, loaded by the caller. None sends no configured auth. Client construction rejects active values that cannot form the provider’s auth header.

forward_auth: bool

Whether this backend forwards the caller’s provider credential and application headers.

The forwarding backends in one route must use one credential family (OpenAI or Anthropic) unless they all use the same scheme, host, and port, such as one LLM gateway. Backends that send a configured key are not restricted.

extra_headers: BTreeMap<String, String>

Custom headers added to every outbound call to this backend.

Provider-owned headers are rejected so a static value cannot replace configured or forwarded auth. Names and values must be valid HTTP header bytes; header names are case-insensitive.

extra_body: BTreeMap<String, Value>

Default top-level request fields, applied only when the request omits the key.

omit_body_fields: BTreeSet<String>

Top-level request fields dropped from the outbound body before extra_body is merged, so a target can reinstate one deliberately. For providers that reject an otherwise standard field (e.g. an internal API that 400s on max_output_tokens).

reasoning_effort: Option<String>

Reasoning effort forced on every request to this backend, replacing whatever the caller sent. Responses carries it as reasoning.effort, Chat Completions as reasoning_effort; Anthropic has no equivalent and rejects the setting at configuration time.

max_retries: u32

Additional attempts after the initial upstream request.

failure_cooldown: Duration

Cooldown after an exhausted transient completion failure. Zero disables it.

timeout: Option<Duration>

Deadline for one complete response, including retries, retry delays, and stream reads. None leaves the wait unbounded.

Trait Implementations

impl Clone for HttpBackendConfig

fn clone(&self) -> HttpBackendConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for HttpBackendConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.