Struct ChromeFetchParams 

pub struct ChromeFetchParams<'a> {

    pub wait_for: &'a Option<WaitFor>,
    pub screenshot: &'a Option<ScreenShotConfig>,
    pub openai_config: &'a Option<Box<GPTConfigs>>,
    pub execution_scripts: &'a Option<ExecutionScripts>,
    pub automation_scripts: &'a Option<AutomationScripts>,
    pub viewport: &'a Option<Viewport>,
    pub request_timeout: &'a Option<Duration>,
    pub track_events: &'a Option<ChromeEventTracker>,
    pub cache_policy: &'a Option<BasicCachePolicy>,
    pub remote_multimodal: &'a Option<Box<RemoteMultimodalConfigs>>,
    pub remote_cache_read_only: bool,
    pub remote_cache_main_doc_only: bool,
    pub first_byte_timeout: &'a Option<Duration>,
    pub first_byte_timeout_jitter: &'a Option<Duration>,
    pub browser_dead: Option<&'a Arc<AtomicBool>>,
    pub chrome_failover: Option<&'a LazyChromeFailover>,
    pub chrome_endpoint_url: Option<&'a str>,
}

Available on crate feature chrome only.

Borrowed parameter bundle for chrome fetch/page methods.

Collapses ten &Option<T> references that otherwise spill across the ABI into one pointer, letting rustc inline the hot fetch pipeline. All fields borrow from a longer-lived Configuration (or locals at the call site) — zero copies, zero allocations.

Fields

wait_for: &'a Option<WaitFor>

Wait-for conditions applied after navigation.

screenshot: &'a Option<ScreenShotConfig>

Screenshot configuration, if any.

openai_config: &'a Option<Box<GPTConfigs>>

OpenAI configuration for content extraction.

execution_scripts: &'a Option<ExecutionScripts>

Per-URL execution scripts trie.

automation_scripts: &'a Option<AutomationScripts>

Per-URL automation scripts trie.

viewport: &'a Option<Viewport>

Viewport used for mouse/smart movement and CDP emulation.

request_timeout: &'a Option<Duration>

Per-request timeout; caps chrome wall time.

track_events: &'a Option<ChromeEventTracker>

CDP event tracking toggles.

cache_policy: &'a Option<BasicCachePolicy>

Disk/remote cache policy.

remote_multimodal: &'a Option<Box<RemoteMultimodalConfigs>>

Remote multimodal automation configuration.

remote_cache_read_only: bool

When true, the remote Chrome cache is read-only: local + session cache serve hits but no responses are uploaded to the remote hybrid_cache_server. Only meaningful when chrome_remote_cache is enabled; ignored otherwise (including no-op at the chromey call site under other feature combinations). Always present so callers don’t have to cfg-gate the field at construction sites.

remote_cache_main_doc_only: bool

When true, only the main (initial) document is published to the remote chrome cache. The per-response listener still populates the local + per-session cache for sub-resources (CSS/JS/manifests) but never uploads them. cache_chrome_response continues to publish the navigated document body — whatever MIME type it is. Orthogonal to remote_cache_read_only: read-only suppresses everything, this flag only suppresses the asset path. Only meaningful when chrome_remote_cache is enabled; ignored otherwise.

first_byte_timeout: &'a Option<Duration>

First-byte watchdog timeout — borrowed from Configuration. None (default) disables the watchdog; the legacy chunk-idle timeout still applies. Some(d) arms the watchdog inside the chrome event-listener block.

first_byte_timeout_jitter: &'a Option<Duration>

Per-fetch jitter window for the first-byte watchdog — borrowed from Configuration. None = no jitter; Some(j) adds a uniform random offset in [0, j) per fetch. Ignored when first_byte_timeout is None.

browser_dead: Option<&'a Arc<AtomicBool>>

Optional browser_dead signal flipped on first-byte timeout. None (default) means “detect and force-stop only” — the watchdog still fires but the AtomicBool is not set. Plumbed through by call sites that have access to a BrowserController. Borrowed pointer to avoid Arc clone overhead at construction; the watchdog clones the inner Arc once if it needs to outlive a spawned future.

chrome_failover: Option<&'a LazyChromeFailover>

Optional handle to the lazy chrome failover. When Some together with chrome_endpoint_url, the first-byte watchdog calls mark_url_bad on the failover so subsequent setup_browser_configuration calls skip the dead URL during the cooldown. Wired automatically from Configuration::chrome_fetch_params — when no multi-endpoint failover is configured the borrowed LazyChromeFailover is empty (uninit OnceLock) and mark_url_bad is a no-op.

chrome_endpoint_url: Option<&'a str>

Optional connected chrome endpoint URL. Set by macros / call sites that have access to a BrowserController.connected_url. The first-byte watchdog passes this URL to LazyChromeFailover::mark_url_bad on timeout.

Implementations

impl<'a> ChromeFetchParams<'a>

pub fn with_browser_dead(self, dead: &'a Arc<AtomicBool>) -> Self

Attach a borrowed browser_dead flag so a first-byte timeout can signal the website-level retry loop to rotate the chrome backend. Builder-style: returns the params with the field overridden.

pub fn with_chrome_endpoint(self, url: Option<&'a str>) -> Self

Attach the connected chrome endpoint URL so the first-byte watchdog can mark it bad on timeout. Pass None to leave it unset (default). Builder-style.

Trait Implementations

impl<'a> Clone for ChromeFetchParams<'a>

fn clone(&self) -> ChromeFetchParams<'a>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<'a> Copy for ChromeFetchParams<'a>