Skip to content

Supercharge your AI agents with data from the web and beyond. Building the library for superintelligence. 🔥

feat(rust-sdk): add DocumentMetadata::provider for Exchange scrapes (#5001)

Gergő Móricz7599de41e7a3
diff --git a/apps/rust-sdk/CHANGELOG.md b/apps/rust-sdk/CHANGELOG.md
index abf5e2a47..1b5661f4c 100644
--- a/apps/rust-sdk/CHANGELOG.md
+++ b/apps/rust-sdk/CHANGELOG.md
@@ -1,5 +1,14 @@
## CHANGELOG
+## [2.21.5] - 2026-10-07
+
+### Added
+
+- `DocumentMetadata::provider` (`ScrapeProvider`) reports the third-party
+ provider that served an Exchange scrape: its `id`, the access price in
+ `credits_cost`, and every provider tried in `steps` (`ScrapeProviderStep`).
+ It is `None` for any other scrape.
+
## [2.21.4] - 2026-10-07
### Fixed
diff --git a/apps/rust-sdk/Cargo.lock b/apps/rust-sdk/Cargo.lock
index 29328766e..6c425b6db 100644
--- a/apps/rust-sdk/Cargo.lock
+++ b/apps/rust-sdk/Cargo.lock
@@ -250,7 +250,7 @@ checksum = "37909eebbb50d72f9059c3b6d82c0463f2ff062c9e95845c43a6c9c0355411be"
[[package]]
name = "firecrawl"
-version = "2.21.4"
+version = "2.21.5"
dependencies = [
"mockito",
"reqwest",
diff --git a/apps/rust-sdk/Cargo.toml b/apps/rust-sdk/Cargo.toml
index ef8b27f3f..69908bece 100644
--- a/apps/rust-sdk/Cargo.toml
+++ b/apps/rust-sdk/Cargo.toml
@@ -1,6 +1,6 @@
[package]
name = "firecrawl"
-version = "2.21.4"
+version = "2.21.5"
edition = "2021"
license = "MIT"
homepage = "https://www.firecrawl.dev/"
diff --git a/apps/rust-sdk/src/serde_helpers.rs b/apps/rust-sdk/src/serde_helpers.rs
index 21fb59dc4..36550d1b7 100644
--- a/apps/rust-sdk/src/serde_helpers.rs
+++ b/apps/rust-sdk/src/serde_helpers.rs
@@ -30,6 +30,17 @@ where
}
}
+/// Reads a field as `T`, or as `None` when the value has another shape, such as
+/// a page's own meta tag of the same name.
+pub(crate) fn deserialize_or_none<'de, D, T>(deserializer: D) -> Result<Option<T>, D::Error>
+where
+ D: serde::Deserializer<'de>,
+ T: serde::de::DeserializeOwned,
+{
+ Ok(Option::<Value>::deserialize(deserializer)?
+ .and_then(|value| serde_json::from_value(value).ok()))
+}
+
#[cfg(test)]
mod tests {
use serde::Deserialize;
diff --git a/apps/rust-sdk/src/types.rs b/apps/rust-sdk/src/types.rs
index 93ade4cf7..791f74818 100644
--- a/apps/rust-sdk/src/types.rs
+++ b/apps/rust-sdk/src/types.rs
@@ -4,7 +4,7 @@ use serde::{de, Deserialize, Deserializer, Serialize, Serializer};
use serde_json::Value;
use std::collections::HashMap;
-use crate::serde_helpers::deserialize_string_or_array;
+use crate::serde_helpers::{deserialize_or_none, deserialize_string_or_array};
/// Available output formats for scraping operations.
#[derive(Clone, Debug, PartialEq, Eq)]
@@ -636,6 +636,33 @@ pub struct DocumentMetadata {
pub cached_at: Option<String>,
pub credits_used: Option<u32>,
pub concurrency_limited: Option<bool>,
+ /// The third-party provider that served an Exchange scrape. `None` for
+ /// any other scrape.
+ #[serde(default, deserialize_with = "deserialize_or_none")]
+ pub provider: Option<ScrapeProvider>,
+}
+
+/// The third-party provider that served an Exchange scrape, what the access
+/// cost in credits, and every provider tried for it in order.
+#[serde_with::skip_serializing_none]
+#[derive(Deserialize, Serialize, Debug, Default, Clone, PartialEq)]
+#[serde(rename_all = "camelCase")]
+pub struct ScrapeProvider {
+ /// The provider that returned the data.
+ pub id: String,
+ pub credits_cost: u32,
+ pub steps: Vec<ScrapeProviderStep>,
+}
+
+/// One provider tried for an Exchange scrape.
+#[serde_with::skip_serializing_none]
+#[derive(Deserialize, Serialize, Debug, Default, Clone, PartialEq)]
+#[serde(rename_all = "camelCase")]
+pub struct ScrapeProviderStep {
+ pub provider: String,
+ /// `matched`, `not_found` or `error`.
+ pub status: String,
+ pub credits_cost: Option<u32>,
}
/// Extracted attribute result.
@@ -1265,6 +1292,63 @@ mod tests {
assert_eq!(meta.keywords, Some("rust, sdk, firecrawl".to_string()));
}
+ #[test]
+ fn test_document_metadata_provider() {
+ let json = json!({
+ "metadata": {
+ "sourceURL": "https://profiles.example/in/example-person",
+ "statusCode": 200,
+ "provider": {
+ "id": "globex",
+ "creditsCost": 30,
+ "steps": [
+ { "provider": "acme", "status": "not_found", "creditsCost": 10 },
+ { "provider": "globex", "status": "matched", "creditsCost": 20 },
+ { "provider": "initech", "status": "error" }
+ ]
+ }
+ }
+ });
+ let doc: Document = serde_json::from_value(json).unwrap();
+ let provider = doc.metadata.unwrap().provider.unwrap();
+ assert_eq!(provider.id, "globex");
+ assert_eq!(provider.credits_cost, 30);
+ assert_eq!(
+ provider.steps,
+ vec![
+ ScrapeProviderStep {
+ provider: "acme".to_string(),
+ status: "not_found".to_string(),
+ credits_cost: Some(10),
+ },
+ ScrapeProviderStep {
+ provider: "globex".to_string(),
+ status: "matched".to_string(),
+ credits_cost: Some(20),
+ },
+ ScrapeProviderStep {
+ provider: "initech".to_string(),
+ status: "error".to_string(),
+ credits_cost: None,
+ },
+ ]
+ );
+ }
+
+ #[test]
+ fn test_document_metadata_provider_absent_or_other_shape() {
+ for metadata in [
+ json!({ "statusCode": 200 }),
+ json!({ "statusCode": 200, "provider": "Example Provider" }),
+ json!({ "statusCode": 200, "provider": ["a", "b"] }),
+ ] {
+ let doc: Document = serde_json::from_value(json!({ "metadata": metadata })).unwrap();
+ let meta = doc.metadata.unwrap();
+ assert_eq!(meta.status_code, Some(200));
+ assert_eq!(meta.provider, None);
+ }
+ }
+
#[test]
fn test_format_menu_round_trip() {
let format = Format::Menu;