test(harness): add OCR parity with migration strategy runners

.github/ci-coverage-allowlist.yml

@@ -4,6 +4,11 @@ description: >-
   by a job nor listed here, so every entry below is a decision on the record.
 
 test_paths:
+  - reason: >-
+      The Rust/Python parity harness is run manually through its local CLI. Recorded replay,
+      fixture generation, and harness checks are intentionally outside pull request CI
+    paths:
+      - tests/rust-python-harness
   - reason: >-
       What is left of the caching suite in tests/local_testing that runs nowhere. Every job that
       globs that directory either deselects it (local_testing_part1 and part2 carry `-k "... and

litellm-rust/Cargo.toml

@@ -27,7 +27,7 @@ rand = "0.8"
 reqwest = { version = "0.12", default-features = false, features = ["blocking", "json", "rustls-tls", "http2", "stream"] }
 rstest = "0.26.1"
 serde = { version = "1.0", features = ["derive"] }
-serde_json = "1.0"
+serde_json = { version = "1.0", features = ["float_roundtrip"] }
 sha2 = "0.10"
 subtle = "2"
 thiserror = "2.0"

litellm-rust/crates/core/src/providers/azure_ai/ocr/transformation.rs

@@ -14,7 +14,7 @@ const AZURE_DOCUMENT_INTELLIGENCE_ENDPOINT_ENV: &str = "AZURE_DOCUMENT_INTELLIGE
 const AZURE_DOCUMENT_INTELLIGENCE_API_VERSION: &str = "2024-11-30";
 const AZURE_DOCUMENT_INTELLIGENCE_DEFAULT_DPI: i64 = 96;
 
-const AZURE_DOCUMENT_INTELLIGENCE_SUPPORTED_OCR_PARAMS: &[&str] = &["pages"];
+const AZURE_DOCUMENT_INTELLIGENCE_SUPPORTED_OCR_PARAMS: &[&str] = &["pages", "features"];
 
 pub struct AzureAiOcrConfig;
 pub struct AzureDocumentIntelligenceOcrConfig;
@@ -192,6 +192,46 @@ fn normalize_pages_param(pages: &Value) -> Result<Option<String>, Error> {
     }
 }
 
+fn feature_token_is_valid(token: &str) -> bool {
+    let Some((first, rest)) = token.as_bytes().split_first() else {
+        return false;
+    };
+    first.is_ascii_alphabetic() && rest.iter().all(u8::is_ascii_alphanumeric)
+}
+
+fn invalid_features_error(features: &Value) -> Error {
+    Error::InvalidRequest(format!(
+        "Invalid `features` for Azure Document Intelligence: {features:?}. Expected a list of feature names or a comma-separated string like 'keyValuePairs' or 'keyValuePairs,languages'."
+    ))
+}
+
+fn normalize_features_param(features: &Value) -> Result<Option<String>, Error> {
+    let normalized = match features {
+        Value::String(value) => value
+            .split(',')
+            .map(str::trim)
+            .collect::<Vec<_>>()
+            .join(","),
+        Value::Array(values) if values.is_empty() => return Ok(None),
+        Value::Array(values) => values
+            .iter()
+            .map(Value::as_str)
+            .collect::<Option<Vec<_>>>()
+            .ok_or_else(|| invalid_features_error(features))?
+            .into_iter()
+            .map(str::trim)
+            .collect::<Vec<_>>()
+            .join(","),
+        _ => return Err(invalid_features_error(features)),
+    };
+
+    if normalized.split(',').all(feature_token_is_valid) {
+        Ok(Some(normalized))
+    } else {
+        Err(invalid_features_error(features))
+    }
+}
+
 pub fn complete_document_intelligence_url(
     api_base: Option<&str>,
     model: &str,
@@ -213,6 +253,13 @@ pub fn complete_document_intelligence_url(
         url.push_str(&normalized);
     }
 
+    if let Some(features) = optional_params.get("features")
+        && let Some(normalized) = normalize_features_param(features)?
+    {
+        url.push_str("&features=");
+        url.push_str(&normalized);
+    }
+
     Ok(url)
 }
 
@@ -475,6 +522,103 @@ mod tests {
         );
     }
 
+    #[test]
+    fn document_intelligence_url_normalizes_features() {
+        let params = serde_json::Map::from_iter([(
+            "features".to_string(),
+            json!("keyValuePairs, languages"),
+        )]);
+        let url = complete_document_intelligence_url(
+            Some("https://example.cognitiveservices.azure.com"),
+            "prebuilt-layout",
+            &params,
+            &|_| None,
+        )
+        .expect("url builds");
+
+        assert_eq!(
+            url,
+            "https://example.cognitiveservices.azure.com/documentintelligence/documentModels/prebuilt-layout:analyze?api-version=2024-11-30&features=keyValuePairs,languages"
+        );
+    }
+
+    #[test]
+    fn document_intelligence_url_combines_pages_and_feature_list() {
+        let params = serde_json::Map::from_iter([
+            ("pages".to_string(), json!([0, 1, 2])),
+            (
+                "features".to_string(),
+                json!([" keyValuePairs ", "languages"]),
+            ),
+        ]);
+        let url = complete_document_intelligence_url(
+            Some("https://example.cognitiveservices.azure.com"),
+            "prebuilt-layout",
+            &params,
+            &|_| None,
+        )
+        .expect("url builds");
+
+        assert_eq!(
+            url,
+            "https://example.cognitiveservices.azure.com/documentintelligence/documentModels/prebuilt-layout:analyze?api-version=2024-11-30&pages=1,2,3&features=keyValuePairs,languages"
+        );
+    }
+
+    #[test]
+    fn document_intelligence_url_omits_empty_feature_list() {
+        let params = serde_json::Map::from_iter([("features".to_string(), json!([]))]);
+        let url = complete_document_intelligence_url(
+            Some("https://example.cognitiveservices.azure.com"),
+            "prebuilt-layout",
+            &params,
+            &|_| None,
+        )
+        .expect("url builds");
+
+        assert_eq!(
+            url,
+            "https://example.cognitiveservices.azure.com/documentintelligence/documentModels/prebuilt-layout:analyze?api-version=2024-11-30"
+        );
+    }
+
+    #[test]
+    fn document_intelligence_url_rejects_invalid_features() {
+        for features in [
+            json!("keyValuePairs&pages=9"),
+            json!(""),
+            json!(["keyValuePairs", 1]),
+            json!({"feature": "keyValuePairs"}),
+        ] {
+            let params = serde_json::Map::from_iter([("features".to_string(), features.clone())]);
+            let error = complete_document_intelligence_url(
+                Some("https://example.cognitiveservices.azure.com"),
+                "prebuilt-layout",
+                &params,
+                &|_| None,
+            )
+            .expect_err("invalid features must fail");
+
+            assert!(
+                matches!(error, Error::InvalidRequest(message) if message.contains("Invalid `features`")),
+                "features={features:?}"
+            );
+        }
+    }
+
+    #[test]
+    fn document_intelligence_maps_features() {
+        let params = Map::from_iter([
+            ("features".to_string(), json!(["keyValuePairs"])),
+            ("unsupported".to_string(), json!(true)),
+        ]);
+
+        assert_eq!(
+            AZURE_DOCUMENT_INTELLIGENCE_OCR_CONFIG.map_ocr_params(&params),
+            Map::from_iter([("features".to_string(), json!(["keyValuePairs"]))])
+        );
+    }
+
     #[test]
     fn document_intelligence_request_uses_base64_source_for_data_uri() {
         let body = AZURE_DOCUMENT_INTELLIGENCE_OCR_CONFIG

litellm-rust/crates/python-bridge/src/errors.rs

@@ -59,3 +59,41 @@ pub(crate) fn register(module: &Bound<'_, PyModule>) -> PyResult<()> {
     module.add("RustBridgeDeclined", py.get_type::<RustBridgeDeclined>())?;
     module.add("RustUpstreamError", py.get_type::<RustUpstreamError>())
 }
+
+pub(crate) fn ocr_error_to_pyerr(err: Error) -> PyErr {
+    match err {
+        Error::MissingField("document_url" | "image_url") => {
+            PyValueError::new_err("Document URL is required")
+        }
+        Error::Http { status, body } => RustUpstreamError::new_err((status, body)),
+        other => core_error_to_pyerr(other),
+    }
+}
+
+#[cfg(test)]
+mod ocr_error_tests {
+    use super::*;
+
+    #[test]
+    fn ocr_errors_preserve_python_validation_and_provider_details() {
+        Python::initialize();
+        Python::attach(|py| {
+            for field in ["document_url", "image_url"] {
+                let mapped = ocr_error_to_pyerr(Error::MissingField(field));
+                assert!(mapped.is_instance_of::<PyValueError>(py));
+                assert_eq!(mapped.value(py).to_string(), "Document URL is required");
+            }
+            let mapped = ocr_error_to_pyerr(Error::Http {
+                status: 429,
+                body: r#"{"message":"rate limited"}"#.to_string(),
+            });
+            assert!(mapped.is_instance_of::<RustUpstreamError>(py));
+            let args: (u16, String) = mapped
+                .value(py)
+                .getattr("args")
+                .and_then(|args| args.extract())
+                .expect("OCR failures retain status and unprefixed provider message");
+            assert_eq!(args, (429, r#"{"message":"rate limited"}"#.to_string()));
+        });
+    }
+}

litellm-rust/crates/python-bridge/src/routes/ocr.rs

@@ -5,7 +5,7 @@ use litellm_ai_gateway::io::ocr::{OcrRequest, ocr as run_ocr};
 use pyo3::prelude::*;
 use serde_json::Value;
 
-use crate::errors::core_error_to_pyerr;
+use crate::errors::ocr_error_to_pyerr;
 use crate::marshal::{RouteOptions, RouteOptionsInputs, object_or_empty};
 
 fn prepare_ocr(
@@ -69,5 +69,5 @@ bridge_route! {
         timeout_seconds: Option<f64>,
     },
     prepare = prepare_ocr,
-    errors = core_error_to_pyerr,
+    errors = ocr_error_to_pyerr,
 }

pyproject.toml

@@ -174,6 +174,8 @@ litellm-proxy = "litellm.proxy.client.cli:cli"
 [dependency-groups]
 dev = [
     "diff-cover==9.7.2",
+    "hypothesis==6.165.10",
+    "reportlab==5.0.1",
     "basedpyright==1.39.7",
     "keyring==25.7.0",
     "pytest==9.0.3",

tests/code_coverage_tests/liccheck.ini

@@ -172,6 +172,7 @@ pylint: >=3.3.9 # GPLv2 license
 langchain-mcp-adapters: >=0.2.1 # MIT License
 langgraph: >=1.0.10 # MIT License
 langgraph-prebuilt: >=1.0.8 # MIT License - https://github.com/langchain-ai/langgraph/blob/main/LICENSE
+hypothesis: >=6.165.10 # MPL 2.0 license
 pytest-rerunfailures: >=15.1 # MPL 2.0 license
 pytest-recording: >=0.13.4 # MIT license
 expression: >=5.6.0 # MIT License - https://github.com/cognitedata/Expression/blob/main/LICENSE