fix: jina base url pass as params

integrations/jina/src/haystack_integrations/components/embedders/jina/document_embedder.py

@@ -50,6 +50,8 @@ def __init__(
         task: str | None = None,
         dimensions: int | None = None,
         late_chunking: bool | None = None,
+        *,
+        base_url: str = JINA_API_URL,
     ) -> None:
         """
         Create a JinaDocumentEmbedder component.
@@ -72,13 +74,15 @@ def __init__(
         :param late_chunking: A boolean to enable or disable late chunking.
             Apply the late chunking technique to leverage the model's long-context capabilities for
             generating contextual chunk embeddings.
+        :param base_url: The base URL of the Jina API.
 
             The support of `task` and `late_chunking` parameters is only available for jina-embeddings-v3.
         """
         resolved_api_key = api_key.resolve_value()
 
         self.api_key = api_key
         self.model_name = model
+        self.base_url = base_url
         self.prefix = prefix
         self.suffix = suffix
         self.batch_size = batch_size
@@ -113,6 +117,7 @@ def to_dict(self) -> dict[str, Any]:
         kwargs = {
             "api_key": self.api_key.to_dict(),
             "model": self.model_name,
+            "base_url": self.base_url,
             "prefix": self.prefix,
             "suffix": self.suffix,
             "batch_size": self.batch_size,
@@ -173,7 +178,7 @@ def _embed_batch(
         ):
             batch = texts_to_embed[i : i + batch_size]
             response = self._session.post(
-                JINA_API_URL,
+                self.base_url,
                 json={"input": batch, "model": self.model_name, **(parameters or {})},
             ).json()
             if "data" not in response:

integrations/jina/src/haystack_integrations/components/embedders/jina/document_image_embedder.py

@@ -57,6 +57,7 @@ def __init__(
         *,
         api_key: Secret = Secret.from_env_var("JINA_API_KEY"),  # noqa: B008
         model: str = "jina-clip-v2",
+        base_url: str = JINA_API_URL,
         file_path_meta_field: str = "file_path",
         root_path: str | None = None,
         embedding_dimension: int | None = None,
@@ -74,6 +75,7 @@ def __init__(
             - "jina-clip-v2" (default)
             - "jina-embeddings-v4"
             Check the list of available models on [Jina documentation](https://jina.ai/embeddings/).
+        :param base_url: The base URL of the Jina API.
         :param file_path_meta_field: The metadata field in the Document that contains the file path to the image or PDF.
         :param root_path: The root directory path where document files are located. If provided, file paths in
             document metadata will be resolved relative to this path. If None, file paths are treated as absolute paths.
@@ -89,6 +91,7 @@ def __init__(
 
         self.api_key = api_key
         self.model_name = model
+        self.base_url = base_url
         self.file_path_meta_field = file_path_meta_field
         self.root_path = root_path or ""
         self.embedding_dimension = embedding_dimension
@@ -120,6 +123,7 @@ def to_dict(self) -> dict[str, Any]:
             self,
             api_key=self.api_key.to_dict(),
             model=self.model_name,
+            base_url=self.base_url,
             file_path_meta_field=self.file_path_meta_field,
             root_path=self.root_path,
             embedding_dimension=self.embedding_dimension,
@@ -235,7 +239,7 @@ def run(self, documents: list[Document]) -> dict[str, list[Document]]:
 
             try:
                 response = self._session.post(
-                    JINA_API_URL,
+                    self.base_url,
                     json={
                         "input": batch_images,
                         "model": self.model_name,

integrations/jina/src/haystack_integrations/components/embedders/jina/text_embedder.py

@@ -42,6 +42,8 @@ def __init__(
         task: str | None = None,
         dimensions: int | None = None,
         late_chunking: bool | None = None,
+        *,
+        base_url: str = JINA_API_URL,
     ) -> None:
         """
         Create a JinaTextEmbedder component.
@@ -60,6 +62,7 @@ def __init__(
         :param late_chunking: A boolean to enable or disable late chunking.
             Apply the late chunking technique to leverage the model's long-context capabilities for
             generating contextual chunk embeddings.
+        :param base_url: The base URL of the Jina API.
 
             The support of `task` and `late_chunking` parameters is only available for jina-embeddings-v3.
         """
@@ -68,6 +71,7 @@ def __init__(
 
         self.api_key = api_key
         self.model_name = model
+        self.base_url = base_url
         self.prefix = prefix
         self.suffix = suffix
         self._session = requests.Session()
@@ -98,6 +102,7 @@ def to_dict(self) -> dict[str, Any]:
         kwargs: dict[str, Any] = {
             "api_key": self.api_key.to_dict(),
             "model": self.model_name,
+            "base_url": self.base_url,
             "prefix": self.prefix,
             "suffix": self.suffix,
         }
@@ -152,7 +157,7 @@ def run(self, text: str) -> dict[str, Any]:
             parameters["late_chunking"] = self.late_chunking
 
         resp = self._session.post(
-            JINA_API_URL,
+            self.base_url,
             json={"input": [text_to_embed], "model": self.model_name, **parameters},
         ).json()
 

integrations/jina/src/haystack_integrations/components/rankers/jina/ranker.py

@@ -36,6 +36,8 @@ def __init__(
         api_key: Secret = Secret.from_env_var("JINA_API_KEY"),  # noqa: B008,
         top_k: int | None = None,
         score_threshold: float | None = None,
+        *,
+        base_url: str = JINA_API_URL,
     ) -> None:
         """
         Creates an instance of JinaRanker.
@@ -47,6 +49,7 @@ def __init__(
             The maximum number of Documents to return per query. If `None`, all documents are returned
         :param score_threshold:
             If provided only returns documents with a score above this threshold.
+        :param base_url: The base URL of the Jina API.
 
         :raises ValueError:
             If `top_k` is not > 0.
@@ -57,6 +60,7 @@ def __init__(
         self.model = model
         self.top_k = top_k
         self.score_threshold = score_threshold
+        self.base_url = base_url
 
         if self.top_k is not None and self.top_k <= 0:
             msg = f"top_k must be > 0, but got {top_k}"
@@ -82,6 +86,7 @@ def to_dict(self) -> dict[str, Any]:
             self,
             api_key=self.api_key.to_dict(),
             model=self.model,
+            base_url=self.base_url,
             top_k=self.top_k,
             score_threshold=self.score_threshold,
         )
@@ -149,7 +154,7 @@ def run(
         }
 
         resp = self._session.post(
-            JINA_API_URL,
+            self.base_url,
             json=data,
         ).json()
 

integrations/jina/tests/test_document_embedder.py

@@ -33,6 +33,7 @@ def test_init_default(self, monkeypatch):
 
         assert embedder.api_key == Secret.from_env_var("JINA_API_KEY")
         assert embedder.model_name == "jina-embeddings-v3"
+        assert embedder.base_url == "https://api.jina.ai/v1/embeddings"
         assert embedder.prefix == ""
         assert embedder.suffix == ""
         assert embedder.batch_size == 32
@@ -44,6 +45,7 @@ def test_init_with_parameters(self):
         embedder = JinaDocumentEmbedder(
             api_key=Secret.from_token("fake-api-key"),
             model="model",
+            base_url="https://my.custom.url/v1/embeddings",
             prefix="prefix",
             suffix="suffix",
             batch_size=64,
@@ -57,6 +59,7 @@ def test_init_with_parameters(self):
 
         assert embedder.api_key == Secret.from_token("fake-api-key")
         assert embedder.model_name == "model"
+        assert embedder.base_url == "https://my.custom.url/v1/embeddings"
         assert embedder.prefix == "prefix"
         assert embedder.suffix == "suffix"
         assert embedder.batch_size == 64
@@ -81,6 +84,7 @@ def test_to_dict(self, monkeypatch):
             "init_parameters": {
                 "api_key": {"env_vars": ["JINA_API_KEY"], "strict": True, "type": "env_var"},
                 "model": "jina-embeddings-v3",
+                "base_url": "https://api.jina.ai/v1/embeddings",
                 "prefix": "",
                 "suffix": "",
                 "batch_size": 32,
@@ -94,6 +98,7 @@ def test_to_dict_with_custom_init_parameters(self, monkeypatch):
         monkeypatch.setenv("JINA_API_KEY", "fake-api-key")
         component = JinaDocumentEmbedder(
             model="model",
+            base_url="https://my.custom.url/v1/embeddings",
             prefix="prefix",
             suffix="suffix",
             batch_size=64,
@@ -109,6 +114,7 @@ def test_to_dict_with_custom_init_parameters(self, monkeypatch):
             "init_parameters": {
                 "api_key": {"env_vars": ["JINA_API_KEY"], "strict": True, "type": "env_var"},
                 "model": "model",
+                "base_url": "https://my.custom.url/v1/embeddings",
                 "prefix": "prefix",
                 "suffix": "suffix",
                 "batch_size": 64,

integrations/jina/tests/test_document_image_embedder.py

@@ -20,6 +20,7 @@ def test_init_default(self, monkeypatch):
         monkeypatch.setenv("JINA_API_KEY", "fake-api-key")
         embedder = JinaDocumentImageEmbedder()
         assert embedder.model_name == "jina-clip-v2"
+        assert embedder.base_url == "https://api.jina.ai/v1/embeddings"
         assert embedder.file_path_meta_field == "file_path"
         assert embedder.root_path == ""
         assert embedder.embedding_dimension is None
@@ -31,13 +32,15 @@ def test_init_with_parameters(self):
         embedder = JinaDocumentImageEmbedder(
             api_key=Secret.from_token("fake-api-token"),
             model="jina-embeddings-v4",
+            base_url="https://my.custom.url/v1/embeddings",
             file_path_meta_field="custom_file_path",
             root_path="/custom/root",
             embedding_dimension=256,
             image_size=(512, 512),
             batch_size=5,
         )
         assert embedder.model_name == "jina-embeddings-v4"
+        assert embedder.base_url == "https://my.custom.url/v1/embeddings"
         assert embedder.file_path_meta_field == "custom_file_path"
         assert embedder.root_path == "/custom/root"
         assert embedder.embedding_dimension == 256
@@ -50,6 +53,7 @@ def test_to_dict(self, monkeypatch):
         component = JinaDocumentImageEmbedder(
             api_key=Secret.from_env_var("JINA_API_KEY"),
             model="jina-clip-v2",
+            base_url="https://api.jina.ai/v1/embeddings",
             file_path_meta_field="image_path",
             root_path="/images",
             embedding_dimension=512,
@@ -62,6 +66,7 @@ def test_to_dict(self, monkeypatch):
             "init_parameters": {
                 "api_key": {"env_vars": ["JINA_API_KEY"], "strict": True, "type": "env_var"},
                 "model": "jina-clip-v2",
+                "base_url": "https://api.jina.ai/v1/embeddings",
                 "file_path_meta_field": "image_path",
                 "root_path": "/images",
                 "embedding_dimension": 512,
@@ -78,6 +83,7 @@ def test_from_dict(self, monkeypatch):
             "init_parameters": {
                 "api_key": {"env_vars": ["JINA_API_KEY"], "strict": True, "type": "env_var"},
                 "model": "jina-clip-v2",
+                "base_url": "https://api.jina.ai/v1/embeddings",
                 "file_path_meta_field": "image_path",
                 "root_path": "/images",
                 "embedding_dimension": 512,
@@ -87,6 +93,7 @@ def test_from_dict(self, monkeypatch):
         }
         component = JinaDocumentImageEmbedder.from_dict(data)
         assert component.model_name == "jina-clip-v2"
+        assert component.base_url == "https://api.jina.ai/v1/embeddings"
         assert component.file_path_meta_field == "image_path"
         assert component.root_path == "/images"
         assert component.embedding_dimension == 512

integrations/jina/tests/test_ranker.py

@@ -38,10 +38,17 @@ def test_init_default(self, monkeypatch):
         assert embedder.model == "jina-reranker-v1-base-en"
 
     def test_init_with_parameters(self):
-        embedder = JinaRanker(api_key=Secret.from_token("fake-api-key"), model="model", top_k=64, score_threshold=0.5)
+        embedder = JinaRanker(
+            api_key=Secret.from_token("fake-api-key"),
+            model="model",
+            base_url="https://my.custom.url/v1/rerank",
+            top_k=64,
+            score_threshold=0.5,
+        )
 
         assert embedder.api_key == Secret.from_token("fake-api-key")
         assert embedder.model == "model"
+        assert embedder.base_url == "https://my.custom.url/v1/rerank"
         assert embedder.top_k == 64
         assert embedder.score_threshold == 0.5
 
@@ -59,20 +66,22 @@ def test_to_dict(self, monkeypatch):
             "init_parameters": {
                 "api_key": {"env_vars": ["JINA_API_KEY"], "strict": True, "type": "env_var"},
                 "model": "jina-reranker-v1-base-en",
+                "base_url": "https://api.jina.ai/v1/rerank",
                 "top_k": None,
                 "score_threshold": None,
             },
         }
 
     def test_to_dict_with_custom_init_parameters(self, monkeypatch):
         monkeypatch.setenv("JINA_API_KEY", "fake-api-key")
-        component = JinaRanker(model="model", top_k=64, score_threshold=0.5)
+        component = JinaRanker(model="model", top_k=64, score_threshold=0.5, base_url="https://my.custom.url/v1/rerank")
         data = component.to_dict()
         assert data == {
             "type": "haystack_integrations.components.rankers.jina.ranker.JinaRanker",
             "init_parameters": {
                 "api_key": {"env_vars": ["JINA_API_KEY"], "strict": True, "type": "env_var"},
                 "model": "model",
+                "base_url": "https://my.custom.url/v1/rerank",
                 "top_k": 64,
                 "score_threshold": 0.5,
             },

integrations/jina/tests/test_text_embedder.py

@@ -19,19 +19,22 @@ def test_init_default(self, monkeypatch):
 
         assert embedder.api_key == Secret.from_env_var("JINA_API_KEY")
         assert embedder.model_name == "jina-embeddings-v3"
+        assert embedder.base_url == "https://api.jina.ai/v1/embeddings"
         assert embedder.prefix == ""
         assert embedder.suffix == ""
 
     def test_init_with_parameters(self):
         embedder = JinaTextEmbedder(
             api_key=Secret.from_token("fake-api-key"),
             model="model",
+            base_url="https://my.custom.url/v1/embeddings",
             prefix="prefix",
             suffix="suffix",
             late_chunking=True,
         )
         assert embedder.api_key == Secret.from_token("fake-api-key")
         assert embedder.model_name == "model"
+        assert embedder.base_url == "https://my.custom.url/v1/embeddings"
         assert embedder.prefix == "prefix"
         assert embedder.suffix == "suffix"
         assert embedder.late_chunking is True
@@ -50,6 +53,7 @@ def test_to_dict(self, monkeypatch):
             "init_parameters": {
                 "api_key": {"env_vars": ["JINA_API_KEY"], "strict": True, "type": "env_var"},
                 "model": "jina-embeddings-v3",
+                "base_url": "https://api.jina.ai/v1/embeddings",
                 "prefix": "",
                 "suffix": "",
             },
@@ -59,6 +63,7 @@ def test_to_dict_with_custom_init_parameters(self, monkeypatch):
         monkeypatch.setenv("JINA_API_KEY", "fake-api-key")
         component = JinaTextEmbedder(
             model="model",
+            base_url="https://my.custom.url/v1/embeddings",
             prefix="prefix",
             suffix="suffix",
             task="retrieval.query",
@@ -70,6 +75,7 @@ def test_to_dict_with_custom_init_parameters(self, monkeypatch):
             "init_parameters": {
                 "api_key": {"env_vars": ["JINA_API_KEY"], "strict": True, "type": "env_var"},
                 "model": "model",
+                "base_url": "https://my.custom.url/v1/embeddings",
                 "prefix": "prefix",
                 "suffix": "suffix",
                 "task": "retrieval.query",