fix: make base url as part of params for jina document embedder

Author: srini047

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

@@ -41,6 +41,7 @@ def __init__(
         self,
         api_key: Secret = Secret.from_env_var("JINA_API_KEY"),  # noqa: B008
         model: str = "jina-embeddings-v3",
+        base_url: str = JINA_API_URL,
         prefix: str = "",
         suffix: str = "",
         batch_size: int = 32,
@@ -57,6 +58,7 @@ def __init__(
         :param api_key: The Jina API key.
         :param model: The name of the Jina model to use.
             Check the list of available models on [Jina documentation](https://jina.ai/embeddings/).
+        :param base_url: The base URL of the Jina API.
         :param prefix: A string to add to the beginning of each text.
         :param suffix: A string to add to the end of each text.
         :param batch_size: Number of Documents to encode at once.
@@ -79,6 +81,7 @@ def __init__(
 
         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 +116,7 @@ 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 +177,7 @@ 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/text_embedder.py

@@ -51,6 +51,7 @@ def __init__(
             environment variable `JINA_API_KEY` (recommended).
         :param model: The name of the Jina model to use.
             Check the list of available models on [Jina documentation](https://jina.ai/embeddings/).
+        :param base_url: The base URL of the Jina API.
         :param prefix: A string to add to the beginning of each text.
         :param suffix: A string to add to the end of each text.
         :param task: The downstream task for which the embeddings will be used.

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="http://api.jina.ai/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 == "http://api.jina.ai/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="http://api.jina.ai/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": "http://api.jina.ai/v1/embeddings",
                 "prefix": "prefix",
                 "suffix": "suffix",
                 "batch_size": 64,