fix: Make env based api url overrideable (#881)

Author: adrianlyjak

py/llama_cloud_services/parse/base.py

@@ -17,7 +17,6 @@
 from llama_index.core.bridge.pydantic import (
     Field,
     PrivateAttr,
-    field_validator,
     model_validator,
 )
 from llama_index.core.constants import DEFAULT_BASE_URL
@@ -105,17 +104,29 @@ class BackoffPattern(str, Enum):
     EXPONENTIAL = "exponential"
 
 
+def _get_default_api_key() -> str:
+    env_key = os.getenv("LLAMA_CLOUD_API_KEY")
+    if env_key is None:
+        raise ValueError("The API key is required.")
+    return env_key
+
+
+def _get_default_base_url() -> str:
+    env_url = os.getenv("LLAMA_CLOUD_BASE_URL")
+    return env_url or DEFAULT_BASE_URL
+
+
 class LlamaParse(BasePydanticReader):
     """A smart-parser for files."""
 
     # Library / access specific configurations
     api_key: str = Field(
-        default="",
+        default_factory=_get_default_api_key,
         description="The API key for the LlamaParse API.",
         validate_default=True,
     )
     base_url: str = Field(
-        default=DEFAULT_BASE_URL,
+        default_factory=_get_default_base_url,
         description="The base URL of the Llama Parsing API.",
     )
     organization_id: Optional[str] = Field(
@@ -561,27 +572,6 @@ def warn_extra_params(cls, data: Dict[str, Any]) -> Dict[str, Any]:
 
         return data
 
-    @field_validator("api_key", mode="before", check_fields=True)
-    @classmethod
-    def validate_api_key(cls, v: str) -> str:
-        """Validate the API key."""
-        if not v:
-            import os
-
-            api_key = os.getenv("LLAMA_CLOUD_API_KEY", None)
-            if api_key is None:
-                raise ValueError("The API key is required.")
-            return api_key
-
-        return v
-
-    @field_validator("base_url", mode="before", check_fields=True)
-    @classmethod
-    def validate_base_url(cls, v: str) -> str:
-        """Validate the base URL."""
-        url = os.getenv("LLAMA_CLOUD_BASE_URL", None)
-        return url or v or DEFAULT_BASE_URL
-
     _aclient: Union[httpx.AsyncClient, None] = PrivateAttr(default=None, init=False)
 
     @property

py/llama_parse/pyproject.toml

@@ -11,13 +11,13 @@ dev = [
 
 [project]
 name = "llama-parse"
-version = "0.6.60"
+version = "0.6.62"
 description = "Parse files into RAG-Optimized formats."
 authors = [{name = "Logan Markewich", email = "[email protected]"}]
 requires-python = ">=3.9,<4.0"
 readme = "README.md"
 license = "MIT"
-dependencies = ["llama-cloud-services>=0.6.60"]
+dependencies = ["llama-cloud-services>=0.6.62"]
 
 [project.scripts]
 llama-parse = "llama_parse.cli.main:parse"

py/pyproject.toml

@@ -19,7 +19,7 @@ dev = [
 
 [project]
 name = "llama-cloud-services"
-version = "0.6.61"
+version = "0.6.62"
 description = "Tailored SDK clients for LlamaCloud services."
 authors = [{name = "Logan Markewich", email = "[email protected]"}]
 requires-python = ">=3.9,<4.0"

py/unit_tests/parse/test_base.py

@@ -0,0 +1,57 @@
+from collections.abc import Generator
+import pytest
+import os
+
+from llama_cloud_services.parse.base import LlamaParse
+
+
+@pytest.fixture(autouse=True)
+def clear_env() -> Generator[None, None, None]:
+    """entirely clear the environment, and then reset after test completion"""
+    original_env = os.environ.copy()
+    os.environ.clear()
+    try:
+        yield
+    finally:
+        os.environ.clear()
+        os.environ.update(original_env)
+
+
+def test_should_obtain_api_key_base_url_from_env_vars():
+    os.environ["LLAMA_CLOUD_API_KEY"] = "test-api-key"
+    os.environ["LLAMA_CLOUD_BASE_URL"] = "https://example.test"
+
+    client = LlamaParse()
+    assert client.api_key == "test-api-key"
+    assert client.base_url == "https://example.test"
+
+
+def test_should_be_able_to_pass_api_key_base_url_as_kwargs():
+    os.environ["LLAMA_CLOUD_API_KEY"] = "not-this-one"
+    os.environ["LLAMA_CLOUD_BASE_URL"] = "https://wrong.site"
+
+    client = LlamaParse(
+        api_key="test-api-key",
+        base_url="https://example.test",
+    )
+    assert client.api_key == "test-api-key"
+    assert client.base_url == "https://example.test"
+
+
+def test_should_raise_error_if_api_key_is_not_provided():
+    with pytest.raises(ValueError, match="The API key is required."):
+        LlamaParse()
+
+
+def test_should_default_to_llama_cloud_base_url_if_not_provided():
+    client = LlamaParse(
+        api_key="test-api-key",
+    )
+    assert client.base_url == "https://api.cloud.llamaindex.ai"
+
+
+def test_json_parseable():
+    os.environ["LLAMA_CLOUD_API_KEY"] = "test-api-key"
+    client2 = LlamaParse.model_validate_json('{"base_url":"https://example.test"}')
+    assert client2.api_key == "test-api-key"
+    assert client2.base_url == "https://example.test"