More robust extract tests with pytest xdist

.github/workflows/hourly_extract_e2e.yml

@@ -1,8 +1,8 @@
-name: Hourly Extract E2E Tests
+name: Extract E2E Tests (every 4 hours)
 
 on:
   schedule:
-    - cron: "18 * * * *"
+    - cron: "0 */4 * * *"
   workflow_dispatch:
     # Allows manual triggering
     inputs:
@@ -29,7 +29,7 @@ env:
 
 jobs:
   extract-e2e:
-    name: "Hourly Extract E2E Tests (${{ matrix.environment }})"
+    name: "Extract E2E Tests (${{ matrix.environment }})"
     runs-on: ubuntu-latest
     timeout-minutes: 30
     concurrency:

py/CHANGELOG.md

@@ -1,5 +1,11 @@
 # llama-cloud-services-py
 
+## 0.6.93
+
+### Patch Changes
+
+- da1916c: Add more warnings
+
 ## 0.6.92
 
 ### Patch Changes

py/README.md

@@ -4,6 +4,16 @@
 
 # Llama Cloud Services
 
+> **⚠️ DEPRECATION NOTICE**
+>
+> This repository and its packages are deprecated and will be maintained until **May 1, 2026**.
+>
+> **Please migrate to the new packages:**
+> - **Python**: `pip install llama-cloud>=1.0` ([GitHub](https://github.com/run-llama/llama-cloud-py))
+> - **TypeScript**: `npm install @llamaindex/llama-cloud` ([GitHub](https://github.com/run-llama/llama-cloud-ts))
+>
+> The new packages provide the same functionality with improved performance, better support, and active development.
+
 This repository contains the code for hand-written SDKs and clients for interacting with LlamaCloud.
 
 This includes:

py/llama_cloud_services/extract/extract.py

@@ -806,6 +806,7 @@ class LlamaExtract(BaseComponent):
             # Document files
             ".pdf": "application/pdf",
             ".docx": "application/vnd.openxmlformats-officedocument.wordprocessingml.document",
+            ".xlsx": "application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
             # Image files
             ".png": "image/png",
             ".jpg": "image/jpeg",

py/llama_parse/CHANGELOG.md

@@ -1,5 +1,13 @@
 # llama_parse
 
+## 0.6.93
+
+### Patch Changes
+
+- da1916c: Add more warnings
+- Updated dependencies [da1916c]
+  - llama-cloud-services-py@0.6.93
+
 ## 0.6.92
 
 ### Patch Changes

py/llama_parse/README.md

@@ -1,5 +1,15 @@
 # LlamaParse
 
+> **⚠️ DEPRECATION NOTICE**
+>
+> This repository and its packages are deprecated and will be maintained until **May 1, 2026**.
+>
+> **Please migrate to the new packages:**
+> - **Python**: `pip install llama-cloud>=1.0` ([GitHub](https://github.com/run-llama/llama-cloud-py))
+> - **TypeScript**: `npm install @llamaindex/llama-cloud` ([GitHub](https://github.com/run-llama/llama-cloud-ts))
+>
+> The new packages provide the same functionality with improved performance, better support, and active development.
+
 [![PyPI - Downloads](https://img.shields.io/pypi/dm/llama-parse)](https://pypi.org/project/llama-parse/)
 [![GitHub contributors](https://img.shields.io/github/contributors/run-llama/llama_parse)](https://github.com/run-llama/llama_parse/graphs/contributors)
 [![Discord](https://img.shields.io/discord/1059199217496772688)](https://discord.gg/dGcwcsnxhU)

py/llama_parse/llama_parse/__init__.py

@@ -1,8 +1,18 @@
-from llama_cloud_services.parse import (
+import warnings
+from llama_cloud_services.parse import (  # type: ignore[attr-defined]
     LlamaParse,
     ResultType,
     ParsingMode,
     FailedPageMode,
 )
 
+warnings.warn(
+    "The 'llama-parse' package is deprecated and will no longer receive updates. "
+    "Please migrate to the new unified SDK. "
+    "See https://developers.llamaindex.ai/python/cloud/llamaparse/getting_started/ "
+    "and https://github.com/run-llama/llama-cloud-py/blob/main/README.md for migration instructions.",
+    DeprecationWarning,
+    stacklevel=2,
+)
+
 __all__ = ["LlamaParse", "ResultType", "ParsingMode", "FailedPageMode"]

py/llama_parse/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llama_parse",
-  "version": "0.6.92",
+  "version": "0.6.93",
   "description": "",
   "main": "index.js",
   "private": false,

py/llama_parse/pyproject.toml

@@ -11,13 +11,13 @@ dev = [
 
 [project]
 name = "llama-parse"
-version = "0.6.92"
+version = "0.6.93"
 description = "Parse files into RAG-Optimized formats."
 authors = [{name = "Logan Markewich", email = "logan@llamaindex.ai"}]
 requires-python = ">=3.9,<4.0"
 readme = "README.md"
 license = "MIT"
-dependencies = ["llama-cloud-services>=0.6.92"]
+dependencies = ["llama-cloud-services>=0.6.93"]
 
 [project.scripts]
 llama-parse = "llama_parse.cli.main:parse"

py/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llama-cloud-services-py",
-  "version": "0.6.92",
+  "version": "0.6.93",
   "private": false,
   "license": "MIT",
   "scripts": {},

py/pyproject.toml

@@ -23,7 +23,7 @@ dev = [
 
 [project]
 name = "llama-cloud-services"
-version = "0.6.92"
+version = "0.6.93"
 description = "Tailored SDK clients for LlamaCloud services."
 authors = [{name = "Logan Markewich", email = "logan@runllama.ai"}]
 requires-python = ">=3.9,<4.0"

py/tests/extract/conftest.py

@@ -1,5 +1,4 @@
-import os
-from typing import Any, Dict, List, Optional, Union
+from typing import Any, Dict, Optional, Union
 
 from llama_cloud.core.api_error import ApiError
 from llama_cloud.types import ExtractConfig
@@ -13,9 +12,6 @@ from tenacity import (
 
 from llama_cloud_services.extract import ExtractionAgent, LlamaExtract
 
-# Global storage for agents to cleanup
-_TEST_AGENTS_TO_CLEANUP: List[str] = []
-
 
 def _is_rate_limit_error(exception: BaseException) -> bool:
     """Check if the exception is a rate limit error (429)."""
@@ -42,38 +38,3 @@ def pytest_configure(config):
     """Register custom markers for extract tests."""
     config.addinivalue_line("markers", "agent_name: custom agent name for test")
     config.addinivalue_line("markers", "agent_schema: custom agent schema for test")
-
-
-def pytest_sessionfinish(session, exitstatus):
-    """Hook that runs after all tests complete - cleanup agents here"""
-    print(
-        f"pytest_sessionfinish hook called! Agents to cleanup: {_TEST_AGENTS_TO_CLEANUP}"
-    )
-
-    if _TEST_AGENTS_TO_CLEANUP:
-        print("Creating cleanup client...")
-        # Create a fresh client just for cleanup
-        cleanup_client = LlamaExtract(
-            api_key=os.getenv("LLAMA_CLOUD_API_KEY"),
-            base_url=os.getenv("LLAMA_CLOUD_BASE_URL"),
-            project_id=os.getenv("LLAMA_CLOUD_PROJECT_ID"),
-            verbose=True,
-        )
-
-        for agent_id in _TEST_AGENTS_TO_CLEANUP:
-            try:
-                print(f"Deleting agent {agent_id}...")
-                cleanup_client.delete_agent(agent_id)
-                print(f"Cleaned up agent {agent_id}")
-            except Exception as e:
-                print(f"Warning: Failed to delete agent {agent_id}: {e}")
-
-        _TEST_AGENTS_TO_CLEANUP.clear()
-        print("Agent cleanup completed")
-    else:
-        print("No agents to cleanup")
-
-
-def register_agent_for_cleanup(agent_id: str):
-    """Register an agent ID for cleanup at the end of the test session"""
-    _TEST_AGENTS_TO_CLEANUP.append(agent_id)

py/tests/extract/test_extract_api.py

@@ -1,4 +1,6 @@
 import os
+import shutil
+import uuid
 import pytest
 from pathlib import Path
 from pydantic import BaseModel
@@ -6,7 +8,7 @@ from pydantic import BaseModel
 from llama_cloud_services.extract import LlamaExtract, ExtractionAgent, SourceText
 from llama_cloud.types import ExtractConfig, ExtractMode, ExtractRun
 from tests.extract.util import load_test_dotenv
-from .conftest import register_agent_for_cleanup, create_agent_with_retry
+from .conftest import create_agent_with_retry
 
 load_test_dotenv()
 
@@ -59,17 +61,27 @@ def test_schema_dict():
 
 
 @pytest.fixture
-def test_agent(llama_extract, test_agent_name, test_schema_dict, request):
-    """Creates a test agent and collects it for cleanup at the end of all tests"""
-    test_id = request.node.nodeid
-    test_hash = hex(hash(test_id))[-8:]
-    base_name = test_agent_name
+def unique_test_pdf(tmp_path):
+    """Copy test PDF to a unique path to avoid file deduplication across parallel tests.
 
+    Uses a UUID in the filename so that external_file_id is unique regardless of
+    whether the full path or just the filename is sent to the backend.
+    """
+    unique_name = f"{TEST_PDF.stem}-{uuid.uuid4().hex[:8]}{TEST_PDF.suffix}"
+    unique_pdf = tmp_path / unique_name
+    shutil.copy2(TEST_PDF, unique_pdf)
+    return unique_pdf
+
+
+@pytest.fixture
+def test_agent(llama_extract, test_agent_name, test_schema_dict, request):
+    """Creates a test agent with a unique name and cleans it up after the test."""
+    unique_id = uuid.uuid4().hex[:8]
     base_name = next(
         (marker.args[0] for marker in request.node.iter_markers("agent_name")),
-        base_name,
+        test_agent_name,
     )
-    name = f"{base_name}_{test_hash}"
+    name = f"{base_name}_{unique_id}"
 
     schema = next(
         (
@@ -79,25 +91,20 @@ def test_agent(llama_extract, test_agent_name, test_schema_dict, request):
         test_schema_dict,
     )
 
-    # Cleanup existing agent
-    try:
-        for agent in llama_extract.list_agents():
-            if agent.name == name:
-                llama_extract.delete_agent(agent.id)
-    except Exception as e:
-        print(f"Warning: Failed to cleanup existing agent: {e}")
-
     # Use config with cache invalidation to ensure fresh results in tests
     config = ExtractConfig(invalidate_cache=True)
     agent = create_agent_with_retry(
         llama_extract, name=name, data_schema=schema, config=config
     )
 
-    # Add agent to cleanup list via conftest helper
-    register_agent_for_cleanup(agent.id)
-
     yield agent
 
+    # Inline cleanup -- each worker cleans up its own agents
+    try:
+        llama_extract.delete_agent(agent.id)
+    except Exception as e:
+        print(f"Warning: Failed to cleanup agent {agent.id}: {e}")
+
 
 class TestLlamaExtract:
     def test_init_without_api_key(self):
@@ -138,34 +145,38 @@ class TestLlamaExtract:
 
 class TestExtractionAgent:
     @pytest.mark.asyncio
-    async def test_extract_single_file(self, test_agent):
-        result = await test_agent.aextract(TEST_PDF)
+    async def test_extract_single_file(self, test_agent, unique_test_pdf):
+        result = await test_agent.aextract(unique_test_pdf)
         assert result.status == "SUCCESS"
         assert result.data is not None
         assert isinstance(result.data, dict)
         assert "title" in result.data
         assert "summary" in result.data
 
-    def test_sync_extract_single_file(self, test_agent):
-        result = test_agent.extract(TEST_PDF)
+    def test_sync_extract_single_file(self, test_agent, unique_test_pdf):
+        result = test_agent.extract(unique_test_pdf)
         assert result.status == "SUCCESS"
         assert result.data is not None
         assert isinstance(result.data, dict)
         assert "title" in result.data
         assert "summary" in result.data
 
-    def test_extract_file_from_buffered_io(self, test_agent):
-        result = test_agent.extract(SourceText(file=open(TEST_PDF, "rb")))
+    def test_extract_file_from_buffered_io(self, test_agent, unique_test_pdf):
+        result = test_agent.extract(
+            SourceText(file=open(unique_test_pdf, "rb"), filename=unique_test_pdf.name)
+        )
         assert result.status == "SUCCESS"
         assert result.data is not None
         assert isinstance(result.data, dict)
         assert "title" in result.data
         assert "summary" in result.data
 
-    def test_extract_file_from_bytes(self, test_agent):
-        with open(TEST_PDF, "rb") as f:
+    def test_extract_file_from_bytes(self, test_agent, unique_test_pdf):
+        with open(unique_test_pdf, "rb") as f:
             file_bytes = f.read()
-        result = test_agent.extract(SourceText(file=file_bytes, filename=TEST_PDF.name))
+        result = test_agent.extract(
+            SourceText(file=file_bytes, filename=unique_test_pdf.name)
+        )
         assert result.status == "SUCCESS"
         assert result.data is not None
         assert isinstance(result.data, dict)
@@ -181,7 +192,10 @@ class TestExtractionAgent:
         weight for 8 to 13 km (5–8 miles).[3] The name llama (also historically spelled
         "glama") was adopted by European settlers from native Peruvians.
         """
-        result = test_agent.extract(SourceText(text_content=TEST_TEXT))
+        unique_name = f"text-{uuid.uuid4().hex[:8]}.txt"
+        result = test_agent.extract(
+            SourceText(text_content=TEST_TEXT, filename=unique_name)
+        )
         assert result.status == "SUCCESS"
         assert result.data is not None
         assert isinstance(result.data, dict)
@@ -189,8 +203,8 @@ class TestExtractionAgent:
         assert "summary" in result.data
 
     @pytest.mark.asyncio
-    async def test_extract_multiple_files(self, test_agent):
-        files = [TEST_PDF, TEST_PDF]  # Using same file twice for testing
+    async def test_extract_multiple_files(self, test_agent, unique_test_pdf):
+        files = [unique_test_pdf, unique_test_pdf]  # Using same file twice for testing
         response = await test_agent.aextract(files)
 
         assert len(response) == 2
@@ -219,15 +233,15 @@ class TestExtractionAgent:
         updated_agent = llama_extract.get_agent(name=test_agent.name)
         assert "new_field" in updated_agent.data_schema["properties"]
 
-    def test_list_extraction_runs(self, test_agent: ExtractionAgent):
+    def test_list_extraction_runs(self, test_agent: ExtractionAgent, unique_test_pdf):
         assert test_agent.list_extraction_runs().total == 0
-        test_agent.extract(TEST_PDF)
+        test_agent.extract(unique_test_pdf)
         runs = test_agent.list_extraction_runs()
         assert runs.total > 0
 
-    def test_delete_extraction_run(self, test_agent: ExtractionAgent):
+    def test_delete_extraction_run(self, test_agent: ExtractionAgent, unique_test_pdf):
         assert test_agent.list_extraction_runs().total == 0
-        run: ExtractRun = test_agent.extract(TEST_PDF)
+        run: ExtractRun = test_agent.extract(unique_test_pdf)
         test_agent.delete_extraction_run(run.id)
         runs = test_agent.list_extraction_runs()
         assert runs.total == 0

py/tests/extract/test_extract_e2e.py

@@ -10,7 +10,7 @@ import uuid
 from llama_cloud.types import ExtractConfig, ExtractMode
 from deepdiff import DeepDiff
 from tests.extract.util import json_subset_match_score, load_test_dotenv
-from .conftest import register_agent_for_cleanup, create_agent_with_retry
+from .conftest import create_agent_with_retry
 
 load_test_dotenv()
 
@@ -109,32 +109,24 @@ def extractor():
 @pytest.fixture
 def extraction_agent(test_case: ExtractionTestCase, extractor: LlamaExtract):
     """Fixture to create and cleanup extraction agent for each test."""
-    # Create unique name with random UUID (important for CI to avoid conflicts)
     unique_id = uuid.uuid4().hex[:8]
     agent_name = f"{test_case.name}_{unique_id}"
 
     with open(test_case.schema_path, "r") as f:
         schema = json.load(f)
 
-    # Clean up any existing agents with this name
-    try:
-        agents = extractor.list_agents()
-        for agent in agents:
-            if agent.name == agent_name:
-                extractor.delete_agent(agent.id)
-    except Exception as e:
-        print(f"Warning: Failed to cleanup existing agent: {str(e)}")
-
-    # Create new agent with retry logic for rate limiting
     agent = create_agent_with_retry(
         extractor, name=agent_name, data_schema=schema, config=test_case.config
     )
 
-    # Register agent for cleanup at the end of the test session
-    register_agent_for_cleanup(agent.id)
-
     yield agent
 
+    # Inline cleanup -- each worker cleans up its own agents
+    try:
+        extractor.delete_agent(agent.id)
+    except Exception as e:
+        print(f"Warning: Failed to cleanup agent {agent.id}: {e}")
+
 
 @pytest.mark.skipif(
     os.environ.get("LLAMA_CLOUD_API_KEY", "") == "",