From 2faa45162c50eef3c65f3ce7c45980030cdcc5ae Mon Sep 17 00:00:00 2001
From: rafaelmmiller <8574157+rafaelmmiller@users.noreply.github.com>
Date: Fri, 30 May 2025 10:16:44 -0300
Subject: [PATCH] sdk(v3): all tests complete

---
 apps/python-sdk/firecrawl/types.py            |   2 +-
 ...scrape.py => test_a_async_batch_scrape.py} |   0
 ...t_async_crawl.py => test_a_async_crawl.py} |   0
 ...batch_scrape.py => test_a_batch_scrape.py} |   0
 .../async/{test_crawl.py => test_a_crawl.py}  |   0
 .../tests/e2e/async/test_a_extract.py         | 520 ++++++++++++++++++
 .../e2e/async/{test_map.py => test_a_map.py}  |   0
 .../{test_scrape.py => test_a_scrape.py}      |   0
 .../tests/e2e/async/test_extract.py           |  99 ----
 apps/python-sdk/tests/e2e/test_extract.py     | 404 +++++++++++++-
 10 files changed, 904 insertions(+), 121 deletions(-)
 rename apps/python-sdk/tests/e2e/async/{test_async_batch_scrape.py => test_a_async_batch_scrape.py} (100%)
 rename apps/python-sdk/tests/e2e/async/{test_async_crawl.py => test_a_async_crawl.py} (100%)
 rename apps/python-sdk/tests/e2e/async/{test_batch_scrape.py => test_a_batch_scrape.py} (100%)
 rename apps/python-sdk/tests/e2e/async/{test_crawl.py => test_a_crawl.py} (100%)
 create mode 100644 apps/python-sdk/tests/e2e/async/test_a_extract.py
 rename apps/python-sdk/tests/e2e/async/{test_map.py => test_a_map.py} (100%)
 rename apps/python-sdk/tests/e2e/async/{test_scrape.py => test_a_scrape.py} (100%)
 delete mode 100644 apps/python-sdk/tests/e2e/async/test_extract.py

diff --git a/apps/python-sdk/firecrawl/types.py b/apps/python-sdk/firecrawl/types.py
index ab6e22e3..e0ee7b8d 100644
--- a/apps/python-sdk/firecrawl/types.py
+++ b/apps/python-sdk/firecrawl/types.py
@@ -279,7 +279,7 @@ class ExtractResponse(pydantic.BaseModel, Generic[T]):
     data: Optional[T] = None
     error: Optional[str] = None
     warning: Optional[str] = None
-    sources: Optional[List[str]] = None
+    sources: Optional[Union[List[str], Dict[str, List[str]]]] = None
 
 class SearchParams(pydantic.BaseModel):
     query: str
diff --git a/apps/python-sdk/tests/e2e/async/test_async_batch_scrape.py b/apps/python-sdk/tests/e2e/async/test_a_async_batch_scrape.py
similarity index 100%
rename from apps/python-sdk/tests/e2e/async/test_async_batch_scrape.py
rename to apps/python-sdk/tests/e2e/async/test_a_async_batch_scrape.py
diff --git a/apps/python-sdk/tests/e2e/async/test_async_crawl.py b/apps/python-sdk/tests/e2e/async/test_a_async_crawl.py
similarity index 100%
rename from apps/python-sdk/tests/e2e/async/test_async_crawl.py
rename to apps/python-sdk/tests/e2e/async/test_a_async_crawl.py
diff --git a/apps/python-sdk/tests/e2e/async/test_batch_scrape.py b/apps/python-sdk/tests/e2e/async/test_a_batch_scrape.py
similarity index 100%
rename from apps/python-sdk/tests/e2e/async/test_batch_scrape.py
rename to apps/python-sdk/tests/e2e/async/test_a_batch_scrape.py
diff --git a/apps/python-sdk/tests/e2e/async/test_crawl.py b/apps/python-sdk/tests/e2e/async/test_a_crawl.py
similarity index 100%
rename from apps/python-sdk/tests/e2e/async/test_crawl.py
rename to apps/python-sdk/tests/e2e/async/test_a_crawl.py
diff --git a/apps/python-sdk/tests/e2e/async/test_a_extract.py b/apps/python-sdk/tests/e2e/async/test_a_extract.py
new file mode 100644
index 00000000..25a8b654
--- /dev/null
+++ b/apps/python-sdk/tests/e2e/async/test_a_extract.py
@@ -0,0 +1,520 @@
+import os
+import sys
+import pytest
+from dotenv import load_dotenv
+from pydantic import BaseModel
+from typing import List
+
+sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '../../')))
+from firecrawl.firecrawl import AsyncFirecrawlApp
+
+load_dotenv()
+
+API_URL = os.getenv("TEST_API_URL") or "https://api.firecrawl.dev"
+API_KEY = os.getenv("TEST_API_KEY")
+
+app = AsyncFirecrawlApp(api_url=API_URL, api_key=API_KEY)
+
+class ExtractSchema(BaseModel):
+    title: str
+    description: str
+    links: List[str]
+
+
+@pytest.mark.asyncio
+async def test_extract_simple_schema_class():
+    result = await app.extract(
+        ["https://example.com"],
+        prompt="Extract the title, description, and links from the website",
+        schema=ExtractSchema
+    )
+    
+    # Basic response assertions
+    assert result is not None
+    assert result.success
+    assert hasattr(result, "data")
+    assert result.data is not None
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Warning handling assertions
+    assert hasattr(result, "warning")
+    # Warning can be present or None, so we just check it exists
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    
+    # Response metadata validation
+    assert hasattr(result, "id")
+    assert hasattr(result, "status")
+    if result.status is not None:
+        assert result.status in ["processing", "completed", "failed"]
+
+    # Sources validation
+    assert hasattr(result, "sources")
+    if result.sources is not None:
+        if isinstance(result.sources, dict):
+            # Sources returned as dictionary with keys like 'links[0]'
+            for key, source_list in result.sources.items():
+                assert isinstance(key, str)
+                assert isinstance(source_list, list)
+                for source_url in source_list:
+                    assert isinstance(source_url, str)
+                    assert len(source_url) > 0
+                    # Sources should be valid URLs
+                    assert source_url.startswith(("http://", "https://"))
+        elif isinstance(result.sources, list):
+            # Sources returned as simple list (fallback)
+            for source in result.sources:
+                assert isinstance(source, str)
+                assert len(source) > 0
+
+@pytest.mark.asyncio
+async def test_extract_simple_schema_dict():
+    result = await app.extract(
+        ["https://example.com"],
+        prompt="Extract the title, description, and links from the website",
+        schema=ExtractSchema.schema()
+    )
+    
+    # Basic response assertions
+    assert result is not None
+    assert result.success
+    assert hasattr(result, "data")
+    assert result.data is not None
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Warning handling assertions
+    assert hasattr(result, "warning")
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    
+    # Response metadata validation
+    assert hasattr(result, "id")
+    assert hasattr(result, "status")
+    if result.status is not None:
+        assert result.status in ["processing", "completed", "failed"]
+
+    # Sources validation
+    assert hasattr(result, "sources")
+    if result.sources is not None:
+        if isinstance(result.sources, dict):
+            # Sources returned as dictionary with keys like 'links[0]'
+            for key, source_list in result.sources.items():
+                assert isinstance(key, str)
+                assert isinstance(source_list, list)
+                for source_url in source_list:
+                    assert isinstance(source_url, str)
+                    assert len(source_url) > 0
+                    # Sources should be valid URLs
+                    assert source_url.startswith(("http://", "https://"))
+        elif isinstance(result.sources, list):
+            # Sources returned as simple list (fallback)
+            for source in result.sources:
+                assert isinstance(source, str)
+                assert len(source) > 0
+
+@pytest.mark.asyncio
+async def test_extract_simple_schema_json():
+    schema = None
+    if hasattr(ExtractSchema, "model_json_schema"):
+        schema = ExtractSchema.model_json_schema()
+    elif hasattr(ExtractSchema, "schema_json"):
+        schema = ExtractSchema.schema_json()
+    else:
+        pytest.skip("No JSON schema export method available on ExtractSchema")
+    
+    result = await app.extract(
+        ["https://example.com"],
+        prompt="Extract the title, description, and links from the website",
+        schema=schema
+    )
+    
+    # Basic response assertions
+    assert result is not None
+    assert result.success
+    assert hasattr(result, "data")
+    assert result.data is not None
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Warning handling assertions
+    assert hasattr(result, "warning")
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    
+    # Response metadata validation
+    assert hasattr(result, "id")
+    assert hasattr(result, "status")
+    if result.status is not None:
+        assert result.status in ["processing", "completed", "failed"]
+
+    # Sources validation
+    assert hasattr(result, "sources")
+    if result.sources is not None:
+        if isinstance(result.sources, dict):
+            # Sources returned as dictionary with keys like 'links[0]'
+            for key, source_list in result.sources.items():
+                assert isinstance(key, str)
+                assert isinstance(source_list, list)
+                for source_url in source_list:
+                    assert isinstance(source_url, str)
+                    assert len(source_url) > 0
+                    # Sources should be valid URLs
+                    assert source_url.startswith(("http://", "https://"))
+        elif isinstance(result.sources, list):
+            # Sources returned as simple list (fallback)
+            for source in result.sources:
+                assert isinstance(source, str)
+                assert len(source) > 0
+
+@pytest.mark.asyncio
+async def test_extract_all_params():
+    class AllParamsSchema(BaseModel):
+        title: str
+        description: str
+        links: List[str]
+        author: str
+    
+    schema = AllParamsSchema.schema()
+    result = await app.extract(
+        ["https://www.iana.org"],
+        prompt="Extract the title, description, links, and author from the website",
+        schema=schema,
+        system_prompt="You are a helpful extraction agent.",
+        allow_external_links=False,
+        enable_web_search=False,
+        show_sources=True,
+        # agent={"model": "FIRE-1"}
+    )
+    
+    # Basic response assertions
+    assert result is not None
+    assert result.success
+    assert hasattr(result, "data")
+    assert result.data is not None
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Warning handling assertions
+    assert hasattr(result, "warning")
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert "author" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert isinstance(result.data["author"], str)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert hasattr(result.data, "author")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert isinstance(result.data.author, str)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    
+    # Sources validation (show_sources=True)
+    assert hasattr(result, "sources")
+    if result.sources is not None:
+        if isinstance(result.sources, dict):
+            # Sources returned as dictionary with keys like 'links[0]'
+            for key, source_list in result.sources.items():
+                assert isinstance(key, str)
+                assert isinstance(source_list, list)
+                for source_url in source_list:
+                    assert isinstance(source_url, str)
+                    assert len(source_url) > 0
+                    # Sources should be valid URLs
+                    assert source_url.startswith(("http://", "https://"))
+        elif isinstance(result.sources, list):
+            # Sources returned as simple list (fallback)
+            for source in result.sources:
+                assert isinstance(source, str)
+                assert len(source) > 0
+    
+    # Response metadata validation
+    assert hasattr(result, "id")
+    assert hasattr(result, "status")
+    if result.status is not None:
+        assert result.status in ["processing", "completed", "failed"]
+    
+    # Expires at validation
+    assert hasattr(result, "expires_at")
+    # expires_at can be None or a datetime, so we just check it exists
+
+@pytest.mark.asyncio
+async def test_extract_multiple_urls():
+    """Test extraction with multiple URLs."""
+    class MultiUrlSchema(BaseModel):
+        title: str
+        description: str
+        links: List[str]
+    
+    urls = ["https://example.com", "https://www.iana.org"]
+    result = await app.extract(
+        urls,
+        prompt="Extract the title, description, and links from the websites",
+        schema=MultiUrlSchema
+    )
+    
+    # Basic response assertions
+    assert result is not None
+    assert result.success
+    assert hasattr(result, "data")
+    assert result.data is not None
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Sources validation
+    assert hasattr(result, "sources")
+    if result.sources is not None:
+        if isinstance(result.sources, dict):
+            # Sources returned as dictionary with keys like 'links[0]'
+            for key, source_list in result.sources.items():
+                assert isinstance(key, str)
+                assert isinstance(source_list, list)
+                for source_url in source_list:
+                    assert isinstance(source_url, str)
+                    assert len(source_url) > 0
+                    # Sources should be valid URLs
+                    assert source_url.startswith(("http://", "https://"))
+        elif isinstance(result.sources, list):
+            # Sources returned as simple list (fallback)
+            for source in result.sources:
+                assert isinstance(source, str)
+                assert len(source) > 0
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+
+@pytest.mark.asyncio
+async def test_extract_with_system_prompt():
+    """Test extraction with custom system prompt."""
+    class SystemPromptSchema(BaseModel):
+        title: str
+        summary: str
+    
+    result = await app.extract(
+        ["https://example.com"],
+        prompt="Extract the title and create a brief summary",
+        schema=SystemPromptSchema,
+        system_prompt="You are an expert content analyzer. Focus on accuracy and brevity."
+    )
+    
+    # Basic response assertions
+    assert result is not None
+    assert result.success
+    assert hasattr(result, "data")
+    assert result.data is not None
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Sources validation
+    assert hasattr(result, "sources")
+    if result.sources is not None:
+        if isinstance(result.sources, dict):
+            # Sources returned as dictionary with keys like 'links[0]'
+            for key, source_list in result.sources.items():
+                assert isinstance(key, str)
+                assert isinstance(source_list, list)
+                for source_url in source_list:
+                    assert isinstance(source_url, str)
+                    assert len(source_url) > 0
+                    # Sources should be valid URLs
+                    assert source_url.startswith(("http://", "https://"))
+        elif isinstance(result.sources, list):
+            # Sources returned as simple list (fallback)
+            for source in result.sources:
+                assert isinstance(source, str)
+                assert len(source) > 0
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "summary" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["summary"], str)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["summary"]) > 0
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "summary")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.summary, str)
+        assert len(result.data.title) > 0
+        assert len(result.data.summary) > 0 
\ No newline at end of file
diff --git a/apps/python-sdk/tests/e2e/async/test_map.py b/apps/python-sdk/tests/e2e/async/test_a_map.py
similarity index 100%
rename from apps/python-sdk/tests/e2e/async/test_map.py
rename to apps/python-sdk/tests/e2e/async/test_a_map.py
diff --git a/apps/python-sdk/tests/e2e/async/test_scrape.py b/apps/python-sdk/tests/e2e/async/test_a_scrape.py
similarity index 100%
rename from apps/python-sdk/tests/e2e/async/test_scrape.py
rename to apps/python-sdk/tests/e2e/async/test_a_scrape.py
diff --git a/apps/python-sdk/tests/e2e/async/test_extract.py b/apps/python-sdk/tests/e2e/async/test_extract.py
deleted file mode 100644
index 37d4b224..00000000
--- a/apps/python-sdk/tests/e2e/async/test_extract.py
+++ /dev/null
@@ -1,99 +0,0 @@
-import os
-import sys
-import pytest
-from dotenv import load_dotenv
-from pydantic import BaseModel
-from typing import List
-
-sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), '../../../')))
-from firecrawl.firecrawl import AsyncFirecrawlApp
-
-load_dotenv()
-
-API_URL = "https://api.firecrawl.dev"
-API_KEY = os.getenv("TEST_API_KEY")
-
-app = AsyncFirecrawlApp(api_url=API_URL, api_key=API_KEY)
-
-class ExtractSchema(BaseModel):
-    title: str
-    description: str
-    links: List[str]
-
-@pytest.mark.asyncio
-async def test_extract_async_simple_schema_class():
-    result = await app.extract(
-        ["https://example.com"],
-        prompt="Extract the title, description, and links from the website",
-        schema=ExtractSchema
-    )
-    assert result.success
-    assert result.data is not None
-    assert hasattr(result.data, "title")
-    assert hasattr(result.data, "description")
-    assert hasattr(result.data, "links")
-    assert isinstance(result.data.links, list)
-
-@pytest.mark.asyncio
-async def test_extract_async_simple_schema_dict():
-    result = await app.extract(
-        ["https://example.com"],
-        prompt="Extract the title, description, and links from the website",
-        schema=ExtractSchema.schema()
-    )
-    assert result.success
-    assert result.data is not None
-    assert hasattr(result.data, "title")
-    assert hasattr(result.data, "description")
-    assert hasattr(result.data, "links")
-    assert isinstance(result.data.links, list)
-
-@pytest.mark.asyncio
-async def test_extract_async_simple_schema_json():
-    # Try both model_json_schema (Pydantic v2) and schema_json (Pydantic v1)
-    schema = None
-    if hasattr(ExtractSchema, "model_json_schema"):
-        schema = ExtractSchema.model_json_schema()
-    elif hasattr(ExtractSchema, "schema_json"):
-        schema = ExtractSchema.schema_json()
-    else:
-        pytest.skip("No JSON schema export method available on ExtractSchema")
-    result = await app.extract(
-        ["https://example.com"],
-        prompt="Extract the title, description, and links from the website",
-        schema=schema
-    )
-    assert result.success
-    assert result.data is not None
-    assert hasattr(result.data, "title")
-    assert hasattr(result.data, "description")
-    assert hasattr(result.data, "links")
-    assert isinstance(result.data.links, list)
-
-@pytest.mark.asyncio
-async def test_extract_async_all_params():
-    class AllParamsSchema(BaseModel):
-        title: str
-        description: str
-        links: List[str]
-        author: str
-    schema = AllParamsSchema.schema()
-    result = await app.extract(
-        ["https://www.iana.org"],
-        prompt="Extract the title, description, links, and author from the website",
-        schema=schema,
-        system_prompt="You are a helpful extraction agent.",
-        allow_external_links=False,
-        enable_web_search=True,
-        show_sources=True,
-        agent={"model": "FIRE-1"}
-    )
-    assert result.success
-    assert result.data is not None
-    assert hasattr(result.data, "title")
-    assert hasattr(result.data, "description")
-    assert hasattr(result.data, "links")
-    assert hasattr(result.data, "author")
-    assert isinstance(result.data.links, list)
-    assert hasattr(result, "sources")
-    assert isinstance(result.sources, list) 
\ No newline at end of file
diff --git a/apps/python-sdk/tests/e2e/test_extract.py b/apps/python-sdk/tests/e2e/test_extract.py
index f3c5f1e7..fc6e9973 100644
--- a/apps/python-sdk/tests/e2e/test_extract.py
+++ b/apps/python-sdk/tests/e2e/test_extract.py
@@ -10,7 +10,7 @@ from firecrawl.firecrawl import FirecrawlApp
 
 load_dotenv()
 
-API_URL = "https://api.firecrawl.dev"
+API_URL = os.getenv("TEST_API_URL") or "https://api.firecrawl.dev"
 API_KEY = os.getenv("TEST_API_KEY")
 
 app = FirecrawlApp(api_url=API_URL, api_key=API_KEY)
@@ -26,12 +26,62 @@ def test_extract_simple_schema_class():
         prompt="Extract the title, description, and links from the website",
         schema=ExtractSchema
     )
+    
+    # Basic response assertions
+    assert result is not None
     assert result.success
+    assert hasattr(result, "data")
     assert result.data is not None
-    assert hasattr(result.data, "title")
-    assert hasattr(result.data, "description")
-    assert hasattr(result.data, "links")
-    assert isinstance(result.data.links, list)
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Warning handling assertions
+    assert hasattr(result, "warning")
+    # Warning can be present or None, so we just check it exists
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    
+    # Response metadata validation
+    assert hasattr(result, "id")
+    assert hasattr(result, "status")
+    if result.status is not None:
+        assert result.status in ["processing", "completed", "failed"]
 
 def test_extract_simple_schema_dict():
     result = app.extract(
@@ -39,12 +89,80 @@ def test_extract_simple_schema_dict():
         prompt="Extract the title, description, and links from the website",
         schema=ExtractSchema.schema()
     )
+    
+    # Basic response assertions
+    assert result is not None
     assert result.success
+    assert hasattr(result, "data")
     assert result.data is not None
-    assert hasattr(result.data, "title")
-    assert hasattr(result.data, "description")
-    assert hasattr(result.data, "links")
-    assert isinstance(result.data.links, list)
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Warning handling assertions
+    assert hasattr(result, "warning")
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    
+    # Response metadata validation
+    assert hasattr(result, "id")
+    assert hasattr(result, "status")
+    if result.status is not None:
+        assert result.status in ["processing", "completed", "failed"]
+
+    # Sources validation
+    assert hasattr(result, "sources")
+    if result.sources is not None:
+        if isinstance(result.sources, dict):
+            # Sources returned as dictionary with keys like 'links[0]'
+            for key, source_list in result.sources.items():
+                assert isinstance(key, str)
+                assert isinstance(source_list, list)
+                for source_url in source_list:
+                    assert isinstance(source_url, str)
+                    assert len(source_url) > 0
+                    # Sources should be valid URLs
+                    assert source_url.startswith(("http://", "https://"))
+        elif isinstance(result.sources, list):
+            # Sources returned as simple list (fallback)
+            for source in result.sources:
+                assert isinstance(source, str)
+                assert len(source) > 0
 
 def test_extract_simple_schema_json():
     schema = None
@@ -54,17 +172,86 @@ def test_extract_simple_schema_json():
         schema = ExtractSchema.schema_json()
     else:
         pytest.skip("No JSON schema export method available on ExtractSchema")
+    
     result = app.extract(
         ["https://example.com"],
         prompt="Extract the title, description, and links from the website",
         schema=schema
     )
+    
+    # Basic response assertions
+    assert result is not None
     assert result.success
+    assert hasattr(result, "data")
     assert result.data is not None
-    assert hasattr(result.data, "title")
-    assert hasattr(result.data, "description")
-    assert hasattr(result.data, "links")
-    assert isinstance(result.data.links, list)
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Warning handling assertions
+    assert hasattr(result, "warning")
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    
+    # Response metadata validation
+    assert hasattr(result, "id")
+    assert hasattr(result, "status")
+    if result.status is not None:
+        assert result.status in ["processing", "completed", "failed"]
+
+    # Sources validation
+    assert hasattr(result, "sources")
+    if result.sources is not None:
+        if isinstance(result.sources, dict):
+            # Sources returned as dictionary with keys like 'links[0]'
+            for key, source_list in result.sources.items():
+                assert isinstance(key, str)
+                assert isinstance(source_list, list)
+                for source_url in source_list:
+                    assert isinstance(source_url, str)
+                    assert len(source_url) > 0
+                    # Sources should be valid URLs
+                    assert source_url.startswith(("http://", "https://"))
+        elif isinstance(result.sources, list):
+            # Sources returned as simple list (fallback)
+            for source in result.sources:
+                assert isinstance(source, str)
+                assert len(source) > 0
 
 def test_extract_all_params():
     class AllParamsSchema(BaseModel):
@@ -72,6 +259,7 @@ def test_extract_all_params():
         description: str
         links: List[str]
         author: str
+    
     schema = AllParamsSchema.schema()
     result = app.extract(
         ["https://www.iana.org"],
@@ -79,16 +267,190 @@ def test_extract_all_params():
         schema=schema,
         system_prompt="You are a helpful extraction agent.",
         allow_external_links=False,
-        enable_web_search=True,
+        enable_web_search=False,
         show_sources=True,
-        agent={"model": "FIRE-1"}
+        # agent={"model": "FIRE-1"}
     )
+    
+    # Basic response assertions
+    assert result is not None
     assert result.success
+    assert hasattr(result, "data")
     assert result.data is not None
-    assert hasattr(result.data, "title")
-    assert hasattr(result.data, "description")
-    assert hasattr(result.data, "links")
-    assert hasattr(result.data, "author")
-    assert isinstance(result.data.links, list)
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Warning handling assertions
+    assert hasattr(result, "warning")
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert "author" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert isinstance(result.data["author"], str)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert hasattr(result.data, "author")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert isinstance(result.data.author, str)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    
+    # Sources validation (show_sources=True)
     assert hasattr(result, "sources")
-    assert isinstance(result.sources, list) 
\ No newline at end of file
+    if result.sources is not None:
+        if isinstance(result.sources, dict):
+            # Sources returned as dictionary with keys like 'links[0]'
+            for key, source_list in result.sources.items():
+                assert isinstance(key, str)
+                assert isinstance(source_list, list)
+                for source_url in source_list:
+                    assert isinstance(source_url, str)
+                    assert len(source_url) > 0
+                    # Sources should be valid URLs
+                    assert source_url.startswith(("http://", "https://"))
+        elif isinstance(result.sources, list):
+            # Sources returned as simple list (fallback)
+            for source in result.sources:
+                assert isinstance(source, str)
+                assert len(source) > 0
+    
+    # Response metadata validation
+    assert hasattr(result, "id")
+    assert hasattr(result, "status")
+    if result.status is not None:
+        assert result.status in ["processing", "completed", "failed"]
+    
+    # Expires at validation
+    assert hasattr(result, "expires_at")
+    # expires_at can be None or a datetime, so we just check it exists
+
+def test_extract_multiple_urls():
+    """Test extraction with multiple URLs."""
+    class MultiUrlSchema(BaseModel):
+        title: str
+        description: str
+        links: List[str]
+    
+    urls = ["https://example.com", "https://www.iana.org"]
+    result = app.extract(
+        urls,
+        prompt="Extract the title, description, and links from the websites",
+        schema=MultiUrlSchema
+    )
+    
+    # Basic response assertions
+    assert result is not None
+    assert result.success
+    assert hasattr(result, "data")
+    assert result.data is not None
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "description" in result.data
+        assert "links" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["description"], str)
+        assert isinstance(result.data["links"], list)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["description"]) > 0
+        
+        # Links validation
+        for link in result.data["links"]:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "description")
+        assert hasattr(result.data, "links")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.description, str)
+        assert isinstance(result.data.links, list)
+        assert len(result.data.title) > 0
+        assert len(result.data.description) > 0
+        
+        # Links validation
+        for link in result.data.links:
+            assert isinstance(link, str)
+            assert len(link) > 0
+            # Links should be valid URLs or relative paths
+            assert link.startswith(("http://", "https://", "/", "#")) or "." in link
+
+def test_extract_with_system_prompt():
+    """Test extraction with custom system prompt."""
+    class SystemPromptSchema(BaseModel):
+        title: str
+        summary: str
+    
+    result = app.extract(
+        ["https://example.com"],
+        prompt="Extract the title and create a brief summary",
+        schema=SystemPromptSchema,
+        system_prompt="You are an expert content analyzer. Focus on accuracy and brevity."
+    )
+    
+    # Basic response assertions
+    assert result is not None
+    assert result.success
+    assert hasattr(result, "data")
+    assert result.data is not None
+    
+    # Error handling assertions
+    assert hasattr(result, "error")
+    assert result.error is None
+    
+    # Handle both dictionary and object responses
+    if isinstance(result.data, dict):
+        # Data returned as dictionary
+        assert "title" in result.data
+        assert "summary" in result.data
+        assert isinstance(result.data["title"], str)
+        assert isinstance(result.data["summary"], str)
+        assert len(result.data["title"]) > 0
+        assert len(result.data["summary"]) > 0
+    else:
+        # Data returned as Pydantic model object
+        assert hasattr(result.data, "title")
+        assert hasattr(result.data, "summary")
+        assert isinstance(result.data.title, str)
+        assert isinstance(result.data.summary, str)
+        assert len(result.data.title) > 0
+        assert len(result.data.summary) > 0 
\ No newline at end of file