firecrawl-haystack 1.1.0__tar.gz → 2.0.0__tar.gz

firecrawl_haystack-2.0.0/CHANGELOG.md

@@ -0,0 +1,50 @@
+# Changelog
+
+## [integrations/firecrawl-v1.2.0] - 2026-09-04
+
+### 🐛 Bug Fixes
+
+- Fix new issues raised by ruff 0.16.0 (#3670)
+- Standardize license files (#3771)
+
+### 🚜 Refactor
+
+- Firecrawl - decouple sync/async clients creation (#3914)
+
+### 📚 Documentation
+
+- Fix FirecrawlCrawler wrong name in docstrings (#3560)
+
+### 🧪 Testing
+
+- Test compatible integrations with python 3.14; update pyproject (#3001)
+- Track test coverage for all integrations (#3065)
+
+### 🧹 Chores
+
+- Add ANN type annotations to faiss, fastembed, firecrawl, github, google_ai (#2988)
+- Enforce ruff docstring rules (D102/D103/D205/D209/D213/D417/D419) in integrations 11-20 (#3009)
+- Improve consistency of integrations folder structure (#3430)
+- Re-standardize READMEs (#3768)
+
+
+## [integrations/firecrawl-v1.1.0] - 2026-03-09
+
+### 🚀 Features
+
+- Add FirecrawlWebSearch component (#2888)
+
+
+## [integrations/firecrawl-v1.0.1] - 2026-02-24
+
+### 🌀 Miscellaneous
+
+- Docs: Update Firecrawl module path in Docusaurus configuration (#2869)
+
+## [integrations/firecrawl-v1.0.0] - 2026-02-24
+
+### 🚀 Features
+
+- Add FirecrawlCrawler (#2859)
+
+<!-- generated by git-cliff -->

firecrawl_haystack-1.1.0/LICENSE → firecrawl_haystack-2.0.0/LICENSE.txt

@@ -186,7 +186,7 @@
       same "printed page" as the copyright notice for easier
       identification within third-party archives.
 
-   Copyright 2023 deepset GmbH
+   Copyright 2023-present deepset GmbH
 
    Licensed under the Apache License, Version 2.0 (the "License");
    you may not use this file except in compliance with the License.

{firecrawl_haystack-1.1.0 → firecrawl_haystack-2.0.0}/PKG-INFO

@@ -1,23 +1,24 @@
-Metadata-Version: 2.4
+Metadata-Version: 2.5
 Name: firecrawl-haystack
-Version: 1.1.0
+Version: 2.0.0
 Summary: Haystack integration for Firecrawl
 Project-URL: Documentation, https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/firecrawl#readme
 Project-URL: Issues, https://github.com/deepset-ai/haystack-core-integrations/issues
 Project-URL: Source, https://github.com/deepset-ai/haystack-core-integrations/tree/main/integrations/firecrawl
 Author-email: deepset GmbH <info@deepset.ai>
 License-Expression: Apache-2.0
-License-File: LICENSE
+License-File: LICENSE.txt
 Keywords: Firecrawl,Haystack,Web Crawler,Web Scraping,Web Search
 Classifier: Development Status :: 4 - Beta
 Classifier: License :: OSI Approved :: Apache Software License
 Classifier: Programming Language :: Python
-Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Programming Language :: Python :: 3.15
 Classifier: Programming Language :: Python :: Implementation :: CPython
-Requires-Python: >=3.10
+Requires-Python: >=3.11
 Requires-Dist: firecrawl-py>=4.0.0
 Requires-Dist: haystack-ai>=2.24.1
 Description-Content-Type: text/markdown
@@ -27,6 +28,7 @@ Description-Content-Type: text/markdown
 [![PyPI - Version](https://img.shields.io/pypi/v/firecrawl-haystack.svg)](https://pypi.org/project/firecrawl-haystack)
 [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/firecrawl-haystack.svg)](https://pypi.org/project/firecrawl-haystack)
 
+- [Integration page](https://haystack.deepset.ai/integrations/firecrawl)
 - [Changelog](https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/firecrawl/CHANGELOG.md)
 
 ---
@@ -35,5 +37,4 @@ Description-Content-Type: text/markdown
 
 Refer to the general [Contribution Guidelines](https://github.com/deepset-ai/haystack-core-integrations/blob/main/CONTRIBUTING.md).
 
-
-To run integration tests locally, you need to export the `FIRECRAWL_API_KEY` environment variable. 
+To run integration tests locally, you need to export the `FIRECRAWL_API_KEY` environment variable. 

{firecrawl_haystack-1.1.0 → firecrawl_haystack-2.0.0}/README.md

@@ -3,6 +3,7 @@
 [![PyPI - Version](https://img.shields.io/pypi/v/firecrawl-haystack.svg)](https://pypi.org/project/firecrawl-haystack)
 [![PyPI - Python Version](https://img.shields.io/pypi/pyversions/firecrawl-haystack.svg)](https://pypi.org/project/firecrawl-haystack)
 
+- [Integration page](https://haystack.deepset.ai/integrations/firecrawl)
 - [Changelog](https://github.com/deepset-ai/haystack-core-integrations/blob/main/integrations/firecrawl/CHANGELOG.md)
 
 ---
@@ -11,5 +12,4 @@
 
 Refer to the general [Contribution Guidelines](https://github.com/deepset-ai/haystack-core-integrations/blob/main/CONTRIBUTING.md).
 
-
-To run integration tests locally, you need to export the `FIRECRAWL_API_KEY` environment variable. 
+To run integration tests locally, you need to export the `FIRECRAWL_API_KEY` environment variable. 

{firecrawl_haystack-1.1.0 → firecrawl_haystack-2.0.0}/pyproject.toml

@@ -7,7 +7,7 @@ name = "firecrawl-haystack"
 dynamic = ["version"]
 description = "Haystack integration for Firecrawl"
 readme = "README.md"
-requires-python = ">=3.10"
+requires-python = ">=3.11"
 license = "Apache-2.0"
 keywords = [
     "Firecrawl",
@@ -23,10 +23,11 @@ classifiers = [
     "License :: OSI Approved :: Apache Software License",
     "Development Status :: 4 - Beta",
     "Programming Language :: Python",
-    "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
     "Programming Language :: Python :: 3.13",
+    "Programming Language :: Python :: 3.14",
+    "Programming Language :: Python :: 3.15",
     "Programming Language :: Python :: Implementation :: CPython",
 ]
 dependencies = [
@@ -73,7 +74,8 @@ dependencies = [
 unit = 'pytest -m "not integration" {args:tests}'
 integration = 'pytest -m "integration" {args:tests}'
 all = 'pytest {args:tests}'
-cov-retry = 'pytest --cov=haystack_integrations --reruns 3 --reruns-delay 30 -x {args:tests}'
+unit-cov-retry = 'pytest --cov=haystack_integrations --reruns 3 --reruns-delay 30 -x -m "not integration" {args:tests}'
+integration-cov-append-retry = 'pytest --cov=haystack_integrations --cov-append --reruns 3 --reruns-delay 30 -x -m "integration" {args:tests}'
 types = "mypy -p haystack_integrations.components.fetchers.firecrawl -p haystack_integrations.components.websearch.firecrawl {args}"
 
 [tool.mypy]
@@ -88,9 +90,17 @@ line-length = 120
 [tool.ruff.lint]
 select = [
     "A",
+    "ANN",
     "ARG",
     "B",
     "C",
+    "D102",   # Missing docstring in public method
+    "D103",   # Missing docstring in public function
+    "D205",   # 1 blank line required between summary line and description
+    "D209",   # Closing triple quotes go to new line
+    "D213",   # summary lines must be positioned on the second physical line of the docstring
+    "D417",   # Missing argument descriptions in the docstring
+    "D419",   # Docstring is empty
     "DTZ",
     "E",
     "EM",
@@ -123,12 +133,15 @@ ignore = [
     "S105",
     "S106",
     "S107",
+    # Allow `Any` - used legitimately for dynamic types and SDK boundaries
+    "ANN401",
     # Ignore complexity
     "C901",
     "PLR0911",
     "PLR0912",
     "PLR0913",
     "PLR0915",
+    "PLR0917",
 ]
 unfixable = [
     # Don't touch unused imports
@@ -143,11 +156,12 @@ ban-relative-imports = "parents"
 
 [tool.ruff.lint.per-file-ignores]
 # Tests can use magic values, assertions, and relative imports
-"tests/**/*" = ["PLR2004", "S101", "TID252"]
+"tests/**/*" = ["D", "PLR2004", "S101", "TID252", "ANN"]
 
 [tool.coverage.run]
 source = ["haystack_integrations"]
 branch = true
+relative_files = true
 parallel = false
 
 [tool.coverage.report]

{firecrawl_haystack-1.1.0 → firecrawl_haystack-2.0.0}/src/haystack_integrations/components/fetchers/firecrawl/firecrawl_crawler.py

@@ -2,7 +2,7 @@
 #
 # SPDX-License-Identifier: Apache-2.0
 
-from datetime import datetime, timezone
+from datetime import UTC, datetime
 from typing import Any
 
 from haystack import Document, component, logging
@@ -27,15 +27,15 @@ class FirecrawlCrawler:
     ### Usage example
 
     ```python
-    from haystack_integrations.components.fetchers.firecrawl import FirecrawlFetcher
+    from haystack_integrations.components.fetchers.firecrawl import FirecrawlCrawler
 
-    fetcher = FirecrawlFetcher(
+    crawler = FirecrawlCrawler(
         api_key=Secret.from_env_var("FIRECRAWL_API_KEY"),
         params={"limit": 5},
     )
-    fetcher.warm_up()
+    crawler.warm_up()
 
-    result = fetcher.run(urls=["https://docs.haystack.deepset.ai/docs/intro"])
+    result = crawler.run(urls=["https://docs.haystack.deepset.ai/docs/intro"])
     documents = result["documents"]
     ```
     """
@@ -46,7 +46,7 @@ class FirecrawlCrawler:
         params: dict[str, Any] | None = None,
     ) -> None:
         """
-        Initialize the FirecrawlFetcher.
+        Initialize the FirecrawlCrawler.
 
         :param api_key:
             API key for Firecrawl.
@@ -83,14 +83,18 @@ class FirecrawlCrawler:
         :returns: A dictionary with the following keys:
             - `documents`: List of documents, one for each URL crawled.
         """
-        if self._firecrawl_client is None:
-            self.warm_up()
+        self.warm_up()
+        assert self._firecrawl_client is not None  # noqa: S101
 
         current_params = params if params is not None else self._params
         documents: list[Document] = []
         for url in urls:
-            docs = self._crawl_url(url=url, params=current_params)
-            documents.extend(docs)
+            try:
+                crawl_response = self._firecrawl_client.crawl(url=url, **current_params)
+            except Exception as error:
+                logger.exception(f"Failed to crawl website {url}: {error}")
+                continue
+            documents.extend(self._documents_from_crawl_response(url=url, crawl_response=crawl_response))
 
         return {"documents": documents}
 
@@ -111,67 +115,31 @@ class FirecrawlCrawler:
         :returns: A dictionary with the following keys:
             - `documents`: List of documents, one for each URL crawled.
         """
-        if self._async_firecrawl_client is None:
-            self.warm_up()
+        await self.warm_up_async()
+        assert self._async_firecrawl_client is not None  # noqa: S101
 
         current_params = params if params is not None else self._params
         documents: list[Document] = []
         for url in urls:
-            docs = await self._crawl_url_async(url=url, params=current_params)
-            documents.extend(docs)
+            try:
+                crawl_response = await self._async_firecrawl_client.crawl(url=url, **current_params)
+            except Exception as error:
+                logger.exception(f"Failed to crawl website {url}: {error}")
+                continue
+            documents.extend(self._documents_from_crawl_response(url=url, crawl_response=crawl_response))
 
         return {"documents": documents}
 
     def warm_up(self) -> None:
-        """
-        Warm up the Firecrawl client by initializing the clients.
-        This is useful to avoid cold start delays when crawling many URLs.
-        """
+        """Warm up the synchronous Firecrawl client."""
         if self._firecrawl_client is None:
             self._firecrawl_client = Firecrawl(api_key=self.api_key.resolve_value())
+
+    async def warm_up_async(self) -> None:
+        """Warm up the asynchronous Firecrawl client."""
         if self._async_firecrawl_client is None:
             self._async_firecrawl_client = AsyncFirecrawl(api_key=self.api_key.resolve_value())
 
-    def _crawl_url(self, url: str, params: dict[str, Any]) -> list[Document]:
-        """
-        Crawl a single URL and return Documents.
-
-        :param url: URL to crawl.
-        :param params: Crawl request parameters.
-        :return: List of Documents from the crawl result.
-        """
-        try:
-            # Ignoring type because the firecrawl client is initialized as None and only set during warm_up
-            crawl_response = self._firecrawl_client.crawl(  # type: ignore[union-attr]
-                url=url,
-                **params,
-            )
-        except Exception as error:
-            logger.exception(f"Failed to crawl website {url}: {error}")
-            return []
-
-        return self._documents_from_crawl_response(url=url, crawl_response=crawl_response)
-
-    async def _crawl_url_async(self, url: str, params: dict[str, Any]) -> list[Document]:
-        """
-        Asynchronously crawl a single URL and return Documents.
-
-        :param url: URL to crawl.
-        :param params: Crawl request parameters.
-        :return: List of Documents from the crawl result.
-        """
-        try:
-            # Ignoring type because the firecrawl client is initialized as None and only set during warm_up
-            crawl_response = await self._async_firecrawl_client.crawl(  # type: ignore[union-attr]
-                url=url,
-                **params,
-            )
-        except Exception as error:
-            logger.exception(f"Failed to crawl website {url}: {error}")
-            return []
-
-        return self._documents_from_crawl_response(url=url, crawl_response=crawl_response)
-
     def _documents_from_crawl_response(self, url: str, crawl_response: Any) -> list[Document]:
         """
         Convert a Firecrawl crawl response to Haystack Documents.
@@ -189,7 +157,7 @@ class FirecrawlCrawler:
             doc = Document(
                 content=page.markdown,
                 meta={
-                    "timestamp": datetime.now(tz=timezone.utc).isoformat(),
+                    "timestamp": datetime.now(tz=UTC).isoformat(),
                     **metadata,
                 },
             )

{firecrawl_haystack-1.1.0 → firecrawl_haystack-2.0.0}/src/haystack_integrations/components/websearch/firecrawl/firecrawl_websearch.py

@@ -4,11 +4,12 @@
 
 from typing import Any
 
-from firecrawl import AsyncFirecrawl, Firecrawl  # type: ignore[import-untyped]
-from firecrawl.types import SearchData  # type: ignore[import-untyped]
 from haystack import Document, component, logging
 from haystack.utils import Secret
 
+from firecrawl import AsyncFirecrawl, Firecrawl  # type: ignore[import-untyped]
+from firecrawl.types import SearchData  # type: ignore[import-untyped]
+
 logger = logging.getLogger(__name__)
 
 
@@ -69,12 +70,12 @@ class FirecrawlWebSearch:
         self._async_firecrawl_client: AsyncFirecrawl | None = None
 
     def warm_up(self) -> None:
-        """
-        Warm up the Firecrawl clients by initializing the sync and async clients.
-        This is useful to avoid cold start delays when performing searches.
-        """
+        """Warm up the synchronous Firecrawl client."""
         if self._firecrawl_client is None:
             self._firecrawl_client = Firecrawl(api_key=self.api_key.resolve_value())
+
+    async def warm_up_async(self) -> None:
+        """Warm up the asynchronous Firecrawl client."""
         if self._async_firecrawl_client is None:
             self._async_firecrawl_client = AsyncFirecrawl(api_key=self.api_key.resolve_value())
 
@@ -95,8 +96,8 @@ class FirecrawlWebSearch:
             - `documents`: List of documents with search result content.
             - `links`: List of URLs from the search results.
         """
-        if self._firecrawl_client is None:
-            self.warm_up()
+        self.warm_up()
+        assert self._firecrawl_client is not None  # noqa: S101
 
         current_params = search_params if search_params is not None else self._search_params
         params = current_params.copy()
@@ -104,7 +105,7 @@ class FirecrawlWebSearch:
             params["limit"] = self.top_k
 
         try:
-            search_response = self._firecrawl_client.search(  # type: ignore[union-attr]
+            search_response = self._firecrawl_client.search(
                 query=query,
                 **params,
             )
@@ -132,8 +133,8 @@ class FirecrawlWebSearch:
             - `documents`: List of documents with search result content.
             - `links`: List of URLs from the search results.
         """
-        if self._async_firecrawl_client is None:
-            self.warm_up()
+        await self.warm_up_async()
+        assert self._async_firecrawl_client is not None  # noqa: S101
 
         current_params = search_params if search_params is not None else self._search_params
         params = current_params.copy()
@@ -141,7 +142,7 @@ class FirecrawlWebSearch:
             params["limit"] = self.top_k
 
         try:
-            search_response = await self._async_firecrawl_client.search(  # type: ignore[union-attr]
+            search_response = await self._async_firecrawl_client.search(
                 query=query,
                 **params,
             )

{firecrawl_haystack-1.1.0 → firecrawl_haystack-2.0.0}/tests/test_firecrawl_crawler.py

@@ -14,38 +14,44 @@ from haystack.utils import Secret
 from haystack_integrations.components.fetchers.firecrawl import FirecrawlCrawler
 
 
-class TestFirecrawlCrawler:
-    @pytest.fixture
-    def page(self) -> MagicMock:
-        page = MagicMock()
-        page.markdown = "Page content"
-        page.metadata_dict = {"url": "https://example.com"}
-        return page
-
-    @pytest.fixture
-    def crawl_response(self, page) -> MagicMock:
-        crawl_response = MagicMock()
-        crawl_response.status = "completed"
-        crawl_response.data = [page]
-        return crawl_response
-
-    @pytest.fixture
-    def mock_client(self, crawl_response) -> MagicMock:
-        mock_client = MagicMock()
-        mock_client.crawl.return_value = crawl_response
-        return mock_client
-
-    @pytest.fixture
-    def mock_async_client(self, crawl_response) -> MagicMock:
-        mock_async_client = MagicMock()
-        mock_async_client.crawl = AsyncMock(return_value=crawl_response)
-        return mock_async_client
+@pytest.fixture
+def page() -> MagicMock:
+    page = MagicMock()
+    page.markdown = "Page content"
+    page.metadata_dict = {"url": "https://example.com"}
+    return page
 
+
+@pytest.fixture
+def crawl_response(page) -> MagicMock:
+    crawl_response = MagicMock()
+    crawl_response.status = "completed"
+    crawl_response.data = [page]
+    return crawl_response
+
+
+@pytest.fixture
+def mock_client(crawl_response) -> MagicMock:
+    mock_client = MagicMock()
+    mock_client.crawl.return_value = crawl_response
+    return mock_client
+
+
+@pytest.fixture
+def mock_async_client(crawl_response) -> MagicMock:
+    mock_async_client = MagicMock()
+    mock_async_client.crawl = AsyncMock(return_value=crawl_response)
+    return mock_async_client
+
+
+class TestInit:
     def test_init_default(self, monkeypatch: pytest.MonkeyPatch) -> None:
         monkeypatch.setenv("FIRECRAWL_API_KEY", "test-key")
         fetcher = FirecrawlCrawler()
         assert fetcher._params == {"limit": 1, "scrape_options": {"formats": ["markdown"]}}
         assert fetcher.api_key.resolve_value() == "test-key"
+        assert fetcher._firecrawl_client is None
+        assert fetcher._async_firecrawl_client is None
 
     def test_init_with_params(self) -> None:
         fetcher = FirecrawlCrawler(
@@ -55,13 +61,14 @@ class TestFirecrawlCrawler:
         assert fetcher._params == {"limit": 10, "scrape_options": {"formats": ["markdown"]}}
         assert fetcher.api_key.resolve_value() == "custom-key"
 
-    def test_to_dict(self, monkeypatch: pytest.MonkeyPatch) -> None:
-        monkeypatch.setenv("FIRECRAWL_API_KEY", "test-key")
+
+class TestSerialization:
+    def test_to_dict(self) -> None:
         fetcher = FirecrawlCrawler(
             api_key=Secret.from_env_var("FIRECRAWL_API_KEY"),
             params={"limit": 5},
         )
-        data = component_to_dict(fetcher, "FirecrawlFetcher")
+        data = component_to_dict(fetcher, "FirecrawlCrawler")
         assert data["type"] == "haystack_integrations.components.fetchers.firecrawl.firecrawl_crawler.FirecrawlCrawler"
         assert data["init_parameters"]["params"] == {"limit": 5}
         assert "api_key" in data["init_parameters"]
@@ -75,10 +82,46 @@ class TestFirecrawlCrawler:
                 "api_key": {"env_vars": ["FIRECRAWL_API_KEY"], "strict": True, "type": "env_var"},
             },
         }
-        fetcher = component_from_dict(FirecrawlCrawler, data, "FirecrawlFetcher")
+        fetcher = component_from_dict(FirecrawlCrawler, data, "FirecrawlCrawler")
         assert fetcher.params == {"limit": 3}
         assert fetcher.api_key.resolve_value() == "test-key"
 
+
+class TestComponentLifecycle:
+    def test_key_resolved_at_warm_up_not_init(self, monkeypatch: pytest.MonkeyPatch) -> None:
+        monkeypatch.delenv("FIRECRAWL_API_KEY", raising=False)
+        fetcher = FirecrawlCrawler()
+
+        with pytest.raises(ValueError, match="None of the following authentication environment variables are set"):
+            fetcher.warm_up()
+
+    def test_warm_up(self) -> None:
+        fetcher = FirecrawlCrawler(api_key=Secret.from_token("test-key"))
+
+        with patch("haystack_integrations.components.fetchers.firecrawl.firecrawl_crawler.Firecrawl") as mock_client:
+            fetcher.warm_up()
+            fetcher.warm_up()
+
+        assert fetcher._firecrawl_client is mock_client.return_value
+        assert fetcher._async_firecrawl_client is None
+        mock_client.assert_called_once_with(api_key="test-key")
+
+    @pytest.mark.asyncio
+    async def test_warm_up_async(self) -> None:
+        fetcher = FirecrawlCrawler(api_key=Secret.from_token("test-key"))
+
+        with patch(
+            "haystack_integrations.components.fetchers.firecrawl.firecrawl_crawler.AsyncFirecrawl"
+        ) as mock_async_client:
+            await fetcher.warm_up_async()
+            await fetcher.warm_up_async()
+
+        assert fetcher._firecrawl_client is None
+        assert fetcher._async_firecrawl_client is mock_async_client.return_value
+        mock_async_client.assert_called_once_with(api_key="test-key")
+
+
+class TestRun:
     def test_run_with_mocked_firecrawl_client(self, mock_client) -> None:
         fetcher = FirecrawlCrawler(api_key=Secret.from_token("test-key"))
 
@@ -135,12 +178,9 @@ class TestFirecrawlCrawler:
         )
 
     def test_run_calls_warm_up(self, crawl_response) -> None:
-        with (
-            patch(
-                "haystack_integrations.components.fetchers.firecrawl.firecrawl_crawler.Firecrawl"
-            ) as mock_firecrawl_client,
-            patch("haystack_integrations.components.fetchers.firecrawl.firecrawl_crawler.AsyncFirecrawl"),
-        ):
+        with patch(
+            "haystack_integrations.components.fetchers.firecrawl.firecrawl_crawler.Firecrawl"
+        ) as mock_firecrawl_client:
             mock_firecrawl_client.return_value.crawl.return_value = crawl_response
 
             fetcher = FirecrawlCrawler(api_key=Secret.from_token("test-key"))
@@ -151,12 +191,9 @@ class TestFirecrawlCrawler:
 
     @pytest.mark.asyncio
     async def test_run_async_calls_warm_up(self, crawl_response) -> None:
-        with (
-            patch("haystack_integrations.components.fetchers.firecrawl.firecrawl_crawler.Firecrawl"),
-            patch(
-                "haystack_integrations.components.fetchers.firecrawl.firecrawl_crawler.AsyncFirecrawl"
-            ) as mock_async_firecrawl_client,
-        ):
+        with patch(
+            "haystack_integrations.components.fetchers.firecrawl.firecrawl_crawler.AsyncFirecrawl"
+        ) as mock_async_firecrawl_client:
             mock_async_firecrawl_client.return_value.crawl = AsyncMock(return_value=crawl_response)
 
             fetcher = FirecrawlCrawler(api_key=Secret.from_token("test-key"))
@@ -222,21 +259,13 @@ class TestFirecrawlCrawler:
 
         assert "Failed to crawl website https://example.com: failed" in caplog.text
 
-    def test_warm_up_initializes_clients(self) -> None:
-        fetcher = FirecrawlCrawler(api_key=Secret.from_token("test-key"))
-        assert fetcher._firecrawl_client is None
-        assert fetcher._async_firecrawl_client is None
-
-        fetcher.warm_up()
-
-        assert fetcher._firecrawl_client is not None
-        assert fetcher._async_firecrawl_client is not None
 
-    @pytest.mark.skipif(
-        not os.environ.get("FIRECRAWL_API_KEY"),
-        reason="Export FIRECRAWL_API_KEY to run integration tests.",
-    )
-    @pytest.mark.integration
+@pytest.mark.skipif(
+    not os.environ.get("FIRECRAWL_API_KEY"),
+    reason="Export FIRECRAWL_API_KEY to run integration tests.",
+)
+@pytest.mark.integration
+class TestIntegration:
     def test_run_integration(self) -> None:
         fetcher = FirecrawlCrawler(api_key=Secret.from_env_var("FIRECRAWL_API_KEY"), params={"limit": 1})
         result = fetcher.run(urls=["https://docs.haystack.deepset.ai/docs/intro"], params={"limit": 1})
@@ -247,11 +276,6 @@ class TestFirecrawlCrawler:
         assert "Haystack" in result["documents"][0].content
         assert "https://docs.haystack.deepset.ai/docs/intro" == result["documents"][0].meta["url"]
 
-    @pytest.mark.skipif(
-        not os.environ.get("FIRECRAWL_API_KEY"),
-        reason="Export FIRECRAWL_API_KEY to run integration tests.",
-    )
-    @pytest.mark.integration
     @pytest.mark.asyncio
     async def test_run_async_integration(self) -> None:
         fetcher = FirecrawlCrawler(api_key=Secret.from_env_var("FIRECRAWL_API_KEY"), params={"limit": 1})

{firecrawl_haystack-1.1.0 → firecrawl_haystack-2.0.0}/tests/test_firecrawl_websearch.py

@@ -13,50 +13,57 @@ from haystack.utils import Secret
 from haystack_integrations.components.websearch.firecrawl import FirecrawlWebSearch
 
 
-class TestFirecrawlWebSearch:
-    @pytest.fixture
-    def search_result_web(self) -> MagicMock:
-        """A SearchResultWeb-like object (no markdown, just search metadata)."""
-        result = MagicMock(spec=["url", "title", "description"])
-        result.url = "https://example.com"
-        result.title = "Example Title"
-        result.description = "Example description snippet"
-        return result
-
-    @pytest.fixture
-    def search_result_document(self) -> MagicMock:
-        """A Document-like object (with markdown, from scrapeOptions)."""
-        result = MagicMock(spec=["markdown", "metadata_dict", "url"])
-        result.markdown = "# Full page content\nSome markdown text."
-        result.metadata_dict = {"url": "https://example.com/page", "title": "Page Title"}
-        result.url = "https://example.com/page"
-        return result
-
-    @pytest.fixture
-    def search_response(self, search_result_web) -> MagicMock:
-        """Standard search response with web results."""
-        response = MagicMock()
-        response.web = [search_result_web]
-        return response
-
-    @pytest.fixture
-    def mock_client(self, search_response) -> MagicMock:
-        client = MagicMock()
-        client.search.return_value = search_response
-        return client
-
-    @pytest.fixture
-    def mock_async_client(self, search_response) -> MagicMock:
-        client = MagicMock()
-        client.search = AsyncMock(return_value=search_response)
-        return client
+@pytest.fixture
+def search_result_web() -> MagicMock:
+    """A SearchResultWeb-like object (no markdown, just search metadata)."""
+    result = MagicMock(spec=["url", "title", "description"])
+    result.url = "https://example.com"
+    result.title = "Example Title"
+    result.description = "Example description snippet"
+    return result
 
+
+@pytest.fixture
+def search_result_document() -> MagicMock:
+    """A Document-like object (with markdown, from scrapeOptions)."""
+    result = MagicMock(spec=["markdown", "metadata_dict", "url"])
+    result.markdown = "# Full page content\nSome markdown text."
+    result.metadata_dict = {"url": "https://example.com/page", "title": "Page Title"}
+    result.url = "https://example.com/page"
+    return result
+
+
+@pytest.fixture
+def search_response(search_result_web) -> MagicMock:
+    """Standard search response with web results."""
+    response = MagicMock()
+    response.web = [search_result_web]
+    return response
+
+
+@pytest.fixture
+def mock_client(search_response) -> MagicMock:
+    client = MagicMock()
+    client.search.return_value = search_response
+    return client
+
+
+@pytest.fixture
+def mock_async_client(search_response) -> MagicMock:
+    client = MagicMock()
+    client.search = AsyncMock(return_value=search_response)
+    return client
+
+
+class TestInit:
     def test_init_default(self, monkeypatch: pytest.MonkeyPatch) -> None:
         monkeypatch.setenv("FIRECRAWL_API_KEY", "test-key")
         ws = FirecrawlWebSearch()
         assert ws.top_k == 10
         assert ws._search_params == {}
         assert ws.api_key.resolve_value() == "test-key"
+        assert ws._firecrawl_client is None
+        assert ws._async_firecrawl_client is None
 
     def test_init_with_params(self) -> None:
         ws = FirecrawlWebSearch(
@@ -68,8 +75,9 @@ class TestFirecrawlWebSearch:
         assert ws._search_params == {"tbs": "qdr:d", "location": "US"}
         assert ws.api_key.resolve_value() == "custom-key"
 
-    def test_to_dict(self, monkeypatch: pytest.MonkeyPatch) -> None:
-        monkeypatch.setenv("FIRECRAWL_API_KEY", "test-key")
+
+class TestSerialization:
+    def test_to_dict(self) -> None:
         ws = FirecrawlWebSearch(top_k=5, search_params={"tbs": "qdr:d"})
         data = component_to_dict(ws, "FirecrawlWebSearch")
         assert (
@@ -94,6 +102,42 @@ class TestFirecrawlWebSearch:
         assert ws.search_params == {"location": "UK"}
         assert ws.api_key.resolve_value() == "test-key"
 
+
+class TestComponentLifecycle:
+    def test_key_resolved_at_warm_up_not_init(self, monkeypatch: pytest.MonkeyPatch) -> None:
+        monkeypatch.delenv("FIRECRAWL_API_KEY", raising=False)
+        ws = FirecrawlWebSearch()
+
+        with pytest.raises(ValueError, match="None of the following authentication environment variables are set"):
+            ws.warm_up()
+
+    def test_warm_up(self) -> None:
+        ws = FirecrawlWebSearch(api_key=Secret.from_token("test-key"))
+
+        with patch("haystack_integrations.components.websearch.firecrawl.firecrawl_websearch.Firecrawl") as mock_client:
+            ws.warm_up()
+            ws.warm_up()
+
+        assert ws._firecrawl_client is mock_client.return_value
+        assert ws._async_firecrawl_client is None
+        mock_client.assert_called_once_with(api_key="test-key")
+
+    @pytest.mark.asyncio
+    async def test_warm_up_async(self) -> None:
+        ws = FirecrawlWebSearch(api_key=Secret.from_token("test-key"))
+
+        with patch(
+            "haystack_integrations.components.websearch.firecrawl.firecrawl_websearch.AsyncFirecrawl"
+        ) as mock_async_client:
+            await ws.warm_up_async()
+            await ws.warm_up_async()
+
+        assert ws._firecrawl_client is None
+        assert ws._async_firecrawl_client is mock_async_client.return_value
+        mock_async_client.assert_called_once_with(api_key="test-key")
+
+
+class TestRun:
     def test_run_returns_documents_and_links(self, mock_client) -> None:
         ws = FirecrawlWebSearch(api_key=Secret.from_token("test-key"), top_k=10)
         ws._firecrawl_client = mock_client
@@ -175,12 +219,9 @@ class TestFirecrawlWebSearch:
         assert result["links"] == []
 
     def test_run_calls_warm_up(self, search_response) -> None:
-        with (
-            patch(
-                "haystack_integrations.components.websearch.firecrawl.firecrawl_websearch.Firecrawl"
-            ) as mock_firecrawl_client,
-            patch("haystack_integrations.components.websearch.firecrawl.firecrawl_websearch.AsyncFirecrawl"),
-        ):
+        with patch(
+            "haystack_integrations.components.websearch.firecrawl.firecrawl_websearch.Firecrawl"
+        ) as mock_firecrawl_client:
             mock_firecrawl_client.return_value.search.return_value = search_response
 
             ws = FirecrawlWebSearch(api_key=Secret.from_token("test-key"))
@@ -191,12 +232,9 @@ class TestFirecrawlWebSearch:
 
     @pytest.mark.asyncio
     async def test_run_async_calls_warm_up(self, search_response) -> None:
-        with (
-            patch("haystack_integrations.components.websearch.firecrawl.firecrawl_websearch.Firecrawl"),
-            patch(
-                "haystack_integrations.components.websearch.firecrawl.firecrawl_websearch.AsyncFirecrawl"
-            ) as mock_async_firecrawl_client,
-        ):
+        with patch(
+            "haystack_integrations.components.websearch.firecrawl.firecrawl_websearch.AsyncFirecrawl"
+        ) as mock_async_firecrawl_client:
             mock_async_firecrawl_client.return_value.search = AsyncMock(return_value=search_response)
 
             ws = FirecrawlWebSearch(api_key=Secret.from_token("test-key"))
@@ -205,16 +243,6 @@ class TestFirecrawlWebSearch:
             assert ws._async_firecrawl_client is mock_async_firecrawl_client.return_value
             mock_async_firecrawl_client.assert_called_once_with(api_key="test-key")
 
-    def test_warm_up_initializes_clients(self) -> None:
-        ws = FirecrawlWebSearch(api_key=Secret.from_token("test-key"))
-        assert ws._firecrawl_client is None
-        assert ws._async_firecrawl_client is None
-
-        ws.warm_up()
-
-        assert ws._firecrawl_client is not None
-        assert ws._async_firecrawl_client is not None
-
     def test_run_empty_web_results(self, mock_client) -> None:
         empty_response = MagicMock()
         empty_response.web = []
@@ -237,11 +265,13 @@ class TestFirecrawlWebSearch:
         assert result["documents"] == []
         assert result["links"] == []
 
-    @pytest.mark.skipif(
-        not os.environ.get("FIRECRAWL_API_KEY"),
-        reason="Export FIRECRAWL_API_KEY to run integration tests.",
-    )
-    @pytest.mark.integration
+
+@pytest.mark.skipif(
+    not os.environ.get("FIRECRAWL_API_KEY"),
+    reason="Export FIRECRAWL_API_KEY to run integration tests.",
+)
+@pytest.mark.integration
+class TestIntegration:
     def test_run_integration(self) -> None:
         ws = FirecrawlWebSearch(
             api_key=Secret.from_env_var("FIRECRAWL_API_KEY"),
@@ -257,11 +287,6 @@ class TestFirecrawlWebSearch:
         assert isinstance(result["documents"][0], Document)
         assert result["documents"][0].content
 
-    @pytest.mark.skipif(
-        not os.environ.get("FIRECRAWL_API_KEY"),
-        reason="Export FIRECRAWL_API_KEY to run integration tests.",
-    )
-    @pytest.mark.integration
     @pytest.mark.asyncio
     async def test_run_async_integration(self) -> None:
         ws = FirecrawlWebSearch(

firecrawl_haystack-1.1.0/CHANGELOG.md

@@ -1,15 +0,0 @@
-# Changelog
-
-## [integrations/firecrawl-v1.0.1] - 2026-02-24
-
-### 🌀 Miscellaneous
-
-- Docs: Update Firecrawl module path in Docusaurus configuration (#2869)
-
-## [integrations/firecrawl-v1.0.0] - 2026-02-24
-
-### 🚀 Features
-
-- Add FirecrawlCrawler (#2859)
-
-<!-- generated by git-cliff -->

firecrawl_haystack-1.1.0/tests/__init__.py

@@ -1,3 +0,0 @@
-# SPDX-FileCopyrightText: 2026-present deepset GmbH <info@deepset.ai>
-#
-# SPDX-License-Identifier: Apache-2.0