firecrawl 2.16.5__py3-none-any.whl → 3.0.3__py3-none-any.whl

firecrawl/__tests__/e2e/v2/test_search.py

@@ -0,0 +1,265 @@
+from firecrawl import Firecrawl
+import os
+from dotenv import load_dotenv
+from firecrawl.types import SearchData, SearchResult, Document, ScrapeFormats, ScrapeOptions
+
+load_dotenv()
+
+firecrawl = Firecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+
+def _collect_texts(entries):
+    texts = []
+    for r in entries or []:
+        title = getattr(r, 'title', None) if hasattr(r, 'title') else None
+        desc = getattr(r, 'description', None) if hasattr(r, 'description') else None
+        if title:
+            texts.append(str(title).lower())
+        if desc:
+            texts.append(str(desc).lower())
+    return texts
+
+def _is_document(entry) -> bool:
+    try:
+        from firecrawl.v2.types import Document
+        return isinstance(entry, Document) or \
+               hasattr(entry, 'markdown') or \
+               hasattr(entry, 'html') or \
+               hasattr(entry, 'raw_html') or \
+               hasattr(entry, 'json') or \
+               hasattr(entry, 'screenshot') or \
+               hasattr(entry, 'change_tracking') or \
+               hasattr(entry, 'summary')
+    except Exception:
+        return hasattr(entry, 'markdown') or \
+               hasattr(entry, 'html') or \
+               hasattr(entry, 'raw_html') or \
+               hasattr(entry, 'json') or \
+               hasattr(entry, 'screenshot') or \
+               hasattr(entry, 'change_tracking') or \
+               hasattr(entry, 'summary')
+
+def test_search_minimal_request():
+    results = firecrawl.search(
+        query="What is the capital of France?"
+    )
+    
+    assert isinstance(results, SearchData)
+    assert hasattr(results, 'web')
+    assert results.web is not None
+    assert len(results.web) > 0
+    assert hasattr(results, 'news')
+    assert results.news is None
+    assert hasattr(results, 'images')
+    assert results.images is None
+    
+    for result in results.web:
+        assert isinstance(result, SearchResult)
+        assert hasattr(result, 'url')
+        assert hasattr(result, 'title')
+        assert hasattr(result, 'description')
+        assert result.url.startswith('http')
+        assert result.title is not None
+        assert result.description is not None
+
+    all_text = ' '.join(_collect_texts(results.web))
+    
+    assert 'paris' in all_text
+    
+    assert results.news is None
+    assert results.images is None
+
+
+def test_search_with_sources():
+    """Test search with specific sources."""
+    results = firecrawl.search(
+        query="firecrawl",
+        sources=["web", "news"],
+        limit=3
+    )
+    
+    assert isinstance(results, SearchData)
+    
+    assert results.web is not None
+    assert len(results.web) <= 3
+    
+    if results.news is not None:
+        assert len(results.news) <= 3
+    
+    assert results.images is None
+    
+    web_titles = [result.title.lower() for result in results.web]
+    web_descriptions = [result.description.lower() for result in results.web]
+    all_web_text = ' '.join(web_titles + web_descriptions)
+    
+    assert 'firecrawl' in all_web_text
+
+def test_search_result_structure():
+    """Test that SearchResult objects have the correct structure."""
+    results = firecrawl.search(
+        query="test query",
+        limit=1
+    )
+    
+    if results.web and len(results.web) > 0:
+        result = results.web[0]
+        
+        assert hasattr(result, 'url')
+        assert hasattr(result, 'title')
+        assert hasattr(result, 'description')
+        
+        assert isinstance(result.url, str)
+        assert isinstance(result.title, str) or result.title is None
+        assert isinstance(result.description, str) or result.description is None
+        
+        # Test URL format
+        assert result.url.startswith('http')
+
+def test_search_all_parameters():
+    """Test search with all available parameters (comprehensive e2e test)."""
+    from firecrawl.types import ScrapeOptions, JsonFormat, Location, WaitAction
+    
+    # Define a schema for JSON extraction
+    schema = {
+        "type": "object",
+        "properties": {
+            "title": {"type": "string"},
+            "description": {"type": "string"},
+            "url": {"type": "string"}
+        },
+        "required": ["title", "description"]
+    }
+    
+    results = firecrawl.search(
+        query="artificial intelligence",
+        sources=[
+            {"type": "web"},
+            {"type": "news"}
+        ],
+        limit=3,
+        tbs="qdr:m",  # Last month
+        location="US",
+        ignore_invalid_urls=True,
+        timeout=60000,
+        scrape_options=ScrapeOptions(
+            formats=[
+                "markdown",
+                "html",
+                {
+                    "type": "json",
+                    "prompt": "Extract the title and description from the page",
+                    "schema": schema
+                },
+                {"type": "summary"}
+            ],
+            headers={"User-Agent": "Firecrawl-Test/1.0"},
+            include_tags=["h1", "h2", "p"],
+            exclude_tags=["nav", "footer"],
+            only_main_content=True,
+            wait_for=2000,
+            mobile=False,
+            skip_tls_verification=False,
+            remove_base64_images=True,
+            block_ads=True,
+            proxy="basic",
+            max_age=3600000,  # 1 hour cache
+            store_in_cache=True,
+            location=Location(
+                country="US",
+                languages=["en"]
+            ),
+            actions=[
+                WaitAction(milliseconds=1000)
+            ]
+            # Note: raw_html and screenshot_full_page are not supported by v2 API yet
+        )
+    )
+    
+    # Test structure
+    assert isinstance(results, SearchData)
+    assert hasattr(results, 'web')
+    assert hasattr(results, 'news')
+    assert hasattr(results, 'images')
+    
+    # Test that web results exist
+    assert results.web is not None
+    assert len(results.web) <= 3  # Should respect limit
+    
+    # Test that results contain expected content for non-document entries only
+    non_doc_entries = [r for r in (results.web or []) if not _is_document(r)]
+    if non_doc_entries:
+        all_web_text = ' '.join(_collect_texts(non_doc_entries))
+        ai_terms = ['artificial', 'intelligence', 'ai', 'machine', 'learning']
+        assert any(term in all_web_text for term in ai_terms)
+    
+    # Test that each result has proper structure
+    for result in results.web:
+        assert isinstance(result, (SearchResult, Document))
+        if isinstance(result, Document):
+            # Document path: ensure content present
+            assert (result.markdown is not None) or (result.html is not None)
+        else:
+            # LinkResult path
+            assert hasattr(result, 'url')
+            assert isinstance(result.url, str) and result.url.startswith('http')
+    
+    # Test that news results exist (if API supports it)
+    if results.news is not None:
+        assert len(results.news) <= 3
+        for result in results.news:
+            assert isinstance(result, (SearchResult, Document))
+            if isinstance(result, Document):
+                assert (result.markdown is not None) or (result.html is not None)
+            else:
+                assert hasattr(result, 'url')
+                assert isinstance(result.url, str) and result.url.startswith('http')
+    
+    # Test that unspecified sources are None
+    assert results.images is None
+
+
+def test_search_formats_flexibility():
+    """Test that both list and ScrapeFormats work for formats."""
+    from firecrawl.types import ScrapeFormats
+    
+    # Test with list format
+    results1 = firecrawl.search(
+        query="python programming",
+        limit=1,
+        scrape_options=ScrapeOptions(
+            formats=["markdown"]
+        )
+    )
+    
+    # Test with ScrapeFormats object
+    results2 = firecrawl.search(
+        query="python programming", 
+        limit=1,
+        scrape_options=ScrapeOptions(
+            formats=ScrapeFormats(markdown=True)
+        )
+    )
+    
+    # Both should work without errors
+    assert isinstance(results1, SearchData)
+    assert isinstance(results2, SearchData)
+    assert results1.web is not None
+    assert results2.web is not None
+
+def test_search_with_json_format_object():
+    """Search with scrape_options including a JSON format object (prompt + schema)."""
+    json_schema = {
+        "type": "object",
+        "properties": {
+            "title": {"type": "string"}
+        },
+        "required": ["title"],
+    }
+    results = firecrawl.search(
+        query="site:docs.firecrawl.dev",
+        limit=1,
+        scrape_options=ScrapeOptions(
+            formats=[{"type": "json", "prompt": "Extract page title", "schema": json_schema}]
+        ),
+    )
+    assert isinstance(results, SearchData)
+    assert results.web is not None and len(results.web) >= 0

firecrawl/__tests__/e2e/v2/test_usage.py

@@ -0,0 +1,26 @@
+import os
+from dotenv import load_dotenv
+from firecrawl import Firecrawl
+
+load_dotenv()
+
+
+class TestUsageE2E:
+    def setup_method(self):
+        # Environment is exported by conftest at import time
+        self.client = Firecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+
+    def test_get_concurrency(self):
+        resp = self.client.get_concurrency()
+        # Shape assertions (endpoint not live yet, but types are defined)
+        assert hasattr(resp, "concurrency")
+        assert hasattr(resp, "max_concurrency")
+
+    def test_get_credit_usage(self):
+        resp = self.client.get_credit_usage()
+        assert hasattr(resp, "remaining_credits")
+
+    def test_get_token_usage(self):
+        resp = self.client.get_token_usage()
+        assert hasattr(resp, "remaining_tokens")
+

firecrawl/__tests__/e2e/v2/test_watcher.py

@@ -0,0 +1,65 @@
+import os
+import time
+from dotenv import load_dotenv
+from firecrawl import Firecrawl
+
+load_dotenv()
+
+if not os.getenv("API_KEY"):
+    raise ValueError("API_KEY is not set")
+
+if not os.getenv("API_URL"):
+    raise ValueError("API_URL is not set")
+
+
+class TestWatcherE2E:
+    def setup_method(self):
+        from firecrawl import Firecrawl
+        self.client = Firecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+
+    def test_crawl_watcher(self):
+        # Start a small crawl job
+        start_job = self.client.start_crawl("https://docs.firecrawl.dev", limit=2)
+        job_id = start_job.id
+
+        statuses = []
+        w = self.client.watcher(job_id, kind="crawl", poll_interval=1, timeout=120)
+        w.add_listener(lambda s: statuses.append(s.status))
+        w.start()
+
+        # Wait for terminal state up to 180 seconds
+        deadline = time.time() + 180
+        while time.time() < deadline:
+            if statuses and statuses[-1] in ["completed", "failed"]:
+                break
+            time.sleep(1)
+
+        w.stop()
+
+        assert len(statuses) > 0
+        assert statuses[-1] in ["completed", "failed"]
+
+    def test_batch_watcher(self):
+        urls = [
+            "https://docs.firecrawl.dev",
+            "https://firecrawl.dev",
+        ]
+        start_resp = self.client.start_batch_scrape(urls, formats=["markdown"], max_concurrency=1)
+        job_id = start_resp.id
+
+        statuses = []
+        w = self.client.watcher(job_id, kind="batch", poll_interval=1, timeout=180)
+        w.add_listener(lambda s: statuses.append(s.status))
+        w.start()
+
+        deadline = time.time() + 240
+        while time.time() < deadline:
+            if statuses and statuses[-1] in ["completed", "failed", "cancelled"]:
+                break
+            time.sleep(1)
+
+        w.stop()
+
+        assert len(statuses) > 0
+        assert statuses[-1] in ["completed", "failed", "cancelled"]
+

firecrawl/__tests__/unit/v2/methods/aio/test_aio_crawl_params.py

@@ -0,0 +1,12 @@
+import pytest
+from firecrawl.v2.types import CrawlParamsRequest
+from firecrawl.v2.methods.aio import crawl as aio_crawl
+
+
+@pytest.mark.asyncio
+async def test_crawl_params_request_validation():
+    with pytest.raises(ValueError):
+        await aio_crawl.crawl_params_preview(None, CrawlParamsRequest(url="", prompt="x"))
+    with pytest.raises(ValueError):
+        await aio_crawl.crawl_params_preview(None, CrawlParamsRequest(url="https://x", prompt=""))
+

firecrawl/__tests__/unit/v2/methods/aio/test_aio_crawl_request_preparation.py

@@ -0,0 +1,61 @@
+from firecrawl.v2.types import CrawlRequest, ScrapeOptions, WebhookConfig
+from firecrawl.v2.methods.aio.crawl import _prepare_crawl_request
+
+
+class TestAsyncCrawlRequestPreparation:
+    def test_basic_request(self):
+        req = CrawlRequest(url="https://example.com")
+        payload = _prepare_crawl_request(req)
+        assert payload["url"] == "https://example.com"
+
+    def test_field_mappings(self):
+        req = CrawlRequest(
+            url="https://example.com",
+            include_paths=["/docs/*"],
+            exclude_paths=["/admin/*"],
+            max_discovery_depth=2,
+            ignore_sitemap=True,
+            ignore_query_parameters=True,
+            crawl_entire_domain=True,
+            allow_external_links=False,
+            allow_subdomains=True,
+            max_concurrency=5,
+            zero_data_retention=True,
+        )
+        payload = _prepare_crawl_request(req)
+        assert payload["includePaths"] == ["/docs/*"]
+        assert payload["excludePaths"] == ["/admin/*"]
+        assert payload["maxDiscoveryDepth"] == 2
+        assert payload["ignoreSitemap"] is True
+        assert payload["ignoreQueryParameters"] is True
+        assert payload["crawlEntireDomain"] is True
+        assert payload["allowExternalLinks"] is False
+        assert payload["allowSubdomains"] is True
+        assert payload["maxConcurrency"] == 5
+        assert payload["zeroDataRetention"] is True
+
+    def test_webhook_preparation(self):
+        # string webhook
+        req = CrawlRequest(url="https://example.com", webhook="https://example.com/hook")
+        payload = _prepare_crawl_request(req)
+        assert payload["webhook"] == "https://example.com/hook"
+
+        # object webhook
+        req2 = CrawlRequest(url="https://example.com", webhook=WebhookConfig(url="https://x/h", headers={"X": "1"}, events=["completed"]))
+        payload2 = _prepare_crawl_request(req2)
+        assert isinstance(payload2["webhook"], dict)
+        assert payload2["webhook"]["url"] == "https://x/h"
+        assert payload2["webhook"]["headers"] == {"X": "1"}
+
+    def test_webhook_none_values_excluded(self):
+        req = CrawlRequest(
+            url="https://example.com",
+            webhook=WebhookConfig(url="https://example.com/webhook", headers=None, metadata=None, events=None),
+        )
+        payload = _prepare_crawl_request(req)
+        webhook = payload["webhook"]
+        assert webhook["url"] == "https://example.com/webhook"
+        assert "headers" not in webhook
+        assert "metadata" not in webhook
+        assert "events" not in webhook
+

firecrawl/__tests__/unit/v2/methods/aio/test_aio_crawl_validation.py

@@ -0,0 +1,12 @@
+from firecrawl.v2.types import CrawlRequest, ScrapeOptions
+from firecrawl.v2.methods.aio.crawl import _prepare_crawl_request
+import pytest
+
+
+class TestAsyncCrawlValidation:
+    def test_invalid_url(self):
+        with pytest.raises(ValueError):
+            _prepare_crawl_request(CrawlRequest(url=""))
+        with pytest.raises(ValueError):
+            _prepare_crawl_request(CrawlRequest(url="   "))
+

firecrawl/__tests__/unit/v2/methods/aio/test_aio_map_request_preparation.py

@@ -0,0 +1,19 @@
+import pytest
+from firecrawl.v2.types import MapOptions
+from firecrawl.v2.methods.aio.map import _prepare_map_request
+
+
+class TestAsyncMapRequestPreparation:
+    def test_basic(self):
+        payload = _prepare_map_request("https://example.com")
+        assert payload["url"] == "https://example.com"
+
+    def test_fields(self):
+        opts = MapOptions(search="docs", include_subdomains=True, limit=10, sitemap="only", timeout=15000)
+        payload = _prepare_map_request("https://example.com", opts)
+        assert payload["search"] == "docs"
+        assert payload["includeSubdomains"] is True
+        assert payload["limit"] == 10
+        assert payload["sitemap"] == "only"
+        assert payload["timeout"] == 15000
+

firecrawl/__tests__/unit/v2/methods/aio/test_aio_scrape_request_preparation.py

@@ -0,0 +1,50 @@
+import pytest
+from firecrawl.v2.types import ScrapeOptions, Location
+from firecrawl.v2.methods.aio.scrape import _prepare_scrape_request
+
+
+class TestAsyncScrapeRequestPreparation:
+    @pytest.mark.asyncio
+    async def test_basic_request_preparation(self):
+        payload = await _prepare_scrape_request("https://example.com", None)
+        assert payload["url"] == "https://example.com"
+
+    @pytest.mark.asyncio
+    async def test_options_conversion(self):
+        opts = ScrapeOptions(
+            formats=["markdown", {"type": "screenshot", "full_page": True, "quality": 80}],
+            include_tags=["main"],
+            exclude_tags=["nav"],
+            only_main_content=True,
+            wait_for=500,
+            timeout=30000,
+            mobile=True,
+            parsers=["pdf"],
+            location=Location(country="us", languages=["en"]),
+            skip_tls_verification=False,
+            remove_base64_images=False,
+            fast_mode=True,
+            use_mock="test",
+            block_ads=False,
+            proxy="basic",
+            max_age=1000,
+            store_in_cache=False,
+        )
+        payload = await _prepare_scrape_request("https://example.com", opts)
+        assert payload["url"] == "https://example.com"
+        assert isinstance(payload.get("formats"), list) and "markdown" in payload["formats"]
+        assert payload["includeTags"] == ["main"]
+        assert payload["excludeTags"] == ["nav"]
+        assert payload["onlyMainContent"] is True
+        assert payload["waitFor"] == 500
+        assert payload["timeout"] == 30000
+        assert payload["mobile"] is True
+        assert payload["skipTlsVerification"] is False
+        assert payload["removeBase64Images"] is False
+        assert payload["fastMode"] is True
+        assert payload["useMock"] == "test"
+        assert payload["blockAds"] is False
+        assert payload["proxy"] == "basic"
+        assert payload["maxAge"] == 1000
+        assert payload["storeInCache"] is False
+

firecrawl/__tests__/unit/v2/methods/aio/test_aio_search_request_preparation.py

@@ -0,0 +1,63 @@
+import pytest
+from firecrawl.v2.types import SearchRequest, ScrapeOptions
+from firecrawl.v2.methods.aio.search import _prepare_search_request
+
+
+class TestAsyncSearchRequestPreparation:
+    def test_basic_request_preparation(self):
+        request = SearchRequest(query="test query")
+        data = _prepare_search_request(request)
+        assert data["query"] == "test query"
+        assert "ignore_invalid_urls" not in data
+        assert "scrape_options" not in data
+
+    def test_all_fields_conversion(self):
+        scrape_opts = ScrapeOptions(
+            formats=["markdown"],
+            headers={"User-Agent": "Test"},
+            include_tags=["h1", "h2"],
+            exclude_tags=["nav"],
+            only_main_content=False,
+            timeout=15000,
+            wait_for=2000,
+            mobile=True,
+            skip_tls_verification=True,
+            remove_base64_images=False,
+        )
+        request = SearchRequest(
+            query="test query",
+            sources=["web", "news"],
+            limit=10,
+            tbs="qdr:w",
+            location="US",
+            ignore_invalid_urls=False,
+            timeout=30000,
+            scrape_options=scrape_opts,
+        )
+        data = _prepare_search_request(request)
+        assert data["ignoreInvalidURLs"] is False
+        assert "scrapeOptions" in data
+
+    def test_exclude_none_behavior(self):
+        request = SearchRequest(
+            query="test",
+            sources=None,
+            limit=None,
+            tbs=None,
+            location=None,
+            ignore_invalid_urls=None,
+            timeout=None,
+            scrape_options=None,
+        )
+        data = _prepare_search_request(request)
+        assert "query" in data
+        assert len(data) == 1
+
+    def test_empty_scrape_options(self):
+        request = SearchRequest(query="test", scrape_options=ScrapeOptions())
+        data = _prepare_search_request(request)
+        assert "scrapeOptions" in data
+        scrape_data = data["scrapeOptions"]
+        assert "onlyMainContent" in scrape_data
+        assert "mobile" in scrape_data
+

firecrawl/__tests__/unit/v2/methods/aio/test_batch_request_preparation_async.py

@@ -0,0 +1,28 @@
+from firecrawl.v2.types import ScrapeOptions, Location
+from firecrawl.v2.methods.aio.batch import _prepare as _prepare_batch
+
+
+class TestAsyncBatchRequestPreparation:
+    def test_urls_validation_and_conversion(self):
+        payload = _prepare_batch(["https://example.com", "http://foo.bar"], options=None)
+        assert payload["urls"] == ["https://example.com", "http://foo.bar"]
+
+    def test_options_and_batch_fields(self):
+        opts = ScrapeOptions(formats=["markdown"], only_main_content=True)
+        payload = _prepare_batch(
+            ["https://example.com"],
+            options=opts,
+            webhook="https://hook.example",
+            append_to_id="00000000-0000-0000-0000-000000000000",
+            ignore_invalid_urls=True,
+            max_concurrency=3,
+            zero_data_retention=True,
+            integration="zapier",
+        )
+        assert payload["webhook"] == "https://hook.example"
+        assert payload["appendToId"] == "00000000-0000-0000-0000-000000000000"
+        assert payload["ignoreInvalidURLs"] is True
+        assert payload["maxConcurrency"] == 3
+        assert payload["zeroDataRetention"] is True
+        assert payload["integration"] == "zapier"
+