firecrawl 4.12.0__py3-none-any.whl

firecrawl/__tests__/e2e/v2/aio/test_aio_search.py

@@ -0,0 +1,249 @@
+import os
+import pytest
+from dotenv import load_dotenv
+from firecrawl import AsyncFirecrawl
+from firecrawl.types import (
+    SearchData,
+    Document,
+    ScrapeOptions,
+    ScrapeFormats,
+    SearchResultWeb,
+    SearchResultNews,
+    SearchResultImages,
+)
+
+load_dotenv()
+
+firecrawl = AsyncFirecrawl(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')
+
+@pytest.mark.asyncio
+async def test_async_search_minimal_request():
+    results = await 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, SearchResultWeb)
+        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
+
+@pytest.mark.asyncio
+async def test_async_search_with_sources():
+    results = await firecrawl.search(
+        query="firecrawl",
+        sources=["web", "news", "images"],
+        limit=3
+    )
+    assert isinstance(results, SearchData)
+    assert results.web is not None
+    assert len(results.web) <= 3
+    assert isinstance(results.web[0], SearchResultWeb)
+
+    if results.news is not None:
+        assert len(results.news) <= 3
+        assert isinstance(results.news[0], SearchResultNews)
+
+    if results.images is not None:
+        assert len(results.images) <= 3
+        assert isinstance(results.images[0], SearchResultImages)
+
+    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
+
+@pytest.mark.asyncio
+async def test_async_search_result_structure():
+    results = await 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
+        assert result.url.startswith('http')
+
+@pytest.mark.asyncio
+async def test_async_search_all_parameters():
+    from firecrawl.types import ScrapeOptions, Location, WaitAction
+    schema = {
+        "type": "object",
+        "properties": {
+            "title": {"type": "string"},
+            "description": {"type": "string"},
+            "url": {"type": "string"}
+        },
+        "required": ["title", "description"]
+    }
+    results = await firecrawl.search(
+        query="artificial intelligence",
+        sources=[
+            {"type": "web"},
+            {"type": "news"}
+        ],
+        limit=3,
+        tbs="qdr:m",
+        location="US",
+        ignore_invalid_urls=True,
+        timeout=60000,
+        integration="_e2e-test",
+        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,
+            store_in_cache=True,
+            location=Location(
+                country="US",
+                languages=["en"]
+            ),
+            actions=[
+                WaitAction(milliseconds=1000)
+            ]
+        )
+    )
+    assert isinstance(results, SearchData)
+    assert hasattr(results, 'web')
+    assert hasattr(results, 'news')
+    assert hasattr(results, 'images')
+    assert results.web is not None
+    assert len(results.web) <= 3
+
+    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)
+
+    for result in results.web:
+        assert isinstance(result, (SearchResultWeb, 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')
+
+    if results.news is not None:
+        assert len(results.news) <= 3
+        for result in results.news:
+            assert isinstance(result, (SearchResultNews, 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')
+
+    assert results.images is None
+
+@pytest.mark.asyncio
+async def test_async_search_formats_flexibility():
+    # Test with list format
+    results1 = await firecrawl.search(
+        query="python programming",
+        limit=1,
+        scrape_options=ScrapeOptions(
+            formats=["markdown"]
+        )
+    )
+    # Test with ScrapeFormats object
+    results2 = await firecrawl.search(
+        query="python programming",
+        limit=1,
+        scrape_options=ScrapeOptions(
+            formats=ScrapeFormats(markdown=True)
+        )
+    )
+    assert isinstance(results1, SearchData)
+    assert isinstance(results2, SearchData)
+    assert results1.web is not None
+    assert results2.web is not None
+
+@pytest.mark.asyncio
+async def test_async_search_with_json_format_object():
+    json_schema = {
+        "type": "object",
+        "properties": {
+            "title": {"type": "string"}
+        },
+        "required": ["title"],
+    }
+    results = await 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/aio/test_aio_usage.py

@@ -0,0 +1,42 @@
+import os
+import pytest
+from dotenv import load_dotenv
+from firecrawl import AsyncFirecrawl
+
+
+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")
+
+
+@pytest.mark.asyncio
+async def test_async_get_concurrency():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    conc = await client.get_concurrency()
+    assert hasattr(conc, "concurrency") and hasattr(conc, "max_concurrency")
+
+
+@pytest.mark.asyncio
+async def test_async_get_credit_usage():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    credits = await client.get_credit_usage()
+    assert hasattr(credits, "remaining_credits")
+
+
+@pytest.mark.asyncio
+async def test_async_get_token_usage():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    tokens = await client.get_token_usage()
+    assert hasattr(tokens, "remaining_tokens")
+
+
+@pytest.mark.asyncio
+async def test_async_get_queue_status():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    status = await client.get_queue_status()
+    assert hasattr(status, "jobs_in_queue")
+

firecrawl/__tests__/e2e/v2/aio/test_aio_watcher.py

@@ -0,0 +1,43 @@
+import os
+import asyncio
+import pytest
+from dotenv import load_dotenv
+from firecrawl import AsyncFirecrawl
+from firecrawl.v2.watcher_async import AsyncWatcher
+
+
+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")
+
+
+@pytest.mark.asyncio
+async def test_async_watcher_crawl_progresses():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    start = await client.start_crawl("https://docs.firecrawl.dev", limit=2)
+    statuses = []
+    async for snapshot in AsyncWatcher(client, start.id, kind="crawl", timeout=180):
+        statuses.append(snapshot.status)
+        if snapshot.status in ("completed", "failed"):
+            break
+    assert statuses and statuses[-1] in ("completed", "failed")
+
+
+@pytest.mark.asyncio
+async def test_async_watcher_batch_progresses():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    start = await client.start_batch_scrape([
+        "https://docs.firecrawl.dev",
+        "https://firecrawl.dev",
+    ], formats=["markdown"], max_concurrency=1)
+    statuses = []
+    async for snapshot in AsyncWatcher(client, start.id, kind="batch", timeout=240):
+        statuses.append(snapshot.status)
+        if snapshot.status in ("completed", "failed", "cancelled"):
+            break
+    assert statuses and statuses[-1] in ("completed", "failed", "cancelled")
+

firecrawl/__tests__/e2e/v2/conftest.py

@@ -0,0 +1,73 @@
+import os
+import json
+import pytest
+import requests
+from dotenv import load_dotenv
+
+load_dotenv()
+
+def _idmux(identity_request: dict) -> dict:
+    idmux_url = os.getenv("IDMUX_URL")
+    if not idmux_url:
+        raise EnvironmentError("IDMUX_URL is not set. E2E tests must use idmux for credentials.")
+    run_number = int(os.getenv("GITHUB_RUN_NUMBER") or 0)
+    payload = {
+        "refName": os.getenv("GITHUB_REF_NAME") or "local",
+        "runNumber": run_number,
+        "concurrency": identity_request.get("concurrency", 100),
+        **identity_request,
+    }
+    resp = requests.post(idmux_url + "/", json=payload)
+    resp.raise_for_status()
+    return resp.json()
+
+@pytest.fixture(scope="session")
+def api_url():
+    # Prefer TEST_URL, then FIRECRAWL_API_URL (for parity with JS), then legacy API_URL
+    return (
+        os.getenv("TEST_URL")
+        or os.getenv("FIRECRAWL_API_URL")
+        or os.getenv("API_URL")
+        or "https://api.firecrawl.dev"
+    )
+
+# Resolve identity and export environment at import time so tests that read env at module import succeed
+_IDENTITY = None
+_API_URL = (
+    os.getenv("TEST_URL")
+    or os.getenv("FIRECRAWL_API_URL")
+    or os.getenv("API_URL")
+    or "https://api.firecrawl.dev"
+)
+
+_IDMUX_URL = os.getenv("IDMUX_URL")
+if _IDMUX_URL:
+    run_name = os.getenv("PYTEST_RUN_NAME") or "py-e2e"
+    # If IDMUX_URL is set, idmux MUST succeed; do not silently fall back
+    _IDENTITY = _idmux({"name": run_name})
+    os.environ["API_KEY"] = _IDENTITY.get("apiKey", "")
+    os.environ["API_URL"] = _API_URL
+
+@pytest.fixture(scope="session")
+def api_identity():
+    return _IDENTITY or {"apiKey": os.getenv("API_KEY") or "", "teamId": os.getenv("TEST_TEAM_ID") or os.getenv("TEAM_ID") or ""}
+
+@pytest.fixture(autouse=True)
+def _inject_client(request, api_identity, api_url):
+    # For class-based tests that rely on self.client, inject a client if missing
+    inst = getattr(request, "instance", None)
+    if inst is not None and not hasattr(inst, "client"):
+        try:
+            from firecrawl import Firecrawl
+            inst.client = Firecrawl(api_key=api_identity.get("apiKey", ""), api_url=api_url)
+        except Exception:
+            pass
+    # For function-based modules that expect a module-level `firecrawl` symbol
+    mod = getattr(request, "module", None)
+    if mod is not None and not hasattr(mod, "firecrawl"):
+        try:
+            from firecrawl import Firecrawl
+            setattr(mod, "firecrawl", Firecrawl(api_key=api_identity.get("apiKey", ""), api_url=api_url))
+        except Exception:
+            pass
+

firecrawl/__tests__/e2e/v2/test_async.py

@@ -0,0 +1,73 @@
+import os
+import asyncio
+import pytest
+from dotenv import load_dotenv
+
+from firecrawl import AsyncFirecrawl
+from firecrawl.v2.types import Document
+
+
+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")
+
+
+@pytest.mark.asyncio
+async def test_async_scrape_minimal():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    doc = await client.scrape("https://docs.firecrawl.dev")
+    assert isinstance(doc, Document)
+    # Accept any primary content or alternate outputs
+    assert doc.markdown is not None and doc.markdown and len(doc.markdown) > 0
+
+
+@pytest.mark.asyncio
+async def test_async_crawl_start_and_status():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    start = await client.start_crawl("https://docs.firecrawl.dev", limit=2)
+    job_id = start.id
+
+    # Poll status until terminal or timeout
+    deadline = asyncio.get_event_loop().time() + 180
+    status = await client.get_crawl_status(job_id)
+    while status.status not in ("completed", "failed") and asyncio.get_event_loop().time() < deadline:
+        await asyncio.sleep(2)
+        status = await client.get_crawl_status(job_id)
+
+    assert status.status in ("completed", "failed")
+
+
+@pytest.mark.asyncio
+async def test_async_batch_start_and_status():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    start = await client.start_batch_scrape([
+        "https://docs.firecrawl.dev",
+        "https://firecrawl.dev",
+    ], formats=["markdown"], max_concurrency=1)
+    job_id = start.id
+
+    deadline = asyncio.get_event_loop().time() + 240
+    status = await client.get_batch_scrape_status(job_id)
+    while status.status not in ("completed", "failed", "cancelled") and asyncio.get_event_loop().time() < deadline:
+        await asyncio.sleep(2)
+        status = await client.get_batch_scrape_status(job_id)
+
+    assert status.status in ("completed", "failed", "cancelled")
+
+
+@pytest.mark.asyncio
+async def test_async_usage_minimal():
+    client = AsyncFirecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+    conc = await client.get_concurrency()
+    assert hasattr(conc, "concurrency") and hasattr(conc, "max_concurrency")
+
+    credits = await client.get_credit_usage()
+    assert hasattr(credits, "remaining_credits")
+
+    tokens = await client.get_token_usage()
+    assert hasattr(tokens, "remaining_tokens")
+

firecrawl/__tests__/e2e/v2/test_batch_scrape.py

@@ -0,0 +1,106 @@
+import os
+import pytest
+from dotenv import load_dotenv
+from firecrawl import Firecrawl
+from firecrawl.v2.types import ScrapeOptions
+
+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 TestBatchScrapeE2E:
+    """End-to-end tests for batch scrape (v2)."""
+
+    def setup_method(self):
+        self.client = Firecrawl(api_key=os.getenv("API_KEY"), api_url=os.getenv("API_URL"))
+
+    def test_batch_scrape_minimal(self):
+        """Start a small batch and wait for completion."""
+        urls = [
+            "https://docs.firecrawl.dev",
+            "https://firecrawl.dev",
+        ]
+
+        job = self.client.batch_scrape(urls, formats=["markdown"], poll_interval=1, wait_timeout=120)
+
+        assert job.status in ["completed", "failed"]
+        assert job.completed >= 0
+        assert job.total >= 0
+        assert isinstance(job.data, list)
+
+    def test_start_batch_minimal_and_status(self):
+        """Start via start_batch_scrape (minimal), then fetch status once."""
+        urls = [
+            "https://docs.firecrawl.dev",
+            "https://firecrawl.dev",
+        ]
+
+        start_resp = self.client.start_batch_scrape(urls, formats=["markdown"], ignore_invalid_urls=True)
+        assert start_resp.id is not None
+        assert start_resp.url is not None
+
+        job = self.client.get_batch_scrape_status(start_resp.id)
+        assert job.status in ["scraping", "completed", "failed"]
+        assert job.total >= 0
+
+    def test_wait_batch_with_all_params(self):
+        """Blocking waiter with JSON and changeTracking formats plus many options."""
+        urls = [
+            "https://docs.firecrawl.dev",
+            "https://firecrawl.dev",
+        ]
+
+        json_schema = {
+            "type": "object",
+            "properties": {
+                "title": {"type": "string"}
+            },
+            "required": ["title"],
+        }
+
+        opts = ScrapeOptions(
+            formats=[
+                "markdown",
+                {"type": "json", "prompt": "Extract page title", "schema": json_schema},
+                {"type": "changeTracking", "prompt": "Track changes", "modes": ["json"]},
+            ],
+            only_main_content=True,
+            mobile=False,
+        )
+
+        job = self.client.batch_scrape(
+            urls,
+            formats=opts.formats,
+            only_main_content=opts.only_main_content,
+            mobile=opts.mobile,
+            ignore_invalid_urls=True,
+            max_concurrency=2,
+            zero_data_retention=False,
+            poll_interval=1,
+            wait_timeout=180,
+            integration="_e2e-test",
+        )
+
+        assert job.status in ["completed", "failed"]
+        assert job.completed >= 0
+        assert job.total >= 0
+        assert isinstance(job.data, list)
+
+    def test_cancel_batch(self):
+        """Start a batch and cancel it."""
+        urls = [
+            "https://docs.firecrawl.dev",
+            "https://firecrawl.dev",
+        ]
+
+        start_resp = self.client.start_batch_scrape(urls, formats=["markdown"], max_concurrency=1)
+        assert start_resp.id is not None
+
+        cancelled = self.client.cancel_batch_scrape(start_resp.id)
+        assert cancelled is True
+