firecrawl 3.4.0__py3-none-any.whl → 4.1.0__py3-none-any.whl

firecrawl/v2/client.py

@@ -18,6 +18,7 @@ from .types import (
     CrawlResponse,
     CrawlJob,
     CrawlParamsRequest,
+    PDFParser,
     CrawlParamsData,
     WebhookConfig,
     CrawlErrorsResponse,
@@ -35,6 +36,7 @@ from .types import (
     ExecuteJavascriptAction,
     PDFAction,
     Location,
+    PaginationConfig,
 )
 from .utils.http_client import HttpClient
 from .utils.error_handler import FirecrawlError
@@ -104,7 +106,7 @@ class FirecrawlClient:
         timeout: Optional[int] = None,
         wait_for: Optional[int] = None,
         mobile: Optional[bool] = None,
-        parsers: Optional[List[str]] = None,
+        parsers: Optional[Union[List[str], List[Union[str, PDFParser]]]] = None,
         actions: Optional[List[Union['WaitAction', 'ScreenshotAction', 'ClickAction', 'WriteAction', 'PressAction', 'ScrollAction', 'ScrapeAction', 'ExecuteJavascriptAction', 'PDFAction']]] = None,
         location: Optional['Location'] = None,
         skip_tls_verification: Optional[bool] = None,
@@ -356,12 +358,17 @@ class FirecrawlClient:
         
         return crawl_module.start_crawl(self.http_client, request)
     
-    def get_crawl_status(self, job_id: str) -> CrawlJob:
+    def get_crawl_status(
+        self, 
+        job_id: str,
+        pagination_config: Optional[PaginationConfig] = None
+    ) -> CrawlJob:
         """
         Get the status of a crawl job.
         
         Args:
             job_id: ID of the crawl job
+            pagination_config: Optional configuration for pagination behavior
             
         Returns:
             CrawlJob with current status and data
@@ -369,7 +376,11 @@ class FirecrawlClient:
         Raises:
             Exception: If the status check fails
         """
-        return crawl_module.get_crawl_status(self.http_client, job_id)
+        return crawl_module.get_crawl_status(
+            self.http_client, 
+            job_id,
+            pagination_config=pagination_config
+        )
     
     def get_crawl_errors(self, crawl_id: str) -> CrawlErrorsResponse:
         """
@@ -561,7 +572,7 @@ class FirecrawlClient:
         timeout: Optional[int] = None,
         wait_for: Optional[int] = None,
         mobile: Optional[bool] = None,
-        parsers: Optional[List[str]] = None,
+        parsers: Optional[Union[List[str], List[Union[str, PDFParser]]]] = None,
         actions: Optional[List[Union['WaitAction', 'ScreenshotAction', 'ClickAction', 'WriteAction', 'PressAction', 'ScrollAction', 'ScrapeAction', 'ExecuteJavascriptAction', 'PDFAction']]] = None,
         location: Optional['Location'] = None,
         skip_tls_verification: Optional[bool] = None,
@@ -651,16 +662,25 @@ class FirecrawlClient:
             idempotency_key=idempotency_key,
         )
 
-    def get_batch_scrape_status(self, job_id: str):
+    def get_batch_scrape_status(
+        self, 
+        job_id: str,
+        pagination_config: Optional[PaginationConfig] = None
+    ):
         """Get current status and any scraped data for a batch job.
 
         Args:
             job_id: Batch job ID
+            pagination_config: Optional configuration for pagination behavior
 
         Returns:
             Status payload including counts and partial data
         """
-        return batch_module.get_batch_scrape_status(self.http_client, job_id)
+        return batch_module.get_batch_scrape_status(
+            self.http_client, 
+            job_id,
+            pagination_config=pagination_config
+        )
 
     def cancel_batch_scrape(self, job_id: str) -> bool:
         """Cancel a running batch scrape job.
@@ -740,7 +760,7 @@ class FirecrawlClient:
         timeout: Optional[int] = None,
         wait_for: Optional[int] = None,
         mobile: Optional[bool] = None,
-        parsers: Optional[List[str]] = None,
+        parsers: Optional[Union[List[str], List[Union[str, PDFParser]]]] = None,
         actions: Optional[List[Union['WaitAction', 'ScreenshotAction', 'ClickAction', 'WriteAction', 'PressAction', 'ScrollAction', 'ScrapeAction', 'ExecuteJavascriptAction', 'PDFAction']]] = None,
         location: Optional['Location'] = None,
         skip_tls_verification: Optional[bool] = None,

firecrawl/v2/client_async.py

@@ -31,6 +31,7 @@ from .types import (
     ExecuteJavascriptAction,
     PDFAction,
     Location,
+    PaginationConfig,
 )
 from .utils.http_client import HttpClient
 from .utils.http_client_async import AsyncHttpClient
@@ -94,8 +95,16 @@ class AsyncFirecrawlClient:
         timeout = kwargs.get("timeout")
         return await self.wait_crawl(resp.id, poll_interval=poll_interval, timeout=timeout)
 
-    async def get_crawl_status(self, job_id: str) -> CrawlJob:
-        return await async_crawl.get_crawl_status(self.async_http_client, job_id)
+    async def get_crawl_status(
+        self, 
+        job_id: str,
+        pagination_config: Optional[PaginationConfig] = None
+    ) -> CrawlJob:
+        return await async_crawl.get_crawl_status(
+            self.async_http_client, 
+            job_id,
+            pagination_config=pagination_config
+        )
 
     async def cancel_crawl(self, job_id: str) -> bool:
         return await async_crawl.cancel_crawl(self.async_http_client, job_id)
@@ -154,8 +163,16 @@ class AsyncFirecrawlClient:
         timeout = kwargs.get("timeout")
         return await self.wait_batch_scrape(job_id, poll_interval=poll_interval, timeout=timeout)
 
-    async def get_batch_scrape_status(self, job_id: str):
-        return await async_batch.get_batch_scrape_status(self.async_http_client, job_id)
+    async def get_batch_scrape_status(
+        self, 
+        job_id: str,
+        pagination_config: Optional[PaginationConfig] = None
+    ):
+        return await async_batch.get_batch_scrape_status(
+            self.async_http_client, 
+            job_id,
+            pagination_config=pagination_config
+        )
 
     async def cancel_batch_scrape(self, job_id: str) -> bool:
         return await async_batch.cancel_batch_scrape(self.async_http_client, job_id)

firecrawl/v2/methods/aio/batch.py

@@ -1,8 +1,10 @@
 from typing import Optional, List, Dict, Any
-from ...types import ScrapeOptions, WebhookConfig, Document, BatchScrapeResponse, BatchScrapeJob
+from ...types import ScrapeOptions, WebhookConfig, Document, BatchScrapeResponse, BatchScrapeJob, PaginationConfig
 from ...utils.http_client_async import AsyncHttpClient
 from ...utils.validation import prepare_scrape_options
 from ...utils.error_handler import handle_response_error
+from ...utils.normalize import normalize_document_input
+import time
 
 
 def _prepare(urls: List[str], *, options: Optional[ScrapeOptions] = None, **kwargs) -> Dict[str, Any]:
@@ -39,7 +41,25 @@ async def start_batch_scrape(client: AsyncHttpClient, urls: List[str], **kwargs)
     return BatchScrapeResponse(id=body.get("id"), url=body.get("url"), invalid_urls=body.get("invalidURLs"))
 
 
-async def get_batch_scrape_status(client: AsyncHttpClient, job_id: str) -> BatchScrapeJob:
+async def get_batch_scrape_status(
+    client: AsyncHttpClient, 
+    job_id: str,
+    pagination_config: Optional[PaginationConfig] = None
+) -> BatchScrapeJob:
+    """
+    Get the status of a batch scrape job.
+    
+    Args:
+        client: Async HTTP client instance
+        job_id: ID of the batch scrape job
+        pagination_config: Optional configuration for pagination behavior
+        
+    Returns:
+        BatchScrapeJob containing job status and data
+        
+    Raises:
+        Exception: If the status check fails
+    """
     response = await client.get(f"/v2/batch/scrape/{job_id}")
     if response.status_code >= 400:
         handle_response_error(response, "get batch scrape status")
@@ -49,23 +69,102 @@ async def get_batch_scrape_status(client: AsyncHttpClient, job_id: str) -> Batch
     docs: List[Document] = []
     for doc in body.get("data", []) or []:
         if isinstance(doc, dict):
-            normalized = dict(doc)
-            if 'rawHtml' in normalized and 'raw_html' not in normalized:
-                normalized['raw_html'] = normalized.pop('rawHtml')
-            if 'changeTracking' in normalized and 'change_tracking' not in normalized:
-                normalized['change_tracking'] = normalized.pop('changeTracking')
+            normalized = normalize_document_input(doc)
             docs.append(Document(**normalized))
+    
+    # Handle pagination if requested
+    auto_paginate = pagination_config.auto_paginate if pagination_config else True
+    if auto_paginate and body.get("next"):
+        docs = await _fetch_all_batch_pages_async(
+            client, 
+            body.get("next"), 
+            docs, 
+            pagination_config
+        )
+    
     return BatchScrapeJob(
         status=body.get("status"),
         completed=body.get("completed", 0),
         total=body.get("total", 0),
         credits_used=body.get("creditsUsed"),
         expires_at=body.get("expiresAt"),
-        next=body.get("next"),
+        next=body.get("next") if not auto_paginate else None,
         data=docs,
     )
 
 
+async def _fetch_all_batch_pages_async(
+    client: AsyncHttpClient,
+    next_url: str,
+    initial_documents: List[Document],
+    pagination_config: Optional[PaginationConfig] = None
+) -> List[Document]:
+    """
+    Fetch all pages of batch scrape results asynchronously.
+    
+    Args:
+        client: Async HTTP client instance
+        next_url: URL for the next page
+        initial_documents: Documents from the first page
+        pagination_config: Optional configuration for pagination limits
+        
+    Returns:
+        List of all documents from all pages
+    """
+    documents = initial_documents.copy()
+    current_url = next_url
+    page_count = 0
+    
+    # Apply pagination limits
+    max_pages = pagination_config.max_pages if pagination_config else None
+    max_results = pagination_config.max_results if pagination_config else None
+    max_wait_time = pagination_config.max_wait_time if pagination_config else None
+    
+    start_time = time.monotonic()
+    
+    while current_url:
+        # Check pagination limits
+        if (max_pages is not None) and (page_count >= max_pages):
+            break
+            
+        if (max_wait_time is not None) and (time.monotonic() - start_time) > max_wait_time:
+            break
+        
+        # Fetch next page
+        response = await client.get(current_url)
+        
+        if response.status_code >= 400:
+            # Log error but continue with what we have
+            import logging
+            logger = logging.getLogger("firecrawl")
+            logger.warning(f"Failed to fetch next page: {response.status_code}")
+            break
+        
+        page_data = response.json()
+        
+        if not page_data.get("success"):
+            break
+        
+        # Add documents from this page
+        for doc in page_data.get("data", []) or []:
+            if isinstance(doc, dict):
+                # Check max_results limit
+                if (max_results is not None) and (len(documents) >= max_results):
+                    break
+                normalized = normalize_document_input(doc)
+                documents.append(Document(**normalized))
+        
+        # Check if we hit max_results limit
+        if (max_results is not None) and (len(documents) >= max_results):
+            break
+        
+        # Get next URL
+        current_url = page_data.get("next")
+        page_count += 1
+    
+    return documents
+
+
 async def cancel_batch_scrape(client: AsyncHttpClient, job_id: str) -> bool:
     response = await client.delete(f"/v2/batch/scrape/{job_id}")
     if response.status_code >= 400:

firecrawl/v2/methods/aio/crawl.py

@@ -1,4 +1,4 @@
-from typing import Optional, Dict, Any
+from typing import Optional, Dict, Any, List
 from ...types import (
     CrawlRequest,
     CrawlJob,
@@ -10,11 +10,13 @@ from ...types import (
     CrawlErrorsResponse,
     ActiveCrawlsResponse,
     ActiveCrawl,
+    PaginationConfig,
 )
 from ...utils.error_handler import handle_response_error
 from ...utils.validation import prepare_scrape_options
 from ...utils.http_client_async import AsyncHttpClient
 from ...utils.normalize import normalize_document_input
+import time
 
 
 def _prepare_crawl_request(request: CrawlRequest) -> dict:
@@ -58,6 +60,20 @@ def _prepare_crawl_request(request: CrawlRequest) -> dict:
 
 
 async def start_crawl(client: AsyncHttpClient, request: CrawlRequest) -> CrawlResponse:
+    """
+    Start a crawl job for a website.
+    
+    Args:
+        client: Async HTTP client instance
+        request: CrawlRequest containing URL and options
+        
+    Returns:
+        CrawlResponse with job information
+        
+    Raises:
+        ValueError: If request is invalid
+        Exception: If the crawl operation fails to start
+    """
     payload = _prepare_crawl_request(request)
     response = await client.post("/v2/crawl", payload)
     if response.status_code >= 400:
@@ -68,7 +84,25 @@ async def start_crawl(client: AsyncHttpClient, request: CrawlRequest) -> CrawlRe
     raise Exception(body.get("error", "Unknown error occurred"))
 
 
-async def get_crawl_status(client: AsyncHttpClient, job_id: str) -> CrawlJob:
+async def get_crawl_status(
+    client: AsyncHttpClient, 
+    job_id: str,
+    pagination_config: Optional[PaginationConfig] = None
+) -> CrawlJob:
+    """
+    Get the status of a crawl job.
+    
+    Args:
+        client: Async HTTP client instance
+        job_id: ID of the crawl job
+        pagination_config: Optional configuration for pagination limits
+        
+    Returns:
+        CrawlJob with job information
+        
+    Raises:
+        Exception: If the status check fails
+    """
     response = await client.get(f"/v2/crawl/{job_id}")
     if response.status_code >= 400:
         handle_response_error(response, "get crawl status")
@@ -79,19 +113,115 @@ async def get_crawl_status(client: AsyncHttpClient, job_id: str) -> CrawlJob:
             if isinstance(doc_data, dict):
                 normalized = normalize_document_input(doc_data)
                 documents.append(Document(**normalized))
+        
+        # Handle pagination if requested
+        auto_paginate = pagination_config.auto_paginate if pagination_config else True
+        if auto_paginate and body.get("next"):
+            documents = await _fetch_all_pages_async(
+                client, 
+                body.get("next"), 
+                documents, 
+                pagination_config
+            )
+        
         return CrawlJob(
             status=body.get("status"),
             completed=body.get("completed", 0),
             total=body.get("total", 0),
             credits_used=body.get("creditsUsed", 0),
             expires_at=body.get("expiresAt"),
-            next=body.get("next"),
+            next=body.get("next") if not auto_paginate else None,
             data=documents,
         )
     raise Exception(body.get("error", "Unknown error occurred"))
 
 
+async def _fetch_all_pages_async(
+    client: AsyncHttpClient,
+    next_url: str,
+    initial_documents: List[Document],
+    pagination_config: Optional[PaginationConfig] = None
+) -> List[Document]:
+    """
+    Fetch all pages of crawl results asynchronously.
+    
+    Args:
+        client: Async HTTP client instance
+        next_url: URL for the next page
+        initial_documents: Documents from the first page
+        pagination_config: Optional configuration for pagination limits
+        
+    Returns:
+        List of all documents from all pages
+    """
+    documents = initial_documents.copy()
+    current_url = next_url
+    page_count = 0
+    
+    # Apply pagination limits
+    max_pages = pagination_config.max_pages if pagination_config else None
+    max_results = pagination_config.max_results if pagination_config else None
+    max_wait_time = pagination_config.max_wait_time if pagination_config else None
+    
+    start_time = time.monotonic()
+    
+    while current_url:
+        # Check pagination limits (treat 0 as a valid limit)
+        if (max_pages is not None) and page_count >= max_pages:
+            break
+
+        if (max_wait_time is not None) and (time.monotonic() - start_time) > max_wait_time:
+            break
+        
+        # Fetch next page
+        response = await client.get(current_url)
+        
+        if response.status_code >= 400:
+            # Log error but continue with what we have
+            import logging
+            logger = logging.getLogger("firecrawl")
+            logger.warning("Failed to fetch next page", extra={"status_code": response.status_code})
+            break
+        
+        page_data = response.json()
+        
+        if not page_data.get("success"):
+            break
+        
+        # Add documents from this page
+        for doc_data in page_data.get("data", []):
+            if isinstance(doc_data, dict):
+                # Check max_results limit
+                if (max_results is not None) and (len(documents) >= max_results):
+                    break
+                normalized = normalize_document_input(doc_data)
+                documents.append(Document(**normalized))
+        
+        # Check if we hit max_results limit
+        if (max_results is not None) and (len(documents) >= max_results):
+            break
+        
+        # Get next URL
+        current_url = page_data.get("next")
+        page_count += 1
+    
+    return documents
+
+
 async def cancel_crawl(client: AsyncHttpClient, job_id: str) -> bool:
+    """
+    Cancel a crawl job.
+    
+    Args:
+        client: Async HTTP client instance
+        job_id: ID of the crawl job
+        
+    Returns:
+        True if cancellation was successful
+        
+    Raises:
+        Exception: If the cancellation operation fails
+    """
     response = await client.delete(f"/v2/crawl/{job_id}")
     if response.status_code >= 400:
         handle_response_error(response, "cancel crawl")
@@ -100,6 +230,20 @@ async def cancel_crawl(client: AsyncHttpClient, job_id: str) -> bool:
 
 
 async def crawl_params_preview(client: AsyncHttpClient, request: CrawlParamsRequest) -> CrawlParamsData:
+    """
+    Preview crawl parameters before starting a crawl job.
+    
+    Args:
+        client: Async HTTP client instance
+        request: CrawlParamsRequest containing URL and prompt
+        
+    Returns:
+        CrawlParamsData containing crawl configuration
+        
+    Raises:
+        ValueError: If request is invalid
+        Exception: If the parameter preview fails
+    """
     if not request.url or not request.url.strip():
         raise ValueError("URL cannot be empty")
     if not request.prompt or not request.prompt.strip():
@@ -138,6 +282,19 @@ async def crawl_params_preview(client: AsyncHttpClient, request: CrawlParamsRequ
 
 
 async def get_crawl_errors(client: AsyncHttpClient, crawl_id: str) -> CrawlErrorsResponse:
+    """
+    Get errors from a crawl job.
+    
+    Args:
+        client: Async HTTP client instance
+        crawl_id: ID of the crawl job
+        
+    Returns:
+        CrawlErrorsResponse with errors and robots blocked
+        
+    Raises:
+        Exception: If the error check operation fails
+    """
     response = await client.get(f"/v2/crawl/{crawl_id}/errors")
     if response.status_code >= 400:
         handle_response_error(response, "check crawl errors")
@@ -151,6 +308,18 @@ async def get_crawl_errors(client: AsyncHttpClient, crawl_id: str) -> CrawlError
 
 
 async def get_active_crawls(client: AsyncHttpClient) -> ActiveCrawlsResponse:
+    """
+    Get active crawl jobs.
+    
+    Args:
+        client: Async HTTP client instance
+        
+    Returns:
+        ActiveCrawlsResponse with active crawl jobs
+        
+    Raises:
+        Exception: If the active crawl jobs operation fails
+    """
     response = await client.get("/v2/crawl/active")
     if response.status_code >= 400:
         handle_response_error(response, "get active crawls")

firecrawl/v2/methods/batch.py

@@ -11,6 +11,7 @@ from ..types import (
     ScrapeOptions,
     Document,
     WebhookConfig,
+    PaginationConfig,
 )
 from ..utils import HttpClient, handle_response_error, validate_scrape_options, prepare_scrape_options
 from ..utils.normalize import normalize_document_input
@@ -77,7 +78,8 @@ def start_batch_scrape(
 
 def get_batch_scrape_status(
     client: HttpClient,
-    job_id: str
+    job_id: str,
+    pagination_config: Optional[PaginationConfig] = None
 ) -> BatchScrapeJob:
     """
     Get the status of a batch scrape job.
@@ -85,9 +87,10 @@ def get_batch_scrape_status(
     Args:
         client: HTTP client instance
         job_id: ID of the batch scrape job
+        pagination_config: Optional configuration for pagination behavior
         
     Returns:
-        BatchScrapeStatusResponse containing job status and data
+        BatchScrapeJob containing job status and data
         
     Raises:
         FirecrawlError: If the status check fails
@@ -111,17 +114,99 @@ def get_batch_scrape_status(
             normalized = normalize_document_input(doc)
             documents.append(Document(**normalized))
 
+    # Handle pagination if requested
+    auto_paginate = pagination_config.auto_paginate if pagination_config else True
+    if auto_paginate and body.get("next"):
+        documents = _fetch_all_batch_pages(
+            client, 
+            body.get("next"), 
+            documents, 
+            pagination_config
+        )
+
     return BatchScrapeJob(
         status=body.get("status"),
         completed=body.get("completed", 0),
         total=body.get("total", 0),
         credits_used=body.get("creditsUsed"),
         expires_at=body.get("expiresAt"),
-        next=body.get("next"),
+        next=body.get("next") if not auto_paginate else None,
         data=documents,
     )
 
 
+def _fetch_all_batch_pages(
+    client: HttpClient,
+    next_url: str,
+    initial_documents: List[Document],
+    pagination_config: Optional[PaginationConfig] = None
+) -> List[Document]:
+    """
+    Fetch all pages of batch scrape results.
+    
+    Args:
+        client: HTTP client instance
+        next_url: URL for the next page
+        initial_documents: Documents from the first page
+        pagination_config: Optional configuration for pagination limits
+        
+    Returns:
+        List of all documents from all pages
+    """
+    documents = initial_documents.copy()
+    current_url = next_url
+    page_count = 0
+    
+    # Apply pagination limits
+    max_pages = pagination_config.max_pages if pagination_config else None
+    max_results = pagination_config.max_results if pagination_config else None
+    max_wait_time = pagination_config.max_wait_time if pagination_config else None
+    
+    start_time = time.monotonic()
+    
+    while current_url:
+        # Check pagination limits (treat 0 as a valid limit)
+        if (max_pages is not None) and page_count >= max_pages:
+            break
+        
+        if (max_wait_time is not None) and (time.monotonic() - start_time) > max_wait_time:
+            break
+        
+        # Fetch next page
+        response = client.get(current_url)
+        
+        if not response.ok:
+            # Log error but continue with what we have
+            import logging
+            logger = logging.getLogger("firecrawl")
+            logger.warning("Failed to fetch next page", extra={"status_code": response.status_code})
+            break
+        
+        page_data = response.json()
+        
+        if not page_data.get("success"):
+            break
+        
+        # Add documents from this page
+        for doc in page_data.get("data", []) or []:
+            if isinstance(doc, dict):
+                # Check max_results limit
+                if max_results is not None and len(documents) >= max_results:
+                    break
+                normalized = normalize_document_input(doc)
+                documents.append(Document(**normalized))
+        
+        # Check if we hit max_results limit after adding all docs from this page
+        if max_results is not None and len(documents) >= max_results:
+            break
+        
+        # Get next URL
+        current_url = page_data.get("next")
+        page_count += 1
+    
+    return documents
+
+
 def cancel_batch_scrape(
     client: HttpClient,
     job_id: str
@@ -173,7 +258,7 @@ def wait_for_batch_completion(
         FirecrawlError: If the job fails or timeout is reached
         TimeoutError: If timeout is reached
     """
-    start_time = time.time()
+    start_time = time.monotonic()
     
     while True:
         status_job = get_batch_scrape_status(client, job_id)
@@ -183,7 +268,7 @@ def wait_for_batch_completion(
             return status_job
         
         # Check timeout
-        if timeout and (time.time() - start_time) > timeout:
+        if timeout and (time.monotonic() - start_time) > timeout:
             raise TimeoutError(f"Batch scrape job {job_id} did not complete within {timeout} seconds")
         
         # Wait before next poll