feat(batch): support Scrape Anything batches up to 100k URLs (#1883)

Stainless-Generated-From: 1c12fc91530bca169d3625153736b147c679d944

Author: stlc-bot

src/context/dev/resources/batch.py

@@ -95,6 +95,7 @@ def list(
         limit: int | Omit = omit,
         q: str | Omit = omit,
         search_type: Literal["exact", "prefix"] | Omit = omit,
+        should_include_unified: bool | Omit = omit,
         status: Literal["queued", "running", "cancelling", "completed", "cancelled", "failed"] | Omit = omit,
         tags: Union[str, SequenceNotStr[str]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -118,6 +119,8 @@ def list(
           search_type: `prefix` for as-you-type prefix matching (default), `exact` for full-token
               matching.
 
+          should_include_unified: Include batches submitted with Scrape Anything formats. Defaults to false.
+
           status: Filter by status.
 
           tags: Tags to filter by (matches batches having any of them). Pass repeated `tags`
@@ -144,6 +147,7 @@ def list(
                         "limit": limit,
                         "q": q,
                         "search_type": search_type,
+                        "should_include_unified": should_include_unified,
                         "status": status,
                         "tags": tags,
                     },
@@ -242,8 +246,9 @@ def get_results(
     ) -> BatchGetResultsResponse:
         """Page through a finished batch’s results as JSON.
 
-        Results remain available for
-        180 days.
+        Each record contains result
+        with the same outputs and per-format errors as POST /web/scrape. Results remain
+        available for 180 days.
 
         Args:
           batch_id: Batch ID.
@@ -296,10 +301,11 @@ def submit(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> BatchSubmitResponse:
-        """Scrape up to 25,000 URLs, or crawl a site, asynchronously.
-
-        Poll the batch ID or
-        receive a webhook when it finishes.
+        """
+        Scrape up to 100,000 URLs, or crawl a site, asynchronously using the same
+        formats and scrape parameters as POST /web/scrape. Choose Markdown, HTML,
+        screenshots, images, bytes, parsed fields, products, highlights, and JSON. Poll
+        the batch ID or receive a webhook when it finishes.
 
         Args:
           input: Choose a URL list or a site crawl.
@@ -408,6 +414,7 @@ async def list(
         limit: int | Omit = omit,
         q: str | Omit = omit,
         search_type: Literal["exact", "prefix"] | Omit = omit,
+        should_include_unified: bool | Omit = omit,
         status: Literal["queued", "running", "cancelling", "completed", "cancelled", "failed"] | Omit = omit,
         tags: Union[str, SequenceNotStr[str]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
@@ -431,6 +438,8 @@ async def list(
           search_type: `prefix` for as-you-type prefix matching (default), `exact` for full-token
               matching.
 
+          should_include_unified: Include batches submitted with Scrape Anything formats. Defaults to false.
+
           status: Filter by status.
 
           tags: Tags to filter by (matches batches having any of them). Pass repeated `tags`
@@ -457,6 +466,7 @@ async def list(
                         "limit": limit,
                         "q": q,
                         "search_type": search_type,
+                        "should_include_unified": should_include_unified,
                         "status": status,
                         "tags": tags,
                     },
@@ -555,8 +565,9 @@ async def get_results(
     ) -> BatchGetResultsResponse:
         """Page through a finished batch’s results as JSON.
 
-        Results remain available for
-        180 days.
+        Each record contains result
+        with the same outputs and per-format errors as POST /web/scrape. Results remain
+        available for 180 days.
 
         Args:
           batch_id: Batch ID.
@@ -609,10 +620,11 @@ async def submit(
         extra_body: Body | None = None,
         timeout: float | httpx.Timeout | None | NotGiven = not_given,
     ) -> BatchSubmitResponse:
-        """Scrape up to 25,000 URLs, or crawl a site, asynchronously.
-
-        Poll the batch ID or
-        receive a webhook when it finishes.
+        """
+        Scrape up to 100,000 URLs, or crawl a site, asynchronously using the same
+        formats and scrape parameters as POST /web/scrape. Choose Markdown, HTML,
+        screenshots, images, bytes, parsed fields, products, highlights, and JSON. Poll
+        the batch ID or receive a webhook when it finishes.
 
         Args:
           input: Choose a URL list or a site crawl.

src/context/dev/types/batch_cancel_response.py

@@ -3,12 +3,14 @@
 from typing import List, Optional
 from typing_extensions import Literal
 
+from pydantic import Field as FieldInfo
+
 from .intake import Intake
 from .._models import BaseModel
 from .crawl_controls import CrawlControls
 from .page_error_count import PageErrorCount
 
-__all__ = ["BatchCancelResponse", "Credits", "Progress", "Timing", "KeyMetadata"]
+__all__ = ["BatchCancelResponse", "Credits", "Formats", "Progress", "Timing", "KeyMetadata"]
 
 
 class Credits(BaseModel):
@@ -18,6 +20,37 @@ class Credits(BaseModel):
     """Credits held at submission; unused credits are refunded when the batch settles."""
 
 
+class Formats(BaseModel):
+    """Outputs requested for each page. Matches input.data.formats on submission."""
+
+    bytes: Optional[bool] = None
+    """The original HTTP response body."""
+
+    highlights: Optional[bool] = None
+    """Markdown excerpts relevant to `highlightsParams.query`."""
+
+    html: Optional[bool] = None
+    """Rendered HTML."""
+
+    images: Optional[bool] = None
+    """Images found on the page."""
+
+    json_: Optional[bool] = FieldInfo(alias="json", default=None)
+    """An object matching `jsonParams.schema`, extracted from the page."""
+
+    markdown: Optional[bool] = None
+    """Page content as Markdown."""
+
+    parse: Optional[bool] = None
+    """Fields extracted with custom rules or a prebuilt parser, returned as `parsed`."""
+
+    product: Optional[bool] = None
+    """Product details such as name, price, and availability."""
+
+    screenshot: Optional[bool] = None
+    """A screenshot of the page."""
+
+
 class Progress(BaseModel):
     """How far the batch got before cancellation."""
 
@@ -61,8 +94,8 @@ class BatchCancelResponse(BaseModel):
     credits: Credits
     """What this batch cost so far."""
 
-    format: Literal["markdown", "html"]
-    """What each page is returned as."""
+    formats: Formats
+    """Outputs requested for each page. Matches input.data.formats on submission."""
 
     input: Intake
     """What the submission accepted."""