Skip to content

v2.38.0

Choose a tag to compare

@context-cross-repo-prod-writer context-cross-repo-prod-writer released this 17 Nov 18:26
137ebb1

2.38.0 (2026-11-17)

Features

  • access-policies: limit which websites API requests can reach (#2213) (9c97d2d)
  • access-policies: limit which websites API requests can reach (#2213) (e2290f4)
  • access-policies: limit which websites API requests can reach (#2213) (c31170c)
  • access-policies: limit which websites API requests can reach (#2213) (2374d4c)
  • access-policies: limit which websites API requests can reach (#2213) (7b7d307)
  • api: serve a public OpenAPI spec and generate the SDKs from it (#2075) (622d0c3)
  • batch: accept lightweight submissions and queue pending work (#2683) (94f185d)
  • batches: let sitemap crawls choose which paths count as non-pages (#2796) (e153dce)
  • batch: raise batch submissions to 2,500 per minute per organization (#2734) (82264dc)
  • billing: report plan facts in usage and limit errors (#2897) (6f99353)
  • brand: support log-free brand and styleguide requests (#2231) (9c97d2d)
  • brand: support log-free brand and styleguide requests (#2231) (e2290f4)
  • brand: support log-free brand and styleguide requests (#2231) (c31170c)
  • brand: support log-free brand and styleguide requests (#2231) (2374d4c)
  • monitors: add search target to watch the web for new developments (#2215) (9c97d2d)
  • monitors: add search target to watch the web for new developments (#2215) (e2290f4)
  • parse: mark PDF page numbers inline in Markdown with includeNumberedPages (#2110) (c60456e)
  • parse: support WARC archives in parsing and scraping (#2761) (d09556d)
  • people: add ultra mode to people search (#2149) (f9d3393)
  • scrape: add reddit-post prebuilt parser with full comment trees (#2217) (95a3632)
  • scrape: add Tripadvisor reviews parser (#2508) (8f10abd)
  • scrape: add Yelp reviews parser (ea66000)
  • scrape: add Yelp search parser (bdd2db7)
  • scrape: read framework-embedded page data in the json format (#2405) (12707d0)
  • scrape: return Yelp reviewer id and profile url in yelp-reviews (#2402) (4596198)
  • zdr: accept zero data retention on people search and company search (#2852) (84988b5)

Bug Fixes

  • brand: keep metadata-only usage logs for ZDR brand requests (#2234) (9c97d2d)
  • brand: keep metadata-only usage logs for ZDR brand requests (#2234) (e2290f4)
  • brand: keep metadata-only usage logs for ZDR brand requests (#2234) (c31170c)
  • openapi: keep legacy batch inputs in the spec as deprecated (#2225) (9c97d2d)
  • openapi: keep legacy batch inputs in the spec as deprecated (#2225) (e2290f4)
  • openapi: keep legacy batch inputs in the spec as deprecated (#2225) (c31170c)
  • openapi: keep legacy batch inputs in the spec as deprecated (#2225) (2374d4c)
  • openapi: keep legacy batch inputs in the spec as deprecated (#2225) (7b7d307)
  • openapi: keep legacy batch inputs in the spec as deprecated (#2225) (f45c764)
  • people: split ultra people ranking on its strongest phrase signals (#2209) (de365f8)
  • scrape: derive the /web/scrape request spec from its zod schema (#2044) (5193bc1)
  • scrape: restore Instagram profile posts from public pages (#2233) (9c97d2d)

Documentation

  • logs: say that a request log appears a few seconds after the request (#2564) (c60a6f0)
  • people: note that complex ultra people searches can take up to 60 seconds (#2206) (9c483ca)
  • scrape: mention transcripts on the Markdown format (#2903) (6f99353)

Refactors

  • fonts: share font link schemas and inferred types (6495548)