feat(scrape): add Yelp reviews parser

Merge pull request #2307 from context-dot-dev/feat/yelp-reviews-parser

Stainless-Generated-From: 0e315e503e18ef2869595fc64880edb2674536ad

Author: stlc-bot

src/context/dev/types/batch_get_results_response.py

@@ -294,7 +294,7 @@ class DataOkResultMetadata(BaseModel):
 
 class DataOkResultParsed(BaseModel):
     """
-    Custom rules return fields after content filters; unmatched fields are null ([] for lists). The google-maps parser returns business records in data.results, google-maps-reviews returns data.reviews, and the google-play parser returns one app record as data, google-play-reviews returns data.reviews, app-store and app-store-reviews do the same for the App Store, g2-reviews returns data.reviews for a G2 product, instagram-profile and instagram-post return one record as data, instagram-feed returns data.posts, tiktok-profile and tiktok-video return one record as data, tiktok-feed returns data.videos, tiktok-comments returns data.comments, tiktok-followers, tiktok-following, and tiktok-user-search return data.users, yelp-search returns up to 20 ranked organic businesses in data.businesses, plus total and pagination, from a Yelp search URL; emails, phone-numbers, and links return data.emails, data.phone_numbers, and data.links from the rendered page; the store, Maps, G2, Instagram, and TikTok parsers read their data directly, independently of DOM filters and other outputs.
+    Custom rules return fields after content filters; unmatched fields are null ([] for lists). The google-maps parser returns business records in data.results, google-maps-reviews returns data.reviews, and the google-play parser returns one app record as data, google-play-reviews returns data.reviews, app-store and app-store-reviews do the same for the App Store, g2-reviews returns data.reviews for a G2 product, instagram-profile and instagram-post return one record as data, instagram-feed returns data.posts, tiktok-profile and tiktok-video return one record as data, tiktok-feed returns data.videos, tiktok-comments returns data.comments, tiktok-followers, tiktok-following, and tiktok-user-search return data.users, yelp-search returns up to 20 ranked organic businesses in data.businesses, plus total and pagination, from a Yelp search URL; yelp-reviews returns up to 20 latest public reviews in data.reviews, with business details and next_offset, from a Yelp business URL; emails, phone-numbers, and links return data.emails, data.phone_numbers, and data.links from the rendered page; the store, Maps, G2, Instagram, and TikTok parsers read their data directly, independently of DOM filters and other outputs.
     """
 
     data: Optional[Dict[str, object]] = None
@@ -324,6 +324,7 @@ class DataOkResultParsed(BaseModel):
         "tiktok-following",
         "tiktok-user-search",
         "yelp-search",
+        "yelp-reviews",
         "emails",
         "phone-numbers",
         "links",
@@ -524,10 +525,12 @@ class DataOkResult(BaseModel):
     one record as data, tiktok-feed returns data.videos, tiktok-comments returns
     data.comments, tiktok-followers, tiktok-following, and tiktok-user-search return
     data.users, yelp-search returns up to 20 ranked organic businesses in
-    data.businesses, plus total and pagination, from a Yelp search URL; emails,
-    phone-numbers, and links return data.emails, data.phone_numbers, and data.links
-    from the rendered page; the store, Maps, G2, Instagram, and TikTok parsers read
-    their data directly, independently of DOM filters and other outputs.
+    data.businesses, plus total and pagination, from a Yelp search URL; yelp-reviews
+    returns up to 20 latest public reviews in data.reviews, with business details
+    and next_offset, from a Yelp business URL; emails, phone-numbers, and links
+    return data.emails, data.phone_numbers, and data.links from the rendered page;
+    the store, Maps, G2, Instagram, and TikTok parsers read their data directly,
+    independently of DOM filters and other outputs.
     """
 
     product: DataOkResultProduct
@@ -829,7 +832,7 @@ class DataErrorResultMetadata(BaseModel):
 
 class DataErrorResultParsed(BaseModel):
     """
-    Custom rules return fields after content filters; unmatched fields are null ([] for lists). The google-maps parser returns business records in data.results, google-maps-reviews returns data.reviews, and the google-play parser returns one app record as data, google-play-reviews returns data.reviews, app-store and app-store-reviews do the same for the App Store, g2-reviews returns data.reviews for a G2 product, instagram-profile and instagram-post return one record as data, instagram-feed returns data.posts, tiktok-profile and tiktok-video return one record as data, tiktok-feed returns data.videos, tiktok-comments returns data.comments, tiktok-followers, tiktok-following, and tiktok-user-search return data.users, yelp-search returns up to 20 ranked organic businesses in data.businesses, plus total and pagination, from a Yelp search URL; emails, phone-numbers, and links return data.emails, data.phone_numbers, and data.links from the rendered page; the store, Maps, G2, Instagram, and TikTok parsers read their data directly, independently of DOM filters and other outputs.
+    Custom rules return fields after content filters; unmatched fields are null ([] for lists). The google-maps parser returns business records in data.results, google-maps-reviews returns data.reviews, and the google-play parser returns one app record as data, google-play-reviews returns data.reviews, app-store and app-store-reviews do the same for the App Store, g2-reviews returns data.reviews for a G2 product, instagram-profile and instagram-post return one record as data, instagram-feed returns data.posts, tiktok-profile and tiktok-video return one record as data, tiktok-feed returns data.videos, tiktok-comments returns data.comments, tiktok-followers, tiktok-following, and tiktok-user-search return data.users, yelp-search returns up to 20 ranked organic businesses in data.businesses, plus total and pagination, from a Yelp search URL; yelp-reviews returns up to 20 latest public reviews in data.reviews, with business details and next_offset, from a Yelp business URL; emails, phone-numbers, and links return data.emails, data.phone_numbers, and data.links from the rendered page; the store, Maps, G2, Instagram, and TikTok parsers read their data directly, independently of DOM filters and other outputs.
     """
 
     data: Optional[Dict[str, object]] = None
@@ -859,6 +862,7 @@ class DataErrorResultParsed(BaseModel):
         "tiktok-following",
         "tiktok-user-search",
         "yelp-search",
+        "yelp-reviews",
         "emails",
         "phone-numbers",
         "links",
@@ -1060,10 +1064,12 @@ class DataErrorResult(BaseModel):
     one record as data, tiktok-feed returns data.videos, tiktok-comments returns
     data.comments, tiktok-followers, tiktok-following, and tiktok-user-search return
     data.users, yelp-search returns up to 20 ranked organic businesses in
-    data.businesses, plus total and pagination, from a Yelp search URL; emails,
-    phone-numbers, and links return data.emails, data.phone_numbers, and data.links
-    from the rendered page; the store, Maps, G2, Instagram, and TikTok parsers read
-    their data directly, independently of DOM filters and other outputs.
+    data.businesses, plus total and pagination, from a Yelp search URL; yelp-reviews
+    returns up to 20 latest public reviews in data.reviews, with business details
+    and next_offset, from a Yelp business URL; emails, phone-numbers, and links
+    return data.emails, data.phone_numbers, and data.links from the rendered page;
+    the store, Maps, G2, Instagram, and TikTok parsers read their data directly,
+    independently of DOM filters and other outputs.
     """
 
     product: DataErrorResultProduct

src/context/dev/types/batch_submit_params.py

@@ -895,6 +895,7 @@ class InputScrapeDataBatchUnifiedScrapeDataParseParamsPrebuiltParser(TypedDict,
             "tiktok-following",
             "tiktok-user-search",
             "yelp-search",
+            "yelp-reviews",
             "emails",
             "phone-numbers",
             "links",
@@ -962,16 +963,22 @@ class InputScrapeDataBatchUnifiedScrapeDataParseParamsPrebuiltParser(TypedDict,
     links (vm.tiktok.com, vt.tiktok.com, tiktok.com/t/…). yelp-search accepts Yelp
     search URLs with find_desc and find_loc, reads start as the result offset, and
     returns up to 20 ranked organic businesses in data.businesses with total and
-    pagination. emails accepts any page and returns its email addresses as
-    data.emails ([{ email }]), read from mailto links, Cloudflare-protected
-    addresses, and visible text. phone-numbers accepts any page and returns its
-    valid phone numbers in E.164 as data.phone_numbers ([{ phone }]), read from tel
-    links and visible text; numbers without a country code are read for the country
-    of the site TLD, defaulting to US. links accepts any page and returns its unique
-    http(s) links as data.links ([{ url, text }]), with absolute URLs; text falls
-    back to aria-label or image alt and is null when a link has none. The emails
-    parser reads the rendered page and respects content filters; the other prebuilt
-    parsers read their data independently of DOM filters.
+    pagination. yelp-reviews accepts a Yelp business URL (yelp.com/biz/{alias}),
+    uses optional start for the offset, and always sorts newest first; unsupported
+    sort and review filters are rejected, while tracking params are ignored. It
+    returns up to 20 latest public reviews in data.reviews with business details and
+    pagination.next_offset. Each review has rating, full text, updated_at, and
+    author; owner replies and photos are not included. emails accepts any page and
+    returns its email addresses as data.emails ([{ email }]), read from mailto
+    links, Cloudflare-protected addresses, and visible text. phone-numbers accepts
+    any page and returns its valid phone numbers in E.164 as data.phone_numbers ([{
+    phone }]), read from tel links and visible text; numbers without a country code
+    are read for the country of the site TLD, defaulting to US. links accepts any
+    page and returns its unique http(s) links as data.links ([{ url, text }]), with
+    absolute URLs; text falls back to aria-label or image alt and is null when a
+    link has none. The emails parser reads the rendered page and respects content
+    filters; the other prebuilt parsers read their data independently of DOM
+    filters.
     """
 
     type: Required[Literal["prebuilt"]]
@@ -2538,6 +2545,7 @@ class InputCrawlDataBatchUnifiedCrawlDataParseParamsPrebuiltParser(TypedDict, to
             "tiktok-following",
             "tiktok-user-search",
             "yelp-search",
+            "yelp-reviews",
             "emails",
             "phone-numbers",
             "links",
@@ -2605,16 +2613,22 @@ class InputCrawlDataBatchUnifiedCrawlDataParseParamsPrebuiltParser(TypedDict, to
     links (vm.tiktok.com, vt.tiktok.com, tiktok.com/t/…). yelp-search accepts Yelp
     search URLs with find_desc and find_loc, reads start as the result offset, and
     returns up to 20 ranked organic businesses in data.businesses with total and
-    pagination. emails accepts any page and returns its email addresses as
-    data.emails ([{ email }]), read from mailto links, Cloudflare-protected
-    addresses, and visible text. phone-numbers accepts any page and returns its
-    valid phone numbers in E.164 as data.phone_numbers ([{ phone }]), read from tel
-    links and visible text; numbers without a country code are read for the country
-    of the site TLD, defaulting to US. links accepts any page and returns its unique
-    http(s) links as data.links ([{ url, text }]), with absolute URLs; text falls
-    back to aria-label or image alt and is null when a link has none. The emails
-    parser reads the rendered page and respects content filters; the other prebuilt
-    parsers read their data independently of DOM filters.
+    pagination. yelp-reviews accepts a Yelp business URL (yelp.com/biz/{alias}),
+    uses optional start for the offset, and always sorts newest first; unsupported
+    sort and review filters are rejected, while tracking params are ignored. It
+    returns up to 20 latest public reviews in data.reviews with business details and
+    pagination.next_offset. Each review has rating, full text, updated_at, and
+    author; owner replies and photos are not included. emails accepts any page and
+    returns its email addresses as data.emails ([{ email }]), read from mailto
+    links, Cloudflare-protected addresses, and visible text. phone-numbers accepts
+    any page and returns its valid phone numbers in E.164 as data.phone_numbers ([{
+    phone }]), read from tel links and visible text; numbers without a country code
+    are read for the country of the site TLD, defaulting to US. links accepts any
+    page and returns its unique http(s) links as data.links ([{ url, text }]), with
+    absolute URLs; text falls back to aria-label or image alt and is null when a
+    link has none. The emails parser reads the rendered page and respects content
+    filters; the other prebuilt parsers read their data independently of DOM
+    filters.
     """
 
     type: Required[Literal["prebuilt"]]

src/context/dev/types/web_scrape_params.py

@@ -232,6 +232,7 @@ class ParseParamsPrebuiltParser(TypedDict, total=False):
             "tiktok-following",
             "tiktok-user-search",
             "yelp-search",
+            "yelp-reviews",
             "emails",
             "phone-numbers",
             "links",
@@ -299,16 +300,22 @@ class ParseParamsPrebuiltParser(TypedDict, total=False):
     links (vm.tiktok.com, vt.tiktok.com, tiktok.com/t/…). yelp-search accepts Yelp
     search URLs with find_desc and find_loc, reads start as the result offset, and
     returns up to 20 ranked organic businesses in data.businesses with total and
-    pagination. emails accepts any page and returns its email addresses as
-    data.emails ([{ email }]), read from mailto links, Cloudflare-protected
-    addresses, and visible text. phone-numbers accepts any page and returns its
-    valid phone numbers in E.164 as data.phone_numbers ([{ phone }]), read from tel
-    links and visible text; numbers without a country code are read for the country
-    of the site TLD, defaulting to US. links accepts any page and returns its unique
-    http(s) links as data.links ([{ url, text }]), with absolute URLs; text falls
-    back to aria-label or image alt and is null when a link has none. The emails
-    parser reads the rendered page and respects content filters; the other prebuilt
-    parsers read their data independently of DOM filters.
+    pagination. yelp-reviews accepts a Yelp business URL (yelp.com/biz/{alias}),
+    uses optional start for the offset, and always sorts newest first; unsupported
+    sort and review filters are rejected, while tracking params are ignored. It
+    returns up to 20 latest public reviews in data.reviews with business details and
+    pagination.next_offset. Each review has rating, full text, updated_at, and
+    author; owner replies and photos are not included. emails accepts any page and
+    returns its email addresses as data.emails ([{ email }]), read from mailto
+    links, Cloudflare-protected addresses, and visible text. phone-numbers accepts
+    any page and returns its valid phone numbers in E.164 as data.phone_numbers ([{
+    phone }]), read from tel links and visible text; numbers without a country code
+    are read for the country of the site TLD, defaulting to US. links accepts any
+    page and returns its unique http(s) links as data.links ([{ url, text }]), with
+    absolute URLs; text falls back to aria-label or image alt and is null when a
+    link has none. The emails parser reads the rendered page and respects content
+    filters; the other prebuilt parsers read their data independently of DOM
+    filters.
     """
 
     type: Required[Literal["prebuilt"]]

src/context/dev/types/web_scrape_response.py

@@ -267,7 +267,7 @@ class Metadata(BaseModel):
 
 class Parsed(BaseModel):
     """
-    Custom rules return fields after content filters; unmatched fields are null ([] for lists). The google-maps parser returns business records in data.results, google-maps-reviews returns data.reviews, and the google-play parser returns one app record as data, google-play-reviews returns data.reviews, app-store and app-store-reviews do the same for the App Store, g2-reviews returns data.reviews for a G2 product, instagram-profile and instagram-post return one record as data, instagram-feed returns data.posts, tiktok-profile and tiktok-video return one record as data, tiktok-feed returns data.videos, tiktok-comments returns data.comments, tiktok-followers, tiktok-following, and tiktok-user-search return data.users, yelp-search returns up to 20 ranked organic businesses in data.businesses, plus total and pagination, from a Yelp search URL; emails, phone-numbers, and links return data.emails, data.phone_numbers, and data.links from the rendered page; the store, Maps, G2, Instagram, and TikTok parsers read their data directly, independently of DOM filters and other outputs.
+    Custom rules return fields after content filters; unmatched fields are null ([] for lists). The google-maps parser returns business records in data.results, google-maps-reviews returns data.reviews, and the google-play parser returns one app record as data, google-play-reviews returns data.reviews, app-store and app-store-reviews do the same for the App Store, g2-reviews returns data.reviews for a G2 product, instagram-profile and instagram-post return one record as data, instagram-feed returns data.posts, tiktok-profile and tiktok-video return one record as data, tiktok-feed returns data.videos, tiktok-comments returns data.comments, tiktok-followers, tiktok-following, and tiktok-user-search return data.users, yelp-search returns up to 20 ranked organic businesses in data.businesses, plus total and pagination, from a Yelp search URL; yelp-reviews returns up to 20 latest public reviews in data.reviews, with business details and next_offset, from a Yelp business URL; emails, phone-numbers, and links return data.emails, data.phone_numbers, and data.links from the rendered page; the store, Maps, G2, Instagram, and TikTok parsers read their data directly, independently of DOM filters and other outputs.
     """
 
     data: Optional[Dict[str, object]] = None
@@ -297,6 +297,7 @@ class Parsed(BaseModel):
         "tiktok-following",
         "tiktok-user-search",
         "yelp-search",
+        "yelp-reviews",
         "emails",
         "phone-numbers",
         "links",
@@ -493,10 +494,12 @@ class WebScrapeResponse(BaseModel):
     one record as data, tiktok-feed returns data.videos, tiktok-comments returns
     data.comments, tiktok-followers, tiktok-following, and tiktok-user-search return
     data.users, yelp-search returns up to 20 ranked organic businesses in
-    data.businesses, plus total and pagination, from a Yelp search URL; emails,
-    phone-numbers, and links return data.emails, data.phone_numbers, and data.links
-    from the rendered page; the store, Maps, G2, Instagram, and TikTok parsers read
-    their data directly, independently of DOM filters and other outputs.
+    data.businesses, plus total and pagination, from a Yelp search URL; yelp-reviews
+    returns up to 20 latest public reviews in data.reviews, with business details
+    and next_offset, from a Yelp business URL; emails, phone-numbers, and links
+    return data.emails, data.phone_numbers, and data.links from the rendered page;
+    the store, Maps, G2, Instagram, and TikTok parsers read their data directly,
+    independently of DOM filters and other outputs.
     """
 
     product: Product