context.dev 2.2.0 → 2.3.0

data/rbi/context_dev/models/parse_handle_params.rbi

@@ -0,0 +1,163 @@
+# typed: strong
+
+module ContextDev
+  module Models
+    class ParseHandleParams < ContextDev::Internal::Type::BaseModel
+      extend ContextDev::Internal::Type::RequestParameters::Converter
+      include ContextDev::Internal::Type::RequestParameters
+
+      OrHash =
+        T.type_alias do
+          T.any(ContextDev::ParseHandleParams, ContextDev::Internal::AnyHash)
+        end
+
+      sig { returns(ContextDev::Internal::FileInput) }
+      attr_accessor :body
+
+      # Optional HTTP(S) source document URL used to resolve relative links and image
+      # references. Relative references remain relative when omitted.
+      sig { returns(T.nilable(String)) }
+      attr_reader :base_url
+
+      sig { params(base_url: String).void }
+      attr_writer :base_url
+
+      # Optional file extension hint, such as pdf, docx, xlsx, pptx, html, json, csv,
+      # md, py, rtf, jpg, png, or txt.
+      sig { returns(T.nilable(String)) }
+      attr_reader :extension
+
+      sig { params(extension: String).void }
+      attr_writer :extension
+
+      # Optional filename hint used to infer the extension when extension is omitted.
+      sig { returns(T.nilable(String)) }
+      attr_reader :filename
+
+      sig { params(filename: String).void }
+      attr_writer :filename
+
+      # Include image references in Markdown output
+      sig { returns(T.nilable(T::Boolean)) }
+      attr_reader :include_images
+
+      sig { params(include_images: T::Boolean).void }
+      attr_writer :include_images
+
+      # Preserve hyperlinks in Markdown output
+      sig { returns(T.nilable(T::Boolean)) }
+      attr_reader :include_links
+
+      sig { params(include_links: T::Boolean).void }
+      attr_writer :include_links
+
+      # When true for PDF inputs, detect and OCR images embedded in the selected pages,
+      # inserting recognized text at each image's position in page reading order while
+      # preserving the PDF text layer. pdfStart/pdfEnd limit the inclusive page range.
+      # This is separate from automatic scanned-PDF OCR fallback.
+      sig { returns(T.nilable(T::Boolean)) }
+      attr_reader :ocr
+
+      sig { params(ocr: T::Boolean).void }
+      attr_writer :ocr
+
+      # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
+      # Must be greater than or equal to pdfStart when both are provided.
+      sig { returns(T.nilable(Integer)) }
+      attr_reader :pdf_end
+
+      sig { params(pdf_end: Integer).void }
+      attr_writer :pdf_end
+
+      # First 1-based PDF page to parse. When omitted, parsing starts at the first page.
+      sig { returns(T.nilable(Integer)) }
+      attr_reader :pdf_start
+
+      sig { params(pdf_start: Integer).void }
+      attr_writer :pdf_start
+
+      # Shorten base64-encoded image data in the Markdown output
+      sig { returns(T.nilable(T::Boolean)) }
+      attr_reader :shorten_base64_images
+
+      sig { params(shorten_base64_images: T::Boolean).void }
+      attr_writer :shorten_base64_images
+
+      # Extract only the main content from HTML-like inputs
+      sig { returns(T.nilable(T::Boolean)) }
+      attr_reader :use_main_content_only
+
+      sig { params(use_main_content_only: T::Boolean).void }
+      attr_writer :use_main_content_only
+
+      sig do
+        params(
+          body: ContextDev::Internal::FileInput,
+          base_url: String,
+          extension: String,
+          filename: String,
+          include_images: T::Boolean,
+          include_links: T::Boolean,
+          ocr: T::Boolean,
+          pdf_end: Integer,
+          pdf_start: Integer,
+          shorten_base64_images: T::Boolean,
+          use_main_content_only: T::Boolean,
+          request_options: ContextDev::RequestOptions::OrHash
+        ).returns(T.attached_class)
+      end
+      def self.new(
+        body:,
+        # Optional HTTP(S) source document URL used to resolve relative links and image
+        # references. Relative references remain relative when omitted.
+        base_url: nil,
+        # Optional file extension hint, such as pdf, docx, xlsx, pptx, html, json, csv,
+        # md, py, rtf, jpg, png, or txt.
+        extension: nil,
+        # Optional filename hint used to infer the extension when extension is omitted.
+        filename: nil,
+        # Include image references in Markdown output
+        include_images: nil,
+        # Preserve hyperlinks in Markdown output
+        include_links: nil,
+        # When true for PDF inputs, detect and OCR images embedded in the selected pages,
+        # inserting recognized text at each image's position in page reading order while
+        # preserving the PDF text layer. pdfStart/pdfEnd limit the inclusive page range.
+        # This is separate from automatic scanned-PDF OCR fallback.
+        ocr: nil,
+        # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
+        # Must be greater than or equal to pdfStart when both are provided.
+        pdf_end: nil,
+        # First 1-based PDF page to parse. When omitted, parsing starts at the first page.
+        pdf_start: nil,
+        # Shorten base64-encoded image data in the Markdown output
+        shorten_base64_images: nil,
+        # Extract only the main content from HTML-like inputs
+        use_main_content_only: nil,
+        request_options: {}
+      )
+      end
+
+      sig do
+        override.returns(
+          {
+            body: ContextDev::Internal::FileInput,
+            base_url: String,
+            extension: String,
+            filename: String,
+            include_images: T::Boolean,
+            include_links: T::Boolean,
+            ocr: T::Boolean,
+            pdf_end: Integer,
+            pdf_start: Integer,
+            shorten_base64_images: T::Boolean,
+            use_main_content_only: T::Boolean,
+            request_options: ContextDev::RequestOptions
+          }
+        )
+      end
+      def to_hash
+      end
+    end
+  end
+end

data/rbi/context_dev/models/parse_handle_response.rbi

@@ -0,0 +1,377 @@
+# typed: strong
+
+module ContextDev
+  module Models
+    class ParseHandleResponse < ContextDev::Internal::Type::BaseModel
+      OrHash =
+        T.type_alias do
+          T.any(
+            ContextDev::Models::ParseHandleResponse,
+            ContextDev::Internal::AnyHash
+          )
+        end
+
+      # Input bytes converted to GitHub Flavored Markdown
+      sig { returns(String) }
+      attr_accessor :markdown
+
+      # Indicates success
+      sig do
+        returns(ContextDev::Models::ParseHandleResponse::Success::TaggedBoolean)
+      end
+      attr_accessor :success
+
+      # Detected content type used for parsing
+      sig do
+        returns(ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol)
+      end
+      attr_accessor :type
+
+      # Metadata about the API key used for the request. Included in every response
+      # whenever a valid API key is provided, even when the response status is not 200.
+      sig do
+        returns(T.nilable(ContextDev::Models::ParseHandleResponse::KeyMetadata))
+      end
+      attr_reader :key_metadata
+
+      sig do
+        params(
+          key_metadata:
+            ContextDev::Models::ParseHandleResponse::KeyMetadata::OrHash
+        ).void
+      end
+      attr_writer :key_metadata
+
+      sig do
+        params(
+          markdown: String,
+          success: ContextDev::Models::ParseHandleResponse::Success::OrBoolean,
+          type: ContextDev::Models::ParseHandleResponse::Type::OrSymbol,
+          key_metadata:
+            ContextDev::Models::ParseHandleResponse::KeyMetadata::OrHash
+        ).returns(T.attached_class)
+      end
+      def self.new(
+        # Input bytes converted to GitHub Flavored Markdown
+        markdown:,
+        # Indicates success
+        success:,
+        # Detected content type used for parsing
+        type:,
+        # Metadata about the API key used for the request. Included in every response
+        # whenever a valid API key is provided, even when the response status is not 200.
+        key_metadata: nil
+      )
+      end
+
+      sig do
+        override.returns(
+          {
+            markdown: String,
+            success:
+              ContextDev::Models::ParseHandleResponse::Success::TaggedBoolean,
+            type: ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol,
+            key_metadata: ContextDev::Models::ParseHandleResponse::KeyMetadata
+          }
+        )
+      end
+      def to_hash
+      end
+
+      # Indicates success
+      module Success
+        extend ContextDev::Internal::Type::Enum
+
+        TaggedBoolean =
+          T.type_alias do
+            T.all(T::Boolean, ContextDev::Models::ParseHandleResponse::Success)
+          end
+        OrBoolean = T.type_alias { T::Boolean }
+
+        TRUE =
+          T.let(
+            true,
+            ContextDev::Models::ParseHandleResponse::Success::TaggedBoolean
+          )
+
+        sig do
+          override.returns(
+            T::Array[
+              ContextDev::Models::ParseHandleResponse::Success::TaggedBoolean
+            ]
+          )
+        end
+        def self.values
+        end
+      end
+
+      # Detected content type used for parsing
+      module Type
+        extend ContextDev::Internal::Type::Enum
+
+        TaggedSymbol =
+          T.type_alias do
+            T.all(Symbol, ContextDev::Models::ParseHandleResponse::Type)
+          end
+        OrSymbol = T.type_alias { T.any(Symbol, String) }
+
+        HTML =
+          T.let(
+            :html,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        XML =
+          T.let(
+            :xml,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        JSON =
+          T.let(
+            :json,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        JSONL =
+          T.let(
+            :jsonl,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        TEXT =
+          T.let(
+            :text,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        CSV =
+          T.let(
+            :csv,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        TSV =
+          T.let(
+            :tsv,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        MARKDOWN =
+          T.let(
+            :markdown,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        YAML =
+          T.let(
+            :yaml,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PYTHON =
+          T.let(
+            :python,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        JAVA =
+          T.let(
+            :java,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        JAVASCRIPT =
+          T.let(
+            :javascript,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PHP =
+          T.let(
+            :php,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        SHELL =
+          T.let(
+            :shell,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        RUBY =
+          T.let(
+            :ruby,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        TYPESCRIPT =
+          T.let(
+            :typescript,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        RTF =
+          T.let(
+            :rtf,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        SRT =
+          T.let(
+            :srt,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        CSS =
+          T.let(
+            :css,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        SCSS =
+          T.let(
+            :scss,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        LESS =
+          T.let(
+            :less,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        STYLUS =
+          T.let(
+            :stylus,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        SASS =
+          T.let(
+            :sass,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        SVG =
+          T.let(
+            :svg,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PDF =
+          T.let(
+            :pdf,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        DOCX =
+          T.let(
+            :docx,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        DOC =
+          T.let(
+            :doc,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        XLSX =
+          T.let(
+            :xlsx,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        XLS =
+          T.let(
+            :xls,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PPTX =
+          T.let(
+            :pptx,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PPT =
+          T.let(
+            :ppt,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        JPG =
+          T.let(
+            :jpg,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PNG =
+          T.let(
+            :png,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        GIF =
+          T.let(
+            :gif,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        BMP =
+          T.let(
+            :bmp,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        TIFF =
+          T.let(
+            :tiff,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        WEBP =
+          T.let(
+            :webp,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PPM =
+          T.let(
+            :ppm,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PBM =
+          T.let(
+            :pbm,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PGM =
+          T.let(
+            :pgm,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+        PNM =
+          T.let(
+            :pnm,
+            ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+          )
+
+        sig do
+          override.returns(
+            T::Array[
+              ContextDev::Models::ParseHandleResponse::Type::TaggedSymbol
+            ]
+          )
+        end
+        def self.values
+        end
+      end
+
+      class KeyMetadata < ContextDev::Internal::Type::BaseModel
+        OrHash =
+          T.type_alias do
+            T.any(
+              ContextDev::Models::ParseHandleResponse::KeyMetadata,
+              ContextDev::Internal::AnyHash
+            )
+          end
+
+        # The number of credits consumed by this request.
+        sig { returns(Integer) }
+        attr_accessor :credits_consumed
+
+        # The number of credits remaining for your organization after this request.
+        sig { returns(Integer) }
+        attr_accessor :credits_remaining
+
+        # Metadata about the API key used for the request. Included in every response
+        # whenever a valid API key is provided, even when the response status is not 200.
+        sig do
+          params(credits_consumed: Integer, credits_remaining: Integer).returns(
+            T.attached_class
+          )
+        end
+        def self.new(
+          # The number of credits consumed by this request.
+          credits_consumed:,
+          # The number of credits remaining for your organization after this request.
+          credits_remaining:
+        )
+        end
+
+        sig do
+          override.returns(
+            { credits_consumed: Integer, credits_remaining: Integer }
+          )
+        end
+        def to_hash
+        end
+      end
+    end
+  end
+end

data/rbi/context_dev/models/web_web_crawl_md_params.rbi

@@ -101,8 +101,8 @@ module ContextDev
       sig { params(max_pages: Integer).void }
       attr_writer :max_pages
 
-      # PDF parsing controls. Use start/end to limit text extraction and OCR to an
-      # inclusive 1-based page range.
+      # PDF parsing controls. Use start/end to limit text extraction and embedded-image
+      # detection/OCR to an inclusive 1-based page range.
       sig { returns(T.nilable(ContextDev::WebWebCrawlMdParams::Pdf)) }
       attr_reader :pdf
 
@@ -226,8 +226,8 @@ module ContextDev
         max_depth: nil,
         # Maximum number of pages to crawl. Hard cap: 500.
         max_pages: nil,
-        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
-        # inclusive 1-based page range.
+        # PDF parsing controls. Use start/end to limit text extraction and embedded-image
+        # detection/OCR to an inclusive 1-based page range.
         pdf: nil,
         # When true, waits briefly for CSS and transition animations to settle before
         # extracting each crawled page. Defaults to false. This adds a bit of latency in
@@ -528,6 +528,15 @@ module ContextDev
         sig { params(end_: Integer).void }
         attr_writer :end_
 
+        # When true, detect and OCR images embedded in the selected PDF pages, inserting
+        # recognized text at each image's position in page reading order while preserving
+        # the PDF text layer. This is separate from automatic scanned-PDF OCR fallback.
+        sig { returns(T.nilable(T::Boolean)) }
+        attr_reader :ocr
+
+        sig { params(ocr: T::Boolean).void }
+        attr_writer :ocr
+
         # When true, PDF pages are fetched and parsed. When false, PDF pages are skipped
         # entirely (not included in results and not counted as failures).
         sig { returns(T.nilable(T::Boolean)) }
@@ -543,11 +552,12 @@ module ContextDev
         sig { params(start: Integer).void }
         attr_writer :start
 
-        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
-        # inclusive 1-based page range.
+        # PDF parsing controls. Use start/end to limit text extraction and embedded-image
+        # detection/OCR to an inclusive 1-based page range.
         sig do
           params(
             end_: Integer,
+            ocr: T::Boolean,
             should_parse: T::Boolean,
             start: Integer
           ).returns(T.attached_class)
@@ -556,6 +566,10 @@ module ContextDev
           # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
           # Must be greater than or equal to start when both are provided.
           end_: nil,
+          # When true, detect and OCR images embedded in the selected PDF pages, inserting
+          # recognized text at each image's position in page reading order while preserving
+          # the PDF text layer. This is separate from automatic scanned-PDF OCR fallback.
+          ocr: nil,
           # When true, PDF pages are fetched and parsed. When false, PDF pages are skipped
           # entirely (not included in results and not counted as failures).
           should_parse: nil,
@@ -566,7 +580,12 @@ module ContextDev
 
         sig do
           override.returns(
-            { end_: Integer, should_parse: T::Boolean, start: Integer }
+            {
+              end_: Integer,
+              ocr: T::Boolean,
+              should_parse: T::Boolean,
+              start: Integer
+            }
           )
         end
         def to_hash

data/rbi/context_dev/models/web_web_scrape_html_params.rbi

@@ -77,8 +77,8 @@ module ContextDev
       sig { params(max_age_ms: Integer).void }
       attr_writer :max_age_ms
 
-      # PDF parsing controls. Use start/end to limit text extraction and OCR to an
-      # inclusive 1-based page range.
+      # PDF parsing controls. Use start/end to limit text extraction and embedded-image
+      # detection/OCR to an inclusive 1-based page range.
       sig { returns(T.nilable(ContextDev::WebWebScrapeHTMLParams::Pdf)) }
       attr_reader :pdf
 
@@ -160,8 +160,8 @@ module ContextDev
         # younger than this many milliseconds. Defaults to 1 day (86400000 ms) when
         # omitted. Max is 30 days (2592000000 ms). Set to 0 to always scrape fresh.
         max_age_ms: nil,
-        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
-        # inclusive 1-based page range.
+        # PDF parsing controls. Use start/end to limit text extraction and embedded-image
+        # detection/OCR to an inclusive 1-based page range.
         pdf: nil,
         # When true, waits briefly for CSS and transition animations to settle before
         # extracting HTML. Defaults to false. This adds a bit of latency in exchange for
@@ -649,6 +649,15 @@ module ContextDev
         sig { params(end_: Integer).void }
         attr_writer :end_
 
+        # When true, detect and OCR images embedded in the selected PDF pages, inserting
+        # recognized text at each image's position in page reading order while preserving
+        # the PDF text layer. This is separate from automatic scanned-PDF OCR fallback.
+        sig { returns(T.nilable(T::Boolean)) }
+        attr_reader :ocr
+
+        sig { params(ocr: T::Boolean).void }
+        attr_writer :ocr
+
         # When true, PDF URLs are fetched and parsed. When false, PDF URLs are skipped and
         # a 400 WEBSITE_ACCESS_ERROR is returned.
         sig { returns(T.nilable(T::Boolean)) }
@@ -664,11 +673,12 @@ module ContextDev
         sig { params(start: Integer).void }
         attr_writer :start
 
-        # PDF parsing controls. Use start/end to limit text extraction and OCR to an
-        # inclusive 1-based page range.
+        # PDF parsing controls. Use start/end to limit text extraction and embedded-image
+        # detection/OCR to an inclusive 1-based page range.
         sig do
           params(
             end_: Integer,
+            ocr: T::Boolean,
             should_parse: T::Boolean,
             start: Integer
           ).returns(T.attached_class)
@@ -677,6 +687,10 @@ module ContextDev
           # Last 1-based PDF page to parse. When omitted, parsing ends at the last page.
           # Must be greater than or equal to start when both are provided.
           end_: nil,
+          # When true, detect and OCR images embedded in the selected PDF pages, inserting
+          # recognized text at each image's position in page reading order while preserving
+          # the PDF text layer. This is separate from automatic scanned-PDF OCR fallback.
+          ocr: nil,
           # When true, PDF URLs are fetched and parsed. When false, PDF URLs are skipped and
           # a 400 WEBSITE_ACCESS_ERROR is returned.
           should_parse: nil,
@@ -687,7 +701,12 @@ module ContextDev
 
         sig do
           override.returns(
-            { end_: Integer, should_parse: T::Boolean, start: Integer }
+            {
+              end_: Integer,
+              ocr: T::Boolean,
+              should_parse: T::Boolean,
+              start: Integer
+            }
           )
         end
         def to_hash