xberg 1.3.0 → 1.3.3

data/lib/xberg/version.rb

@@ -1,10 +1,10 @@
 # This file is auto-generated by alef — DO NOT EDIT.
-# alef:hash:05ecab5f3a98752913f33a49a0d8e2003efdf37c21a36463c366e780e9f9f882
+# alef:hash:fe82079c10797b9f3415c2e9fe70094470cb68e2c7bb3d301d89afea37551beb
 # To regenerate: alef generate
 # To verify freshness: alef verify
 # frozen_string_literal: true
 
 module Xberg
   ## The version string for this package.
-  VERSION = "1.3.0"
+  VERSION = "1.3.3"
 end

data/sig/types.rbs

@@ -1,5 +1,5 @@
 # This file is auto-generated by alef — DO NOT EDIT.
-# alef:hash:04f362999b06cc48b7e517b58690ed05ddc13831ae9ef1b20fac65475ac1f94e
+# alef:hash:4d6e82c5b0834d91c2575ca21fe65ed7dc5bb424ab80623cb550f60e8cd268ad
 # To regenerate: alef generate
 # To verify freshness: alef verify
 
@@ -125,8 +125,9 @@ def initialize: (?mode: enum_BrowserMode, ?backend: enum_BrowserBackend, ?endpoi
         attr_reader llm: LlmConfig
         attr_reader prompt: String?
         attr_reader min_image_area: Integer
+        attr_reader alt_text: enum_CaptionAltTextMode
 
-    def initialize: (llm: LlmConfig, ?prompt: String, min_image_area: Integer) -> void
+    def initialize: (llm: LlmConfig, ?prompt: String, min_image_area: Integer, alt_text: enum_CaptionAltTextMode) -> void
     end
 
     class CellChange
@@ -354,6 +355,7 @@ def initialize: (?output_format: String, ?preprocessing_preset: String, ?remove_
     attr_reader strip_tags: Array[String]
     attr_reader preserve_tags: Array[String]
         attr_reader skip_images: bool
+        attr_reader inline_data_media: enum_InlineDataMedia
         attr_reader url_escape_style: enum_UrlEscapeStyle
         attr_reader link_style: enum_LinkStyle
         attr_reader output_format: enum_HtmlToMarkdownRsOutputFormat
@@ -367,7 +369,7 @@ def initialize: (?output_format: String, ?preprocessing_preset: String, ?remove_
         attr_reader tier_strategy: enum_TierStrategy
         attr_reader base_url: String?
 
-def initialize: (?heading_style: enum_HeadingStyle, ?list_indent_type: enum_ListIndentType, ?list_indent_width: Integer, ?bullets: String, ?strong_em_symbol: String, ?escape_asterisks: bool, ?escape_underscores: bool, ?escape_misc: bool, ?escape_ascii: bool, ?code_language: String, ?autolinks: bool, ?default_title: bool, ?br_in_tables: bool, ?compact_tables: bool, ?highlight_style: enum_HighlightStyle, ?extract_metadata: bool, ?whitespace_mode: enum_WhitespaceMode, ?strip_newlines: bool, ?wrap: bool, ?wrap_width: Integer, ?convert_as_inline: bool, ?sub_symbol: String, ?sup_symbol: String, ?newline_style: enum_NewlineStyle, ?code_block_style: enum_CodeBlockStyle, ?keep_inline_images_in: Array[String], ?preprocessing: PreprocessingOptions, ?encoding: String, ?debug: bool, ?strip_tags: Array[String], ?preserve_tags: Array[String], ?skip_images: bool, ?url_escape_style: enum_UrlEscapeStyle, ?link_style: enum_LinkStyle, ?output_format: enum_HtmlToMarkdownRsOutputFormat, ?include_document_structure: bool, ?extract_images: bool, ?max_image_size: Integer, ?capture_svg: bool, ?infer_dimensions: bool, ?max_depth: Integer, ?exclude_selectors: Array[String], ?tier_strategy: enum_TierStrategy, ?base_url: String) -> void
+def initialize: (?heading_style: enum_HeadingStyle, ?list_indent_type: enum_ListIndentType, ?list_indent_width: Integer, ?bullets: String, ?strong_em_symbol: String, ?escape_asterisks: bool, ?escape_underscores: bool, ?escape_misc: bool, ?escape_ascii: bool, ?code_language: String, ?autolinks: bool, ?default_title: bool, ?br_in_tables: bool, ?compact_tables: bool, ?highlight_style: enum_HighlightStyle, ?extract_metadata: bool, ?whitespace_mode: enum_WhitespaceMode, ?strip_newlines: bool, ?wrap: bool, ?wrap_width: Integer, ?convert_as_inline: bool, ?sub_symbol: String, ?sup_symbol: String, ?newline_style: enum_NewlineStyle, ?code_block_style: enum_CodeBlockStyle, ?keep_inline_images_in: Array[String], ?preprocessing: PreprocessingOptions, ?encoding: String, ?debug: bool, ?strip_tags: Array[String], ?preserve_tags: Array[String], ?skip_images: bool, ?inline_data_media: enum_InlineDataMedia, ?url_escape_style: enum_UrlEscapeStyle, ?link_style: enum_LinkStyle, ?output_format: enum_HtmlToMarkdownRsOutputFormat, ?include_document_structure: bool, ?extract_images: bool, ?max_image_size: Integer, ?capture_svg: bool, ?infer_dimensions: bool, ?max_depth: Integer, ?exclude_selectors: Array[String], ?tier_strategy: enum_TierStrategy, ?base_url: String) -> void
     end
 
     class CoreProperties
@@ -845,6 +847,16 @@ def initialize: (name: String, markdown: String, row_count: Integer, col_count:
 def initialize: (sheets: Array[ExcelSheet], metadata: Hash[String, String], ?revisions: Array[DocumentRevision]) -> void
     end
 
+    class ExternalRedactionFinding
+        attr_reader label: String
+        attr_reader text: String?
+        attr_reader start: Integer?
+        attr_reader end: Integer?
+        attr_reader score: Float?
+
+    def initialize: (?label: String, ?text: String, ?start: Integer, ?end: Integer, ?score: Float) -> void
+    end
+
     class ExtractInput
         attr_reader kind: enum_ExtractInputKind
         attr_reader bytes: String?
@@ -1692,6 +1704,7 @@ def initialize: (name: String, supported_languages: Array[String]) -> void
         attr_reader tesseract_config: TesseractConfig?
         attr_reader output_format: OutputFormat?
         attr_reader paddle_ocr_config: String?
+        attr_reader paddle_ocr_settings: PaddleOcrConfig?
         attr_reader backend_options: String?
         attr_reader element_config: OcrElementConfig?
         attr_reader quality_thresholds: OcrQualityThresholds?
@@ -1706,7 +1719,7 @@ def initialize: (name: String, supported_languages: Array[String]) -> void
         attr_reader tessdata_path: String?
         attr_reader numeric_repair: bool
 
-def initialize: (?enabled: bool, ?backend: String, ?language: Array[String], ?tesseract_config: TesseractConfig, ?output_format: OutputFormat, ?paddle_ocr_config: String, ?backend_options: String, ?element_config: OcrElementConfig, ?quality_thresholds: OcrQualityThresholds, ?pipeline: OcrPipelineConfig, ?auto_rotate: bool, ?vlm_fallback: VlmFallbackPolicy, ?vlm_config: LlmConfig, ?vlm_prompt: String, ?acceleration: AccelerationConfig, ?security_limits: SecurityLimits, ?tessdata_bytes: Hash[String, String], ?tessdata_path: String, ?numeric_repair: bool) -> void
+def initialize: (?enabled: bool, ?backend: String, ?language: Array[String], ?tesseract_config: TesseractConfig, ?output_format: OutputFormat, ?paddle_ocr_config: String, ?paddle_ocr_settings: PaddleOcrConfig, ?backend_options: String, ?element_config: OcrElementConfig, ?quality_thresholds: OcrQualityThresholds, ?pipeline: OcrPipelineConfig, ?auto_rotate: bool, ?vlm_fallback: VlmFallbackPolicy, ?vlm_config: LlmConfig, ?vlm_prompt: String, ?acceleration: AccelerationConfig, ?security_limits: SecurityLimits, ?tessdata_bytes: Hash[String, String], ?tessdata_path: String, ?numeric_repair: bool) -> void
     end
 
     class OcrElement
@@ -1765,10 +1778,11 @@ def initialize: (stages: Array[OcrPipelineStage], quality_thresholds: OcrQuality
     attr_reader language: Array[String]?
         attr_reader tesseract_config: TesseractConfig?
         attr_reader paddle_ocr_config: String?
+        attr_reader paddle_ocr_settings: PaddleOcrConfig?
         attr_reader vlm_config: LlmConfig?
         attr_reader backend_options: String?
 
-def initialize: (backend: String, priority: Integer, ?language: Array[String], ?tesseract_config: TesseractConfig, ?paddle_ocr_config: String, ?vlm_config: LlmConfig, ?backend_options: String) -> void
+def initialize: (backend: String, priority: Integer, ?language: Array[String], ?tesseract_config: TesseractConfig, ?paddle_ocr_config: String, ?paddle_ocr_settings: PaddleOcrConfig, ?vlm_config: LlmConfig, ?backend_options: String) -> void
     end
 
     class OcrPoint
@@ -2264,8 +2278,12 @@ def initialize: (detection_bbox: BBox, cells: Array[Array[String]], markdown: St
         attr_reader preserve_offsets: bool
     attr_reader custom_terms: Array[RedactionTerm]
     attr_reader custom_patterns: Array[RedactionPattern]
+    attr_reader findings: Array[ExternalRedactionFinding]
+        attr_reader min_score: Float?
+        attr_reader findings_path: String?
+        attr_reader findings_offset_encoding: enum_RedactionOffsetEncoding
 
-def initialize: (?categories: Array[PiiCategory], ?strategy: enum_RedactionStrategy, ?ner: NerConfig, ?preserve_offsets: bool, ?custom_terms: Array[RedactionTerm], ?custom_patterns: Array[RedactionPattern]) -> void
+def initialize: (?categories: Array[PiiCategory], ?strategy: enum_RedactionStrategy, ?ner: NerConfig, ?preserve_offsets: bool, ?custom_terms: Array[RedactionTerm], ?custom_patterns: Array[RedactionPattern], ?findings: Array[ExternalRedactionFinding], ?min_score: Float, ?findings_path: String, ?findings_offset_encoding: enum_RedactionOffsetEncoding) -> void
     def validate: () -> void
     end
 
@@ -2771,6 +2789,8 @@ def initialize: (?min: Integer, ?max: Integer, years: Array[Integer]) -> void
 
     type enum_CandleTrocrVariant = :"base-printed" | :"large-printed" | :"base-handwritten" | :"large-handwritten"
 
+    type enum_CaptionAltTextMode = :preserve | :combine | :replace
+
     class ChunkSizing
     end
 
@@ -2851,6 +2871,8 @@ def initialize: (?min: Integer, ?max: Integer, years: Array[Integer]) -> void
 
     type enum_ImageType = :"data-uri" | :"inline-svg" | :external | :relative
 
+    type enum_InlineDataMedia = :Keep | :AltTextOnly | :DropElement
+
     type enum_InlineType = :text | :strong | :emphasis | :highlight | :subscript | :superscript | :insert | :delete | :code | :link | :image | :span | :math | :raw_inline | :footnote_ref | :symbol
 
     type enum_JupyterCellRendering = :source | :outputs | :both
@@ -2926,6 +2948,8 @@ def initialize: (?min: Integer, ?max: Integer, years: Array[Integer]) -> void
 
     type enum_ProcessingStage = :Early | :Middle | :Late
 
+    type enum_RedactionOffsetEncoding = :utf8_bytes | :unicode_code_points | :utf16_code_units
+
     type enum_RedactionStrategy = :mask | :hash | :token_replace | :drop
 
     type enum_ReductionLevel = :Off | :Light | :Moderate | :Aggressive | :Maximum
@@ -3002,6 +3026,8 @@ def self.extract: (ExtractInput input, ExtractionConfig config) -> ExtractionRes
 
 def self.extract_batch: (Array[ExtractInput] inputs, ExtractionConfig config) -> ExtractionResult
 
+def self.extract_with_external_redaction: (ExtractInput input, ExtractionConfig config, String findings_json, ?String offset_encoding, ?Integer max_findings) -> ExtractionResult
+
 def self.find_unmarked_claims: (String markdown) -> Array[String]
 
 def self.list_document_extractors: () -> Array[String]
@@ -3038,6 +3064,8 @@ def self.ocr_backend_supports_language_for: (String backend, String language, Oc
 
 def self.pdf_page_count: (String pdf_bytes, ?String password) -> Integer
 
+def self.redact_external: (ExtractedDocument document, RedactionConfig config, String findings_json, ?String offset_encoding, ?Integer max_findings) -> ExtractedDocument
+
 def self.register_document_extractor: (_DocumentExtractor extractor) -> void
 
 def self.register_renderer: (_Renderer renderer) -> void

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: xberg
 version: !ruby/object:Gem::Version
-  version: 1.3.0
+  version: 1.3.3
 platform: ruby
 authors:
 - Na'aman Hirschfeld <naaman@xberg.io>
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-09-29 00:00:00.000000000 Z
+date: 2026-10-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rb_sys