xberg 1.3.7 → 1.3.9

data/ext/xberg_rb/native/Cargo.toml

@@ -1,12 +1,13 @@
 # This file is auto-generated by alef — DO NOT EDIT.
-# alef:hash:a35074a20ca717ee54f07b153b49a9c727f639435558b33d857255e60e309542
+# alef:hash:cefb6a172a39f89f9f6c20af423e60c6ab3016057576cb6b9c1f72589894431c
 # To regenerate: alef generate
 # To verify freshness: alef verify
 
 [package]
 name = "xberg-rb"
-version = "1.3.7"
+version = "1.3.9"
 edition = "2024"
+authors = ["Na'aman Hirschfeld <naaman@xberg.io>", "Tobias Perlstein <tobias@xberg.io>"]
 license = "MIT"
 description = "High-performance document intelligence library"
 readme = false
@@ -82,20 +83,18 @@ tokio = { version = "1", features = ["rt-multi-thread"] }
 tracing = "0.1"
 
 [target.'cfg(all(target_os = "macos", target_arch = "x86_64"))'.dependencies]
-xberg = { version = "1.3.7", default-features = false, features = ["macos-intel-target", "simd-utf8"] }
+xberg = { version = "1.3.9", default-features = false, features = ["macos-intel-target", "simd-utf8"] }
 
 [target.'cfg(not(any(target_os = "windows", all(target_os = "macos", target_arch = "x86_64"))))'.dependencies]
-xberg = { version = "1.3.7", features = ["full", "pdf", "pdf-pdfium", "ocr", "paddle-ocr", "paddle-ocr-types", "layout-detection", "layout-types", "embeddings", "embedding-presets", "reranker", "reranker-presets", "sparse-embeddings", "sparse-embedding-presets", "late-interaction", "late-interaction-presets", "chunking", "keywords-yake", "keywords-rake", "language-detection", "html", "tree-sitter", "office", "email", "archives", "stopwords", "auto-rotate", "auto-rotate-types", "tokio-runtime", "api", "api-types", "mcp", "liter-llm", "quality", "svg", "transcription", "transcription-types", "classification", "captioning", "formula-recognition", "ner-onnx", "ner-llm", "diff", "markdown-footnotes", "redaction", "enrichment", "heuristics", "presets", "structured", "url-ingestion"] }
+xberg = { version = "1.3.9", features = ["full", "pdf", "pdf-pdfium", "ocr", "paddle-ocr", "paddle-ocr-types", "layout-detection", "layout-types", "embeddings", "embedding-presets", "reranker", "reranker-presets", "sparse-embeddings", "sparse-embedding-presets", "late-interaction", "late-interaction-presets", "chunking", "keywords-yake", "keywords-rake", "language-detection", "html", "tree-sitter", "office", "email", "archives", "stopwords", "auto-rotate", "auto-rotate-types", "tokio-runtime", "api", "api-types", "mcp", "liter-llm", "quality", "svg", "transcription", "transcription-types", "classification", "captioning", "formula-recognition", "ner-onnx", "ner-llm", "diff", "markdown-footnotes", "redaction", "enrichment", "heuristics", "presets", "structured", "url-ingestion"] }
 
 [target.'cfg(target_os = "windows")'.dependencies]
-xberg = { version = "1.3.7", default-features = false, features = ["windows-gnu-target", "simd-utf8"] }
+xberg = { version = "1.3.9", default-features = false, features = ["windows-gnu-target", "simd-utf8"] }
 
 # This crate deliberately does not use `[lints]` / `workspace = true`: its C-ABI /
-# PyO3 / napi / ext-php-rs / NIF boundary requires `unsafe` code, and the workspace's
-# `[workspace.lints.rust]` sets `unsafe_code = "deny"` -- an all-or-nothing table that
-# would turn every such boundary into a compile error. The `[lints.clippy]` block below
-# instead carries the subset of the workspace's deny-by-default lint policy this crate
-# can actually satisfy. ~keep
+# PyO3 / napi / ext-php-rs / NIF boundary requires `unsafe` code. Local lint tables
+# avoid inheriting workspace policies that could reject these boundaries, and select
+# the lints this binding can actually satisfy. ~keep
 [lints.clippy]
 dbg_macro = "deny"
 print_stderr = "deny"

data/ext/xberg_rb/src/lib.rs

@@ -1,5 +1,5 @@
 // This file is auto-generated by alef. DO NOT EDIT.
-// alef:hash:d77ecc12b6f0752c5cec96b6f0b205562ea146737a08606f1b824d4cca7eff52
+// alef:hash:8a7f952b1ec204f9582e176aedbd1b79f4f674addb02087b2d868b77db7e5b11
 // Re-generate with: alef generate
 #![allow(dead_code, unused_imports, unused_variables)]
 #![allow(
@@ -12250,6 +12250,7 @@ pub struct ExtractionConfig {
     transcription: Option<TranscriptionConfig>,
     use_layout_for_markdown: bool,
     include_document_structure: bool,
+    strict_parse: bool,
     acceleration: Option<AccelerationConfig>,
     cache_namespace: Option<String>,
     cache_ttl_secs: Option<u64>,
@@ -12661,6 +12662,15 @@ impl ExtractionConfig {
                 })?,
                 None => false,
             },
+            strict_parse: match kwargs.get(ruby.to_symbol("strict_parse")) {
+                Some(v) => bool::try_convert(v).map_err(|e| {
+                    magnus::Error::new(
+                        unsafe { magnus::Ruby::get_unchecked() }.exception_type_error(),
+                        format!("invalid value for `strict_parse`: {}", e),
+                    )
+                })?,
+                None => false,
+            },
             acceleration: match kwargs.get(ruby.to_symbol("acceleration")).filter(|v| !v.is_nil()) {
                 Some(v) => Some(AccelerationConfig::try_convert(v).map_err(|e| {
                     magnus::Error::new(
@@ -12995,6 +13005,10 @@ impl ExtractionConfig {
         self.include_document_structure
     }
 
+    fn strict_parse(&self) -> bool {
+        self.strict_parse
+    }
+
     fn acceleration(&self) -> Option<AccelerationConfig> {
         self.acceleration.clone()
     }
@@ -13135,6 +13149,8 @@ impl ExtractionConfig {
 
             include_document_structure: self.include_document_structure,
 
+            strict_parse: self.strict_parse,
+
             acceleration: self.acceleration.clone().map(Into::into),
 
             cache_namespace: self.cache_namespace.clone(),
@@ -13245,6 +13261,8 @@ impl ExtractionConfig {
 
             include_document_structure: self.include_document_structure,
 
+            strict_parse: self.strict_parse,
+
             acceleration: self.acceleration.clone().map(Into::into),
 
             cache_namespace: self.cache_namespace.clone(),
@@ -13349,6 +13367,8 @@ impl ExtractionConfig {
 
             include_document_structure: self.include_document_structure,
 
+            strict_parse: self.strict_parse,
+
             acceleration: self.acceleration.clone().map(Into::into),
 
             cache_namespace: self.cache_namespace.clone(),
@@ -13453,6 +13473,8 @@ impl ExtractionConfig {
 
             include_document_structure: self.include_document_structure,
 
+            strict_parse: self.strict_parse,
+
             acceleration: self.acceleration.clone().map(Into::into),
 
             cache_namespace: self.cache_namespace.clone(),
@@ -13557,6 +13579,8 @@ impl ExtractionConfig {
 
             include_document_structure: self.include_document_structure,
 
+            strict_parse: self.strict_parse,
+
             acceleration: self.acceleration.clone().map(Into::into),
 
             cache_namespace: self.cache_namespace.clone(),
@@ -14271,6 +14295,7 @@ pub struct FileExtractionConfig {
     result_format: Option<ResultFormat>,
     output_format: Option<OutputFormat>,
     include_document_structure: Option<bool>,
+    strict_parse: Option<bool>,
     #[cfg(feature = "layout-types")]
     layout: Option<LayoutDetectionConfig>,
     #[cfg(feature = "transcription-types")]
@@ -14533,6 +14558,15 @@ impl FileExtractionConfig {
                 })?),
                 None => None,
             },
+            strict_parse: match kwargs.get(ruby.to_symbol("strict_parse")).filter(|v| !v.is_nil()) {
+                Some(v) => Some(bool::try_convert(v).map_err(|e| {
+                    magnus::Error::new(
+                        unsafe { magnus::Ruby::get_unchecked() }.exception_type_error(),
+                        format!("invalid value for `strict_parse`: {}", e),
+                    )
+                })?),
+                None => None,
+            },
             #[cfg(feature = "layout-types")]
             layout: match kwargs.get(ruby.to_symbol("layout")).filter(|v| !v.is_nil()) {
                 Some(v) => Some(LayoutDetectionConfig::try_convert(v).map_err(|e| {
@@ -14762,6 +14796,10 @@ impl FileExtractionConfig {
         self.include_document_structure
     }
 
+    fn strict_parse(&self) -> Option<bool> {
+        self.strict_parse
+    }
+
     #[cfg(feature = "layout-types")]
     fn layout(&self) -> Option<LayoutDetectionConfig> {
         self.layout.clone()
@@ -17378,7 +17416,7 @@ impl ImageExtractionConfig {
                         format!("invalid value for `append_ocr_text`: {}", e),
                     )
                 })?,
-                None => false,
+                None => true,
             },
             output_format: match kwargs.get(ruby.to_symbol("output_format")) {
                 Some(v) => ImageOutputFormat::try_convert(v).map_err(|e| {
@@ -25647,6 +25685,7 @@ pub struct PageConfig {
     extract_pages: bool,
     insert_page_markers: bool,
     marker_format: String,
+    keep_native_content: bool,
 }
 
 unsafe impl IntoValueFromNative for PageConfig {}
@@ -25730,6 +25769,15 @@ impl PageConfig {
     "
                 .to_string(),
             },
+            keep_native_content: match kwargs.get(ruby.to_symbol("keep_native_content")) {
+                Some(v) => bool::try_convert(v).map_err(|e| {
+                    magnus::Error::new(
+                        unsafe { magnus::Ruby::get_unchecked() }.exception_type_error(),
+                        format!("invalid value for `keep_native_content`: {}", e),
+                    )
+                })?,
+                None => false,
+            },
         })
     }
 
@@ -25744,6 +25792,10 @@ impl PageConfig {
     fn marker_format(&self) -> String {
         self.marker_format.clone()
     }
+
+    fn keep_native_content(&self) -> bool {
+        self.keep_native_content
+    }
 }
 
 #[derive(Clone, Debug, serde::Serialize)]
@@ -25761,6 +25813,7 @@ pub struct PageContent {
     section_name: Option<String>,
     sheet_name: Option<String>,
     ocr_confidence: Option<PageOcrConfidence>,
+    native_content: Option<String>,
 }
 
 unsafe impl IntoValueFromNative for PageContent {}
@@ -25814,6 +25867,7 @@ impl Default for PageContent {
             section_name: None,
             sheet_name: None,
             ocr_confidence: None,
+            native_content: None,
         }
     }
 }
@@ -25936,6 +25990,15 @@ impl PageContent {
                 })?),
                 None => None,
             },
+            native_content: match kwargs.get(ruby.to_symbol("native_content")).filter(|v| !v.is_nil()) {
+                Some(v) => Some(String::try_convert(v).map_err(|e| {
+                    magnus::Error::new(
+                        unsafe { magnus::Ruby::get_unchecked() }.exception_type_error(),
+                        format!("invalid value for `native_content`: {}", e),
+                    )
+                })?),
+                None => None,
+            },
         })
     }
 
@@ -25987,6 +26050,10 @@ impl PageContent {
         self.ocr_confidence.clone()
     }
 
+    fn native_content(&self) -> Option<String> {
+        self.native_content.clone()
+    }
+
     #[allow(clippy::should_implement_trait)]
     fn to_s(&self) -> String {
         self.content.clone()
@@ -50683,6 +50750,7 @@ impl From<ExtractionConfig> for xberg::ExtractionConfig {
             transcription: val.transcription.map(Into::into),
             use_layout_for_markdown: val.use_layout_for_markdown,
             include_document_structure: val.include_document_structure,
+            strict_parse: val.strict_parse,
             acceleration: val.acceleration.map(Into::into),
             cache_namespace: val.cache_namespace,
             cache_ttl_secs: val.cache_ttl_secs,
@@ -50754,6 +50822,7 @@ impl From<xberg::ExtractionConfig> for ExtractionConfig {
             transcription: val.transcription.map(Into::into),
             use_layout_for_markdown: val.use_layout_for_markdown,
             include_document_structure: val.include_document_structure,
+            strict_parse: val.strict_parse,
             acceleration: val.acceleration.map(Into::into),
             cache_namespace: val.cache_namespace.map(|v| v.to_string()),
             cache_ttl_secs: val.cache_ttl_secs,
@@ -50931,6 +51000,7 @@ impl From<FileExtractionConfig> for xberg::FileExtractionConfig {
             result_format: val.result_format.map(Into::into),
             output_format: val.output_format.map(Into::into),
             include_document_structure: val.include_document_structure,
+            strict_parse: val.strict_parse,
             #[cfg(feature = "layout-types")]
             layout: val.layout.map(Into::into),
             #[cfg(feature = "transcription-types")]
@@ -50982,6 +51052,7 @@ impl From<xberg::FileExtractionConfig> for FileExtractionConfig {
             result_format: val.result_format.map(Into::into),
             output_format: val.output_format.map(Into::into),
             include_document_structure: val.include_document_structure,
+            strict_parse: val.strict_parse,
             #[cfg(feature = "layout-types")]
             layout: val.layout.map(Into::into),
             #[cfg(feature = "transcription-types")]
@@ -53044,6 +53115,7 @@ impl From<PageConfig> for xberg::PageConfig {
             extract_pages: val.extract_pages,
             insert_page_markers: val.insert_page_markers,
             marker_format: val.marker_format,
+            keep_native_content: val.keep_native_content,
             ..Default::default()
         }
     }
@@ -53055,6 +53127,7 @@ impl From<xberg::PageConfig> for PageConfig {
             extract_pages: val.extract_pages,
             insert_page_markers: val.insert_page_markers,
             marker_format: val.marker_format.to_string(),
+            keep_native_content: val.keep_native_content,
         }
     }
 }
@@ -53075,6 +53148,7 @@ impl From<PageContent> for xberg::PageContent {
             section_name: val.section_name,
             sheet_name: val.sheet_name,
             ocr_confidence: val.ocr_confidence.map(Into::into),
+            native_content: val.native_content,
         }
     }
 }
@@ -53095,6 +53169,7 @@ impl From<xberg::PageContent> for PageContent {
             section_name: val.section_name.map(|v| v.to_string()),
             sheet_name: val.sheet_name.map(|v| v.to_string()),
             ocr_confidence: val.ocr_confidence.map(Into::into),
+            native_content: val.native_content.map(|v| v.to_string()),
         }
     }
 }
@@ -59466,6 +59541,8 @@ fn ruby_init(ruby: &Ruby) -> Result<(), Error> {
         method!(ExtractionConfig::include_document_structure, 0),
     )?;
 
+    class.define_method("strict_parse", method!(ExtractionConfig::strict_parse, 0))?;
+
     class.define_method("acceleration", method!(ExtractionConfig::acceleration, 0))?;
 
     class.define_method("cache_namespace", method!(ExtractionConfig::cache_namespace, 0))?;
@@ -59680,6 +59757,8 @@ fn ruby_init(ruby: &Ruby) -> Result<(), Error> {
         method!(FileExtractionConfig::include_document_structure, 0),
     )?;
 
+    class.define_method("strict_parse", method!(FileExtractionConfig::strict_parse, 0))?;
+
     #[cfg(feature = "layout-types")]
     class.define_method("layout", method!(FileExtractionConfig::layout, 0))?;
 
@@ -61209,6 +61288,8 @@ fn ruby_init(ruby: &Ruby) -> Result<(), Error> {
 
     class.define_method("marker_format", method!(PageConfig::marker_format, 0))?;
 
+    class.define_method("keep_native_content", method!(PageConfig::keep_native_content, 0))?;
+
     let class = module.define_class("PageContent", ruby.class_object())?;
 
     class.define_singleton_method("new", function!(PageContent::new, -1))?;
@@ -61237,6 +61318,8 @@ fn ruby_init(ruby: &Ruby) -> Result<(), Error> {
 
     class.define_method("ocr_confidence", method!(PageContent::ocr_confidence, 0))?;
 
+    class.define_method("native_content", method!(PageContent::native_content, 0))?;
+
     class.define_method("to_s", method!(PageContent::to_s, 0))?;
 
     let class = module.define_class("PageDimensions", ruby.class_object())?;

data/lib/xberg/version.rb

@@ -1,10 +1,10 @@
 # This file is auto-generated by alef — DO NOT EDIT.
-# alef:hash:0687e037476da4ba7e18fe1957eb56d2d154c61c20f178f4c8ee9e1cc7d090c2
+# alef:hash:54266c8ed71cfc0b41f227d99a4c062e533226b81f15e7dc5499221930624881
 # To regenerate: alef generate
 # To verify freshness: alef verify
 # frozen_string_literal: true
 
 module Xberg
   ## The version string for this package.
-  VERSION = "1.3.7"
+  VERSION = "1.3.9"
 end

data/sig/types.rbs

@@ -1,5 +1,5 @@
 # This file is auto-generated by alef — DO NOT EDIT.
-# alef:hash:ebe306889747c0ee64b78fefcf465898aca7871336bda3ba3caa4406a424946d
+# alef:hash:18c7c7ea2cdd0194d2647e8ae2cc52ce3389914a3c1a9b585c79939242ccbef1
 # To regenerate: alef generate
 # To verify freshness: alef verify
 
@@ -990,6 +990,7 @@ def initialize: (?data: String, ?format: String, ?image_index: Integer, ?page_nu
         attr_reader transcription: TranscriptionConfig?
         attr_reader use_layout_for_markdown: bool
         attr_reader include_document_structure: bool
+        attr_reader strict_parse: bool
         attr_reader acceleration: AccelerationConfig?
         attr_reader cache_namespace: String?
         attr_reader cache_ttl_secs: Integer?
@@ -1010,7 +1011,7 @@ def initialize: (?data: String, ?format: String, ?image_index: Integer, ?page_nu
         attr_reader captioning: CaptioningConfig?
         attr_reader qr_codes: bool?
 
-def initialize: (?mime_detection_policy: enum_MimeDetectionPolicy, ?use_cache: bool, ?enable_quality_processing: bool, ?ocr: OcrConfig, ?force_ocr: bool, ?ocr_strategy: OcrStrategy, ?force_ocr_pages: Array[Integer], ?disable_ocr: bool, ?ocr_near_empty_fallback: bool, ?ocr_scanned_page_quality_gate: bool, ?ocr_embedded_images: bool, ?chunking: ChunkingConfig, ?content_filter: ContentFilterConfig, ?images: ImageExtractionConfig, ?pdf_options: PdfConfig, ?token_reduction: TokenReductionOptions, ?language_detection: LanguageDetectionConfig, ?pages: PageConfig, ?keywords: KeywordConfig, ?postprocessor: PostProcessorConfig, ?html_options: ConversionOptions, ?html_output: HtmlOutputConfig, ?extraction_timeout_secs: Integer, ?max_concurrent_extractions: Integer, ?result_format: enum_ResultFormat, ?security_limits: SecurityLimits, ?max_embedded_file_bytes: Integer, ?output_format: OutputFormat, ?escape_markdown: bool, ?table_anchors: bool, ?jupyter_cell_rendering: enum_JupyterCellRendering, ?apply_notebook_cell_tags: bool, ?layout: LayoutDetectionConfig, ?transcription: TranscriptionConfig, ?use_layout_for_markdown: bool, ?include_document_structure: bool, ?acceleration: AccelerationConfig, ?cache_namespace: String, ?cache_ttl_secs: Integer, ?email: EmailConfig, ?csv: CsvConfig, ?geojson: GeoJsonExtractionConfig, ?concurrency: ConcurrencyConfig, ?url: UrlExtractionConfig, ?max_archive_depth: Integer, ?tree_sitter: TreeSitterConfig, ?structured_extraction: StructuredExtractionConfig, ?ner: NerConfig, ?redaction: RedactionConfig, ?summarization: SummarizationConfig, ?translation: TranslationConfig, ?page_classification: PageClassificationConfig, ?chunk_classification: ChunkClassificationConfig, ?captioning: CaptioningConfig, ?qr_codes: bool) -> void
+def initialize: (?mime_detection_policy: enum_MimeDetectionPolicy, ?use_cache: bool, ?enable_quality_processing: bool, ?ocr: OcrConfig, ?force_ocr: bool, ?ocr_strategy: OcrStrategy, ?force_ocr_pages: Array[Integer], ?disable_ocr: bool, ?ocr_near_empty_fallback: bool, ?ocr_scanned_page_quality_gate: bool, ?ocr_embedded_images: bool, ?chunking: ChunkingConfig, ?content_filter: ContentFilterConfig, ?images: ImageExtractionConfig, ?pdf_options: PdfConfig, ?token_reduction: TokenReductionOptions, ?language_detection: LanguageDetectionConfig, ?pages: PageConfig, ?keywords: KeywordConfig, ?postprocessor: PostProcessorConfig, ?html_options: ConversionOptions, ?html_output: HtmlOutputConfig, ?extraction_timeout_secs: Integer, ?max_concurrent_extractions: Integer, ?result_format: enum_ResultFormat, ?security_limits: SecurityLimits, ?max_embedded_file_bytes: Integer, ?output_format: OutputFormat, ?escape_markdown: bool, ?table_anchors: bool, ?jupyter_cell_rendering: enum_JupyterCellRendering, ?apply_notebook_cell_tags: bool, ?layout: LayoutDetectionConfig, ?transcription: TranscriptionConfig, ?use_layout_for_markdown: bool, ?include_document_structure: bool, ?strict_parse: bool, ?acceleration: AccelerationConfig, ?cache_namespace: String, ?cache_ttl_secs: Integer, ?email: EmailConfig, ?csv: CsvConfig, ?geojson: GeoJsonExtractionConfig, ?concurrency: ConcurrencyConfig, ?url: UrlExtractionConfig, ?max_archive_depth: Integer, ?tree_sitter: TreeSitterConfig, ?structured_extraction: StructuredExtractionConfig, ?ner: NerConfig, ?redaction: RedactionConfig, ?summarization: SummarizationConfig, ?translation: TranslationConfig, ?page_classification: PageClassificationConfig, ?chunk_classification: ChunkClassificationConfig, ?captioning: CaptioningConfig, ?qr_codes: bool) -> void
     def validate: () -> void
     def needs_image_data: () -> bool
     def runs_ocr_on_embedded_images: () -> bool
@@ -1091,6 +1092,7 @@ def initialize: (?genres: Array[String], ?sequences: Array[String], ?annotation:
         attr_reader result_format: enum_ResultFormat?
         attr_reader output_format: OutputFormat?
         attr_reader include_document_structure: bool?
+        attr_reader strict_parse: bool?
         attr_reader layout: LayoutDetectionConfig?
         attr_reader transcription: TranscriptionConfig?
         attr_reader timeout_secs: Integer?
@@ -1106,7 +1108,7 @@ def initialize: (?genres: Array[String], ?sequences: Array[String], ?annotation:
         attr_reader captioning: CaptioningConfig?
         attr_reader qr_codes: bool?
 
-def initialize: (?mime_detection_policy: enum_MimeDetectionPolicy, ?enable_quality_processing: bool, ?ocr: OcrConfig, ?force_ocr: bool, ?ocr_strategy: OcrStrategy, ?force_ocr_pages: Array[Integer], ?disable_ocr: bool, ?chunking: ChunkingConfig, ?content_filter: ContentFilterConfig, ?images: ImageExtractionConfig, ?pdf_options: PdfConfig, ?token_reduction: TokenReductionOptions, ?language_detection: LanguageDetectionConfig, ?pages: PageConfig, ?keywords: KeywordConfig, ?postprocessor: PostProcessorConfig, ?html_options: ConversionOptions, ?html_output: HtmlOutputConfig, ?result_format: enum_ResultFormat, ?output_format: OutputFormat, ?include_document_structure: bool, ?layout: LayoutDetectionConfig, ?transcription: TranscriptionConfig, ?timeout_secs: Integer, ?tree_sitter: TreeSitterConfig, ?structured_extraction: StructuredExtractionConfig, ?url: UrlExtractionConfig, ?ner: NerConfig, ?redaction: RedactionConfig, ?summarization: SummarizationConfig, ?translation: TranslationConfig, ?page_classification: PageClassificationConfig, ?chunk_classification: ChunkClassificationConfig, ?captioning: CaptioningConfig, ?qr_codes: bool) -> void
+def initialize: (?mime_detection_policy: enum_MimeDetectionPolicy, ?enable_quality_processing: bool, ?ocr: OcrConfig, ?force_ocr: bool, ?ocr_strategy: OcrStrategy, ?force_ocr_pages: Array[Integer], ?disable_ocr: bool, ?chunking: ChunkingConfig, ?content_filter: ContentFilterConfig, ?images: ImageExtractionConfig, ?pdf_options: PdfConfig, ?token_reduction: TokenReductionOptions, ?language_detection: LanguageDetectionConfig, ?pages: PageConfig, ?keywords: KeywordConfig, ?postprocessor: PostProcessorConfig, ?html_options: ConversionOptions, ?html_output: HtmlOutputConfig, ?result_format: enum_ResultFormat, ?output_format: OutputFormat, ?include_document_structure: bool, ?strict_parse: bool, ?layout: LayoutDetectionConfig, ?transcription: TranscriptionConfig, ?timeout_secs: Integer, ?tree_sitter: TreeSitterConfig, ?structured_extraction: StructuredExtractionConfig, ?url: UrlExtractionConfig, ?ner: NerConfig, ?redaction: RedactionConfig, ?summarization: SummarizationConfig, ?translation: TranslationConfig, ?page_classification: PageClassificationConfig, ?chunk_classification: ChunkClassificationConfig, ?captioning: CaptioningConfig, ?qr_codes: bool) -> void
     end
 
     class Footnote
@@ -1938,8 +1940,9 @@ def initialize: (?prompt_template: String, labels: Array[String], multi_label: b
         attr_reader extract_pages: bool
         attr_reader insert_page_markers: bool
         attr_reader marker_format: String
+        attr_reader keep_native_content: bool
 
-    def initialize: (?extract_pages: bool, ?insert_page_markers: bool, ?marker_format: String) -> void
+    def initialize: (?extract_pages: bool, ?insert_page_markers: bool, ?marker_format: String, ?keep_native_content: bool) -> void
     end
 
     class PageContent
@@ -1955,8 +1958,9 @@ def initialize: (?prompt_template: String, labels: Array[String], multi_label: b
         attr_reader section_name: String?
         attr_reader sheet_name: String?
         attr_reader ocr_confidence: PageOcrConfidence?
+        attr_reader native_content: String?
 
-def initialize: (page_number: Integer, content: String, tables: Array[Table], image_indices: Array[Integer], ?image_preprocessing: ImagePreprocessingMetadata, ?hierarchy: PageHierarchy, ?is_blank: bool, ?layout_regions: Array[LayoutRegion], ?speaker_notes: String, ?section_name: String, ?sheet_name: String, ?ocr_confidence: PageOcrConfidence) -> void
+def initialize: (page_number: Integer, content: String, tables: Array[Table], image_indices: Array[Integer], ?image_preprocessing: ImagePreprocessingMetadata, ?hierarchy: PageHierarchy, ?is_blank: bool, ?layout_regions: Array[LayoutRegion], ?speaker_notes: String, ?section_name: String, ?sheet_name: String, ?ocr_confidence: PageOcrConfidence, ?native_content: String) -> void
     end
 
     class PageDimensions

metadata

@@ -1,14 +1,15 @@
 --- !ruby/object:Gem::Specification
 name: xberg
 version: !ruby/object:Gem::Version
-  version: 1.3.7
+  version: 1.3.9
 platform: ruby
 authors:
-- Na'aman Hirschfeld <naaman@xberg.io>
+- Na'aman Hirschfeld
+- Tobias Perlstein
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2026-10-09 00:00:00.000000000 Z
+date: 2026-10-13 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rb_sys
@@ -36,16 +37,18 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.5'
+        version: '0.6'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.5'
+        version: '0.6'
 description: High-performance document intelligence library
 email:
+- naaman@xberg.io
+- tobias@xberg.io
 executables: []
 extensions:
 - ext/xberg_rb/native/extconf.rb