kreuzberg 4.0.0.pre.rc.6 → 4.0.0.pre.rc.7

data/vendor/kreuzberg/src/extractors/docbook.rs

@@ -23,6 +23,7 @@ use crate::types::{ExtractionResult, Metadata, Table};
 use async_trait::async_trait;
 use quick_xml::Reader;
 use quick_xml::events::Event;
+#[cfg(feature = "tokio-runtime")]
 use std::path::Path;
 
 /// Strip namespace prefix from XML tag names.
@@ -403,6 +404,7 @@ impl DocumentExtractor for DocbookExtractor {
             detected_languages: None,
             chunks: None,
             images: None,
+            pages: None,
         })
     }
 

data/vendor/kreuzberg/src/extractors/docx.rs

@@ -1,3 +1,5 @@
+#![cfg(all(feature = "tokio-runtime", feature = "office"))]
+
 //! DOCX extractor using docx-lite for high-performance text extraction.
 //!
 //! Supports: Microsoft Word (.docx)
@@ -6,7 +8,7 @@ use crate::Result;
 use crate::core::config::ExtractionConfig;
 use crate::extraction::{cells_to_markdown, office_metadata};
 use crate::plugins::{DocumentExtractor, Plugin};
-use crate::types::{ExtractionResult, Metadata, Table};
+use crate::types::{ExtractionResult, Metadata, PageBoundary, PageInfo, PageStructure, PageUnitType, Table};
 use async_trait::async_trait;
 use std::io::Cursor;
 
@@ -116,26 +118,30 @@ impl DocumentExtractor for DocxExtractor {
         mime_type: &str,
         _config: &ExtractionConfig,
     ) -> Result<ExtractionResult> {
-        let (text, tables) = if crate::core::batch_mode::is_batch_mode() {
+        let (text, tables, page_boundaries) = if crate::core::batch_mode::is_batch_mode() {
             let content_owned = content.to_vec();
             let span = tracing::Span::current();
-            tokio::task::spawn_blocking(move || -> crate::error::Result<(String, Vec<Table>)> {
-                let _guard = span.entered();
-                let cursor = Cursor::new(&content_owned);
-                let doc = docx_lite::parse_document(cursor)
-                    .map_err(|e| crate::error::KreuzbergError::parsing(format!("DOCX parsing failed: {}", e)))?;
+            tokio::task::spawn_blocking(
+                move || -> crate::error::Result<(String, Vec<Table>, Option<Vec<PageBoundary>>)> {
+                    let _guard = span.entered();
+                    let cursor = Cursor::new(&content_owned);
+                    let doc = docx_lite::parse_document(cursor)
+                        .map_err(|e| crate::error::KreuzbergError::parsing(format!("DOCX parsing failed: {}", e)))?;
 
-                let text = doc.extract_text();
+                    let text = doc.extract_text();
 
-                let tables: Vec<Table> = doc
-                    .tables
-                    .iter()
-                    .enumerate()
-                    .map(|(idx, table)| convert_docx_table_to_table(table, idx))
-                    .collect();
+                    let tables: Vec<Table> = doc
+                        .tables
+                        .iter()
+                        .enumerate()
+                        .map(|(idx, table)| convert_docx_table_to_table(table, idx))
+                        .collect();
 
-                Ok((text, tables))
-            })
+                    let page_boundaries = crate::extraction::docx::detect_page_breaks_from_docx(&content_owned)?;
+
+                    Ok((text, tables, page_boundaries))
+                },
+            )
             .await
             .map_err(|e| crate::error::KreuzbergError::parsing(format!("DOCX extraction task failed: {}", e)))??
         } else {
@@ -152,7 +158,9 @@ impl DocumentExtractor for DocxExtractor {
                 .map(|(idx, table)| convert_docx_table_to_table(table, idx))
                 .collect();
 
-            (text, tables)
+            let page_boundaries = crate::extraction::docx::detect_page_breaks_from_docx(content)?;
+
+            (text, tables, page_boundaries)
         };
 
         let mut archive = if crate::core::batch_mode::is_batch_mode() {
@@ -260,13 +268,38 @@ impl DocumentExtractor for DocxExtractor {
             }
         }
 
+        let page_structure = if let Some(boundaries) = page_boundaries {
+            let total_count = boundaries.len();
+            Some(PageStructure {
+                total_count,
+                unit_type: PageUnitType::Page,
+                boundaries: Some(boundaries),
+                pages: Some(
+                    (1..=total_count)
+                        .map(|page_num| PageInfo {
+                            number: page_num,
+                            title: None,
+                            dimensions: None,
+                            image_count: None,
+                            table_count: None,
+                            hidden: None,
+                        })
+                        .collect(),
+                ),
+            })
+        } else {
+            None
+        };
+
         Ok(ExtractionResult {
             content: text,
             mime_type: mime_type.to_string(),
             metadata: Metadata {
+                pages: page_structure,
                 additional: metadata_map,
                 ..Default::default()
             },
+            pages: None,
             tables,
             detected_languages: None,
             chunks: None,

data/vendor/kreuzberg/src/extractors/email.rs

@@ -2,9 +2,11 @@
 
 use crate::Result;
 use crate::core::config::ExtractionConfig;
+use crate::extractors::SyncExtractor;
 use crate::plugins::{DocumentExtractor, Plugin};
 use crate::types::{EmailMetadata, ExtractionResult, Metadata};
 use async_trait::async_trait;
+#[cfg(feature = "tokio-runtime")]
 use std::path::Path;
 
 /// Email message extractor.
@@ -42,21 +44,8 @@ impl Plugin for EmailExtractor {
     }
 }
 
-#[async_trait]
-impl DocumentExtractor for EmailExtractor {
-    #[cfg_attr(feature = "otel", tracing::instrument(
-        skip(self, content, _config),
-        fields(
-            extractor.name = self.name(),
-            content.size_bytes = content.len(),
-        )
-    ))]
-    async fn extract_bytes(
-        &self,
-        content: &[u8],
-        mime_type: &str,
-        _config: &ExtractionConfig,
-    ) -> Result<ExtractionResult> {
+impl SyncExtractor for EmailExtractor {
+    fn extract_sync(&self, content: &[u8], mime_type: &str, _config: &ExtractionConfig) -> Result<ExtractionResult> {
         let email_result = crate::extraction::email::extract_email_content(content, mime_type)?;
 
         let text = crate::extraction::email::build_email_text_output(&email_result);
@@ -96,8 +85,28 @@ impl DocumentExtractor for EmailExtractor {
             detected_languages: None,
             chunks: None,
             images: None,
+            pages: None,
         })
     }
+}
+
+#[async_trait]
+impl DocumentExtractor for EmailExtractor {
+    #[cfg_attr(feature = "otel", tracing::instrument(
+        skip(self, content, config),
+        fields(
+            extractor.name = self.name(),
+            content.size_bytes = content.len(),
+        )
+    ))]
+    async fn extract_bytes(
+        &self,
+        content: &[u8],
+        mime_type: &str,
+        config: &ExtractionConfig,
+    ) -> Result<ExtractionResult> {
+        self.extract_sync(content, mime_type, config)
+    }
 
     #[cfg(feature = "tokio-runtime")]
     #[cfg_attr(feature = "otel", tracing::instrument(
@@ -106,6 +115,7 @@ impl DocumentExtractor for EmailExtractor {
             extractor.name = self.name(),
         )
     ))]
+    #[cfg(feature = "tokio-runtime")]
     async fn extract_file(&self, path: &Path, mime_type: &str, config: &ExtractionConfig) -> Result<ExtractionResult> {
         let bytes = tokio::fs::read(path).await?;
         self.extract_bytes(&bytes, mime_type, config).await
@@ -118,6 +128,10 @@ impl DocumentExtractor for EmailExtractor {
     fn priority(&self) -> i32 {
         50
     }
+
+    fn as_sync_extractor(&self) -> Option<&dyn crate::extractors::SyncExtractor> {
+        Some(self)
+    }
 }
 
 #[cfg(test)]

data/vendor/kreuzberg/src/extractors/epub.rs

@@ -574,6 +574,7 @@ impl DocumentExtractor for EpubExtractor {
                 additional: metadata_map,
                 ..Default::default()
             },
+            pages: None,
             tables: vec![],
             detected_languages: None,
             chunks: None,

data/vendor/kreuzberg/src/extractors/excel.rs

@@ -150,6 +150,7 @@ impl DocumentExtractor for ExcelExtractor {
                 additional,
                 ..Default::default()
             },
+            pages: None,
             tables,
             detected_languages: None,
             chunks: None,
@@ -193,6 +194,7 @@ impl DocumentExtractor for ExcelExtractor {
                 additional,
                 ..Default::default()
             },
+            pages: None,
             tables,
             detected_languages: None,
             chunks: None,

data/vendor/kreuzberg/src/extractors/fictionbook.rs

@@ -440,6 +440,7 @@ impl DocumentExtractor for FictionBookExtractor {
             detected_languages: None,
             chunks: None,
             images: None,
+            pages: None,
         })
     }
 

data/vendor/kreuzberg/src/extractors/html.rs

@@ -2,9 +2,11 @@
 
 use crate::Result;
 use crate::core::config::ExtractionConfig;
+use crate::extractors::SyncExtractor;
 use crate::plugins::{DocumentExtractor, Plugin};
 use crate::types::{ExtractionResult, Metadata, Table};
 use async_trait::async_trait;
+#[cfg(feature = "tokio-runtime")]
 use std::path::Path;
 
 // NOTE: scraper dependency has been removed in favor of html-to-markdown-rs
@@ -193,21 +195,8 @@ impl Plugin for HtmlExtractor {
     }
 }
 
-#[async_trait]
-impl DocumentExtractor for HtmlExtractor {
-    #[cfg_attr(feature = "otel", tracing::instrument(
-        skip(self, content, config),
-        fields(
-            extractor.name = self.name(),
-            content.size_bytes = content.len(),
-        )
-    ))]
-    async fn extract_bytes(
-        &self,
-        content: &[u8],
-        mime_type: &str,
-        config: &ExtractionConfig,
-    ) -> Result<ExtractionResult> {
+impl SyncExtractor for HtmlExtractor {
+    fn extract_sync(&self, content: &[u8], mime_type: &str, config: &ExtractionConfig) -> Result<ExtractionResult> {
         let html = std::str::from_utf8(content)
             .map(|s| s.to_string())
             .unwrap_or_else(|_| String::from_utf8_lossy(content).to_string());
@@ -225,12 +214,32 @@ impl DocumentExtractor for HtmlExtractor {
                 format: html_metadata.map(|m| crate::types::FormatMetadata::Html(Box::new(m))),
                 ..Default::default()
             },
+            pages: None,
             tables,
             detected_languages: None,
             chunks: None,
             images: None,
         })
     }
+}
+
+#[async_trait]
+impl DocumentExtractor for HtmlExtractor {
+    #[cfg_attr(feature = "otel", tracing::instrument(
+        skip(self, content, config),
+        fields(
+            extractor.name = self.name(),
+            content.size_bytes = content.len(),
+        )
+    ))]
+    async fn extract_bytes(
+        &self,
+        content: &[u8],
+        mime_type: &str,
+        config: &ExtractionConfig,
+    ) -> Result<ExtractionResult> {
+        self.extract_sync(content, mime_type, config)
+    }
 
     #[cfg(feature = "tokio-runtime")]
     #[cfg_attr(feature = "otel", tracing::instrument(
@@ -239,6 +248,7 @@ impl DocumentExtractor for HtmlExtractor {
             extractor.name = self.name(),
         )
     ))]
+    #[cfg(feature = "tokio-runtime")]
     async fn extract_file(&self, path: &Path, mime_type: &str, config: &ExtractionConfig) -> Result<ExtractionResult> {
         let bytes = tokio::fs::read(path).await?;
         self.extract_bytes(&bytes, mime_type, config).await
@@ -251,6 +261,10 @@ impl DocumentExtractor for HtmlExtractor {
     fn priority(&self) -> i32 {
         50
     }
+
+    fn as_sync_extractor(&self) -> Option<&dyn crate::extractors::SyncExtractor> {
+        Some(self)
+    }
 }
 
 #[cfg(test)]

data/vendor/kreuzberg/src/extractors/image.rs

@@ -20,9 +20,14 @@ impl ImageExtractor {
         Self
     }
 
-    /// Extract text from image using OCR.
+    /// Extract text from image using OCR with optional page tracking for multi-frame TIFFs.
     #[cfg(feature = "ocr")]
-    async fn extract_with_ocr(&self, content: &[u8], config: &ExtractionConfig) -> Result<ExtractionResult> {
+    async fn extract_with_ocr(
+        &self,
+        content: &[u8],
+        mime_type: &str,
+        config: &ExtractionConfig,
+    ) -> Result<ExtractionResult> {
         use crate::plugins::registry::get_ocr_backend_registry;
 
         let ocr_config = config.ocr.as_ref().ok_or_else(|| crate::KreuzbergError::Parsing {
@@ -39,7 +44,21 @@ impl ImageExtractor {
             registry.get(&ocr_config.backend)?
         };
 
-        backend.process_image(content, ocr_config).await
+        let ocr_result = backend.process_image(content, ocr_config).await?;
+
+        let ocr_text = ocr_result.content.clone();
+        let ocr_extraction_result = crate::extraction::image::extract_text_from_image_with_ocr(
+            content,
+            mime_type,
+            ocr_text,
+            config.pages.as_ref(),
+        )?;
+
+        let mut result = ocr_result;
+        result.content = ocr_extraction_result.content;
+        result.pages = ocr_extraction_result.page_contents;
+
+        Ok(result)
     }
 }
 
@@ -102,7 +121,7 @@ impl DocumentExtractor for ImageExtractor {
         if config.ocr.is_some() {
             #[cfg(feature = "ocr")]
             {
-                let mut ocr_result = self.extract_with_ocr(content, config).await?;
+                let mut ocr_result = self.extract_with_ocr(content, mime_type, config).await?;
 
                 ocr_result.metadata.format = Some(crate::types::FormatMetadata::Image(image_metadata));
                 ocr_result.mime_type = mime_type.to_string();
@@ -123,6 +142,7 @@ impl DocumentExtractor for ImageExtractor {
                         format: Some(crate::types::FormatMetadata::Image(image_metadata)),
                         ..Default::default()
                     },
+                    pages: None,
                     tables: vec![],
                     detected_languages: None,
                     chunks: None,
@@ -141,6 +161,7 @@ impl DocumentExtractor for ImageExtractor {
                 format: Some(crate::types::FormatMetadata::Image(image_metadata)),
                 ..Default::default()
             },
+            pages: None,
             tables: vec![],
             detected_languages: None,
             chunks: None,

data/vendor/kreuzberg/src/extractors/jats.rs

@@ -21,6 +21,7 @@ use crate::types::{ExtractionResult, Metadata, Table};
 use async_trait::async_trait;
 use quick_xml::Reader;
 use quick_xml::events::Event;
+#[cfg(feature = "tokio-runtime")]
 use std::path::Path;
 
 /// JATS document extractor.
@@ -569,6 +570,7 @@ impl DocumentExtractor for JatsExtractor {
             detected_languages: None,
             chunks: None,
             images: None,
+            pages: None,
         })
     }
 
@@ -582,6 +584,7 @@ impl DocumentExtractor for JatsExtractor {
             )
         )
     )]
+    #[cfg(feature = "tokio-runtime")]
     async fn extract_file(&self, path: &Path, mime_type: &str, config: &ExtractionConfig) -> Result<ExtractionResult> {
         let bytes = tokio::fs::read(path).await?;
         self.extract_bytes(&bytes, mime_type, config).await

data/vendor/kreuzberg/src/extractors/jupyter.rs

@@ -336,6 +336,7 @@ impl DocumentExtractor for JupyterExtractor {
                 additional: metadata_additional,
                 ..Default::default()
             },
+            pages: None,
             tables: vec![],
             detected_languages: None,
             chunks: None,

data/vendor/kreuzberg/src/extractors/latex.rs

@@ -93,6 +93,7 @@ impl DocumentExtractor for LatexExtractor {
             detected_languages: None,
             chunks: None,
             images: None,
+            pages: None,
         })
     }
 

data/vendor/kreuzberg/src/extractors/markdown.rs

@@ -365,6 +365,7 @@ impl DocumentExtractor for MarkdownExtractor {
             detected_languages: None,
             chunks: None,
             images: None,
+            pages: None,
         })
     }
 

data/vendor/kreuzberg/src/extractors/mod.rs

@@ -4,14 +4,69 @@
 //! All extractors implement the `DocumentExtractor` plugin trait.
 
 use crate::Result;
+use crate::core::config::ExtractionConfig;
 use crate::plugins::registry::get_document_extractor_registry;
+use crate::types::ExtractionResult;
 use once_cell::sync::Lazy;
 use std::sync::Arc;
 
-pub mod security;
+/// Trait for extractors that can work synchronously (WASM-compatible).
+///
+/// This trait defines the synchronous extraction interface for WASM targets and other
+/// environments where async/tokio runtimes are not available or desirable.
+///
+/// # Implementation
+///
+/// Extractors that need to support WASM should implement this trait in addition to
+/// the async `DocumentExtractor` trait. This allows the same extractor to work in both
+/// environments by delegating to the sync implementation.
+///
+/// # MIME Type Validation
+///
+/// The `mime_type` parameter is guaranteed to be already validated.
+///
+/// # Example
+///
+/// ```rust,ignore
+/// impl SyncExtractor for PlainTextExtractor {
+///     fn extract_sync(&self, content: &[u8], config: &ExtractionConfig) -> Result<ExtractionResult> {
+///         let text = String::from_utf8_lossy(content).to_string();
+///         Ok(ExtractionResult {
+///             content: text,
+///             mime_type: "text/plain".to_string(),
+///             metadata: Metadata::default(),
+///             tables: vec![],
+///             detected_languages: None,
+///             chunks: None,
+///             images: None,
+///         })
+///     }
+/// }
+/// ```
+pub trait SyncExtractor {
+    /// Extract content from a byte array synchronously.
+    ///
+    /// This method performs extraction without requiring an async runtime.
+    /// It is called by `extract_bytes_sync()` when the `tokio-runtime` feature is disabled.
+    ///
+    /// # Arguments
+    ///
+    /// * `content` - Raw document bytes
+    /// * `mime_type` - MIME type of the document (already validated)
+    /// * `config` - Extraction configuration
+    ///
+    /// # Returns
+    ///
+    /// An `ExtractionResult` containing the extracted content and metadata.
+    fn extract_sync(&self, content: &[u8], mime_type: &str, config: &ExtractionConfig) -> Result<ExtractionResult>;
+}
+
 pub mod structured;
 pub mod text;
 
+#[cfg(feature = "archives")]
+pub mod security;
+
 #[cfg(feature = "ocr")]
 pub mod image;
 
@@ -30,7 +85,7 @@ pub mod html;
 #[cfg(feature = "office")]
 pub mod bibtex;
 
-#[cfg(feature = "office")]
+#[cfg(all(feature = "tokio-runtime", feature = "office"))]
 pub mod docx;
 
 #[cfg(feature = "office")]
@@ -54,7 +109,7 @@ pub mod jupyter;
 #[cfg(feature = "office")]
 pub mod orgmode;
 
-#[cfg(feature = "office")]
+#[cfg(all(feature = "tokio-runtime", feature = "office"))]
 pub mod odt;
 
 #[cfg(feature = "office")]
@@ -69,7 +124,7 @@ pub mod jats;
 #[cfg(feature = "pdf")]
 pub mod pdf;
 
-#[cfg(feature = "office")]
+#[cfg(all(feature = "tokio-runtime", feature = "office"))]
 pub mod pptx;
 
 #[cfg(feature = "office")]
@@ -102,7 +157,7 @@ pub use html::HtmlExtractor;
 #[cfg(feature = "office")]
 pub use bibtex::BibtexExtractor;
 
-#[cfg(feature = "office")]
+#[cfg(all(feature = "tokio-runtime", feature = "office"))]
 pub use docx::DocxExtractor;
 
 #[cfg(feature = "office")]
@@ -126,7 +181,7 @@ pub use jupyter::JupyterExtractor;
 #[cfg(feature = "office")]
 pub use orgmode::OrgModeExtractor;
 
-#[cfg(feature = "office")]
+#[cfg(all(feature = "tokio-runtime", feature = "office"))]
 pub use odt::OdtExtractor;
 
 #[cfg(feature = "xml")]
@@ -141,7 +196,7 @@ pub use typst::TypstExtractor;
 #[cfg(feature = "pdf")]
 pub use pdf::PdfExtractor;
 
-#[cfg(feature = "office")]
+#[cfg(all(feature = "tokio-runtime", feature = "office"))]
 pub use pptx::PptxExtractor;
 
 #[cfg(feature = "office")]
@@ -230,11 +285,8 @@ pub fn register_default_extractors() -> Result<()> {
     {
         registry.register(Arc::new(EnhancedMarkdownExtractor::new()))?;
         registry.register(Arc::new(BibtexExtractor::new()))?;
-        registry.register(Arc::new(DocxExtractor::new()))?;
         registry.register(Arc::new(EpubExtractor::new()))?;
         registry.register(Arc::new(FictionBookExtractor::new()))?;
-        registry.register(Arc::new(PptxExtractor::new()))?;
-        registry.register(Arc::new(OdtExtractor::new()))?;
         registry.register(Arc::new(RtfExtractor::new()))?;
         registry.register(Arc::new(RstExtractor::new()))?;
         registry.register(Arc::new(LatexExtractor::new()))?;
@@ -244,6 +296,13 @@ pub fn register_default_extractors() -> Result<()> {
         registry.register(Arc::new(TypstExtractor::new()))?;
     }
 
+    #[cfg(all(feature = "tokio-runtime", feature = "office"))]
+    {
+        registry.register(Arc::new(DocxExtractor::new()))?;
+        registry.register(Arc::new(PptxExtractor::new()))?;
+        registry.register(Arc::new(OdtExtractor::new()))?;
+    }
+
     #[cfg(feature = "email")]
     registry.register(Arc::new(EmailExtractor::new()))?;
 
@@ -313,14 +372,11 @@ mod tests {
 
         #[cfg(feature = "office")]
         {
-            expected_count += 13;
+            expected_count += 10;
             assert!(extractor_names.contains(&"markdown-extractor".to_string()));
             assert!(extractor_names.contains(&"bibtex-extractor".to_string()));
-            assert!(extractor_names.contains(&"docx-extractor".to_string()));
             assert!(extractor_names.contains(&"epub-extractor".to_string()));
             assert!(extractor_names.contains(&"fictionbook-extractor".to_string()));
-            assert!(extractor_names.contains(&"pptx-extractor".to_string()));
-            assert!(extractor_names.contains(&"odt-extractor".to_string()));
             assert!(extractor_names.contains(&"rtf-extractor".to_string()));
             assert!(extractor_names.contains(&"rst-extractor".to_string()));
             assert!(extractor_names.contains(&"latex-extractor".to_string()));
@@ -330,6 +386,14 @@ mod tests {
             assert!(extractor_names.contains(&"typst-extractor".to_string()));
         }
 
+        #[cfg(all(feature = "tokio-runtime", feature = "office"))]
+        {
+            expected_count += 3;
+            assert!(extractor_names.contains(&"docx-extractor".to_string()));
+            assert!(extractor_names.contains(&"pptx-extractor".to_string()));
+            assert!(extractor_names.contains(&"odt-extractor".to_string()));
+        }
+
         #[cfg(feature = "email")]
         {
             expected_count += 1;

data/vendor/kreuzberg/src/extractors/odt.rs

@@ -1,3 +1,5 @@
+#![cfg(all(feature = "tokio-runtime", feature = "office"))]
+
 //! ODT (OpenDocument Text) extractor using native Rust parsing.
 //!
 //! Supports: OpenDocument Text (.odt)
@@ -169,13 +171,10 @@ fn extract_content_text(archive: &mut zip::ZipArchive<Cursor<Vec<u8>>>) -> crate
 
     let mut text_parts: Vec<String> = Vec::new();
 
-    // Find the office:text or text body element - this is the main document body
     for body_child in root.children() {
         if body_child.tag_name().name() == "body" {
-            // Process the text element inside body
             for text_elem in body_child.children() {
                 if text_elem.tag_name().name() == "text" {
-                    // Now process only direct children of the text element
                     process_document_elements(text_elem, &mut text_parts);
                 }
             }
@@ -563,6 +562,7 @@ impl DocumentExtractor for OdtExtractor {
                 additional: metadata_map,
                 ..Default::default()
             },
+            pages: None,
             tables,
             detected_languages: None,
             chunks: None,

data/vendor/kreuzberg/src/extractors/opml.rs

@@ -191,6 +191,7 @@ impl DocumentExtractor for OpmlExtractor {
                 additional: metadata_map,
                 ..Default::default()
             },
+            pages: None,
             tables: vec![],
             detected_languages: None,
             chunks: None,

data/vendor/kreuzberg/src/extractors/orgmode.rs

@@ -304,6 +304,7 @@ impl DocumentExtractor for OrgModeExtractor {
             detected_languages: None,
             chunks: None,
             images: None,
+            pages: None,
         })
     }