kreuzberg 4.2.12 → 4.2.13

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

@@ -88,6 +88,9 @@ pub mod html;
 #[cfg(feature = "office")]
 pub mod bibtex;
 
+#[cfg(feature = "office")]
+pub mod citation;
+
 #[cfg(all(feature = "tokio-runtime", feature = "office"))]
 pub mod docx;
 
@@ -146,7 +149,7 @@ pub use text::{MarkdownExtractor, PlainTextExtractor};
 pub use image::ImageExtractor;
 
 #[cfg(feature = "archives")]
-pub use archive::{SevenZExtractor, TarExtractor, ZipExtractor};
+pub use archive::{GzipExtractor, SevenZExtractor, TarExtractor, ZipExtractor};
 
 #[cfg(feature = "email")]
 pub use email::EmailExtractor;
@@ -160,6 +163,9 @@ pub use html::HtmlExtractor;
 #[cfg(feature = "office")]
 pub use bibtex::BibtexExtractor;
 
+#[cfg(feature = "office")]
+pub use citation::CitationExtractor;
+
 #[cfg(all(feature = "tokio-runtime", feature = "office"))]
 pub use docx::DocxExtractor;
 
@@ -278,7 +284,11 @@ pub fn register_default_extractors() -> Result<()> {
     registry.register(Arc::new(ImageExtractor::new()))?;
 
     #[cfg(feature = "xml")]
-    registry.register(Arc::new(XmlExtractor::new()))?;
+    {
+        registry.register(Arc::new(XmlExtractor::new()))?;
+        registry.register(Arc::new(JatsExtractor::new()))?;
+        registry.register(Arc::new(DocbookExtractor::new()))?;
+    }
 
     #[cfg(feature = "pdf")]
     registry.register(Arc::new(PdfExtractor::new()))?;
@@ -292,6 +302,7 @@ pub fn register_default_extractors() -> Result<()> {
     {
         registry.register(Arc::new(EnhancedMarkdownExtractor::new()))?;
         registry.register(Arc::new(BibtexExtractor::new()))?;
+        registry.register(Arc::new(CitationExtractor::new()))?;
         registry.register(Arc::new(EpubExtractor::new()))?;
         registry.register(Arc::new(FictionBookExtractor::new()))?;
         registry.register(Arc::new(RtfExtractor::new()))?;
@@ -321,6 +332,7 @@ pub fn register_default_extractors() -> Result<()> {
         registry.register(Arc::new(ZipExtractor::new()))?;
         registry.register(Arc::new(TarExtractor::new()))?;
         registry.register(Arc::new(SevenZExtractor::new()))?;
+        registry.register(Arc::new(GzipExtractor::new()))?;
     }
 
     Ok(())
@@ -362,8 +374,10 @@ mod tests {
 
         #[cfg(feature = "xml")]
         {
-            expected_count += 1;
+            expected_count += 3;
             assert!(extractor_names.contains(&"xml-extractor".to_string()));
+            assert!(extractor_names.contains(&"jats-extractor".to_string()));
+            assert!(extractor_names.contains(&"docbook-extractor".to_string()));
         }
 
         #[cfg(feature = "pdf")]
@@ -380,9 +394,10 @@ mod tests {
 
         #[cfg(feature = "office")]
         {
-            expected_count += 10;
+            expected_count += 11;
             assert!(extractor_names.contains(&"markdown-extractor".to_string()));
             assert!(extractor_names.contains(&"bibtex-extractor".to_string()));
+            assert!(extractor_names.contains(&"citation-extractor".to_string()));
             assert!(extractor_names.contains(&"epub-extractor".to_string()));
             assert!(extractor_names.contains(&"fictionbook-extractor".to_string()));
             assert!(extractor_names.contains(&"rtf-extractor".to_string()));
@@ -416,10 +431,11 @@ mod tests {
 
         #[cfg(feature = "archives")]
         {
-            expected_count += 3;
+            expected_count += 4;
             assert!(extractor_names.contains(&"zip-extractor".to_string()));
             assert!(extractor_names.contains(&"tar-extractor".to_string()));
             assert!(extractor_names.contains(&"7z-extractor".to_string()));
+            assert!(extractor_names.contains(&"gzip-extractor".to_string()));
         }
 
         assert_eq!(

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

@@ -95,7 +95,7 @@ impl DocumentExtractor for OpmlExtractor {
     }
 
     fn supported_mime_types(&self) -> &[&str] {
-        &["text/x-opml", "application/xml+opml"]
+        &["text/x-opml", "application/xml+opml", "application/x-opml+xml"]
     }
 
     fn priority(&self) -> i32 {
@@ -135,6 +135,7 @@ mod tests {
         let supported = extractor.supported_mime_types();
         assert!(supported.contains(&"text/x-opml"));
         assert!(supported.contains(&"application/xml+opml"));
+        assert!(supported.contains(&"application/x-opml+xml"));
     }
 
     #[tokio::test]

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

@@ -14,7 +14,8 @@ use std::io::{Read, Seek};
 ///
 /// All limits are intentionally conservative to prevent DoS attacks
 /// while still supporting legitimate documents.
-#[derive(Clone, Debug)]
+#[derive(Clone, Debug, serde::Serialize, serde::Deserialize)]
+#[serde(default)]
 pub struct SecurityLimits {
     /// Maximum uncompressed size for archives (500 MB)
     pub max_archive_size: usize,

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

@@ -59,8 +59,12 @@ impl DocumentExtractor for StructuredExtractor {
         _config: &ExtractionConfig,
     ) -> Result<ExtractionResult> {
         let structured_result = match mime_type {
-            "application/json" | "text/json" => crate::extraction::structured::parse_json(content, None)?,
-            "application/x-yaml" | "text/yaml" | "text/x-yaml" => crate::extraction::structured::parse_yaml(content)?,
+            "application/json" | "text/json" | "application/csl+json" => {
+                crate::extraction::structured::parse_json(content, None)?
+            }
+            "application/yaml" | "application/x-yaml" | "text/yaml" | "text/x-yaml" => {
+                crate::extraction::structured::parse_yaml(content)?
+            }
             "application/toml" | "text/toml" => crate::extraction::structured::parse_toml(content)?,
             _ => return Err(crate::KreuzbergError::UnsupportedFormat(mime_type.to_string())),
         };
@@ -112,6 +116,8 @@ impl DocumentExtractor for StructuredExtractor {
         &[
             "application/json",
             "text/json",
+            "application/csl+json",
+            "application/yaml",
             "application/x-yaml",
             "text/yaml",
             "text/x-yaml",
@@ -141,9 +147,10 @@ mod tests {
     fn test_structured_extractor_supported_mime_types() {
         let extractor = StructuredExtractor::new();
         let mime_types = extractor.supported_mime_types();
-        assert_eq!(mime_types.len(), 7);
+        assert_eq!(mime_types.len(), 9);
         assert!(mime_types.contains(&"application/json"));
         assert!(mime_types.contains(&"application/x-yaml"));
         assert!(mime_types.contains(&"application/toml"));
+        assert!(mime_types.contains(&"application/csl+json"));
     }
 }

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

@@ -97,7 +97,15 @@ impl DocumentExtractor for PlainTextExtractor {
     }
 
     fn supported_mime_types(&self) -> &[&str] {
-        &["text/plain", "text/csv", "text/tab-separated-values"]
+        &[
+            "text/plain",
+            "text/csv",
+            "text/tab-separated-values",
+            "text/troff",
+            "text/x-mdoc",
+            "text/x-pod",
+            "text/x-dokuwiki",
+        ]
     }
 
     fn priority(&self) -> i32 {
@@ -192,7 +200,12 @@ impl DocumentExtractor for MarkdownExtractor {
     }
 
     fn supported_mime_types(&self) -> &[&str] {
-        &["text/markdown", "text/x-markdown"]
+        &[
+            "text/markdown",
+            "text/x-markdown",
+            "text/x-markdown-extra",
+            "text/x-multimarkdown",
+        ]
     }
 
     fn priority(&self) -> i32 {
@@ -253,7 +266,15 @@ mod tests {
         assert_eq!(extractor.version(), env!("CARGO_PKG_VERSION"));
         assert_eq!(
             extractor.supported_mime_types(),
-            &["text/plain", "text/csv", "text/tab-separated-values"]
+            &[
+                "text/plain",
+                "text/csv",
+                "text/tab-separated-values",
+                "text/troff",
+                "text/x-mdoc",
+                "text/x-pod",
+                "text/x-dokuwiki",
+            ]
         );
         assert_eq!(extractor.priority(), 50);
     }
@@ -263,7 +284,15 @@ mod tests {
         let extractor = MarkdownExtractor::new();
         assert_eq!(extractor.name(), "markdown-extractor");
         assert_eq!(extractor.version(), env!("CARGO_PKG_VERSION"));
-        assert_eq!(extractor.supported_mime_types(), &["text/markdown", "text/x-markdown"]);
+        assert_eq!(
+            extractor.supported_mime_types(),
+            &[
+                "text/markdown",
+                "text/x-markdown",
+                "text/x-markdown-extra",
+                "text/x-multimarkdown"
+            ]
+        );
         assert_eq!(extractor.priority(), 50);
     }
 }

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

@@ -96,7 +96,12 @@ impl DocumentExtractor for XmlExtractor {
     }
 
     fn supported_mime_types(&self) -> &[&str] {
-        &["application/xml", "text/xml", "image/svg+xml"]
+        &[
+            "application/xml",
+            "text/xml",
+            "image/svg+xml",
+            "application/x-endnote+xml",
+        ]
     }
 
     fn priority(&self) -> i32 {
@@ -142,7 +147,12 @@ mod tests {
         assert_eq!(extractor.version(), env!("CARGO_PKG_VERSION"));
         assert_eq!(
             extractor.supported_mime_types(),
-            &["application/xml", "text/xml", "image/svg+xml"]
+            &[
+                "application/xml",
+                "text/xml",
+                "image/svg+xml",
+                "application/x-endnote+xml"
+            ]
         );
         assert_eq!(extractor.priority(), 50);
     }

data/vendor/kreuzberg/src/ocr/processor/execution.rs

@@ -72,8 +72,21 @@ pub(super) fn perform_ocr(
         )
     });
 
-    let img = image::load_from_memory(image_bytes)
-        .map_err(|e| OcrError::ImageProcessingFailed(format!("Failed to decode image: {}", e)))?;
+    let img = {
+        // Check for JPEG 2000 format which the image crate doesn't support
+        if crate::extraction::image::is_jp2(image_bytes) || crate::extraction::image::is_j2k(image_bytes) {
+            crate::extraction::image::decode_jp2_to_rgb(image_bytes)
+                .map(image::DynamicImage::ImageRgb8)
+                .map_err(|e| OcrError::ImageProcessingFailed(format!("Failed to decode JP2 image: {}", e)))?
+        } else if crate::extraction::image::is_jbig2(image_bytes) {
+            crate::extraction::image::decode_jbig2_to_gray(image_bytes)
+                .map(image::DynamicImage::ImageLuma8)
+                .map_err(|e| OcrError::ImageProcessingFailed(format!("Failed to decode JBIG2 image: {}", e)))?
+        } else {
+            image::load_from_memory(image_bytes)
+                .map_err(|e| OcrError::ImageProcessingFailed(format!("Failed to decode image: {}", e)))?
+        }
+    };
 
     let rgb_image = img.to_rgb8();
     let (width, height) = rgb_image.dimensions();
@@ -224,7 +237,7 @@ pub(super) fn perform_ocr(
         "tsv" => {
             let tsv = tsv_data_for_tables
                 .as_ref()
-                .expect("TSV data should be extracted when output_format is 'tsv'")
+                .ok_or_else(|| OcrError::ProcessingFailed("TSV data not available".to_string()))?
                 .clone();
             (tsv, "text/plain".to_string())
         }

data/vendor/kreuzberg-tesseract/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kreuzberg-tesseract"
-version = "4.2.12"
+version = "4.2.13"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: kreuzberg
 version: !ruby/object:Gem::Version
-  version: 4.2.12
+  version: 4.2.13
 platform: ruby
 authors:
 - Na'aman Hirschfeld
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-02-06 00:00:00.000000000 Z
+date: 2026-02-07 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rb_sys
@@ -422,6 +422,7 @@ files:
 - vendor/kreuzberg/src/core/server_config/validation.rs
 - vendor/kreuzberg/src/embeddings.rs
 - vendor/kreuzberg/src/error.rs
+- vendor/kreuzberg/src/extraction/archive/gzip.rs
 - vendor/kreuzberg/src/extraction/archive/mod.rs
 - vendor/kreuzberg/src/extraction/archive/sevenz.rs
 - vendor/kreuzberg/src/extraction/archive/tar.rs
@@ -463,6 +464,7 @@ files:
 - vendor/kreuzberg/src/extraction/xml.rs
 - vendor/kreuzberg/src/extractors/archive.rs
 - vendor/kreuzberg/src/extractors/bibtex.rs
+- vendor/kreuzberg/src/extractors/citation.rs
 - vendor/kreuzberg/src/extractors/djot_format/attributes.rs
 - vendor/kreuzberg/src/extractors/djot_format/conversion.rs
 - vendor/kreuzberg/src/extractors/djot_format/extractor.rs