edgeparse 0.2.0__tar.gz → 0.2.2__tar.gz

{edgeparse-0.2.0 → edgeparse-0.2.2}/Cargo.lock

@@ -636,7 +636,7 @@ dependencies = [
 
 [[package]]
 name = "edgeparse-cli"
-version = "0.2.0"
+version = "0.2.2"
 dependencies = [
  "anyhow",
  "clap",
@@ -650,7 +650,7 @@ dependencies = [
 
 [[package]]
 name = "edgeparse-core"
-version = "0.2.0"
+version = "0.2.2"
 dependencies = [
  "anyhow",
  "base64",
@@ -674,7 +674,7 @@ dependencies = [
 
 [[package]]
 name = "edgeparse-node"
-version = "0.2.0"
+version = "0.2.2"
 dependencies = [
  "edgeparse-core",
  "napi",
@@ -686,7 +686,7 @@ dependencies = [
 
 [[package]]
 name = "edgeparse-python"
-version = "0.2.0"
+version = "0.2.2"
 dependencies = [
  "edgeparse-core",
  "pyo3",
@@ -695,7 +695,7 @@ dependencies = [
 
 [[package]]
 name = "edgeparse-wasm"
-version = "0.2.0"
+version = "0.2.2"
 dependencies = [
  "console_error_panic_hook",
  "console_log",
@@ -2482,9 +2482,9 @@ dependencies = [
 
 [[package]]
 name = "rustls-webpki"
-version = "0.103.9"
+version = "0.103.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "d7df23109aa6c1567d1c575b9952556388da57401e4ace1d15f79eedad0d8f53"
+checksum = "df33b2b81ac578cabaf06b89b0631153a3f416b0a886e8a7a1707fb51abbd1ef"
 dependencies = [
  "ring",
  "rustls-pki-types",

{edgeparse-0.2.0 → edgeparse-0.2.2}/Cargo.toml

@@ -3,7 +3,7 @@ resolver = "2"
 members = ["crates/edgeparse-core", "crates/pdf-cos", "crates/edgeparse-python"]
 
 [workspace.package]
-version = "0.2.0"
+version = "0.2.2"
 edition = "2021"
 rust-version = "1.85"
 license = "Apache-2.0"

{edgeparse-0.2.0 → edgeparse-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: edgeparse
-Version: 0.2.0
+Version: 0.2.2
 Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: Developers
 Classifier: License :: OSI Approved :: Apache Software License

{edgeparse-0.2.0 → edgeparse-0.2.2}/crates/edgeparse-core/src/api/config.rs

@@ -114,6 +114,8 @@ pub struct ProcessingConfig {
     pub image_format: ImageFormat,
     /// Directory for extracted images
     pub image_dir: Option<String>,
+    /// Enable raster table OCR recovery on image-based tables
+    pub raster_table_ocr: bool,
     /// Pages to extract (e.g., "1,3,5-7")
     pub pages: Option<String>,
     /// Include headers/footers in output
@@ -150,6 +152,7 @@ impl Default for ProcessingConfig {
             image_output: ImageOutput::External,
             image_format: ImageFormat::Png,
             image_dir: None,
+            raster_table_ocr: true,
             pages: None,
             include_header_footer: false,
             hybrid: HybridBackend::Off,
@@ -175,6 +178,7 @@ mod tests {
         assert_eq!(config.table_method, TableMethod::Default);
         assert_eq!(config.image_output, ImageOutput::External);
         assert_eq!(config.image_format, ImageFormat::Png);
+        assert!(config.raster_table_ocr);
         assert_eq!(config.hybrid, HybridBackend::Off);
         assert_eq!(config.hybrid_timeout, 30000);
     }

{edgeparse-0.2.0 → edgeparse-0.2.2}/crates/edgeparse-core/src/lib.rs

@@ -24,7 +24,9 @@ use crate::models::document::PdfDocument;
 use crate::pdf::chunk_parser::extract_page_chunks;
 use crate::pdf::page_info;
 #[cfg(not(target_arch = "wasm32"))]
-use crate::pdf::raster_table_ocr::recover_raster_table_borders;
+use crate::pdf::raster_table_ocr::{
+    recover_page_raster_table_cell_text, recover_raster_table_borders,
+};
 use crate::pipeline::orchestrator::{run_pipeline, PipelineState};
 use crate::tagged::struct_tree::build_mcid_map;
 
@@ -56,17 +58,19 @@ pub fn convert(
     for (&page_num, &page_id) in &pages_map {
         let page_chunks = extract_page_chunks(&raw_doc.document, page_num, page_id)?;
         let mut recovered_tables = Vec::new();
-        if let Some(page_info) = page_info_list
-            .iter()
-            .find(|info| info.page_number == page_num)
-        {
-            recovered_tables = recover_raster_table_borders(
-                input_path,
-                &page_info.crop_box,
-                page_num,
-                &page_chunks.text_chunks,
-                &page_chunks.image_chunks,
-            );
+        if config.raster_table_ocr {
+            if let Some(page_info) = page_info_list
+                .iter()
+                .find(|info| info.page_number == page_num)
+            {
+                recovered_tables = recover_raster_table_borders(
+                    input_path,
+                    &page_info.crop_box,
+                    page_num,
+                    &page_chunks.text_chunks,
+                    &page_chunks.image_chunks,
+                );
+            }
         }
         let mut elements: Vec<ContentElement> = page_chunks
             .text_chunks
@@ -104,7 +108,7 @@ pub fn convert(
     // Run the processing pipeline
     let mcid_map = build_mcid_map(&raw_doc.document);
     let mut pipeline_state = PipelineState::with_mcid_map(page_contents, config.clone(), mcid_map)
-        .with_page_info(page_info_list);
+        .with_page_info(page_info_list.clone());
     run_pipeline(&mut pipeline_state)?;
 
     // Build the output document
@@ -115,12 +119,26 @@ pub fn convert(
         .to_string();
 
     let mut doc = PdfDocument::new(file_name);
+    doc.source_path = Some(input_path.display().to_string());
     doc.number_of_pages = pages_map.len() as u32;
     doc.author = raw_doc.metadata.author;
     doc.title = raw_doc.metadata.title;
     doc.creation_date = raw_doc.metadata.creation_date;
     doc.modification_date = raw_doc.metadata.modification_date;
 
+    if config.raster_table_ocr {
+        for (page_idx, page) in pipeline_state.pages.iter_mut().enumerate() {
+            if let Some(page_info) = page_info_list.get(page_idx) {
+                recover_page_raster_table_cell_text(
+                    input_path,
+                    &page_info.crop_box,
+                    page_info.page_number,
+                    page,
+                );
+            }
+        }
+    }
+
     // Flatten pipeline output into document kids
     for page in pipeline_state.pages {
         doc.kids.extend(page);

{edgeparse-0.2.0 → edgeparse-0.2.2}/crates/edgeparse-core/src/models/document.rs

@@ -9,6 +9,9 @@ use super::content::ContentElement;
 pub struct PdfDocument {
     /// Original file name
     pub file_name: String,
+    /// Original source path when available.
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub source_path: Option<String>,
     /// Number of pages
     pub number_of_pages: u32,
     /// Document author
@@ -36,6 +39,7 @@ impl PdfDocument {
     pub fn new(file_name: String) -> Self {
         Self {
             file_name,
+            source_path: None,
             number_of_pages: 0,
             author: None,
             title: None,
@@ -89,6 +93,7 @@ mod tests {
     fn test_new_document() {
         let doc = PdfDocument::new("test.pdf".to_string());
         assert_eq!(doc.file_name, "test.pdf");
+        assert_eq!(doc.source_path, None);
         assert_eq!(doc.number_of_pages, 0);
         assert!(doc.kids.is_empty());
     }