Updated the third-party dependencies of PDFium and fixed known security vulnerabilities.
Adjusted the minimum system version (iOS 13.0 / Simulator 14.0 / Catalyst 14.0 / macOS 11.0).
Optimized the rendering and loading time during iOS zooming.
修复
Fixed the issue with incorrect rectangular range when editing search text content.
ComPDF SDK for Android 3.0.22026年9月28日
功能
Updated PDFium third-party dependencies and fixed known security vulnerabilities.
修复
Fixed an issue where the rectangle bounds for text search during content editing were incorrect.
ComPDF SDK for iOS 3.0.12026年8月13日
修复
Fixed the issue in the SDK that caused automatic exit from full-screen mode.
Fixed the crash that occurred when editing and saving special document content.
ComPDF SDK for Android 3.0.12026年8月13日
功能
Improved text selection calculations for rotated pages and scanned OCR documents, resulting in more accurate text selection, selection handle dragging, and Markup annotation placement.
修复
Fixed an issue where text selection and highlight regions could have incorrect heights and extend beyond the top of the text on some rotated pages with hidden OCR text layers.
Fixed an issue where CPDFBackground#getOpacity() returned an incorrect value for valid background objects.
ComPDF SDK for iOS 3.0.0主版本2026年8月6日
功能
Ink annotations now support a mixed undo/redo history mode: strokes and document annotation operations are included in the same undo stack, and are uniformly undone and restored according to the actual operation sequence.
A new "pageSameWidth" (equal-width page) preview mode has been added.
It is now possible to set the maximum zoom ratio for PDFView.
Form annotations now support custom background colors when in read-only mode.
Japanese language support has been added.
Optimized the UI layer performance of the drag-and-drop control when scaling in the content editing and annotation mode.
Improved the interaction details between text selection and page scrolling.
Reconstructed the structure of the Stamp annotation rotation method to better adapt to the logic related to annotation rotation.
Optimized the inconsistent rendering effect of the text annotation appearance flow (AP).
修复
Fixed the issues of incorrect watermark rotation angle and ineffective modification of watermark attributes when repeatedly performed.
Fixed the problem of being unable to open network links in certain system versions.
ComPDF SDK for Android 3.0.0主版本2026年8月6日
功能
Raised the minimum supported Android version to API 23 (Android 6.0).
Added mutually exclusive selection support for RadioButton form fields within the same group.
Added accessibility configuration support to the ComPDFKit_Tools module.
Added multithreaded PDF page rendering support.
Added Japanese language support.
Added a hybrid Undo/Redo history mode for Ink annotations, allowing ink strokes and document annotation operations to be restored in their actual undo order.
Improved page editing in the ComPDFKit_Tools module, enhancing page copy stability and thumbnail refresh performance.
Improved storage permission handling and related error messages in the ComPDFKit_Tools module.
Improved network security configuration in the ComPDFKit_Tools module.
Improved current-page callback logic in two-page reading mode to ensure consistent page number callbacks during page transitions.
Improved reader gesture recognition.
修复
Fixed an issue where same-width pages were not displayed correctly in horizontal continuous reading mode, improving layout consistency across different screen sizes.
Fixed an issue where swiping while zoomed in could unintentionally trigger page navigation.
Fixed incorrect page positioning and sizing when switching to full-screen mode in the ComPDFKit_Tools module.
Fixed an issue where scrolling or double-tap zooming could remove text and Markup selections.
ComPDF Conversion SDK V4.0.0主版本2026年6月22日
功能
Multi-threading Support - Faster PDF Conversion:
When processing hundreds or thousands of documents, single-threading is constrained by single-core utilization and sequential processing bottlenecks, resulting in severely insufficient overall throughput. Added official support for external multi-threaded invocation, employing thread pools and task partitioning to split batch documents into multiple parallel workflows, significantly improving system concurrency and resource utilization.
Large-scale OCR services (thread scheduling optimization for CPU/GPU heterogeneous preprocessing).
AI document processing platforms (pipeline parallelism for preprocessing and inference).
Automated archival systems (high-frequency batch processing of small files).
For high-concurrency workloads, upgrading from single‑thread to multi-thread can significantly reduce total processing time (especially for I/O‑intensive or mixed workloads), while reducing the number of server nodes required to meet SLAs, directly lowering infrastructure costs.
PDF Layout Restoration Upgraded by Refined Text and Paragraph Styles:
At its core, high-precision layout restoration represents OCR technology's deep understanding of a document's original structure and visual perception. The new version delivers more refined layout and style restoration through the following key improvements:
Style restoration support – More detailed preservation of original text attributes (font, size, color) and image positions, reduces layout inaccuracies caused by style mismatches.
More natural paragraph reconstruction – Upgrades paragraph alignment restoration, aggregation logic, line height, and more. It reduces issues such as paragraph crowding and inconsistent paragraph alignment.
More accurate line and character spacing - Resolves common OCR issues such as text crowding or excessive looseness, resulting in a cleaner, more professional overall layout.
Output layout closer to the original document – From element positioning to full-page composition, achieves high-fidelity restoration that preserves the original visual hierarchy.
An upgrade from general layout restoration to precision layout restoration when OCR is enabled, delivering more reliable and readable conversion results for document digitization workflows.
Integrated AI Table Recognition - Drastically Improved Table Restoration:
Better restoration of bold, font, size, color, and alignment for text inside tables.
Automatic aggregation of multi-line text.
Improved table hierarchy, header-region detection, and merged-cell recognition.
Supported outputs: PDF to Word / Excel / PPT / RTF / HTML / Markdown / XML.
Highlight: Text paragraph aggregation inside tables automatically merges multi-line content into natural paragraphs while preserving original alignment.
Render Vector Graphics (Path Objects) as Images:
Added independent image rendering for non-table path objects to achieve:
More complete preservation of original visual effects.
Reduced graphic distortion.
Improved restoration quality for complex layouts.
This is especially effective for flowcharts, structural diagrams, and engineering drawings.
Expanded Thai Font Library:
Higher Thai character recognition accuracy.
Fewer font substitution issues.
More natural OCR layout outcomes.
Document Directory Link Navigation Support (PDF to Word):
Added restoration for internal PDF links, including table-of-contents jumps, chapter links, and internal anchors. These links remain usable after PDF to Word conversion.
This is highly valuable for user experience in: eBooks, technical documentation, government standard documents, and long-form reports.
PDF to Searchable PDF Conversion:
Introduced searchable PDF as an output format. Its core mechanism preserves the original scanned image layer while overlaying a transparent OCR text layer, enabling full-text search without altering the document's original visual appearance.
This means users can precisely locate the position of the information they need within the document, achieving efficient information retrieval and content positioning - without text reflow that would disrupt the original layout and visual structure.
This capability is especially valuable in scenarios where preserving the original visual fidelity is critical, such as archival digitization, compliance archiving, legal document storage, and historical document digitalization.
PDF/Image to Searchable OFD:
Added the ability to convert PDFs and images into searchable OFD files, preserving the original typography, styling, and layout while supporting operations such as search, highlight, and annotation.