Version 10.12.0 Changelog (September 4th, 2024)

Version 10.12.1 Hotfix (September 11, 2024)

• [cad] Using the latest CAD Module now requires this version of the SDK or later. If this module is used with an older SDK an exception will be thrown.
• [ocr] Using the latest OCR or IRIS OCR Modules now requires this version of the SDK or later. If either module is used with an older SDK an exception will be thrown.

• [pdf] Improved support for repairing corrupt documents where a space between %%EOF and the following object number was missing. Previously an "Object header not found" exception could be thrown.
• [pdf] Improved support in PDFDoc.MergeXFDF() for merging annotations into a certain type of corrupt PDF page. Previously this could lead to a "SetDirty() can't be invoked on Obj of this type." exception.
• [pdf] Improved the compatibility of compressed object streams created when saving corrupt files. Previously invalid objects could be compressed which could lead to PDF parsing problems in some applications.
• [html] HTML to PDF conversion will now include a document outline (also known as a bookmark tree) that aligns with the header structure of the html input.
• [pdf] Improved support for invalid object references in compressed object streams. Previously when handling these corrupt documents, a 'Bad obj stream' exception could be thrown.
• [pdf] The Data Extraction Module now supports detection of table headers when using the Doc Structure engine.
• [pdf] Improved detection of even-odd page header bounds in the Data Extraction Module's Doc Structure engine.
• [pdf] The Data Extraction Module now includes an object number for annotation graphic groups and textboxes when using the Doc Structure engine.
• [pdf] Improved detection of span bounds for lines with small caps in the Data Extraction Module's Doc Structure engine.

• [cad] CAD to PDF conversion will now ignore irrelevant errors that occur after conversion has finished. Previously these would cause an exception to be thrown even if PDF generation is complete.
• [pdf] Fixed an issue where Annots.FreeText.GetLineColor() could incorrectly return a visible color after the color has been set to fully transparent (0 colorants) with Annots.FreeText.SetLineColor().
• [pdf] Fixed an issue with rendering JPEG2000 images with embedded RGB-based ICC color profiles. Previously these images could be missing entirely.
• [pdf] Fixed an issue with redaction annotation appearance generation that could cause the final appearance to be transparent if the redaction were applied by a third-party application.
• [pdf] Fixed an issue with inserting pages from one document to another where it was possible for some types of malformed button form fields to be missing in the output document.
• [pdf] Fixed an issue with appearance generation of stamp annotations with non-standard rotation (not a multiple of 90 degrees). Previously the generated appearance could be distorted after applying multiple rotations.
• [ocr] Addressed some issues with inconsistent error handling for OCR and IRIS OCR modules. Previously some errors could cause OCR processing to fail silently.
• [pdf] Fixed a PDF rendering error that could result in sharp transitions when rendering extremely large gradients.
• [xps] Fixed an issue where PDF to XPS conversion or printing could throw a "The system cannot find the file specified." exception if the temp folder path contains non-ASCII characters.
• [pdf] Fixed rendering of a specific Identity-encoded font (GGothicBBB) when there is a metrically compatible substitute. Previously some incorrect characters could be displayed.
• [pdf] Fixed an issue with partial redaction of indexed grayscale images that could cause the non-redacted portion of the image to render incorrectly.
• [pdf] The Data Extraction Module's Doc Structure engine now outputs a rect entry for span elements. Previously this would be omitted despite being required in the output JSON specification.

• [doc] Fixed an uncommon issue where the background color of the first row in tables was incorrect in DOC documents.
• [pptx] Fixed an issue with overflowing text in some shapes with the "Shrink text on overflow" option set.
• [docs] Improved accuracy of rendering math equations, particularly vertical positioning of math elements.
• [docx] Documents with nested SDT table rows no longer have extra table rows.
• [docx] Added support for page background images (including tiled images) in Word.
• [docx] Fixed unexpected content displacement by line shapes with "through" wrapping.
• [docx] Improved the balancing of section columns with unequal widths.
• [office] Fixed an issue with the pixelated appearance of some EMF images.
• [xlsx] Implemented support for the "Scale with document" option for Excel headers and footers.
• [docx] Added proper handling of the "Next Column" section break type for Word documents.
• [docx] Improved handling of malformed Word documents with misplaced paragraph properties.
• [docx] Fixed a rare issue with vertically merged tables cells when they span rows which are split across different pages.
• [xlsx] Added implementation of EDATE Excel function.
• [ppt] Fixed a potential crash when processing PPT documents caused by non-standard binary data.
• [docx] Text boxes with rounded corners will now correctly resize to fit the size of the text, if set to do so.
• [office] Fixed an issue with incorrectly applied image cropping in some office documents.
• [docx] Improved positioning of VML shapes.
• [docx] Improved paragraph positioning following a column break.
• [xls] Resolved a crash that could occur when both header/footer images and content images were present in XLS documents.
• [xlsx] Implemented accurate row height calculation for rows containing cells with wrapped text.
• [xlsx] Adjusted checkbox control position to match Excel more accurately.
• [xlsx] Implemented re-calculation of some formula results in Excel sheets when necessary.
• [emf] Fixed an issue where a malformed PDF could be produced as a result of converting a document with a complicated EMF image.
• [xlsx] Fixed an issue with incorrect data labels shown in some charts.
• [xls] Fixed a potential crash that could be caused by incorrect handling of unsupported binary formula records.
• [office] Fixed a rare crash related to missing time unit information in charts.
• [docx] Fixed incorrect line height for some empty lines with floating elements.
• [docx] Added support for hyperlink anchors (hyperlinks with #) in Word.
• [docx] Resolved an issue where the list number could sometimes restart unexpectedly.
• [docx] Inline tables will now correctly avoid floating objects when there is a continuous section break on a page.
• [docx] Fixed multiple layout issues related to nested floating tables.
• [office] Implemented fast image tiling and improved the appearance of tiled shapes and backgrounds.
• [pptx] Implemented the display of fallback images for OLE objects in PPTX documents.
• [ppt] Fixed a rare issue with incorrect shape flipping in binary PPT documents.
• [docx] Fixed an issue with incorrect right indent in some documents with a character grid enabled.
• [docx] Fixed incorrect width of no-wrap cells with preferred width set.
• [docx] Fixed incorrect grid alignment for lines with large non-text elements in Word.
• [docx] Fixed a rare issue where table column widths were inaccurately set.
• [docx] Fixed a rare crash related to vanished paragraphs inside empty table cells.
• [docx] Corrected table widths for tables with percent column widths and without the preferred width set.
• [docx] Added support for custom list number formats with multiple leading zeroes.
• [xls] Improved handling of some cases of malformed XLS binary data.

• [office] Improved the algorithm for averaging text properties of a paragraph that contains unicode groups to describe a single Arabic glyph.
• [docx] Improved list detection.
• [docx] Improved the text line assembly of Arabic content with diacritics.
• [office] Improved detection of small caps text.
• [office] Improved detection of Arabic language when minimal English text is near the Arabic language.
• [docx] Improved detection of header content.
• [docx] Improved z-order placement of graphic in conversion output.
• [docx] Improved detection of narrow columns on borderless tables.
• [docx] Improved conversion result of self-intersecting glyph outlines.
• [office] Improved page margin calculation to be multiples of 1/4 inch for the imperial measurement system and 1/4 centimeter for the metric measurement system.
• [docx] Improved the rendering of Type 3 fonts.
• [docx] Improved usage of tab stops to space content on a single line.
• [docx] Improved conversion results when the encoding of an incoming pdf contains large, broken text areas.
• [docx] Fixed an issue causing the background of an image to become transparent.
• [office] A number of third-party libraries have been updated to include the latest security fixes.

This hotfix addresses two problems:

• Potential Office to PDF conversion instability when handling some edge cases.
• Options being ignored for PDF to Office conversions via the Structured Output Module. (Convert.ToWord(), Convert.ToExcel() and Convert.ToPowerPoint())