PrizmDoc Viewer v13.20 Release - Updated
In This Topic
    Release Notes
    In This Topic

    This section contains information on new features, improvements, fixes, and known issues for each release.

    Release Notes v13.20 Release

    Released Apr 2022

    PrizmDoc Viewer v13.20 introduces the following:

    13.20 New Features and Improvements

    • AutoCAD AC1032 Format Support. The CAD rendering engine in PrizmDoc Server has been updated to now include support for the AC1032 format. In particular, this adds native support for the 2018-2023 versions of the DWG file type as well as improved support for the latest versions of DGN, DWF and DXF files. Additionally, PrizmDoc Server now also supports adding custom CAD fonts for improved fidelity.
    • MS Excel gridlines rendering improved. The PrizmDoc Viewer now provides an option to render Excel gridlines as they are defined in the worksheet; the gridlines will be rendered if they are defined in the document settings and will not be rendered if they are not defined. Previously, there were only 2 options: either forcibly render them or not, regardless of their presence in the document settings. For more information please see the description of the fileTypes.excel.renderGridlines parameter of the central configuration.
    • The Redaction Creators API behavior improved. The PrizmDoc Server now provides an option to control how new line characters of the source document should be handled upon the text fragments matching process. For more information please see the description of the find.newLineHandling parameter of the POST /v2/redactionCreators API.
    • Experimental PDF Viewing Sessions API. PrizmDoc Server has a new experimental "v3" Viewing Sessions API to create and run viewing sessions for all the supported document formats so you can view those documents in the Accusoft PDF Viewer. NOTE: This feature is a work-in-progress that is not officially supported by Accusoft. Its behavior may change at any time in a future release of the product.
    • New E-Signer radio button display option. The E-Signer Viewer now provides support for displaying checkboxes that only allow a single selection as radio buttons. For more information please see the description of the singleSelectionCheckboxDisplay parameter.
    • Support for Kubernetes 1.22 and 1.23 has been added; support for 1.19 and 1.20 has been deprecated and will be dropped in a future release. The list of supported Kubernetes versions now includes 1.21, 1.22 and 1.23.

    13.20 Product Updates and Fixes

    Fidelity

    • Updated the Email Processing Service (EPS) third-party components to the newer versions which removes extra spaces rendered between words and lines in the email files with HTML bodies, thus rendering the content of email documents more precisely. Customers opening email files with HTML bodies that were rendered and redacted in previous versions of PrizmDoc Viewer may be required to adjust or re-create those existing annotation and redaction markups in version 13.20 to adjust to the new fidelity.
    • Updated the PDF Processing Service (PDFPS) third-party components to the newer versions, which improved the fidelity of raster images rendering in resulted PDF documents after applying markups.
    • Addressed an issue in the ViewerControl where the text of TextAreaSignature marks was displayed using a font size slightly less than the maxFontSize property.
    • Updated the Vector Conversion Service (VCS) third-party ODA components to the newer version, which improved the fidelity of CAD file conversion to raster, svg, and PDF formats and allowed support of AutoCAD 2018-2023 (AC1032 format version).
    • Addressed an issue in Vector Conversion Service (VCS) where Unicode characters were not converted or were converted incorrectly.

    Redaction

    • Addressed an issue with PrizmDoc Server markupBurners REST API where the text markup fonts such as Arial, Courier New, and Times New Roman were substituted with Helvetica, Courier, or Times−Roman even when fonts were available on server.

    Stability

    • Improved logging for the PrizmDoc PDF Processing Service to eliminate the redundant "Failed to find General Rotation" warning message.
    • Updated PAS to return charset name in Content-Type headers of its HTTP responses with JSON body for consistency, like this: "application/json;charset=utf-8".
    • The PAS now performs a retry of the viewing session and viewing package creation from URL sources when there is a communication failure.
    • Addressed an issue in PrizmDoc Server where the creation of two viewing sessions simultaneously for the same document might fail.
    • Improved PrizmDoc Server stability when starting with hundreds of thousands of work files in the work file cache folder.

    Security

    • Addressed potential vulnerabilities in the PrizmDoc Application Services (PAS) third-party components by updating them to the newer versions.
    • Addressed potential vulnerabilities in the Email Processing Service (EPS) third-party components by updating them to the newer versions.
    • Updated the Work File Service (WFS) to run on node.js v16 to take advantage of recent security fixes and other changes in the node.js.
    • Addressed a MarkupBurner security concern where the font could be specified using a file path in the markup file instead of the font name. Now the font path is ignored, and the font file name is used to find the installed font on the PrizmDoc Server instance for burning.
    • Addressed potential vulnerabilities in the PDF Processing Service (PDFPS) third-party components by updating them to the newer versions.
    • Updated the PrizmDoc Document Conversion Service (DCS) to run on node.js v16, taking advantage of recent security fixes and other changes in the node.js.
    • Addressed potential vulnerabilities in the Document Conversion Service (DCS) third-party components by updating them to the newer versions.
    • Updated the PrizmDoc Form Extraction Service (FES) to run on node.js v16, taking advantage of recent security fixes and other changes in the node.js.
    • Addressed potential vulnerabilities in the Public Request Service (PRS) third-party components by updating them to the newer versions.
    • Addressed potential vulnerabilities in the Form Extraction Service (FES) third-party components by updating them to the newer versions.
    • Updated the PrizmDoc Email Conversion Service (ECS) to run on node.js v16, taking advantage of recent security fixes and other changes in the node.js.
    • Addressed potential vulnerabilities in the Email Conversion Service (ECS) third-party components by updating them to the newer versions.
    • Updated the PrizmDoc Viewing Service to run on node.js v16, taking advantage of recent security fixes and other changes in the node.js.
    • Addressed XSS vulnerabilities in the PrizmDoc Viewer when displaying the hyperlink menu, the redaction context menu, or the notification dialog.
    • Updated the PrizmDoc Text Service (TS) to run on node.js v16, taking advantage of recent security fixes and other changes in the node.js.
    • Addressed potential vulnerabilities in the Text Service (TS) third-party components by updating them to the newer versions.
    • Updated the PrizmDoc Mongo Manager Service (MMS) to run on node.js v16, taking advantage of recent security fixes and other changes in the node.js.
    • Addressed potential vulnerabilities in the Mongo Manager Service (MMS) third-party components by updating them to the newer versions.
    • Addressed a vulnerability when rendering specific CSV documents in the LibreOffice rendering mode.

    Text Extraction

    • Updated the PDF Processing Service (PDFPS) third-party components to the newer versions, which improved text extraction and search features in PrizmDoc Viewer.

    Viewing

    • The text for proximity search error messages in the PrizmDoc Viewer is now included in the PrizmDoc Viewer language file so it can be configured.

    Additional Fonts

    • Updated the Vector Conversion Service (VCS) to allow additional TrueType and shx fonts.

    13.20 Documentation Updates

    Release Notes v13.19

    Released Jan 2022

    PrizmDoc Viewer v13.19 introduces the following:

    13.19 New Features and Improvements

    • Kubernetes Guidance. We are now providing instructions and recommendations for PrizmDoc Viewer deployment to Kubernetes. Refer to the Kubernetes topic for more information.
    • Cluster Manager for Kubernetes. A new accusoft/prizmdoc-cluster-manager container image is now available, which simplifies the deployment of PrizmDoc Viewer container images in Kubernetes. Please see Deployment to Kubernetes Guidance for more details.
    • New Sample for PrizmDoc Viewer deployment in Kubernetes on GitHub. We've published a new sample project on GitHub. This sample provides manifest files that allow you to get a PrizmDoc Viewer cluster up and running in Kubernetes.
    • MarkupBurner performance improved. The PrizmDoc Viewer component responsible for applying markup was updated and now returns MarkupBurner results up to 30% faster for hi-load scenarios which include MarkupBurner API usage and Hi-End Server configurations. See PrizmDoc Server sizing section.

    13.19 Product Updates and Fixes

    Fidelity

    • Improved fidelity, with certain source files, when converting to PDF after updating to the new Prizm Server Docker image using Ubuntu 20.04.
    • Addressed an issue in PrizmDoc Server where it incorrectly rendered specific MSG documents with UTF-7 encoded HTML body.

    Redaction

    • Addressed an issue with the PrizmDoc Server markupBurners REST API where markup definitions were not applied to PDF files created from "Muhimbi PDF Converter Services".
    • Addressed an issue with the PrizmDoc Server markupBurners REST API taking longer than expected to execute. This occurred when PDF documents had long PDF Document Custom Properties definitions.

    Stability

    • Addressed an issue in the accusoft/prizmdoc-server Docker image where the watchdog_launcher.log and config-normalizer.log were not available in the /logs folder.
    • Addressed an issue in the accusoft/prizmdoc-application-services Docker image where it was not possible to obtain the PAS console log. This log is now available as the stdout/stderr output from the container.
    • Changed the default behavior of the accusoft/prizmdoc-application-services Docker image so that it runs a single instance of PAS, instead of running as many instances as the number of CPU cores detected on the system. This allows for using less system resources when the load is low, and scaling the PAS cluster efficiently by adding or removing prizmdoc-application-services containers. As a result of this update, instead of creating several log files named: pas-0.log, pas-1.log, etc., only one log file named: pas.log will be created by the container. If you want to preserve the legacy behavior, pass the LEGACY_RUN_MULTIPLE_INSTANCES environment variable with the value set to "YES" when starting the container.
    • Updated accusoft/prizmdoc-server Docker image to allow it to start container with overridden user.
    • Updated accusoft/prizmdoc-application-services Docker image to allow it to start container with overridden user.
    • Improved logging for the PrizmDoc Application Services to provide additional information when requesting the Viewing Session source document by URL.
    • Addressed an issue in the accusoft/prizmdoc-server Docker image where it failed to start up on host systems which use the cgroup v2 process controller.
    • Addressed an issue in PrizmDoc Server where its Raster Conversion Service module might enter an incorrect state and fail to convert documents to searchable PDF, if it was restarted while doing such a conversion.
    • Updated accusoft/prizmdoc-server Docker image to control time zone with TZ environment when rendering documents.
    • Addressed an issue in the PrizmDoc Server where the /v2/searchContexts API failed due to using an outdated port configuration of an internal service (this occurred after the expected internal services restarted during a high load period).
    • Addressed an issue in PrizmDoc Server when the LoadBalancer occasionally could not rotate log files and crashed on the Windows platform, making PrizmDoc Server completely unresponsive.
    • The ms-office-conversion-service now respects the Central Configuration cache.directory property in order to produce temporary files in the configurable directory, based on that property. Previously, the temporary files were produced in the current user's profile directory for temporary files (i.e., %TEMP%), thus making it out of the user's control.
    • The ms-office-conversion-service now cleans up the temporary files that were not deleted due to unexpected restarts or other processing failures related to MS Office documents conversions, thus preventing accumulation the temporary folder with outdated temporary files.
    • Addressed an issue in the prizmdoc-server container, when PDF and Office conversions inefficiently utilized resources allocated to the container. This improves performance and stability of the prizmdoc-server containers in cloud environments, as well as when running multiple containers on the same physical host. PDF and Office conversions in the prizmdoc-server container now respect the following CPU limiting options: Shares, Quota/Period and Affinity.
    • Improved stability of PrizmDoc Application Services when uploading Viewing Package artifacts to S3 storage.
    • Improved stability of PrizmDoc Application Services when using the GET /SaveDocument API.
    • Docker image bases have been updated to Ubuntu 20.04 for accusoft/prizmdoc-server, accusoft/prizmdoc-application-services and accusoft/prizmdoc-viewer-eval. Accusoft recommends docker-based deployment option for deploying to Linux platforms.

    Security

    • Updated the PrizmDoc Load Balancer (PLB) to run on node.js v16, taking advantage of recent security fixes and other changes in the node.js.
    • Replaced vulnerable version of log4j2 with the latest version in all Java components of PrizmDoc server to prevent potential vulnerability with the Log4Shell.
    • Updated the PrizmDoc Application Services (PAS) to run on node.js v16, taking advantage of recent security fixes and other changes in the node.js.
    • Updated the PrizmDoc Public Request Service (PRS) to run on node.js v16, taking advantage of recent security fixes and other changes in the node.js.

    13.19 Documentation Updates

    • Added a new "Adding fonts and packages to the container" section to the Using Docker topic to help you add specific fonts and packages to your PrizmDoc Server Docker container.

    Release Notes v13.18

    Released Sep 2021

    PrizmDoc Viewer v13.18 introduces the following:

    13.18 New Features and Improvements

    • Improved the logic of embedding fonts to burned PDF files. Previously, a whole set of font glyphs was added to a burned PDF file, causing an increase in the file size. This was specifically noticeable when embedding CJK fonts. Now only the glyphs that are actually being used are embedded.
    • PrizmDoc Viewer integration with Accusoft PDF Viewer. PrizmDoc Viewer now provides an API to generate PDF Viewing Packages for all the supported document formats, which allows viewing those documents in the Accusoft PDF Viewer.
    • New sample on GitHub. We've published a new sample of a React app which integrates the Accusoft PDF Viewer with PrizmDoc Viewer.

    13.18 Product Updates and Fixes

    Viewing

    • Added support for email files where the charset name is not enclosed in quotes in the HTML body.
    • Addressed an issue in the PrizmDoc Application Services (PAS): when creating a Viewing Session, PAS could not find a local document without a file extension in the filesystem storage.
    • Addressed an issue with loading EML documents containing multipart attachments with missing Content-Type header.
    • Addressed an issue in the Viewer where long comments were hidden when scrolling to the next page.

    Redaction

    Fidelity

    • Addressed an issue with the LibreOffice-based rendering engine (when viewing or converting Excel documents), which produced an incorrect output with a non-default value (false) of the fileTypes.excel.pagination.enabled central configuration parameter.

    Stability

    • Updated bundled version of Java included in our product to a newer LTS version of AdoptOpenJDK (11.0.11+9), taking advantage of recent stability, performance, security fixes and other changes in the JRE.
    • Addressed an issue in the PrizmDoc Server when the Watchdog service was unable to start the PrizmDoc Load Balancer on slow machines, which caused a PrizmDoc Windows Service startup failure.
    • Addressed an issue in the PrizmDoc Server where the PCCIS module of PrizmDoc Viewer failed with a "System.OutOfMemoryException" error upon using the GET request to download a large source document or a large MarkupBurner result of the Viewing Session.

    Security

    • Addressed potential vulnerabilities in some PrizmDoc Viewer third-party components.
    • Updated MongoDB included in our product to version 4.2.15 on all supported platforms, taking advantage of recent security fixes.
    • Addressed a vulnerability when rendering specific RTF documents in the LibreOffice rendering mode.

    13.18 Documentation Updates

    • Added a new "Is this page helpful?" question to each page, allowing you to send "Yes" or "No" feedback with a single click.
    • Updated Security Guidance with information regarding Macros of Microsoft Office formats and some specific Excel formulas in MSO rendering mode.
    • Updated the Troubleshooting section by adding the new subsection to the Document Viewing Issues > Office Files topic to help you workaround possible issues with viewing MS Office documents containing embedded EMF images.
    • Updated the Windows Installation section by adding a new subsection to the Install on Windows topic to help you resolve possible PrizmDoc Viewer service hang issues when Windows Search is enabled.

    Release Notes v13.17

    Released Jun 2021

    PrizmDoc Viewer v13.17 introduces the following:

    13.17 New Features and Improvements

    • Content Conversion Service (CCS) now allows you to control the conversion of MS Word documents with tracked changes when using the MSO rendering engine. If you are using PrizmDoc Server with Microsoft Office, you can now use the Content Conversion Service to convert Microsoft Word documents with accepted or rejected markup changes to view those documents in their final form. See the API guide for more information.
    • Content Conversion Service (CCS) now allows you to control the conversion of MS PowerPoint documents with speaker notes when using the MSO rendering engine. If you are using PrizmDoc Server with Microsoft Office, you can now use the Content Conversion Service to convert Microsoft PowerPoint documents with slides only (the default) or with included speaker notes to view those documents in their final form. See the API guide for more information.
    • Improved support of signed PDF documents. Improved PDF rendering service to better support signatures in encrypted PDF files.
    • CentOS 6 and Red Hat Enterprise Linux 6 are no longer supported. CentOS 6 support was deprecated in PrizmDoc Viewer v13.14. Red Hat Enterprise Linux 6 support was deprecated in PrizmDoc Viewer v13.16. As of PrizmDoc Viewer v13.17, support for CentOS 6 and Red Hat Enterprise Linux 6 has been dropped. Please consider using PrizmDoc Viewer Docker images (PrizmDoc Server or PrizmDoc Application Services) or upgrading your system.
    • Ubuntu 16.04 is no longer supported. Ubuntu 16.04 support was deprecated in PrizmDoc Viewer v13.16. As of PrizmDoc Viewer v13.17, support for Ubuntu 16.04 has been dropped. Please consider using PrizmDoc Viewer Docker images (PrizmDoc Server or PrizmDoc Application Services) or upgrading your system.
    • While we currently continue to offer and support traditional Linux packages for direct installation on a Linux host, these have largely become obsolete now that Docker deployment is an option. We have announced deprecation of our traditional Linux install packages with the release of PrizmDoc Viewer v13.17, and, in a future product release, we intend to only offer our Docker-based deployment option.
    • PrizmDoc Viewer Microsoft Office conversion add-on option is now compatible with Microsoft Office 2019 for PrizmDoc Server running on Windows 2019 platform. Please refer to the Office Issues section to review and understand Microsoft's known performance issue in Excel 2019 when considering migrating your production environment to Microsoft Office 2019.

    13.17 Product Updates and Fixes

    Viewing

    • Addressed an issue where selecting a search result of another page did not bring the search result into view while in single page view mode.
    • Addressed an issue where the redaction search results were not correctly updated when performing a search that excluded the document.
    • Addressed an issue in the Viewer where the wrong search result was selected when selecting the previous or next search result. This occurred while a search with multiple pages of results was in progress.
    • Addressed an issue in the Viewer where certain regular expression searches failed or caused the browser to hang. The Viewer no longer modifies regular expression searches to handle finding phrases that span multiple lines or contain variations of single or double quote characters. Lines of text in the Viewer typically end with a space followed by a \n character, so to include phrases that span multiple lines in your regular expression search results, you will need to provide a regular expression that accounts for words separated by either a space or a space followed by a \n character.
    • Addressed an issue in the Viewer where searching for terms that include single or double quote characters did not return the correct results.
    • Addressed an issue in the Viewer where using the search filter panel to exclude a search term did not unhighlight the search result in a previously selected comment.
    • Addressed an issue when MSG document that contains an attachment with meeting info in it fails to load.
    • Addressed an issue where the Viewer was preserving search highlights on comments in the saved markup.
    • Addressed an issue where EML document with HTML body charset specified as empty string fails to load.

    Redaction

    • Addressed an issue with the PrizmDoc Server markupBurners REST API where it was unable to apply markup definitions to signed PDF files.

    Fidelity

    • Addressed an issue where embedded cyrillic fonts in PDF documents were not being rendered.
    • Addressed a limitation where JBIG2 streams in some specific PDF documents were not being rendered.

    Stability

    • Addressed an issue in the PrizmDoc Server where the Watchdog service was unable to restart the failed microservices after an internal error.
    • Addressed an issue in the PrizmDoc Server where the PCCIS module of PrizmDoc Viewer failed with a "System.OutOfMemoryException" error when processing large documents that were referenced as a URL when creating a Viewing Session or uploaded with PUT /PCCIS/V1/ViewingSession/u{viewingSessionId}/SourceFile API. Also PrizmDoc Server memory consumption was significantly reduced for scenarios mentioned above.
    • Improved the cache clean-up algorithms of PrizmDoc Server. Previously, it became permanently unhealthy due to cache corruption with the MongoDB.
    • When installing or uninstalling the PrizmDoc Server for Windows, the checkbox Restart Now on the final page of the product's Installer is now checked by default if the installer made modifications that require system reboot, such as registry changes.
    • The Docker image accusoft/prizmdoc-application-services base has been updated to Ubuntu 18.04.
    • Addressed an issue in the PrizmDoc Server where the ms-office-conversion-service might not restore Microsoft Word, Excel, or PowerPoint instances that had crashed.
    • Addressed an issue in the PrizmDoc Server where the child LibreOffice instances of the Office Conversion Service (OCS) might continue to consume CPU resources in the background, while the service is idle. This used to happen after viewing or conversion of large documents rendered by the PrizmDoc Viewer as a plain text.
    • Addressed an issue in the PrizmDoc Server where load balancer may crash when GET /admin request is sent during the product start.
    • Addressed an issue in the PrizmDoc Server with enabled MSO rendering engine where it either failed start, or was not fully functional on the machines with more than 32 CPU cores. On such systems, initialization of a large number of MSO instances either took too much time, or did not complete successfully at all - this is related to the system resources usage by MSO. Now the PrizmDoc Server limits the max CPU core usage to 32 for Office documents processing, so if you need to utilize more CPU power for Office documents viewing and conversion, please consider using PrizmDoc Server Clustering.

    Security

    • Updated bundled version of Java included in our product to a newer version of AdoptOpenJDK (1.8.0_282), taking advantage of recent security fixes and other changes in the JRE.
    • Updated MongoDB included in our product to version 4.2.14 on Windows platform, taking advantage of recent security fixes.
    • Updated the client-side viewer and samples to the latest official version of Underscore (v1.13.1), taking advantage of the most-recent updates and bug fixes.
    • Updated the client-side viewer and samples to the latest official version of jQuery (v3.6.0), taking advantage of recent security fixes.

    Release Notes v13.16

    Released Mar 2021

    PrizmDoc Viewer v13.16 introduces the following:

    13.16 New Features and Improvements

    • Metered Licensing customers can now view their usage. Accusoft Customer Portal now provides the ability for PrizmDoc Viewer Metered Licensing customers to download and view their usage statistics for processed documents.
    • Content Conversion Service (CCS) can now convert XLS and CSV files to XLSX. If you are using PrizmDoc Server with Microsoft Office, you can now use the Content Conversion Service to convert XLS and CSV files to XLSX. See the API guide for more information.
    • Experimental option to render meeting information in email files. PrizmDoc Server has a new experimental central configuration option for rendering meeting information associated with email files: __experimental.fileTypes.email.renderMeetingInfo. See central configuration for more information. NOTE: This feature is a work-in-progress that is not officially supported by Accusoft. Its behavior may change at any time in a future release of the product. We are collecting and reviewing any feedback you can provide about this feature at https://ideas.accusoft.com/ideas/PDV-I-745.

    13.16 Product Updates and Fixes

    Viewing

    • Addressed an issue in the viewer UI where using the text selection tool (for redactions, highlights, and text selection) sometimes selected unintended characters when starting the selection in the middle of a word.
    • Addressed an issue in the viewer UI where clearing a search or executing a new search did not always remove the previously active and selected search result.
    • Addressed an issue where some pages failed to display if a large document was opened in alwaysUseRaster mode and scrolled through.
    • Addressed an issue where text was not available for search or selection when viewing PDF files containing TTF fonts with Kerning sub-table format 2.
    • Addressed an issue in the PrizmDoc Viewer API where using ViewerControl convertToHighlight or convertToRedaction method added marks that did not appear correctly and could not be selected, edited, or reloaded when saved. Additionally, convertToHighlight and convertToRedaction have been deprecated; we recommend you use addMarkFromSearchResult instead.
    • Addressed an issue with the PAS GET /v2/viewingSessions/{viewingSessionId}/restrictions REST API where it would fail with an HTTP 500 error if the viewing session had been created from a viewing package.

    Redaction

    • Addressed an issue with the PrizmDoc Server redactionCreators REST API where it was unable to create redaction definitions for PDF files containing TTF fonts with Kerning sub-table format 2.
    • Addressed an issue with the PrizmDoc Server redactionCreators REST API for Linux and Docker platforms where, if the source document contained certain 4-byte-long UTF-8 characters, the redactions produced after those symbols were incorrectly shifted.
    • Addressed an issue in the PAS MarkupBurner and PrizmDoc Server MarkupBurner REST APIs where they were unable to redact specific PDF files. This happened when the PDF document contained either specific inline images, some complex objects, or an empty content stream in the Contents array of the page.
    • Addressed an issue in the PAS MarkupBurner and PrizmDoc Server MarkupBurner REST APIs where sending a JSON body with a request Content-Type of application/json with an explicitly defined charset (such as Content-Type: application/json; charset=utf-8) would fail.

    Text Extraction

    • Addressed an issue where text could not be extracted from PDF files containing TTF fonts with Kerning sub-table format 2.

    Fidelity

    • Addressed an issue when rendering Excel files with MS Office would sometimes render too few or too many pages.
    • Addressed an issue where some Ink Annotations in PDF documents were not being rendered.
    • Addressed an issue in the Content Conversion Service (CCS) where it produced a text-searchable PDF file with an unexpected page orientation. This happened when the source PDF contained raster images with the Rotate property.
    • Addressed an issue in the Content Conversion Service (CCS) API causing it to incorrectly apply DPI when converting a document to TIFF with G4 compression, resulting in low-quality TIFF output.
    • Addressed an issue in MSG rendering when an email was sent by a software system on behalf of someone else. Previously, PrizmDoc Server would render the "From" field with the name of the software system instead of the name the person who had sent the email for MSG files.
    • Addressed an issue where PrizmDoc Server's automatic format detection incorrectly concluded that a text file was a bitmap when the text file began with "BM" or "BA".
    • Addressed fidelity issues when rendering specific DICOM images.

    Stability

    • Improved the MS Office rendering engine to properly detect a broken MS Office installation. Previously, if the MSO rendering mode was enabled, the ms-office-conversion-service declared itself as "running" on the Admin Page even if it was unable to start any MS Word, Excel, or PowerPoint instances.
    • Improved PrizmDoc Server's MS Office rendering engine to detect and gracefully fail when converting Office documents that require some sort of human interaction via a popup window in the MS Office application. Previously, processing of such documents was preventing other concurrent Office document conversions from successful completion.
    • Addressed an issue with the Office Conversion Service (OCS) intermittently crashing and restarting when using a hybrid of Linux and Windows clusters for MSO rendering and there were processing delays of more than 5 minutes in the Windows cluster.
    • Addressed an issue in PrizmDoc Server on Windows where one of the microservices might have mistakenly terminated other microservices during peak load periods, making PrizmDoc Server unhealthy and unable to restart.
    • Addressed an issue where viewing document pages with widths greater than approximately 3500 mm or 21000 pixels in alwaysUseRaster mode caused the Raster Conversion Service worker threads to hang, occupying up to 10 CPU Cores.
    • Addressed an issue where viewing specific GIF images caused PrizmDoc Server to become unhealthy.
    • Addressed an issue with running multiple Redaction Creators in concurrent requests.
    • Addressed an issue where the PrizmDoc Server "ms-office-conversion-service" log files were not being rotated, resulting in large log files which consumed unnecessary disk space.

    Security

    • Addressed vulnerability when rendering specific JPEG, TIFF, DICOM and SGI images.

    13.16 Documentation Updates

    • Updated the Customizing the Styles topic for accuracy and clarity.
    • Updated the Troubleshooting section by adding the new Memory Consumption Issues topic to help you workaround possible unexpected memory consumption by PrizmDoc Java services.

    Release Notes v13.15

    Released Dec 2020

    PrizmDoc Viewer v13.15 introduces the following:

    13.15 New Features and Improvements

    • New Metered Licensing option. We now offer a new kind of Metered License which allows you to use all of the features of PrizmDoc Viewer without any limits as long as your license is current. At runtime, the number of documents you process is automatically reported back to Accusoft. And renewal is easier than ever: simply pay to extend your license and your PrizmDoc Server instances will automatically detect your new license expiration date (no new license key to adopt or servers to redeploy). See Metered Licensing for more information.
    • Deprecation of Cloud Licensing and Node-Locked Licensing. While currently still supported, with the introduction of Metered Licensing we are announcing the deprecation of the older Cloud License and Node-Locked License types. Support for these kinds of licenses will be removed in a future release. See Licensing for more information.
    • New Angular samples on GitHub. We’ve published two new Angular samples to GitHub, one with a .NET backend and one with a Java backend. These samples are deliberately minimal, designed to give developers a clear, concise example of how to use PrizmDoc Viewer in an Angular context.
    • New markup burning REST API option allows you to only include certain kinds of marks. When making a request to the Markup Burner API, you can now request that only annotations, redactions, and/or signatures be included.
    • PrizmDoc Server no longer changes the default Windows printer at runtime. Previously, when using Microsoft Office for rendering, PrizmDoc Server required the Windows default printer be set to "Microsoft XPS Document Writer" and, if necessary, would forcibly change the default printer to be the "Microsoft XPS Document Writer" at runtime. This is no longer the case. While the "Microsoft XPS Document Writer" printer does still need to be installed and available, it no longer needs to be set as the default printer and PrizmDoc Server will no longer change which printer is set as the default.

    13.15 Product Updates and Fixes

    • Improved logging for the PrizmDoc PDF Processing Service to eliminate redundant warnings about a structure of PDF documents and to help reduce log file size.
    • Addressed an issue with the PrizmDoc Viewer public-request-service failing to communicate with external resources (like image references from emails) over HTTPS and making it impossible to view the associated documents.
    • Addressed an issue in the ViewerControl where it displayed the page loading indicator improperly when used in the standalone (chrome-less) mode (without the PrizmDoc Viewer Client.)
    • Changed the level of opacity when previewing redactions in the PrizmDoc Viewer Client UI. The opacity level in the client viewer UI is now 20%, which is consistent with the default opacity level used for draft (transparent) redactions. Previously the opacity level in the client viewer UI was 50%.
    • Addressed an issue in the PrizmDoc Viewer Plain Tеxt Redactor API where redacting a document failed if the markup JSON contained Text Annotations.
    • Addressed low-performance issue on Windows when converting or viewing PDF documents that extensively use embedded fonts.
    • Addressed an issue with the PrizmDoc Viewer Office Conversion Service (OCS) intermittently failing to start and becoming 'Unhealthy' on the Admin Page in AWS Fargate or other Linux environments with NFS shares when there are network connection delays.
    • Addressed an issue in the PrizmDoc Viewer where opening an EML document attachment failed when Content-Disposition header was not defined for the attachment.
    • Addressed an issue with the PrizmDoc Viewer on Windows where it could leave dangling wkhtmltopdf.exe processes when the service was running or after shutdown.
    • Addressed an issue in the ViewerControl where clicking to dismiss a menu would add a mark if a text-based mark tool is selected. When using the text selection, highlight, strikethrough, hyperlink, or text selection redaction tool, clicking will no longer select text or add a mark. You must now click and drag to select text or add a mark.
    • Addressed an issue in the ViewerControl where using the "begins with" or "ends with" search matching option would not find instances of the search term itself. Instances of the search term are now returned when performing a search using the "begins with" or "ends with" search matching option.
    • Addressed an issue in the Viewer where multiple comment search results were selected in the search results panel when filtering output.
    • Addressed an issue with the PrizmDoc Viewer Workfile Service where it sporadically failed to clean up work files and then bloated logs with error messages on being unable to delete those work files.
    • Addressed an issue in the ViewerControl where marks were added in the wrong location on a page. This happened when the addMarkFromSearchResult method was used (to add marks from search results), the page was out-of-view, and the page was a different size than the first page.
    • Addressed an issue in the ViewerControl where using the scrollTo or scrollToAsync method did not scroll to the correct position.
    • Addressed a vulnerability when rendering specific CSV documents in the LibreOffice rendering mode.
    • Addressed an issue with the restart of the PrizmDoc Redaction Service that was caused by an out of memory exception when the plainTextRedactors functionality was used for source files containing hundreds of pages.
    • Addressed an issue in the PrizmDoc Viewer where its micro service restart was delayed on Windows when WMIC was not responding.
    • Improved the PrizmDoc Service health detection to allow it to become healthy again as soon as a backend service is back up and running. Previously, the PrizmDoc service would become permanently unhealthy when a backend service could not be restarted within 2 minutes.
    • Introduced new parameters in Central Configuration to control Java Virtual Machine (JVM) settings when starting PrizmDoc Server Java-based services (PDF Processing Service and Email Processing Service). For more information, refer to the JVM Options section of the Central Configuration topic.
    • Addressed an issue that allowed "Ends With" and "Begins With" to be selected in the Viewer at the same time.

    13.15 Documentation Updates

    Release Notes v13.14

    Released Sep 2020

    PrizmDoc Viewer v13.14 introduces the following:

    13.14 New Features and Improvements

    • PrizmDoc Viewer Markup Burner API and Viewer Client API now provide the ability to apply draft redactions without actually obscuring the content. The Markup Burner API and the Viewer Client API now allow you to produce PDF documents with transparent (draft mode) redactions that display the document content underneath the redaction rectangles.
    • The Content Conversion Service (CCS) now provides the ability to convert documents to raster TIFF using specific DPI (Dots Per Inch) image resolution. When choosing TIFF as the output file format, the Content Conversion Service (CCS) API now allows you to set a specific resolution on the output raster image - it will have the requested DPI value and will be scaled if necessary.
    • PrizmDoc Redaction Creators API now provides support of multiple redaction reasons. When creating redactions using the Redaction Creators API, you can now specify multiple redaction reasons to be associated with redaction marks.
    • New React sample on GitHub. We’ve published a new React sample on GitHub. This sample is deliberately minimal, designed to give developers a clear, concise example of how to use PrizmDoc Viewer in a React context.
    • The ability to retain configuration settings when upgrading PrizmDoc Server. Starting with version 13.14, when upgrading version 13.3 or higher, PrizmDoc Server is preserving the server side configuration available in prizm-services-config.yml and pcc.config configuration files.
    • The ability to retain configuration settings when upgrading PrizmDoc Application Services. Starting with version 13.14, when upgrading version 13.8 or higher, PrizmDoc Application Services are preserving the configuration available in the pcc.nix.yml and pcc.win.yml configuration files.
    • PrizmDoc Viewer Client API now provides support for opening email attachments in the same viewer window. You can now configure PrizmDoc Viewer Client to open email attachments in the same Viewer window where you are viewing the original email.
    • New PrizmDoc Viewer Client UI for browsing email attachments. PrizmDoc Viewer Client UI has been improved to show email attachments in a compact dropdown menu instead of a panel. Click the paperclip icon to view the list of email attachments and switch between the email and its attachments without shrinking the document viewing area.
    • Improved PrizmDoc Viewer performance when viewing specific CAD-like PDF documents. PrizmDoc Server has been updated to produce more optimal SVG content for CAD-like PDF documents which contain a lot of small consecutive elements having the same style and transform. This optimization leads to much higher browser responsiveness while zooming, panning, and annotation drawing operations in PrizmDoc Viewer. It also slightly improves the thickness level of the rendered lines making it more accurate on display.

    13.14 Product Updates and Fixes

    • Addressed an issue in the PrizmDoc Content Conversion Service (CCS) where footer text was misplaced outside of the visible content in the output document. This happened when it was applied to a rotated PDF page that had the Art, Bleed, or Trim boxes defined partially outside of a Media box.
    • Addressed an issue in the Markup Burner API where garbage text showed in the output PDF instead of Unicode text. This happened when text annotations were applied with different encodings (ASCII and Unicode) from one markup file for the same embedded font.
    • Addressed an issue in the Markup Burner API where redacting a document failed when the source PDF document had a newline character as a delimiter for internal PDF stream elements.
    • Addressed a PrizmDoc limitation of the MSO rendering engine causing inability to process Word 95 binary documents and templates.
    • Addressed an issue in the PrizmDoc Content Conversion Service (CCS) where the diagonal watermark in the output PDF document was not correctly centered compared to the rendered content in the Viewer.
    • Аddressed an issue in the Markup Burner API where redacting a PDF document containing specific JBIG2-encoded images would apply over the raster image as expected; however, when the output was saved to PDF, the redacted part of the image was not removed.
    • Improved PrizmDoc Viewer Client immediate menu customization. You can now configure the maximum height of the menu as described in the How to adjust immediate menu size section for a better user experience when working with multiple redaction reasons.
    • Addressed an issue in the Redaction Creators API where the redactions it created were usually too small to show redaction reasons after the burning. The size of rectangle redactions created by the Redaction Creators API is now consistent with the size of the text selection redaction rectangles created in PrizmDoc Viewer.
    • The Microsoft Office Conversion connectivity for PrizmDoc Servers running on Linux now performs a retry of the failed operation when the connection to the Windows server/cluster fails. This type of situation may happen upon the recycling of Windows servers in a cloud environment.
    • Addressed an issue in the Microsoft Office Conversion connectivity for PrizmDoc Servers running on Linux where the conversion of large Office documents failed if it took more than 2 minutes.
    • Addressed an issue in PrizmDoc Server where the PCCIS module of PrizmDoc Viewer consumed a lot of memory and would be terminated by OOM on the Linux platform. This happened when processing large documents (about 380 MB and more) that were referenced as a URL when creating a Viewing Session.
    • Addressed an issue with rendering specific email (EML) files in the Viewer that contained HTML body content with a character set that was not specified in the Content Type field which prevented users from seeing the content of such email files.
    • Addressed an issue with rendering nested MSG attachments from specific MSG files in the Viewer which prevented users from seeing the attachments. Now when you view the emails, you can also view the attachments.
    • Support for Internet Explorer was deprecated with PrizmDoc Viewer v13.14. Support for Internet Explorer will be removed in PrizmDoc Viewer v14.0 which is currently planned for mid-late 2021.
    • Addressed an issue in the PrizmDoc Viewer Client UI where the Redact Full Pages dialog was displaying an incorrect list of default multiple redaction reasons. The incorrect list was based on a previous list of redaction reasons that were used for the Rectangle Redaction or Text Selection Redaction.
    • Addressed an issue in the PrizmDoc Viewer services (when running on Linux and using a PrizmDoc Cloud License), that caused the services to fail after a system reboot.
    • Addressed an issue in ViewerControl that caused an error during burning request generation and didn't allow users to burn in documents with more than 2000 pages from PrizmDoc Viewer Client UI.
    • Addressed PrizmDoc Viewer WorkFile service crash when uploading about one thousand files concurrently. This update has improved the throughput of the concurrent uploads of small files (up to 50 KB) up to 3 times.
    • Аddressed an issue in the Markup Burner API where redacting a document containing internal PDF stream with text and new line characters would apply over the document as expected; however, when the output was saved to PDF, the text was not removed.
    • Support for CentOS 6 is deprecated with the release of PrizmDoc Viewer v13.14, as CentOS End of Lifetime was announced for November 30th, 2020.
    • Addressed an issue in the Markup Burner API where redacting a document with a text selection redaction over text smaller than 8 points would display the redaction reason text in the viewer as expected; however, when the output was saved to PDF, the redaction reason text was removed.
    • Addressed an issue in the PrizmDoc Viewer Client where search result highlight was not restored after Text Selection Redaction, Highlight Annotation or Strikethrough Annotation was created from the toolbar.
    • Addressed an issue in the PrizmDoc Viewer, where Redaction Creators API did not redact all occurrences of a text string in the document, when the extracted text of the document contains NULL (U+0000) characters.
    • Addressed an issue in the PrizmDoc Viewer Client where a search result message was incorrect, when the search was using wildcards and had incorrect term.

    13.14 Documentation Updates

    • We improved the Troubleshooting section by reorganizing the content and adding a list of technical questions and solutions to help you identify and troubleshoot the most frequently occurring issues.

    Release Notes v13.13

    Released Jun 2020

    PrizmDoc Viewer v13.13 introduces the following:

    13.13 New Features and Improvements

    • PrizmDoc Viewer Client API now provides support of multiple redaction reasons. When creating a redaction in the PrizmDoc Viewer Client UI, you can now apply multiple redaction reasons to be associated with the selected redaction. These reasons will be visible in the Viewer and saved to PDF along with the rest of the redaction content.
    • The Content Conversion Service (CCS) now provides the ability to convert documents to 1-bit raster, 8-bit grayscale, 8-bit indexed or 24-bit RGB raster TIFF. When choosing TIFF as the output file format, the Content Conversion Service (CCS) API now allows you to set a Bitonal, a Grayscale, an Indexed or a RGB color mode to convert every page of a document to a 1-bit raster, a 8-bit grayscale, an 8-bit indexed or a 24-bit RGB raster image.
    • The Content Conversion Service (CCS) now provides the ability to convert documents to raster TIFF using LZW, G4 and JPEG compressions. When choosing TIFF as the output file format, the Content Conversion Service (CCS) API now allows you to set a LZW, a G4 or a JPEG compression type to convert every page of a document to a compressed raster image.
    • Improved PrizmDoc Viewer performance when viewing multi-page DWF documents. In this release we significantly reduced the conversion time and memory consumption when viewing multi-page DWF documents containing 10 or more pages. The performance gain to display such documents is roughly proportional to the number of pages and can be 10 times and more comparing to the previous version of the product.
    • Improved PrizmDoc Viewer performance when retrieving revision data for Document Compare feature which produces hundreds or thousands of differences between original and revised documents.
    • Improved PrizmDoc Viewer client responsiveness when viewing and scrolling through document comparison results containing hundreds or thousands of differences between original and revised documents.

    13.13 Product Updates and Fixes

    • Addressed an issue in the PAS create-tables tool which, when exporting an initialization SQL script for MySQL, produced a .sql file which was not directly executable due to a missing semicolon.
    • Addressed an issue for Linux users using the MSO rendering engine. PrizmDoc Viewer was not able to display a document when it was reloaded and viewed again after the workfile lifetime interval expired (the workfile lifetime interval is specified by the workFiles.lifetime central configuration parameter).
    • Addressed a vulnerability when rendering specific Excel documents in the LibreOffice rendering mode.
    • PrizmDoc Application Services (PAS) Amazon S3 storage provider now supports loading credentials by assuming an Identity and Access Management (IAM) role via an OpenID Connect (OIDC) web identity token file.
    • Addressed an issue where environment variable expansion on Windows was not being applied for the logs.path property in the PrizmDoc Application Services configuration file.
    • Addressed an issue with the LibreOffice-based rendering engine where MS Word documents with paragraphs that had the line spacing rule set to "Auto" and also had contextual spacing enabled, were rendered with incorrect spacing between the lines.
    • Addressed a problem with opening specific MSG files in the Viewer that contained HTML embedded into the RTF body (with RTF encoding, code pages: cp20127 and cp50220).
    • Addressed an issue in the ViewerControl where the CSS style that is defined for the HTML body element caused the Rectangle Redaction, Text Annotation, or Text Signature to be displayed incorrectly.
    • Updated the Client-side Viewer to hide the "Select" and "Cancel" immediate menu items by default when creating an annotation or redaction. To add these menu items back, use the immediateActionMenuActionsFilter property.
    • Addressed incorrect behavior within the PrizmDoc Application Services (PAS) Viewing Session API. Previously, when creating a Viewing Session for a source URL which responds with a non HTTP-200 status code, it was rendering HTTP response body content indicating a successful creation of the viewing session, instead of reporting an error. With this update, the PAS will now return an error, ensuring that a Viewing Session will not be successfully created due to incorrect source URL.
    • Addressed an issue where the environment variable expansion on Windows was not being applied correctly to paths in the PrizmDoc Application Services (PAS) config file when a path contained more than one environment variable.
    • Addressed vulnerability issues when rendering specific ICO, PNG, TIFF, DICOM and XBM images.
    • Addressed a race condition in the PCCIS module of PrizmDoc Viewer. Previously, when the original and revised documents were uploaded for the comparison process in parallel, it caused the Viewing Session to return a CouldNotCompareDocuments error.
    • Addressed an issue where a sequential upload of original and revised source files to the PrizmDoc Application Services (PAS) comparison viewing session was taking longer than expected due to a mutual lock.
    • Updated PrizmDoc Cloud Licensing to properly handle the case when the prizmdoc-server container uses a subset of the host's logical cores, specified via the CPU affinity mask. This allows the use of Cloud Licensing for running multiple prizmdoc-server containers on the same host.
    • Addressed an issue with incorrect, less than expected "Content-Length" header value returned by the PrizmDoc Application Services (PAS) Viewing Session restrictions API.
    • Updated the PrizmDoc Viewer startup logic on the Windows platform to check whether the system's non-interactive heap size corresponds to the CPU core count and report the product as 'Unhealthy' on the Admin Page in case of a discrepancy. This will help you know immediately when there is a server configuration error. Please visit the Registry Changes page for more information.
    • Updated the PrizmDoc Viewer Client UI to automatically adjust the redaction reasons dropdown height, and thus minimize the scrolling through the list of reasons.

    13.13 Documentation Updates

    • Improved ability to view methods in the Viewer API by displaying them in the right-hand navigation panel.

    Release Notes v13.12

    Released Apr 2020

    PrizmDoc Viewer v13.12 introduces the following:

    13.12 New Features and Improvements

    • Improved performance of PrizmDoc Server on servers which have 8 or more cores by adjusting the number of concurrent PDF conversion processes according to the number of available cores.
    • Improved stability of PrizmDoc Server when concurrent requests for viewing complex PDF documents exceed the server capacity by making the PDF Conversion Service skip conversions for timed out requests and survive the excess load, instead of becoming unhealthy and restarting.
    • Content Conversion Service (CCS) now provides the ability to apply diagonal text watermarks when converting to PDF. When choosing PDF as the output file format, the Content Conversion Service (CCS) API now allows you to apply a diagonal text watermark to every page of the output document.
    • New Evaluation Docker Image. The new prizmdoc-viewer-eval Docker image provides a simple PrizmDoc Viewer backend and demo, making it easy to evaluate the product on a single machine.

    13.12 Product Updates and Fixes

    • Addressed an issue in PrizmDoc Viewer that caused an unlicensed version of PrizmDoc Viewer Self-Hosted to suppress notifications about the evaluation limitations in IE browsers.
    • Addressed an issue in ViewerControl that caused an error during markup deserialization when creationDateTime or modificationDateTime field didn't have milliseconds (a second fraction) specified.
    • Addressed an issue in PrizmDoc Viewer where an annotation layer selected for merge was not showing a checkmark indicating it was selected.
    • Addressed an issue in PrizmDoc Viewer which caused an error when merging annotation layers if some of the layers referenced pages that were not yet loaded by the viewer.
    • Addressed an issue in the Markup Burner API where redacting a page with the use of the TextSelectionRedaction mark would apply the redaction mark over the area and remove text as expected but did not remove raster, vector, or hyperlinks content from the burned PDF output of the redacted document.
    • Addressed an issue in the MSO rendering engine with inconsistent rendering of non-trustworthy digital signatures as trustworthy.
    • Addressed an issue in PrizmDoc Viewer services that caused PrizmDoc Viewer to duplicate or skip pages when rendering TIFF images with old-style JPEG encoding and inconsistent StripOffsets and JPEGInterchangeFormat tags.
    • Addressed an issue in PrizmDoc Viewer services that caused PrizmDoc Viewer to render 2-bit LZW-compressed TIFF images incorrectly on Windows.
    • Addressed a vulnerability when rendering specific PNG images.
    • Legacy PHP Sample has been removed from PrizmDoc Viewer.
    • Addressed an issue in PrizmDoc Viewer services where PrizmDoc Viewer could not display untitled attachments in EML files.
    • Addressed a DOM-based XSS vulnerability when viewing documents in PrizmDoc Viewer.
    • Addressed an issue in PrizmDoc Viewer that caused a viewing session watermark text to be rendered 30% larger than the expected font size.
    • Addressed an issue in ViewerControl where the character encoding information of the annotation and redaction markup being saved was not specified as UTF-8 in the Content-Type HTTP header for the HTTP request causing the text to be incorrectly interpreted.
    • Addressed a potential Java workflow vulnerability when viewing or processing specific documents in PrizmDoc Viewer.

    13.12 Documentation Updates

    • The online help has been updated with a new look & feel for easier navigation which includes: a left panel which outlines the section you are viewing, a right-side "mini table of contents" in each topic to help you jump to the information you need instead of scrolling, and a web-inspired search bar at the top.
    • The Getting Started section has been updated so it's easier for you to evaluate PrizmDoc.
    • We've added the Initial Integration section to help you easily transition to the next step after evaluation.
    • The Administrator Guide section has been reorganized so that you can find the information you need quickly and easily. The section contains everything you need to install, license, configure, cluster, and troubleshoot the PrizmDoc Server and PAS backend services.
    • Added a table to the PAS Configuration topic that provides a list of storage entities and supported storage providers.
    • Improved introductory content for the PAS and PrizmDoc Server REST APIs.
    • Updated Security Guidance with recommendations on avoiding the Server Side Request Forgery vulnerability.

    Release Notes v13.11

    Released Feb 2020

    PrizmDoc Viewer v13.11 introduces the following:

    13.11 New Features and Improvements

    • Docker Images - In addition to our traditional installers, we are now making PAS and PrizmDoc Server available as Docker images, making setting up a PrizmDoc Viewer backend dramatically easier. Refer to the following topics for more information:
    • PrizmDoc Server .NET SDK - For .NET developers doing backend document processing with PrizmDoc Server, we now offer an official PrizmDoc Server .NET SDK as an open source NuGet package.
    • Installer Updates - The Prizmdoc Server installer now allows in-place upgrades so that you no longer need to uninstall the previous version of PrizmDoc Server before installing the newest version.
    • Evaluate PrizmDoc Viewer Self-Hosted - You can now evaluate PrizmDoc Viewer Self-Hosted without an installed license, making it easier to evaluate the product. PrizmDoc Server now automatically runs in evaluation mode with a fixed feature set if started without a license. For more information, refer to the Product Evaluation section of the documentation.
    • PrizmDoc Viewer Raster Conversion Service - Improved the PrizmDoc Viewer Raster Conversion Service to significantly reduce memory consumption and reduce the time it takes to generate raster tiles when viewing large image files.

    13.11 Product Updates and Fixes

    • Addressed an issue in the Email Processing Service (EPS) causing PrizmDoc Viewer to fail when converting or viewing MSG files with multiple levels of MSG attachments.
    • Addressed an issue with the PCCViewer.Viewer destroy() method throwing an error when it was used before the ViewerReady event was fired.
    • Re-enabled in-place upgrades for the PrizmDoc Server so that you no longer need to uninstall the previous version of PrizmDoc Server before installing the newest version.
    • Addressed an issue in the Raster Conversion Service (RCS) that caused PrizmDoc Viewer to fail when viewing huge 1-bit image files with more than 500,000,000 pixels.
    • Addressed an issue in the Markup Burner API where applying TextSelectionRedaction markup with redactions located on several pages resulted in applying redaction rectangle mark(s) to the first page only in the PDF output as opposed to applying them to all affected pages as specified in the JSON markup.
    • Support for Windows Server 2008 R2 was deprecated with PrizmDoc Viewer v13.7 and will soon be completely removed. PrizmDoc Viewer v13.11 is the final release to claim deprecated support for Windows Server 2008 R2. Future releases will no longer support Windows Server 2008 R2.
    • Support for Ubuntu 14.04 LTS was deprecated with PrizmDoc Viewer v13.6 and will soon be completely removed. PrizmDoc Viewer v13.11 is the final release to claim deprecated support for Ubuntu 14.04 LTS. Future releases will no longer support Ubuntu 14.04 LTS.
    • Addressed an issue in PrizmDoc Viewer services that caused PrizmDoc Viewer to render tiled TIFF images incorrectly, showing only the top left tile.
    • Addressed an issue in PrizmDoc Viewer services that caused PrizmDoc Viewer to render 2-bit LZW-compressed TIFF images incorrectly on Linux.
    • Addressed a vulnerability when rendering specific malformed TIFF, PNG, BMP and GEM images.

    13.11 Documentation Updates

    Release Notes v13.10

    Released Nov 2019

    PrizmDoc Viewer v13.10 introduces the following:

    13.10 New Features and Improvements

    • New viewing option! Accusoft has built an advanced spreadsheet viewer called PrizmDoc Cells to give PrizmDoc Viewer users the ability to review Excel files exactly as they would appear in the native application. You can analyze formulas, view charts and graphs, view multiple spreadsheets in a single workbook, search for content, and navigate without pagination. For more detailed information on how to integrate PrizmDoc Cells, click here.
    • Improved PrizmDoc Viewer performance when viewing specific CAD-like PDF documents. PrizmDoc Server has been updated to produce more optimal SVG content for specific CAD-like PDF documents so that browsers will not freeze while rendering the content. The improvements lead to much higher browser responsiveness while zooming, panning, and annotation drawing operations in PrizmDoc Viewer. This optimization also slightly improves the thickness level of the rendered lines making it more accurate on display.
    • Updated client-side viewer to the latest version of jQuery. We have updated the client-side viewer and samples to the latest official version of jQuery (v3.4.1), taking advantage of the most-recent updates and bug fixes.

    13.10 Product Updates and Fixes

    • Improved the download filename when a PAS viewing session displayName does not include a file extension. PrizmDoc Viewer will now add an automatically generated extension to the name specified by displayName if it does not include a file extension.
    • Removed potential for broken PAS viewing session when displayName does not include a file extension.
    • Addressed an issue in the Text Extraction service causing PrizmDoc Viewer text search to fail when viewing PDF documents containing Type 3 fonts that specify default font metrics and attributes using a font descriptor dictionary.
    • Updated PrizmDoc Viewer services to predictively generate fully optimized SVG content ("svgb") when viewing files from the Chrome browser, instead of the partially-optimized SVG content ("svga"). This improves responsiveness of the viewer in the Chrome browser and reduces disk space usage on the server.
    • Removed PrizmDoc Server RPM package dependency on urw-fonts containing incorrect configuration which led to inappropriate font substitution for the 'fantasy' font family on CentOS 7.

    13.10 Documentation Updates

    Release Notes v13.9

    Released Sep 2019

    PrizmDoc Viewer v13.9 introduces the following:

    13.9 New Features and Improvements

    • The Viewer now gracefully falls back to non-optimized SVG when web fonts cannot be used. PrizmDoc Viewer makes extensive use of dynamically-generated web fonts to optimize the SVG we send to the browser for viewing. In some customer environments, the browser is forbidden from using web fonts. In these environments, our viewer used to render a document with what looked like “garbled text.” In this update, our viewer now automatically detects if font loading is possible and, if not, gracefully falls back to non-optimized SVG which does not require any fonts.

    • Improved text selection boundaries of OCR conversion output. Previously, when using PrizmDoc Server’s Content Conversion Service REST API to OCR a document and produce a text-searchable PDF, the bounding boxes of detected characters in the output document were sometimes too short. If text-based redactions were later applied to the output PDF document, the selected text may not have been entirely removed. With this update, the Content Conversion Service REST API will now OCR a document and produce bounding boxes of the correct height for detected characters, ensuring that subsequent text-based redactions are properly applied.

    • Updated bundled version of Java, now using AdoptOpenJDK instead of Oracle Java. We have updated the bundled version of Java included in our product from Oracle Java 1.8.0_181 to AdoptOpenJDK 1.8.0_212, taking advantage of the most-recent security fixes in the JRE and moving to a Java runtime whose license will continue to permit unrestricted redistribution. Starting with version 13.9, Oracle Java will no longer be bundled with the product.

    13.9 Product Updates and Fixes

    • Addressed an issue in the Markup Burner API where a required font from the PrizmDoc Viewer installation (that is used by the text markup being burned) could not be found. This font initialization issue was specific to a case where the text markup (with that font) was burning for the first time after the PrizmDoc services initialization and resulted in the Markup Burner API failure.
    • Addressed a PrizmDoc limitation of the MSO rendering engine (when viewing or converting specific Excel documents with a defined PrintArea), which produced an incorrect output with the default value of fileTypes.excel.renderOnlyPrintArea central config parameter.
    • Addressed an issue within the Markup Burner API that was causing an unexpected grey background in the output PDF after burning in a Rectangle Redaction. This could occur when a Filled Rectangle redaction was applied over a PDF page that had a JPEG 2000 compressed image and the transparency area was set by the image mask (i.e a mask entry in the image dictionary).
    • Addressed an issue in PrizmDoc Viewer which caused documents to display unreadable text when web fonts were disabled in the browser or ad blocker settings. PrizmDoc Viewer will now switch to a fallback mode in this case, which allows displaying of the document text correctly. Please note that fallback mode for disabled web fonts will result in slower document rendering and scrolling, so we highly recommend to keep the web fonts enabled in the browser.
    • Addressed an issue in the PrizmDoc Content Conversion Service (CCS) (when performing optical character recognition (OCR) to convert a raster file to a searchable PDF document) that resulted in the height of the recognized text within the PDF output to be smaller and not match its image counterpart.
    • Addressed an issue in the PrizmDoc Content Conversion Service (CCS) (when performing optical character recognition (OCR) to convert a GIF file with an unspecified resolution to a searchable PDF document) that caused the resulting PDF document to have incorrect page rotation.
    • The "hosting options" dialog has been removed from the PrizmDoc Viewer client installer. The PrizmDoc Viewer client installer now always uses a Self-Hosted hosting option. Please see Choosing a Backend Hosting Option for more information about different options for hosting the backend.

    13.9 Documentation Updates

    • Updated the topic to clarify How to Enable Content Encryption in the Viewer.
    • Updated the PAS Configuration topic with additional code examples under the Configuring Storage section.
    • Updated the topic Work with Viewing Packages to clarify the Raster Content and Watermark properties.
    • Updated all of the Linux and Windows installation and Licensing topics to include the following: > NOTE: If you have an updated license, you must restart PAS and PrizmDoc Server in order to use the new license.
    • Updated the Adjust Caching Parameters and Implement Caching Strategies topics with new steps on how to manually delete the cache and the following: > NOTE: If you set the cache to 1 day, the timer will start over if someone accesses a file that is in the cache.
    • Updated the Getting Started topic with the following: > IMPORTANT: Scripts must be loaded in the specified order as shown below.

    Release Notes v13.8

    Released Jun 2019

    PrizmDoc Viewer v13.8 introduces the following enhancements/improvements and product updates/fixes:

    13.8 New Features

    • Support for Windows Server 2019.
    • New Plain Text Redactors API. The PrizmDoc Server REST API has long supported the ability to burn-in redactions to a document, producing a redacted PDF via the markupBurners REST API. In this release, we're introducing a plainTextRedactors REST API which allows you to similarly produce plain text output. In the output plain text, a special <Text Redacted> placeholder is used to denote that one or more characters of the original plain text has been redacted. Any document you've been redacting with the markupBurners API can be similarly redacted to plain text with the new plainTextRedactors API.

    13.8 Product Improvements

    • New and improved getting started guide. We’ve completely re-written our getting started guide, making it easier than ever to understand how PrizmDoc Viewer works and how to integrate it into your web application. Additionally, this new getting started guide explains how you can leverage PrizmDoc Cloud to accelerate your evaluation, avoiding the need to install any of the backend server-side software when all you want to do is evaluate the viewer.
    • New samples on GitHub. We’ve published great new introductory samples for node.js, ASP.NET, and Java / Spring on GitHub. These samples are deliberately minimal, designed to give developers a clear, concise example of how PrizmDoc Viewer actually integrates with a web application.
    • Client-side viewer resources now available on GitHub. The client-side viewer resources are now available outside of the “Client Installer.” If you need to download the pre-built viewer assets (JavaScript, CSS, fonts, etc.), or if you need to make deep customizations to the viewer UI and rebuild it yourself, you can now get all of these resources on GitHub at https://github.com/Accusoft/prizmdoc-viewer.
    • Support for Japanese Reiwa era in OpenDocument file formats as well as in LibreOffice rendering mode.

    13.8 Product Updates and Fixes

    • Addressed an issue where the expected cursor did not display when hovering over marks on pages scrolled into view.
    • Addressed a low-performance issue with reading PDF documents having hexadecimal characters in their dictionary objects.
    • Addressed an issue in the Markup Burner API where redacting a rotated raster image would apply the redaction mark over the image as expected but did not remove the image from the burned PDF output of the redacted document.
    • Addressed an issue in the Markup Burner API with the following criteria: if you were redacting a raster image on a PDF page that had been rotated and cropped for display, the redaction would apply over the raster image as expected; however, when the output was saved to PDF, the redacted part of the image was not removed.
    • Addressed an issue where JSON markup generated by PrizmDoc Server Redaction Creator API could not be loaded into the viewer.

    13.8 Beta Features

    • Added support for controlling the default minimum time for the created viewing package content to remain available via a new property defaults.viewingPackageLifetime in PAS Configuration. This is a beta feature that is not officially supported by Accusoft and its behavior can be changed at any time in a future version of the product.
    • Added Microsoft Azure Blob Storage support for documents, image stamps, markups, form definitions and viewing packages in PAS Configuration. This is a beta feature that is not officially supported by Accusoft and its behavior can be changed at any time in a future version of the product.

    13.8 Documentation Updates

    • PrizmDoc Cloud non-default system configuration values are now documented. The PrizmDoc Server Central Configuration documentation and the PAS Configuration documentation now note whenever Accusoft's PrizmDoc Cloud is using a value which differs from the out-of-box product default and what the custom value is.
    • Removed legacy configuration information from the documentation. All legacy configuration topics that were deprecated in v13.7 have been removed in this release: Format Detection Configuration & Use, Adjust Vector Conversion Settings for Optimal Performance, and Customize Text File Encoding for PrizmDoc Server.

    Release Notes v13.7

    Released Apr 2019

    PrizmDoc Viewer v13.7 introduces the following enhancements/improvements and product updates/fixes:

    13.7 New Features

    13.7 Product Improvements

    • Improved stability of PrizmDoc Server running on Linux by updating Mono run-time components to address possible unresponsiveness of PCC Imaging Services when loading configuration settings on start up.
    • Improved MSO conversion service's resistance to faulty conversion transactions by properly detecting and recycling its child processes without having to become unhealthy and restart the service.
    • Improved the consistency of the burn-in operation performed over the text redaction markup for every character regardless of its glyph outline. Now, a character becomes redacted if the redaction area overlaps the character's glyph area (calculated by width * height) by 40% or more.
    • Improved the cursor behavior in the viewer to better indicate the effect of using the mouse. In supported browsers, the grab cursor is now displayed when the hand pan tool is selected. The cursor no longer changes when hovering over a mark if the selected mouse tool does not support mark selection. If the selected mouse tool does support mark selection, the move cursor is displayed over a mark only if the mark is selectable.

    13.7 Product Updates and Fixes

    • Support for Windows Server 2008 R2 was deprecated with PrizmDoc Viewer v13.7 and will soon be completely removed. Future releases will no longer support Windows Server 2008 R2.
    • Addressed an issue where the installed samples are not able to build when following the provided README steps due to a missing folder and misplaced assets.
    • Addressed the Markup Burner issue with text annotations containing tabulation characters that were incorrectly converted to white square characters in the output PDF upon burn in. We are now replacing tabulations with spaces for more accurate fidelity in the output PDF.
    • Addressed a PrizmDoc limitation of the MSO rendering engine causing the conversion process to fail on any PowerPoint document once the MS PowerPoint instance becomes unresponsive.
    • Addressed issues where PrizmDoc's file downloading APIs failed to download files which had non-ASCII characters in file names, or corrupted filenames during the downloading. These APIs now use an RFC 8187 compliant Content-Disposition header in their responses to correctly support such file names. The only exception is GET /PCCIS/V1/WorkFile/{fileId}, which continues to use the simpler ASCII-only syntax when using an automatically generated file name. This was done to preserve backward compatibility for customers who use PrizmDoc Client package v13.6 or older.
    • Addressed incorrect recycling of MSO conversion service's child processes upon the service's startup, which could result in a faulty conversion transaction making the entire MSO conversion service permanently unhealthy.
    • Addressed incorrect location of text redaction markups over bulleted text in PDF output to match what appears in the viewer.
    • Addressed an issue within the Markup Burner API that resulted in a duplicate Redaction Rectangle mark in the output PDF. This would occur when applying a Filled Rectangle redaction over a PDF with the transparency area defined by the image mask (a mask entry in the image dictionary).
    • Addressed PAS host header translation issue which made it impossible to route requests from PAS to PrizmDoc when there is a load balancer in between them. Now, PAS sets the outgoing request host header to PrizmDoc host value.
    • Corrected an issue in the Viewer UI that allowed the user to select the "Exact Phrase" option which is not a viable option for Wildcard search. The Exact Phrase option is now disabled in Wildcard search.
    • Updated the Search Tasks API to accept a JSON body of up to 1MB in size. Previously, this value was set to 100KB which could have resulted in "Invalid JSON" errors for valid JSON bodies that were greater than this value.
    • Addressed the Markup Burner API issue occurring when attempting to redact text in the text block immediately following a previously redacted block in the source PDF document.
    • Addressed an issue where in some cases the last page of a document was blank when printed.

    13.7 Beta Feature for Evaluation

    • Content Conversion Service (CCS) now provides the ability to convert PDF documents to MS Word (DOCX) documents. This functionality requires the Microsoft Office rendering mode to be enabled by the MSO feature in your license key. This is a beta feature that is not officially supported by Accusoft and its behavior can be changed at any time in a future version of the product, but we are happy to collect and review any feedback you can provide about this feature. The API to convert to MS Word (DOCX) documents should be used for evaluation purposes only and should not be used in production deployments. Note that this API may undergo change prior to feature completion. Please refer to the API guide for the details on how to evaluate this feature.

    13.7 Documentation Updates

    • Legacy PrizmDoc Server configuration documentation has been removed. This legacy documentation information had been marked as deprecated for many releases. We strongly recommend that you use Central Configuration. If you need to access legacy configuration documentation, please refer to an archived version of the documentation.

    Release Notes v13.6

    Released Feb 2019

    PrizmDoc Viewer v13.6 introduces the following enhancements/improvements and product updates/fixes:

    13.6 New Features

    13.6 Product Improvements

    • Smoother page scrolling, faster page loading in the viewer. When a user is scrolling through a document, sometimes pages will delay loading causing the user to wait for the next/previous page to load. With this update, the document scrolling is smooth and next/previous pages load immediately providing a more responsive user experience.
    • Faster conversion with Microsoft Office. We've rewritten the task scheduler inside the Microsoft Office Conversion Service (MSOCS) to improve performance. The Office Conversion Service uses MSOCS to process its requests when the MSO option has been licensed, using multiple workers that execute in parallel. There are now a larger number of workers in both services, keeping more documents resident in memory simultaneously, while continuing to have the same amount of execution concurrency (so as to not overwhelm the server). This means that document reloading by the workers happens less frequently, gaining performance overall.
    • Improved PDF fidelity. Improved PDF rendering service to address multiple PDF rendering fidelity issues including, but not limited to: more accurate color rendering, raster images rendering, text rendering (including math symbols), rendering of linearized PDF documents, as well as rendering of write protected PDF documents.

    13.6 Product Updates and Fixes

    • Addressed an issue where the Document Compare API detected an inaccurate number of differences in documents containing tracked changes that were unaccepted.
    • Addressed a PrizmDoc limitation of the MSO rendering engine causing the conversion process to fail when converting an empty Excel document.
    • Modified PrizmDoc logging so that the header and footer data is not recorded in the log files when appending a specified header or footer to a document's pages with the use of PrizmDoc CCS API for enhanced security and user privacy.
    • Modified PrizmDoc PAS logging so that email attachment names are not recorded in the log files (when rendering email documents with attachments) for enhanced security and user privacy.
    • Addressed PrizmDoc PAS issue that caused the viewingPackageCreator process to hang when uploading the output page artifacts to S3 storage (when the size of the artifacts exceeds 20 MB).
    • Support for Ubuntu 14.04 LTS has been deprecated with PrizmDoc Viewer v13.6 and will soon be completely removed. Future releases will no longer support Ubuntu 14.04 LTS.
    • Addressed an XSS vulnerability when displaying the context menu for Text Hyperlink Annotations in the Viewer. Prior to the fix, a malicious script could be injected into the URL value for a Text Hyperlink Annotation and executed when the context menu for the annotation was displayed. The fix ensures that the URL value is properly escaped during construction of the HTML to display the context menu for the Text Hyperlink Annotation in the Viewer.

    Release Notes v13.5

    Released Jan 2019

    PrizmDoc Viewer v13.5 introduces the following enhancements/improvements and product updates/fixes:

    New Product Name

    • Product name changed from PrizmDoc to PrizmDoc Viewer. To accommodate our growing number of products, we have changed the name of this product to PrizmDoc Viewer. And we are excited to announce that PrizmDoc is now a suite of products which includes PrizmDoc Viewer and the all-new PrizmDoc Editor!

    13.5 Enhancements and Improvements

    • Improved visual rendering of 1-bit raster documents by adding upscale image dimensions preventing quality degradation for files with asymmetric resolutions.
    • Smaller PrizmDoc Server log files. We have updated PrizmDoc Server logging to eliminate redundant information from console logs, decreasing their size.
    • Support for Unicode filenames in email attachments. We have improved support for detecting and listing email attachments with names using RFC 8187 encoding in EML documents to provide broader support of email attachment types.
    • Added rendering support for UTF-7-encoded email. Added rendering for UTF-7-encoded email document content that includes email message body, headers, and attachment names to support a broader range of UTF encoding types.
    • PAS now accepts a full database connection string, allowing Microsoft SQL Server customers the ability to enable database-specific features like connection encryption.
    • Security improvements:
      • Addressed XXE vulnerability in PrizmDoc Server.
      • Upgraded PrizmDoc Server to the latest version of Java 8 to take advantage of security fixes within the JRE.

    13.5 Product Updates and Fixes

    • Addressed the issue with all conversions from raster to searchable PDF that failed at the same time (when reaching a certain number of concurrent conversions).
    • Addressed incorrect rendering of Pantone colors on Windows when viewing PDF documents.
    • Updated PrizmDoc Viewer installation instructions to ensure accuracy (see details below under Documentation Updates section).
    • Addressed an issue where PrizmDoc Viewer was unable to load OpenXML Word documents (containing invalid external file references) when running in LibreOffice rendering mode.
    • Resolved an issue where burning a document using XML annotations for hyperlinks and setting the fillColor value to 0, for black, would instead render the hyperlink in a default blue color.
    • Addressed PrizmDoc Viewer email processing service, PDF processing service and MS Office converter logging to rotate the corresponding log files according to the logging.daysToKeep setting in the Central Config file.
    • Addressed an issue to prevent freehand annotations with a width and height of 0 from being created. Previously, a user could create an empty annotation that would cause errors when the document is downloaded and opened in Adobe Viewer.
    • Updated ViewerControl API double-click handling to use the dblclick event (instead of the mouseup event and a timer) to fix issues with double-click handling in IE v11.

    13.5 Documentation Updates

    The "What's New?" section is now called Release Notes

    • The "What's New" section has been renamed and updated to include sections for Product Updates/Fixes, Enhancements, and Known Issues (formerly located in the Release Notes on the website).

    New Topics

    Updated Topics

    Release Notes v13.4

    Released Sep 2018

    PrizmDoc Viewer v13.4 introduces the following enhancements/improvements and product updates/fixes:

    13.4 Enhancements and Improvements

    • Improved stability under load. Conversion performance has been enhanced to avoid system degradation in high-volume periods.
    • Lower CPU usage when converting large Microsoft Excel documents.
    • Faster conversion of Microsoft Office documents. We've optimized the system to more efficiently convert multiple Office documents, even in periods of heavy usage.

    13.4 Product Updates and Fixes

    • Addressed an issue where vector content was not completely removed when a document was redacted and downloaded as a PDF.
    • Addressed a PDF redaction issue that was causing the grid lines (representing tables in the PDF document) to disappear.
    • Addressed potential XSS vulnerability concerns with the Viewer.
    • Addressed potential XSS vulnerability concerns with the PDF Processing Service.
    • Addressed a PDF redaction issue that was causing parts of a document to disappear.
    • Fixed an issue where images within a document were having redactions applied to them when they should not been applied. This occurred when burning redactions into documents on machines using LibreOffice and when the redaction overlaid any portion of a referenced image (on a page within the document) in which the image was not displayed.
    • Addressed an Office document conversion issue (running in the LibreOffice rendering mode) to disallow non HTTP and HTTPS protocols when rendering content of WEBSERVICE formula in Excel and OpenDocument Spreadsheet documents.
    • Addressed a rendering fidelity issue with semi-transparent PDF elements that were previously rendered opaque.
    • Addressed a rendering fidelity issue with PDF highlighter annotations that were previously rendered opaque.
    • Updated Office Conversion Service to fix a bug related to document affinity that would sometimes prevent a worker from switching to a new document at the correct time. This would result in new tasks waiting until all work on a particular document was completed.
    • Addressed a problem with internal links that were not clickable in a recurring Word document footer when rendered using LibreOffice mode.
    • Addressed a PDF redaction issue that caused parts of a document to disappear.
    • Significantly improved loading time for certain PDF documents created by a third-party recognition server previously resulting in a page load timeout.
    • Addressed a rendering fidelity issue with MSG files (with an RTF body) causing an extra line of text to show up in the message body.
    • Addressed an issue with missing libjpeg dependency in PrizmDoc Viewer RPM package when installing on the CentOS 7 platform.
    • Resolved an issue in handling miter operators used within PDFs that could cause some redactions to fail to burn into the document.
    • Addressed an issue with the rendering of email files that contain Rich Text body (with embedded content) stored in raw binary format.

    Release Notes v13.3

    Released Jun 2018

    PrizmDoc Viewer v13.3 introduces the following new features and product updates/fixes:

    13.3 New Features

    Improved Rendering of Microsoft Excel Documents

    • New Excel rendering and pagination options. PrizmDoc Viewer can now paginate and render Microsoft Excel documents while preserving the original page dimensions and margins specified in the document (matching the Page Layout rendering mode of Microsoft Excel). This is a non-default rendering option, which needs to be explicitly turned on. For more information, refer to the Central Configuration topic.

    Improved Customization and Integration

    • Easier initialization of the viewer. We've updated the Viewer's gulp build to produce a new viewerCustomizations.js that contains the required customization objects to instantiate the viewer control. This allows you to build the customizations once and then easily integrate them into your web app, regardless of the server-side language you're using. Previously, our Viewer samples dynamically built these objects at runtime in the particular server-side language of the sample. This approach made it hard to integrate this code into your own web application (especially if you used a language not covered by our samples).

    Improved Office Conversion Service Task Scheduler Performance for LibreOffice Users

    • Faster conversion of Office documents. We've rewritten the task scheduler inside the Office Conversion Service (OCS) to improve performance. OCS processes its requests with multiple workers that execute in parallel. There are now a larger number of workers, keeping more documents resident in memory simultaneously, while continuing to have the same amount of execution concurrency (so as to not overwhelm the server). This means that document reloading by the workers happens less frequently, gaining performance overall.

    13.3 Product Updates and Fixes

    • Improved redaction area text wrapping. Previously, text that did not fit within the redaction area overflowed beneath the redaction upon downloading (burning). Now, when downloading a document, PrizmDoc Viewer will wrap the text, breaking the line on the space characters in the text and attempting to center the text both vertically and horizontally within the redaction area. When there is a single line of text without spaces, it will truncate the characters of the line of text that do not fit. All of these alterations are designed to mirror the behavior of the redaction reason text seen in the Viewer when viewing, previewing, or printing.
    • Added instructions to the online help for how to customize Excel document view settings to match rendering in PrizmDoc Cloud.
    • Implemented rendering for inline attachments embedded in the binary HTML body of MSG email files.
    • Addressed incorrect rendering of email body content and attachments stored as an encapsulated message with the syntax of "RFC 822 message".
    • Updated the MongoDB service, used internally for text searching, to respect the Central Configuration cache.directory property. The old cache location can be safely deleted to free disk space if desired:
      • The Windows directory is: C:\Prizm\services\mongo-manager-service\bin\mongodb\data
      • The Linux directory is: /usr/share/prizm/services/mongo-manager-service/bin/mongodb/data
    • Improved the cache cleanup algorithm to be more predictable.
    • Modified the MongoDB installation that we use internally to no longer use the default password.
    • Addressed incorrect page count calculation specific to the rendering of Excel files (with the disabled "Show Page Breaks" display option) using MSO rendering mode.
    • Increased concurrency of Office Conversion Service when processing tasks with LibreOffice on Linux platforms. Response times now degrade more gradually after the service load reaches full capacity.

    Release Notes v13.2

    Released Apr 2018

    PrizmDoc Viewer v13.2 introduces the following new features and product updates/fixes:

    13.2 New Features

    Security Features

    • PrizmDoc Server now supports TLS 1.1 and TLS 1.2 for outgoing HTTPS requests.
    • Option to disable rendering of externally-referenced HTML content. We've added a server-side configuration option called, security.htmlRendering.blockExternalContent, to control whether or not externally-referenced HTML content, such as images and iframes, will be blocked. This option affects any source document file type which uses HTML, including HTML, EML, and MSG.

    Full Redaction of Vector Images

    • Support for full redaction of vector images by removing the vector content completely from the output document.

    Online Help

    13.2 Product Updates and Fixes

    Fidelity

    • Improved rendering of email. Corrected fidelity issues when rendering plain text EML files with non UTF-8 encoded message body.
    • Improved rendering of Microsoft Word documents using merge fields. Corrected a fidelity issue when rendering Microsoft Word documents with merge fields used for dynamic document creation. Merge field placeholders would be rendered instead of the actual field data.
    • Improved rendering of Microsoft Word Table of Contents. Addressed a fidelity issue where incorrect rendering of Table of Contents for specific Microsoft Word files with table of contents titles that do not have heading styles applied.
    • Improved rendering of Microsoft Word figure numbers. Corrected a fidelity issue when rendering Microsoft Word documents with figure numbers containing automatic sequence number fields.
    • Improved rendering of Microsoft Word documents with chart links to external files. Corrected a rendering issue where Microsoft Word documents that contain charts with links to external files would not render for viewing or convert to JPEG.

    Performance

    • Faster merging of PDF documents. Improved the performance of merging PDF documents (which now take 4-5 seconds instead of taking up to 10 minutes previously).
    • Faster rendering of Microsoft Word documents under load. Improved the performance of rendering and viewing Microsoft Word documents with fields by 12% over PrizmDoc Viewer v13.1 for high volume runs.
    • Faster restart time. Improved PDF Conversion Service restart logic for faster recycle and restoration of its dependency processes which previously could take as long as 5 minutes.

    Security

    • Corrected a security issue where PrizmDoc Viewer would capture information that could be sensitive when processing emails and record that information in the PrizmDoc Viewer service log files.
    • Addressed a service recovery issue when converting an Excel document with a very high number of rows fails to render (millions of rows).
    • Corrected a conversion issue where certain Microsoft Word and PowerPoint documents that were converted to PDF using PrizmDoc Viewer would generate a PDF that had issues with text (so that selecting, copying and pasting into another document would produce partial text of the original document).

    Release Notes v13.1

    Released Jan 2018

    PrizmDoc Viewer v13.1 introduces the following new features and product updates/fixes:

    13.1 New Features

    Document Compare

    Support for Document Comparison feature for Linux platforms. It is important to note that the Microsoft Word document comparison feature uses the Microsoft Office Conversion (MSO) add-on option for PrizmDoc Server running on Windows and therefore requires the PrizmDoc Server running on Linux to be configured to connect to PrizmDoc Server running on Windows.

    Content Conversion Service

    OCR option to produce text-searchable PDF. Content Conversion Service (CCS) now provides the ability to perform optical character recognition (OCR) to convert a raster file or a scanned PDF file to a searchable PDF document. The resulting PDF document will contain the original image and the recognized text in a separate invisible layer, with each text character position matching its image counterpart. This will allow you to search, select and copy the text in the resulting PDF document.

    The Content Conversion Service's feature that performs OCR (to convert a raster file to a searchable PDF document) does not support CentOS 6 and Red Hat Enterprise Linux 6 platforms.

    Rendering Updates

    Email contents now rendered with time zone information. We've updated the rendering of email to support the Date and Time header fields of MSG and EML files with the corresponding time zone. When an end user views emails in the Viewer, they will be able to see what time zone the date and time correspond to.

    Performance Improvements

    • Faster first page load times in the viewer. First page of content in the Viewer loads faster now, especially for large Microsoft Office documents.
    • Faster page loading when scrolling. Page content loads more quickly when scrolling through a document in the Viewer.
    • Faster retrieval of document text. Eliminated the occasional slowness in getting document text.

    Viewer Touch Experience Improvements

    • Improved touch support. The Viewer touch experience (for phones, tablets, and other touch devices) has been improved with product updates and behavioral improvements.

    PrizmDoc Application Services (PAS)

    • S3 support. PAS now supports Amazon S3 for storage of all of its artifacts, including viewing packages and annotation layers.

    Documentation Updates

    With PrizmDoc Viewer v13.1, the PrizmDoc Cloud documentation has been added to the PrizmDoc Viewer help file. You can see the new updates here:

    13.1 Product Updates and Fixes

    Viewer

    • Improved memory usage. Memory usage in the Viewer has been improved in situations where the same web page is used to view multiple documents without recycling the Viewer.

    PAS

    • Stability improvements. The stability of viewing package creation has been improved due to architectural changes.

    PrizmDoc Server

    • Addressed incorrect rendering of Chinese/Taiwanese date format fields within Excel documents when running through LibreOffice rendering mode.
    • Addressed problem with opening specific Outlook Email attachments with explicitly specified "filename" fields as .msg.
    • Addressed problem with opening specific MSG files (with raw HTML body) in the Viewer.

    Release Notes v13.0

    Released Oct 2017

    PrizmDoc Viewer v13.0 introduces the following new features and product updates/fixes:

    13.0 New Features

    Document Comparison Overview

    Comparison rendering of two Microsoft Word documents. We now provide the ability to compare Microsoft Word documents. You can now create a viewing session using two different Microsoft Word documents as input and see a single comparison view of the two documents in the viewer. This comparison view will note whenever text has been added, changed, or removed, when formatting as changed, etc., just as Microsoft Word does with its "Track Changes" feature. For more information on how to use the new Document Compare functionality, refer to the following topics:

    Image Tools

    Image tools in the viewer. We've added new image tools to the viewer, allowing you to adjust things like contrast or line darkness right in the browser.

    Support for New Operating Systems

    Native SVG Icons

    Native SVG icon support. The Viewer has been upgraded to support native SVG icons. This will simplify replacing default icons with your own versions.

    Online Help

    • The Deployment Licensing section has been updated and clarified to help you understand all of your licensing options. A new topic covers the purpose of the Prizm Licensing Utility.

    13.0 Product Updates and Fixes

    PrizmDoc Server

    • Addressed incorrect page count and rendering content issues with MS Word documents (with track changes turned on when using the MSO rendering engine).
    • Resolved incorrect rendering of AAA/AAAA Excel date/time formatting when using the LibreOffice rendering engine.

    Known Issues

    The following items are either currently under investigation by the Accusoft Engineering organization or provide further information regarding PrizmDoc Viewer. Should you require an updated status on any of these items, please contact Accusoft Customer Support.

    Installation and Upgrade Issues

    • Currently, the path PrizmDoc Viewer is installed to on Windows cannot exceed 64 characters total. Longer paths are rejected by the installer.
    • When installing PrizmDoc Viewer on Windows, the account used to start the PrizmDoc service requires a password to be defined. Without a password, the Installer will not be able to proceed.
    • The ext3 file system limits the number of subdirectories within a single directory to 31,998. Because PrizmDoc Viewer creates a directory for each viewing session and/or work file, systems with high traffic combined with longer cache expiration periods may encounter request errors with ERROR_GEN_FAILURE. The ext3 file system is the default for ephemeral drives in AWS as well as many current Linux distributions. Consequently, the possibility of encountering this error exists for these systems unless ext4 was specifically chosen at installation. To enable directories containing greater than the 32K subdirectory limit, ext4 turns on HTree indexes (a specialized version of a B-tree) by default. For systems with extraordinarily high traffic coupled with extended cache expiration times, it is recommended the product be installed on Linux systems with ext4 file systems or at least to configure caching to utilize an ext4 file system.
    • If you are using Viewing Packages in PAS and upgrading from a version prior to 13.1, you must upgrade your database schema. See the setting up your database topic for more information on how to upgrade PAS.
    • PrizmDoc Viewer relies on a fontconfig package that is not shipped with the product and that might be missing from some distributions of Ubuntu. This was resolved by adding automated checks in the PrizmDoc Viewer 13.0 installation scripts. As a workaround for older versions of PrizmDoc Viewer, we recommend installing the fontconfig package manually by using sudo apt-get install fontconfig.
    • You must have Microsoft Office installed on the server when using the Microsoft Office Conversion license. If you do not have Microsoft Office installed, you will get an error when converting Office documents or displaying them in the Viewer.
    • Always keep your PrizmDoc Viewer Windows server updated with the latest Windows and MS Office updates when using the MSO rendering option. If you don't have the latest updates, you may see rendering issues.
    • When updating to the latest PrizmDoc Server Docker image, existing markup files for CSV and HTML files may require updating.

    Browser-Specific Issues

    • The Safari browser may truncate long file names (containing 96 or more ASCII characters or 27 or more non-ASCII characters) when downloading source files from PrizmDoc viewer. This issue only exists when connecting to PrizmDoc Viewer via http/1.1 protocol (with or without https). Setting up an http/2 proxy in front of the PrizmDoc Viewer avoids the issue. See also the Safari bug report 47914517.
    • If you print large documents (100+ pages) from the Viewer, you may run into browser memory constraints that are beyond our control. We recommend downloading large documents as a PDF and then printing them.
    • When printing a document in Firefox or Safari, embedded images may be truncated or missing in some cases.
    • When printing a document in Chrome, images may be printed with a black background when the background should be transparent.
    • When printing documents with the Viewer in the Safari browser on Windows, blank pages are sometimes created, causing extra pages in the document.

    PDF Issues

    • Border of graphic elements may not render for specific PDF documents.
    • Starting from 13.6 release some PDF documents may render slightly shifted compared to previous versions of PrizmDoc.
    • Raster images using CMYK colorspace in PDF documents may be rendered with a slightly different color.
    • Currently, burning markup into PDF documents that contain XFA fields is unsupported. Attempting to burn markup into XFA documents will result in an error.
    • Burning markup into PDF documents that contain AcroForm fields is unsupported. Attempting to burn AcroForm fields in PDF documents will leave those field values unchanged.
    • PDF files with embedded raster images in the Indexed color space using CMYK palette might not display with the correct colors in the Viewing Client after conversion to SVG.
    • Search results returned in the Viewing Client for PDF documents may not be highlighted in cases where the PDF contains image over text results. In this case, content will be returned in the Search results tab, but the highlighted search terms will not be displayed in the page view when navigating to the appropriate page. In this case, a message will be displayed indicating that the page does not support text highlighting. This will be improved in future versions.
    • PrizmDoc Viewer does not currently perform text extraction and search on AcroForm fields content in PDF documents.
    • PrizmDoc Viewer does not currently perform text-extraction and search on PDF annotation markup within PDF documents. This annotation markup includes Text and FreeText annotation types as specified in the PDF reference guide.
    • PrizmDoc Viewer does not support rendering of Dynamic XFA (XML Forms Architecture) PDF forms. Instead of the actual forms content, a message such as "Please wait... If this message is not eventually replaced by the proper contents of the document, your PDF viewer may not be able to display this type of document." will be displayed.
    • If the PDFPS microservice exceeds the open file descriptor limit, then PrizmDoc Viewer will restart the PDFPS microservice.
    • CMYK JPEG files may display as negative (or inverted) color images when converted to PDF or raster formats.
    • Deep Image Redaction may not work as expected when the source PDF contains G32D encoded TIFF images. PDF redactions will be created but any content on the G32D encoded content intended for redaction will not be obscured by black pixel data.
    • When converting HTML to PDF using the Content Conversion Service, you may see poor performance on larger documents when pdfOptions.forceOneFilePerPage is set to true. We recommend that pdfOptions.forceOneFilePerPage be set to false when converting more than 20 pages.

    CAD Issues

    • Embedded OLE objects in CAD files are not currently supported for rendering.
    • Watermarking is not supported for CAD files.
    • The MarkupBurner, responsible for the underlying process of annotation, redaction and e-signature, does not currently support CAD based files.

    Office Issues

    • Excel files may take more time to process when using PrizmDoc Server with Microsoft Office version 2019. Increased processing time depends on the source document content, and may take up to three times longer compared to Excel conversion performed by Microsoft Office version 2016.
    • The Document Compare feature is designed for comparing different revisions originating from the same MS Word document. Using this feature for comparing any random MS Word documents is not recommended and may lead to unexpected results.
    • Internal hyperlinks (like TOC bookmarks) within Office documents are not clickable when rendered or converted to PDF with the use of the Microsoft Office Conversion option.
    • The following TTC font packages on Ubuntu might conflict with Tunga and Latha font substitution implemented in PrizmDoc Viewer Office Converter causing inaccurate rendering. You may need to uninstall those packages for better font substitution fidelity:
      • fonts-gubbi
      • ttf-indic-fonts-core
      • ttf-bengali-fonts
      • ttf-devanagari-fonts
      • ttf-gujarati-fonts
      • ttf-kannada-fonts
      • ttf-malayalam-fonts
      • ttf-oriya-fonts
      • ttf-punjabi-fonts
      • ttf-tamil-fonts
      • ttf-telugu-fonts
    • Excel worksheets will now be rendered with grid lines, headers, footers, and hidden content visible by default causing the existing annotation and redaction markup to be not aligned with the old rendering content. Customers wanting to redact the new output would need to re-create annotation and redaction markups. Customers wanting to go back to the old style rendering can do that by changing the Excel rendering properties available in the central configuration file.
    • Excel pagination causes the Office converter to generate more pages. This puts more stress on the server and may cause the conversion to timeout.
    • Certain VML shapes from Word documents might not render properly to the client viewers.

    E-Signature Issues

    • With some browsers the E-Signature Viewing Client performance could be impacted when the number of fields in the document is more than 1,300.
    • There are known issues with E-Signing when working on iOS devices that can cause the screen to move erratically when moving from field to field. This does not keep the experience from being usable, but it can be disconcerting while using.

    Annotation and Redaction Issues

    • There are known limitations when trying to use Quick Actions to redact large search result sets that are over hundreds of results.
    • Note that we have modified the default properties for several marks when creating them via the API. As long as you are setting mark properties, such as color and line width, this should not affect your code.
    • There are limitations to using the Full-Page Redactions mouse tool on a mobile device.
    • The Email Conversion Service fidelity improvements will result in shifting the rendered content (due to inline image rendering support for more accurate rendering of the HTML body) and making the existing annotation and redaction markup not aligned with the old rendered content. Customers wanting to redact the new output would need to re-create the annotation and redaction markups.
    • There are some special symbols, such as '@', that cannot be properly processed in auto-redaction.
    • Burned annotations are shifted on PDF files with a non-standard PDF page Cropbox or Mediabox boxes.

    Miscellaneous Viewer Issues

    • When requesting raster page content, a '500' error may appear in the network log. As part of performance improvements, the Viewer now always requests SVG content first. If SVG content is not available, a '500' error will be communicated and the client will then request raster content. Although this does not negatively affect the Viewer behavior, this will be changed in a future release to handle the request differently.
    • It is recommended that the document cache be cleared prior to upgrading PrizmDoc Viewer. Failure to clear the cache will result in the inability to search documents in the Viewer that have been cached in prior versions of PrizmDoc Viewer.
    • PrizmDoc Viewer does not currently provide full support for searching and extracting text which reads from right-to-left (like Arabic and Hebrew), which may cause the search and text extraction results to display such text in wrong left-to-right direction.
    • If you print large documents (100+ pages) from the Viewer, you may run into browser memory constraints that are beyond our control. We recommend downloading large documents as a PDF and then printing them.
    • Watermarks appear in bold and are not transparent.
    • You may see space between the image tiles (when rendering raster images that are broken into tiles, within HTML tables) if those tables do not fit into one page.
    • HTML conversion on Windows depends on system DPI. Higher system DPI results in smaller HTML content size when viewing HTML.
    • When printing from within the Viewer using a browser other than Google Chrome, you will have to manually set the page orientation and page size in the system's Print Dialog, even when a particular orientation and/or page size has been set within the Viewer.

    Content Conversion Service Issues

    • The Content Conversion Service does not support conversion of DICOM files to a searchable PDF document format.
    • The Content Conversion Service does not support conversion of transparent TIFF files to a searchable PDF document format.
    • The Content Conversion Service's feature that performs optical character recognition (OCR) to convert a raster file to a searchable PDF document does not support CentOS 6 and Red Hat Enterprise Linux 6 platforms.

    Miscellaneous Server Issues

    • POST /v2/viewingPackageCreators may fail if the request body properties input.source.fileName, input.source.displayName, or input.source.downloadName use unicode characters outside of the Unicode Basic Multilingual Plane.
    • All Server-Side configurations should be identical across all the servers in a cluster because documents are frequently sent to a random server for processing. This also applies for the multi-server Microsoft Office Conversion configuration that now supports connectivity to PrizmDoc Servers running on Linux.
    • The POST /v2/searchContexts/{contextId}/completed URL should not have a body. Though the request may still succeed if a body is used in the v12.2 release, it can begin to fail in the future. Using any body with this request should be avoided.
    • The PrizmDoc Server "GET Page" call requesting a JPEG thumbnail for a certain page hosted on Windows might return HTTP error 500.
    • Abandoned Viewing Sessions - The text extraction process will continue if a user abandons a session before the text extraction process completes. This will be improved in a future release, but it is important to know that a user abandoning a document does not necessarily release conversion resources on the server.