UiPath Release Notes

UiPath Release Notes

2019 FastTrack (2019.4.2)

Release date: 24 April 2019

After a series of monthly rollouts validated by our brave Community, the hard work of our dev and product team has condensed into what we call 2019 Fast Track - the first UiPath Platform Enterprise release of this year.

Some of our components have seen major overhauls, while others were optimized and fine-tuned, making sure we bring top-of-the-line technology at your fingertips. For those of you that are curious about all the changes this release brings, we invite you to delve deep into the page below and feel free to click the documentation links for more information.

What’s New

AI Computer Vision

We proudly present the first stable version of the Computer Vision activities, which implement our UiPath in-house developed neural network into standard UI automation activities such as CV Click or CV Type Into, outfitting them with an AI that emulates the human visual system, capable of recognizing UI elements and mimicking human actions. For more information on the new activities, you can find the docs here.

Our proprietary Machine Learning model is packed for on-prem Enterprise-ready deployments, while our cloud server remains free for our Community users, furthering our RPA democratization efforts. For more information on deploying the on-prem server, you can find the docs here.

Extensibility and Performance

For an enhanced all-around sync of activities and packages in multi-node environments, package metadata is now saved to your SQL database. On top of that, instead of the ordinary File System, you can set up your Orchestrator instance to store all these artifacts in one of the following blob storage solutions: Azure, Amazon, or Minio. More information on how you can set this up is available here. Lastly, rest assured - all setup artifacts can help you easily set this up, be it for a clean installation or an update.


Package integrity and authenticity are vital in terms of security, which is why we’re introducing Package Signing for signing libraries and processes with your own certificate. The feature is available both from Studio’s Publish window and command-line. Learn how to sign packages when publishing processes here.

Furthermore, in support of this feature, you should know it is possible to enforce your Robots to execute only signed packages at install-time, by using the newly added ENFORCE_SIGNED_EXECUTION parameter. More information about this is available here.

In addition, to further enhance the security of the Robot - Orchestrator communication protocol we added the possibility for the Robot to authenticate through a custom HTTP header. In this regard the web.config parameter, Scalability.SignalR.AuthenticationEnabled, is now at your disposal. Please note that if you enable this feature, also referred to as Robot SignalR authentication, you are no longer able to execute jobs on Robots with a version lower than 2018.4.3. More information is available on this page.

Bringing an even better segregation of sensitive information and an enhanced level of security between tenants, you can now use a unique encryption key per tenant, using Microsoft Azure's Key Vault. This has been made possible by adding a few more parameters to the already-popular web.config file. You can view more information on how you can configure these settings on this page.

In Orchestrator v2018.3 we introduced libraries to help you share reusable components at application level. In this release, we've turned our attention to segregating libraries at tenant level to ensure a more secure approach to your data. This means that in addition to the shared host feed, you may configure your libraries to use their own tenant feed.

Last but not least, the system administrator is now able to upload a library which gets propagated to all tenants which use the host feed. How cool is that? If you're interested in reading more about this feature and how to use it explicitly, visit our documentation.


If you always wondered what your Unattended Robot is up to and what exactly is going on with your unsuccessful jobs, you came to the right place. This release ships with a recording functionality that helps you troubleshoot faulted jobs. Simply configure your processes to allow recording and, whenever the corresponding jobs fail, the media files illustrating the last moments of a job are at your disposal. Keep in mind that this functionality comes with a new permissions set. Click here for more information.

Automation processes developed in Studio now come with the option to prevent them from being paused. This means that when the process is running, the Pause button in the Robot Tray is disabled. This feature is useful when pausing an asynchronous execution might cause it to break. Find out how to use this option documentation.


As you're probably used to by now, month after month we draw closer to our final goal of creating the ultimate document processing platform. Alongside the first enterprise release of this year, the IntelligentOCR activities pack has been imbued with some new activities, as follows:

The Taxonomy Manager is the next piece of the document processing puzzle, a wizard created to help you build custom taxonomy files which can then be reused across processes.

We have developed the Load Taxonomy activity, which grants you the ability to load the taxonomy created with the aid of the Taxonomy Manager wizard into a variable which can then be passed on to other activities.

The DegreeOfParallelism property has been added to the Digitize Document activity, enabling you to perform OCR analysis on multiple pages simultaneously. This is not a breaking change, so old workflows still function properly after updating to the latest version of the pack.

The Digitize Document activity has been improved performance-wise with some backend changes.

The IntelligentOCR pack is now upgraded to .NET Framework v4.6.1.

The MatchingDocumentDefinition property of the FCDocument variable has been exposed. Assigning it to a variable generates the same result as a Classify Document activity.

Document Processing Contracts

Our first iteration of the Document Processing Contracts pack enables you to integrate your own data extraction and classifier activities with the IntelligentOCR activities. This means that you can easily match our data extraction interface with your own custom-designed processes by signing them with the data contracts provided in this pack.


We have developed a RegEx Builder wizard which gives you the ability to harness the power of Regular Expressions with a simple and easy to use UI. This wizard can be accessed from the body of the IsMatch, Matches, and Replace activities.

Our goal is to simplify the way you interact with Orchestrator, so we've implemented the possibility of adding multiple queue items into a new activity, Bulk Add Queue Items.

Optimizing the way queue items are processed, the new Wait Queue Item can notify Orchestrator it is ready to receive a queue item, and then wait for Orchestrator to add an item to that queue, eliminating the need for multiple retries.


Variables can now be used when editing a selector in the Selector Editor, so now you can easily store parts of selectors in variables and reuse them without a hassle.

The newly developed Microsoft Azure Computer Vision OCR engine uses the new API from Microsoft, enabling you to keep up with the evolution of technology.

We've added the ability to automatically downsize images to the Google Cloud Vision OCR activity, adding a new property called ResizeToMaxLimitIfNeccesary. This has been done to make sure the size of your images does not break the maximum limit of the engine.

The Abbyy OCR activity can now be configured to use Abbyy FineReader Engine 12 via the FineReaderVersion drop-down property. Also, the PredefinedRecognitionProfile property can be used to use one of the official Abbyy recognition profiles and the CustomRecognitionProfilePath property can be used to load in your own recognition profile into the project.

The new Context Aware Anchor has been created by building upon the old Anchor Base activity. To increase precision, it now takes into account the pairs of anchors and targets, without requiring an anchor to be unique.

Edge (Experimental)

The v19.4.0 UIAutomation package now makes it possible to create automations in the Edge browser! Read all about it here! Oh, since this is an experimental feature, you might also want to check out the troubleshooting guide.


There’s always something more to automate! With this release, we make it possible to natively generate selectors for applications which rely on the JxBrowser library. No extensions required! There are, however, a few limitations, about which you can read on thispage.

RDP Extension

Creating automation projects over RDP connections just got a lot easier. With the help of the Windows Remote Desktop extension, selectors are natively generated. Find out how all of this is possible on this page.


We're proud to say we have added Lotus Notes to the list of email clients we support! This means we've developed four new activities which can handle your inbox with ease. These activities are:

Also, we understood the need to be able to reply to a mail in Outlook, so we have created the Reply To Outlook Mail Message activity, which helps you do just this.


This major release brings five new activities to the PDF activities pack, helping you manipulate more aspects of PDF documents, as follows:

The change to v2.0.0 is also due to the fact that the engine of the PDF activities has been overhauled, and now uses a more efficient method of retrieving text from .pdf and.xps files. This change is reflected in the old activities of this pack, as well, meaning that the Read PDF With OCR and Read XPS With OCR activities have some new properties, as well as some upgraded old ones.

Orchestrator Mobile Apps

This month the Orchestrator Mobile app also received a pretty important update that you might want to check out - you can now start jobs directly from your Android or iOS device. Other improvements have also been made available, about which you can read here.


The Taxonomy Manager displays the Document Type ID of the document type that is being edited.

The Reference and FilterStrategy properties have been added to the Get Transaction Item activity, enabling you to filter the Transaction Item you want to retrieve by its Reference.

The Orchestrator HTTP Request activity has been improved, and is now able to retrieve the headers of the requests it performs.

The Google OCR, Google Cloud OCR, and Microsoft Cloud OCR engine activities have been renamed to Tesseract OCR, Google Cloud Vision OCR, and Microsoft Project Oxford Online OCR, respectively. Don’t worry, as this does not cause a breaking change to your already existing workflows!

The Extract Structured Data has two new properties that enable you to choose what kind of click you want to use when changing pages if the data you want to extract spans multiple pages.

The HTTP Request and Soap Request activities have been improved, and are now able to retrieve the headers of the requests they perform.

The IBM Watson NLU Text Analysis activity has a new property field, Key, which enables compatibility with Identity and Access Management authentication.


You’ll be happy to hear that you can now create browser automations with native selectors for web applications exposed through Citrix Virtual Apps and Desktops.

Since we’re constantly working on new updates for you, we made it possible for the UiPath Remote Runtime component to support multiple versions.

The UIA_IsEnabledPropertyId attribute is now exposed in selectors. This enhances UI element detection in applications which rely on the UIAutomation framework.

User events monitoring faced an overhaul to address different issues when detecting trigger events, while also improving performance.

To ensure backward compatibility, the UiPath Remote Runtime component is compatible with the 18.4.5 and 19.4.1 UIAutomation versions.


For those that work offline, we have some good news: you can disable the default online feeds for Robot during the deployment process. This facilitates a seamless installation process even in restrictive environments without Internet access, from the command line. For more information, access this page.

You can now configure the download location of your automation processes and their dependencies, for Studio and Robot, through the command line of the UiPathStudio.msi installer. This feature provides a customizable architecture to suit both secure deployments and hotseat scenarios. To read more about the newly added PACKAGES_FOLDER parameter and how to use it in your mass-deployment of Robots, see this page.

Moreover, you can add your own activity feeds at install-time by using UiPathStudio.msi's command line. More information about the CUSTOM_NUGET_FEEDS argument and how to use it to set up your environment can be found here.

In our quest for continuous improvement, we collect usage and performance data to help diagnose problems and enhance our products. This feature now also comes with the possibility of turning off telemetry at install-time through the TELEMETRY_ENABLED command line parameter or later on using the dedicated configuration keys. More information is available here for UiPathStudio.msi and here for UiPathOrchestrator.msi.

To help you get a better handle on what happened during the setup process, UiPathStudio.msi now also logs licensing information (regutil.exe) while it is running, at a custom location.

As security has always been one of our biggest concerns, we have now added an extra layer. You can specify the tenant and host admin passwords directly when you install your Orchestrator instance, from the command line or the setup wizard of the UiPathOrchestrator.msi installer. If you want to find out more on this topic, access this pageor this page.

To aid you in silently updating your Orchestrator instance and using the previous database, the APP_ENCRYPTION_KEY parameter of the UiPathOrchestrator.msi installer has been enhanced. More details are available in the guide.

To ensure an even better integration between Azure and Orchestrator, it is now possible to include the local activity feed in your Azure environment directly at install-time, using the newly added activitiesPackagePath parameter. To support this feature, you should also know that we added an archive,, with all the activities in To view more information on how to use this parameter, see this page.

Furthermore, to help you control where to store Robot logs in Azure deployments, a new parameter, -robotsElasticSearchTargets, had been added to the Publish-Orchestrator.ps1 artifact. More information is available here.

UiPathPlatformInstaller.exe now also verifies if you have IIS 7.5+ and URL Rewrite installed on the machine, enabling a smooth set up process.

Bug Fixes


  • Uninstalling Orchestrator now removes the entire contents of the Activities folder, even if it contains custom files.
  • To ensure a smooth experience in offline environments, the list of activity dependencies has been updated in all our installers. Please note that as a result, the size of the artifacts has increased.
  • An error message was displayed if you closed the UiPathPlatformInstaller.exe wizard while the installation process was being executed.
  • Fixes an issue that prevented UiPathOrchestrator.msi from preserving your custom location, website and application pool names. Starting with v2019 Fast Track, all future updates of Orchestrator keep your previous custom values.

Release Notes Per Product

Click below for component release notes and documentation.

Activities Versions

The following activity packages and versions are included in the installers and are added as dependencies, by default, to a newly created project in Studio:

  • UiPath.Excel.Activities - v2.5.3
  • UiPath.Mail.Activities - v1.4.0
  • UiPath.System.Activities - v19.4.0
  • UiPath.UIAutomation.Activities - v19.4.1

Updated about a year ago

2019 FastTrack (2019.4.2)

Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.