Back to home

What's new in 2018.1

Connect 2018.1 new features summary

Introduction

Connect 2018.1 represents a major change in the way we release our products. “.1” represents the first
release in 2018, and the next one will be “.2” in October. Planning releases on a fixed interval has been
underway and has already shown significant benefits for our users. Predictability in the schedule means
everyone can plan for tasks related to the releases.

Here are some major exciting new functionalities covered in this article.

Environment

Connect 2018.1 is the first version to officially support the Windows 2016 Operating Systems. We have been
running the application on our own production servers for the past few months and we are pleased to report
both systems work very well together.

Welcome screen

There were some concerns with the welcome screen previously, so after checking with our customers we came up with a new one. There are some more improvements on the wish list, but we think this will improve the customer experience.
Welcome Screen
Key improvements:

  • Combined all key entrypoints for the different touchpoints in a single menu on the left
  • Changed the previous split between datamapper and designer into a usage split between recent files and creating something new
  • Clarified some names
  • Ensured no critical links fall off the screen

Designer

Revamped Business Graphics

The Business Graphics feature is greatly enhanced and now shows a preview of the data and graph as part of the
insert Pie, Line and Bar chart wizards. This simplifies selecting data and lets you set and preview graph options
upon inserting the graph object. The same functionality is added to the scripting wizard shown when double
clicking a graph script in the Scripts panel.

New Charts

The Properties dialog in Connect lets you configure the graph via the User Interface and stores the configuration following the specifications of this library. A Source tab is introduced to the Graph Properties dialog so that you can view and edit the JSON configuration. The implementation now supports the full Line, Bar and Pie chart API of the underlaying amChart library. It allows you to enable amCharts graph features not available via the user interface of Connect.

One can even copy and paste Line, Bar and Pie chart configuration settings from the online amChart Live Editor directly to this tab. Refer to the amChart documentation for more information on the numerous  configuration options.

More information and samples are found in the Business Graphics revamped and Programmatically configure a Chart object articles.

Email attachments simplified

New features are introduced to create dynamic and static email attachments via the user interface of the Designer. Right mouse click an email section and choose Attachments… to view the new Attachments tab. This tab shows the static file attachments for the selected email section. Click the plus icon to add new attachments from your template resources, file system or a remote location. Alternatively drag and drop a file from the file system or from the Images folder to the icon of an email section in the Resources view.
Email

In addition a new Script wizard is introduced allowing you to construct a path to a file using a Text Script wizard like interface.  Use the prefix, field and suffix fields to dynamically select a file from disk or a remote location.

More information is found in the Email Attachments simplified article.

Custom Patterns in the Text Script wizard

The Format option for data fields  in the Text Script wizard now contains an entry called Custom Pattern. This option invokes a dialog allowing the user to enter a custom formatting pattern for Currencies, Dates and Floats data types.  The Value fields shows the value of the current record, to test alternatives one could enter a custom value. The result  of the pattern is shown in the Example field.

The following images show a custom pattern for a currency value. The pattern has a subpattern for negative values.

Patterns

Documentation for these patterns can be found on the following pages:

Updated internal browser component

The browser component used for the WYIWYG editor in the Designer is updated. This allows you to use new CSS properties like flexbox. Connect 2018.1 is using Gecko 38 as its HTML rendering engine.

Other Designer enhancements

  • Styles view This panel is now capable of showing style information for dynamically added content in Preview mode.
  • Data Model view Fields and tables can be added, removed, renamed in the Designer perspective. This is done via the contextual menu of the Data Model view.
  • Resource() scripting command Parameters to check PDF permissions are added.<
    var pdf = resource("images/stamp.pdf");
    var printing = pdf.permissions.printingAllowed; //return none, highres, lowres, unknown
    var pass = pdf.permissions.hasPassword; //return true or false
  • Template wizards remember the last used settings.
  • Reset the size of an image by clicking the Reset Image Size icon in the Geometry section of the Attributes panel.
  • Options are added to rotate Section Background images using predefined angles (0,90,180,270).
  • Align absolute positioned objects (top, bottom, middle, left, right, center) via the Format menu or use the icons on the toolbar.
  • Support for Remote Section Backgrounds and Stationery images.
  • Images in the Styleheet folder The Stylesheets folder in the Resources panel now accepts images files. This simplifies using images in your styles as they can be referenced relatively to the CSS stylesheet file.

Background

DataMapper

Standalone engine

Ever since the first version of Connect, the DataMapper has been tucked inside the Server. It has now been pulled out of there to form a new, standalone engine much like the Content Creation and Output Engines. Taking it out of the Server improves the robustness of the system: previously, if any major issue occurred in the DataMapper, it could jeopardize the entire server and all its pending jobs. Now, with separate engines, such a major issue can only impact a single job.

In addition to the new engine, we’ve made other significant changes to the DataMapper’s core, and while that series of changes may not necessarily have a big impact on all users, many will experience a drastic performance improvement, especially when processing PDF-based files. Before getting into the details, it should also be said that pulling the DataMapper out of the Server was a required first step in developing the future replacement for Workflow (code-named Automate). And while it’s still too early to discuss those future improvements, it is important to note that by implementing this first step, we are demonstrating our commitment to making Connect a true enterprise-class suite of applications.

Now back to the DataMapper. The only visible change to the module is its ability to be scheduled just like other engines can be so we’ll start by looking at the Connect Server Preferences:

As you can see, this one is pretty straightforward. You may elect to launch several engines simultaneously, you are only limited by the amount of memory you have on your system. Note that in clustered environments, you may even choose to launch no engine at all on the main system and offload all data mapping processes to slave machines. You’ll notice you can still elect to launch the Server’s internal engine, in addition to external engines. This option should only be used under very low RAM conditions. Otherwise, using the internal engine makes the system prone to the potential problems outlined earlier, where an issue with the DataMapper could take down the entire server.

Inside the standalone DataMapper engine, we also added the code from the Weaver engine that’s used to extract data from PDF-based file, as well as the code that is used to rasterize the PDF pages (the DataMapper uses those rasterized images to display the PDF when designing a DM configuration). This means the DataMapper no longer needs to ask a Weaver engine to perform those tasks, which not only makes the data mapping process faster but also benefits output operations by preventing the DataMapper from using up one Weaver instance. Weaver can therefore dedicate all of its resources to output tasks instead of splitting its duties with the DataMapper.

Duplicating records

A simple addition was made to the DataMapper’s record object, allowing users to specify how many copies of any data record should be stored in the database. This allows the number of copies to be driven by the data itself (as shown in the sample below), which is sometimes more convenient than having to do it in the Template or through Presets.

 

Note how the number of copies for this record is displayed at the top of the data model panel and how a condition is used to determine, from the data, how many copies should be extracted.

Performance

The parsing and handling of XML files has been improved in order to automatically remove extra CR/LF characters that greatly impeded performance, especially on very large single records. And speaking of very large records, we have made significant improvements to handle single records that can now contain several thousands of fields without running into memory issues. Note that these improvements do not mean that the DataMapper’s GUI will become snappy when such large records are loaded: at Design time, you should always design your data mapping process using smaller sample data files. But at run time, those very large records will no longer be an issue.

Output

Dynamic Print Control for PostScript printers

A new way to drive PostScript printer features, meant to complement printer definitions, to allow:

  • Using printer features that are not (yet) supported in Connect, such as
    • switching between color and greyscale printing
    • selecting output bins
  • Controlling directly what PostScript to use for standard functionality
    • duplex,
    • media selection,
    • finishing

These printer features can be controlled dynamically, based on business rules.

More information can be found in the sneak preview.

Refreshed and improved Additional Content

All content in one list
Gives a better overview and allows control of what goes on top (Z-order).
Enable/disable control
Turn off certain content without having to delete the item or modify its condition. This is typically for test and debug purposes.
Description
To allow users to note the purpose of content and be able to identify content that is not descriptive by itself. Barcodes and text with lots of variables may have poor readability by themselves.

 

Transparency support for PDF output

A feature that just missed the 1.8 release, but the known issues now have been resolved, so with 2018.1 it is available for general use, and enabled in the standard printer definitions with PDF output. For content with transparency, this can result in smaller PDF’s with higher fidelity. No longer needing flattening or rasterization for that content also improves performance.

This transparency support is of course also present for output formats with embedded PDF, such as AFP and IPDS with multi-page PDF containers.

PDF pass-through

The printer definitions for AFP and IPDS with multi-page PDF containers now have PDF pass-through enabled by default. For PDF output, pass-through remains a user option in the output preset. This feature was already available in 1.8, but had to be turned off by default due to some issues (mentioned in the 1.8 release notes). These known issues have now been resolved.

To recap the 1.8 feature summary on PDF pass-through: this feature includes PDF resources as-is in the output, making sure that the PDF fidelity is exactly retained at the expense of a less optimized output file size. In addition, less processing can improve performance.

Impositioning

A number of improvements were done on impositioning. The most significant one is booklet impositioning of simplex and mixed plex content: before, the simplex pages would result in empty backsides in the resulting booklet. This is no longer the case. Anyone who wants to have an empty backside in their booklet, will now have to create the content in full duplex with explicit empty pages (turn off the “Omit empty backside…” checkbox in the Sheet configuration of the Print Section).

This also means that in a booklet, a section can start on the backside of another section.

Using system fonts in Additional Content

When adding text from an Output Preset, it was not possible to use system fonts when using Create Output from Workflow. This has now been addressed so it is now possible to properly use system fonts in Add additional content. The fonts have to be installed on all servers running Connect, so this is something to note when using system fonts in a clustered environment.

Workflow

Data Repository

The Data Repository Manager includes several improvements that should make navigating and editing keysets easier. When editing keysets, each key is presented in its own page with buttons allowing you to navigate to the next/previous key. This allows the Manager to better handle keys that contain very large values (like a large JSON or HTML file) and it makes it easier for users to edit the contents of each Key. In addition, deleting a keyset no longer collapses the Treeview on the left, which makes it way easier to delete a series of keys without having to expand the Treeview after each deletion.

Another, less visible improvement is the ability to use a different repository through scripting. The new method – repository.ConnectionString() – expects a single parameter that contains the full path/name of the repository file to connect to. If it doesn’t exist, the file is created. This feature allows users to create/use separate repositories, which can be useful in multi-tenancy implementations. Note that the switch is not permanent: it only impacts the script in which the ConnectionString() method is used. This is just a first step: in future versions, we will introduce connection strings for external databases like MySQL or SQL Server, for instance, as well as new preferences that allow the switch to be permanent or on a per process basis.

Finally, we spent a great deal of effort improving the performance of the repository. The results are most obvious when using the batch update methods in scripting, but all operations should generally show some level of performance improvement.

Data mapping/Retrieving Items

Whenever the Data Mapping or the Retrieve Items tasks were instructed to store the extracted fields in the metadata, the operation took much longer than expected, especially as the total number of fields grew. As a result, many users resorted to using scripting to retrieve items in order to bypass the storing of the fields in the metadata. We have now fixed this issue and the performance gain is massive. Remember, however, that this performance gain only applies to instances when the extracted fields are stored in the metadata.

In addition, it is now possible to specify multiple IDs when retrieving items per ID, including IDs stored in variables or in the data:

Miscellaneous

  • The FTP Input task is now able to capture files in subfolders
  • A PDF/VT’s Form Name can be overriden so Search can group documents correctly
  • A user preference has been added to specify the default scripting language to use when adding a new Scripting task
  • The number of attachments in a captured email is now stored in a JobInfo, allowing for easier loops in a process
  • A new option to better handle PHP arrays has been added to the HTTP Server settings, making the selection of repeated elements much easier.

NodeJS Server

As some of you have noticed, the Beta version of Connect 2018.1 installs an instance of NodeJS, and a new NodeJS Server Input task is also available in Workflow. Because we felt the feature was not complete yet, we decided that these modules would be included in the official release of Connect 2018.1, but that they would be hidden by default. Making them visible will simply require creating a Registry Key. The reason for including the modules in the official release is that we felt the feature is already advanced enough to be useful in some implementations. Obviously, the target audience is PSO but some customers may also be interested in taking advantage of some of Node’s very powerful features.

Before the official 2018.1 release, a Tech Article will be published in the OL Tech Center to explain what the NodeJS module can do and in which type of implementation it could be beneficial.

Connect Server

Improved MS SQL Server support

Existing issues when using MS SQL Server as the database engine for Connect have been resolved. This means that customers who have been wanting to run Connect on MS SQL Server, but were running into issues, are now encouraged to test their existing Connect environments with Connect 2018.1 on MS SQL Server, and may be able to migrate their Connect environment to MS SQL Server.

New customers who prefer to run Connect on MS SQL Server, can now do so.

Memory settings for engines

Memory settings can now be done from the Connect Server Configuration instead of having to go into a .ini file somewhere in the installation folder. Changing these memory settings also no longer requires a Server restart. The only exception is for clustered setups: Server Extensions will still have to be restarted.

Leave a Reply

Your email address will not be published. Required fields are marked *