What's New in Spotfire® 7.13

Spotfire® 7.13

Spotfire 7.13 introduces a new automatic responsive page layout feature that adapts Spotfire pages to fit smaller screens. This makes it easier for authors to create Spotfire dashboards and applications that work well on any device, whether it is a desktop, laptop, tablet or phone. In addition, this release continues to improve self-service data wrangling by making the Add Columns (join) feature available in the web authoring client (Spotfire Business Author), and by making it possible to edit the add columns and on-demand settings from the visual data source view. Administrators and power users may appreciate the new ability to easily run more than one version of Spotfire in parallel, for example, during an upgrade process. For developers, the much requested API to trigger Automation Services Jobs is now available.

Note that Spotfire® 7.13 is a mainstream version. Fixes to critical issues discovered after the release will only be made to the most current version and to any long term supported versions . For more information on the difference between mainstream versions and long term supported versions see the documentation. Visual Analytics

Responsive page layout for mobile devices

The Spotfire page layout is now responsive, so that when a page is viewed on a small device like a phone, the layout organizes to suit the screen. The responsive layout enables vertical scrolling if the page is too large to view on the screen directly. This means that Spotfire analysis files now easily can be used on any device, whether it is a desktop computer, a laptop, a tablet, or a phone. 

The responsive behavior is automatic by default, but can be configured and enabled/disabled by the author of the analysis.

Specifically, when right-clicking a page tab and selecting "Page layout options", there is a threshold value for the screen width; if the screen is narrower than this threshold, then the layout reorganizes to a vertically stacked layout with scroll. The threshold value can be configured per page, and the default value is configurable through a server preference.

illustrates how it works by showing the behavior when resizing the browser window.

More responsive marking in OLAP and big data visualizations

Big data visualizations using live queries, especially towards large OLAP data sources like SAP BW and Oracle Essbase, are now much quicker to respond to marking changes you make.

When you mark a selection of data points in a visualization representing billions of rows of data, it can sometimes take a while for the analysis to refresh the visualizations in the analysis file. Spotfire now computes the marking column used behind the scenes in its in-memory engine, thus improving the marking performance significantly.

Spotfire does this by splitting live visualization and live marking queries into two separate queries. The two query results are then joined in the fast in-memory data engine. This allows faster query execution in the external database or cube, allowing visualizations to refresh significantly faster and adapt to marking changes. This feature allows Spotfire end users to do more data discovery and find insight faster, as big data visualizations will be refreshed more quickly and be more responsive, especially when using Spotfire brush linking between many visualizations and different markings.

Aggregations are done once, and then the marking is applied. Before, the aggregations were done once again after marking. This saves one aggregation calculation that could potentially be expensive. The result is visualizations that refresh more quickly.

The image below shows a comparison of a SAP BW analysis used in Spotfire 7.13 and in the previous release Spotfire 7.12. In many cases, you should expect less frequent and long running queries when marking data:

shows a comparison of a SAP BW analysis used in Spotfire 7.13 and in the previous release Spotfire 7.12. In many cases, you should expect less frequent and long running queries when marking data.

Auto-zoom for zoom-sliders that are at the end of the range

Visualizations with zoom-sliders that are "open", that is, when sliders are at the end of their range, now auto-zoom when the data changes (for example, when filtering). If  you have adjusted a zoom slider so that it is not at the end of the range, the zoom slider will keep the position even when filtering.

If a zoom slider is set to a position which is less than the full range, and you move it back towards the end of the range, then the zoom slider will once again be "open", and thus will adjust the zoom when data changes through filtering, data reload or different kinds of data limiting.

See

for an illustration of how it works.

Color picker in Spotfire Analyst

Spotfire Analyst now has a color picker that makes it very easy to create analyses that follow the corporate color scheme, the color scheme of a website or similar. Just click the picker icon under the colored squares (see screenshot below), then pick the color you want from anywhere on your screen, and use it in the custom theme editor or in the color axis of the visualizations.

Data Wrangling

Join data in the Spotfire Business Author web client

Adding data to your analysis by inserting columns (joins) is a core part of data preparation. With this release of Spotfire, you have the option to add columns to a data table using the Spotfire Business Author web client, in addition to using the Windows client, Spotfire Analyst. This means that more users will be able to do their visual data discovery directly in their web browser, without having to install the Windows client. It also means that if you discover issues with your joins while using the web client, you can instantly fix them, without first switching to the Windows client.

A new web-based user interface, based on the recently added Add rows feature in Spotfire Business Author, makes it really easy to understand the different join methods through illustrations. The result is previewed before finishing the operation. The user interface is smart in the sense that it shortlists and recommends ID and categorical columns that share name and data type. This makes it quicker to find the columns you most probably are looking for.

In the image below, the F1 race calendar has been loaded into the web client TIBCO Spotfire Business Author. The data set contains all race locations, but wouldn't it be nice to also include a link to the details of each circuit? With this new feature you can quickly add more columns to your data table. In this example, a column with links to the Wikipedia page of each circuit is added.

The access point for adding columns is found by expanding the data panel and displaying the Data table view.

In this example, the additional data was stored as a data extract file in the Spotfire Cloud library.

Once the new data is loaded you will see a new Add columns dialog. It's based on the Add rows dialog released in Spotfire 7.7, but adapted for joining data.

Note that the new dialog provides the following features that will help you create a successful join:

The recommendation engine of Spotfire has already defined three column matches for the join, based on data heuristics.

You can add and remove column matches and get a live preview of the result directly in the dialog.

The preview includes color coding of columns. Light blue indicates existing columns, dark blue indicates matched columns, and medium blue indicates the added columns.

The Number of input rows for the preview can be changed. This is useful, for example, when you need more rows to get a representative preview of your matched column values.

In this image, two out of three suggested column matches have been removed, and the preview is updated accordingly.

The Columns from new data section lets you specify which columns to add or exclude.

In the image below, you can see that the default join gives duplicate columns for Circuit and Locality. This can be addressed by simply excluding these columns under Columns from new data when doing the match (or by editing the Added columns node afterwards).

The preview also shows that USA in the existing data table (to the left of the dark blue join column) is not present in the new data table (to the right of the dark blue join column). Also, United States of America in the new data table is not present in the existing data table, which results in new rows being added to the final data table. You can work with the Number of input rows setting to find mismatches like these between the two tables. You can also change the join type to achieve a join result better aligned with what you need. If you would like to transform your values to get a better match, you can do so in the Spotfire Analyst client by joining on a calculated column created with the Calculated Column data transformation. You can of course also modify the values directly in the data source, for example in an Excel sheet, if that is what you prefer.

In this example, we are only interested in adding the column containing Wikipedia links for each circuit. Therefore, the other available columns from the new data have been excluded, as seen below.

The Join settings section displays the new preview of the Spotfire join types. Select a join type and hover with the mouse pointer over the join type example to see how each join works.

With the new column added, you can click on the link for a circuit in a Spotfire table visualization and read more about the circuit on Wikipedia.

Edit joins from the source view

The TIBCO Spotfire data wrangling vision to edit everything continues, and with this release you can edit previously specified Add columns (Join) operations. This makes it really easy to adapt your analysis files to changes in your data sources over time. Broken joins can easily be fixed and thanks to the built-in smart indications in the Spotfire source view, you will instantly see when there are issues with your Add columns operations that need to be addressed. A new web-based user interface, based on the recently added Add rows feature in Spotfire Business Author, makes it easy to understand the different join methods through illustrations, and the result is previewed before finishing the edit.

The image below shows the gear icon on the Added columns node, which is the entry point for editing a join.

The new dialog makes it easy to, for example, change the join type as needed.

Edit on-demand settings on data source level

With this release of Spotfire, you can quickly edit on-demand settings for each individual data source in a data table. The setting is available in the source view once you have selected a data source node, as seen in the image below.

Previously, the on-demand setting applied to the final data table, but with this release of Spotfire you can control on-demand settings for each source.

The image below shows the entry point for the setting on a data connection.

This new feature also allows you to switch to on-demand data loading even though it was not specified as such from the beginning. The image below show the On-Demand Settings dialog which allows you to switch from the All data at once mode to the Data on demand mode.

Data Access

Cloudera Impala query timeout setting

With this release of Spotfire, a timeout setting has been added to the Cloudera Impala connector. This means that you can allow Impala queries to run for longer. For example, this allows running queries to complete when you are extracting result data sets into the in-memory data engine of Spotfire.

Amazon EMR support

TIBCO Cloud Spotfire and the Spotfire on-premise platform now support Amazon EMR via Hive and Apache Spark SQL.

This means that you can store analysis files in the Spotfire (Cloud) Library and query Amazon EMR directly from the web-based clients Spotfire Business Author and Consumer.

Use the Hortonworks connector of Spotfire Cloud Analyst and the ODBC driver for Hive from Cloudera to connect to EMR Hive.

Use the Apache Spark SQL connector of Spotfire Cloud Analyst and the TIBCO ODBC Driver for Apache Spark SQL to connect to EMR Spark SQL.

Apache Spark SQL support in TIBCO Cloud Spotfire

TIBCO Cloud Spotfire now supports Databricks Cloud and Apache Spark SQL.

This means that you can store analysis files in the TIBCO Cloud Spotfire library and query Databricks Cloud and Apache Spark SQL directly from the web-based clients Spotfire Business Author and Consumer.

Use the Databricks ODBC driver to connect to Databricks and use the TIBCO ODBC Driver for Apache Spark SQL to connect to generic Apache Spark SQL.

Both drivers are used with the Apache Spark SQL connector of Spotfire Analyst.

Microsoft HDInsight Hive support in TIBCO Cloud Spotfire

TIBCO Cloud Spotfire now connects to Microsoft HDInsight via Hive.

This means that you can store analysis files in the Spotfire Cloud Library and let them query Microsoft HDInsight directly from the web-based clients Spotfire Business Author and Consumer.

Use the Hortonworks connector of Spotfire Cloud Analyst and the ODBC driver for Hive from Cloudera to connect to Microsoft HDInsight Hive.

Administration

Work with multiple versions of Spotfire Analyst

The Spotfire deployment mechanism now supports both upgrading and downgrading of the installed Spotfire client when you connect to a server (and a specific deplyment area). This makes it easier to work with multiple Spotfire versions at the same time.

Server Database Support for Oracle 12c

Spotfire now supports Oracle 12c as the server database.

Developer

REST API to run Automation Services jobs

It is now possible to trigger execution of Automation Services jobs from an external application using a REST API. A job can either be stored in the Spotfire library or passed as an argument. The API uses an OAuth2 based authentication/authorization mechanism.

See REST API Reference for more details.

OAuth2 based authentication for the Web Service (SOAP) API

The Web Service (SOAP) APIs (LibraryService, UserDirectoryService, UpdateAnalysisService, InformationModelService, LicenseService and SecurityService) now uses a OAuth2 based authentication/authorization mechanism. This means that the API client only needs to support a single authentication method that will work with any Spotfire Server authentication configuration.

See Web Services API Reference for more details.

Simplified workflow when building Spotfire .Net extensions

With this release comes an updated and simplified procedure for building .NET extensions for Spotfire. The package building functionality is now integrated with Visual Studio®. Templates are provided so that the configuration needed for a third party developer is kept to a minimum.

See TIBCO Spotfire Developer Documentation for a tutorial on how it works.

Spotfire distribution files

With this release, it is possible to ship a bundled solution containing several Spotfire packages as a single distribution file (.sdn).

See TIBCO Spotfire Developer Documentation for more details.