Bug fixes and minor changes 12.6

Minor changes

03 May 2024 (V 12.6.02.137981):

• New Data source connection timeout parameter for REST APIs- A new parameter has been added to REST API calls defined in a Data source connection to customize the timeout period. The default timeout period is 100 seconds and remains so when this new parameter is not defined. This is helpful, for example, in the event that you anticipate a Data source connection to take longer than 100 seconds
  

July 2023

• Scrollbars Style - The Data View and Heatmap chart Objects offer a new "SCROLLBARS STYLE" menu that allows you to customize the scrollbars of these Objects.

  
  The options for the new Object setting are the following:

  • Hide scrollbars. Enable this option to hide the scrollbars. The option is disabled by default

  • Scrollbar width. You can choose the width of the scrollbars from the dedicated drop-down list. You can choose one of the following options: Small, Medium, Large

  • Contrast level. You can choose the color contrast of the scrollbars from the dedicated drop-down list. You can choose one of the following options: Small, Medium, Large

  • Transparency level. You can choose the transparency level of the scrollbars by sliding the transparency level slider or manually entering a transparency level percentage number in the dedicated field.

  The new settings are also included in the properties that can be copied and pasted between Objects using the "Copy Format" and "Paste format" features in the Top Menu of the Capsules workspace.

• Horizontal scrollbar behavior - When sliding the scrollbar to the end of a Data View or Heatmap chart Object, the displayed data set did not occupy the whole Data View or Heatmap Object space, but displayed only the last column along with a blank space. This behavior has been changed and now the Data View and Heatmap Objects do not display blank spaces when horizontally scrolling to the end of the data set

• API Client authentication in Swagger - Now it is possible to perform an API Client authorization in the Swagger interface, which allows you to test Board public API queries directly from Swagger. The authentication supports only the OAuth2 client credentials flow.
  To authenticate, proceed as follows:

  1. From the Board public API documentation page in Swagger, click on the "Expand operations" of the desired public API endpoint and then click on the red circular button with the exclamation point to open the authorization pop-up window.
     You can access the API documentation page from the API queries page of the desired Data model
      

  2. Select "Request body" from the type field

  3. Enter the Client ID of the required Client API in the "ClientId" field

  4. Enter the Client Secret of the required Client API in the "Secret" field

  5. Enable the "public-api" scope option

  6. Click on "Authorize" to authenticate. The red button with the exclamation point will become blue after a successful authentication. If that happens, you can try out the API queries directly from the Swagger page.
      

• New ODBC drivers for Dynamics 365 - Board ODBC Drivers for Microsoft Dynamics 365 have been updated (Board on-premises installations only)

• New column in the Cubes page - A new column called "In RAM" is now available for the table in the Cubes page. The cells of this column display the configuration of the "In RAM" property for each Cube (enabled or disabled). You can show or hide the new column by clicking the Column chooser button in the upper right corner of the table (hidden by default).
  This is useful for identifying Cubes that, for example, have been migrated from previous versions of Board where it was possible to configure a Cube as "in RAM" or not. A Cube configured as not "in RAM" can be configured as "in RAM" by opening the corresponding Cube configuration panel, selecting the "in RAM" checkbox, and saving.

Please note that, when you set a Cube as "in RAM" and save changes, the checkbox automatically disappears and you cannot go back. This is because, starting from the Board 2021 Spring release (Board 12.0), Cubes can only be "in RAM" and so that option has been removed.

Bug fixes

27 November, 2024 (V 12.6.1.3.246912):

Procedures. When running a Procedure with “Save Data entry”, an invalid data entry input did not prevent the Procedure from completing when it should have. This bug has been fixed.

Sign-out. In corner cases, users were logged out during navigation shortly after logging in when they shouldn't have been. This bug has been fixed.

Procedures (On Premises). When executing a “Go to Screen” Procedure action that opened in a new tab and applied a selection or disabled the Layout execution, the settings were not applied to the opened Screen. Also, when executing “Go to Screen” to a Screen where the “Hide Screen list” was enabled in Capsule settings, the Home Screen would open instead of the desired Screen configured in the Procedure. This bug has been fixed.  

Lock and Spread. When lock and spread was activated on a total and some cells were empty, the total changed instead of the rest of the values when a value cell was modified. This bug has been fixed.  

Dynawrite. The text inserted in the Dynawrite the Dynawrite object using RTF Cubes did not display Bold and Italic formatting. This bug has been fixed.

Labels. When changing the Layout Select of a Label, the size of the corresponding Label changed when it shouldn’t have. This bug has been fixed.

Custom Time Entity. The custom Time Entity with the physical name “E_000002” is not usable whether the Data model was restored or created. When the Data model was restored, the custom Time Entity would not appear in the Custom Relationships page, even after changing the name. When the Data model was created, the custom Time Entity would appear in the Custom Relationships page but not appear listed as a possibility for mapping in a Data reader. This bug has been fixed.

Security selection. In a corner case, when a selection was made on a Security profile using the “Select Entity based on Cube” filter and a Procedure configured to launch a selection was based on the same Entity, the wrong selection displayed. This bug has been fixed.

AzureML. Several fixes regarding AzureML have been fixed, including: 

• Inference step. All inference values were generated from an inference Procedure that was executed after a training scenario ran regardless of an active Screen selection. The Screen selection is now considered.

• Timeout parameters. The request timeout has been increased to match that of the endpoint timeout specified at deployment.

• Inference client call. Some Inference client call settings have changed in AzureML. Adjustments were made to restore communication with Board.

Data entry. When a minimum and maximum filter was made on a Cube in a Data View Layout with Data entry enabled, and data entry was saved within those filters, a Data validation error message appeared when the value should have correctly saved. This bug has been fixed.

03 May, 2024 (V 12.6.02.137981):

• Cubes. When a Data View configured with an unbalanced Entity and a Text Cube with the Lock by feature enabled and referring to another Cube had a selection applied on parent members of the aforementioned unbalanced Entity in the Layout axes, an unhandled error was returned without rendering data in the Data View. This bug has been fixed

• Data model (on-premise only). When a Data model backup Procedure configured with an expiration date was launched, the timestamp in the log file differed from the one that appeared in the Backup folder. This bug has been fixed

• Data source. When an SAP Data source was modified by editing the URL or moving the Data model from one instance to another where the SAP Data source had a different URL, all the configured Data Readers executed even if the endpoint was invalid. This has been fixed, and now, Data Readers will work if the Data Source URL is updated to a new valid endpoint and will not work if the URL is wrong or invalid

• Impact Analysis. In some cases, when accessing the Impact Analysis from a Data model, an error occurred where the page could not be accessed. In this case, the diagnostic log has been enhanced to record information about the file restricting access to the Impact Analysis so the user can address it and then successfully access the Impact Analysis area

23 February, 2024 (V 12.6.02.96848):

• Data entry. The Data entry on a cell with a reverse rule was not saved after unloading from a Database or restarting the Board Engine. This bug has been fixed

• Presentations. When creating a Presentation, if the Presentation name was shorter than 8 characters, changes were not saved. This bug has been fixed

• Procedures. When configuring a Broadcast Procedure using the "From Layout” option, an error message occurred. This bug has been fixed

• System Administration. When more than one Broadcast was set with daily schedule at a specific time, the Broadcastings did not run correctly and an unexpected execution ran instead. This bug has been fixed

1 February, 2024 (V 12.6.0.2.94049):

• Data sources. When a Data Source connection was made with an Amazon S3, the Data Reader only displayed up to 100 text files in the Source folder. This bug has been fixed

• Rules. When applying a Rule with a code defined within square brackets [ ], the Rule was not applied and the reference to the correct member was not highlighted. When using a Rule with a code defined within parentheses ( ), the Rule was applied correctly, but the reference to the correct member was not highlighted. This bug has been fixed

• Subscription Hub. When using the UpdateLicense call to update licenses in the Subscription Hub with a file that contained temporary licenses, the update would not complete. This bug has been fixed

2023 Patch Releases:

• API. When using a Data picker in a POST API call, the JSON output value was always displayed in double quotes, even for numeric values. This bug has been fixed

• API. In case a high number of columns is generated in the response, a soft kill was implemented (added feature) in a test call to abort the remote call and a new logging has been added (in debug mode). The soft kill is now logged in the diagnostic log

• API. When using localhost with Board 12.6, Swagger API documentation does not open as expected. This bug has been fixed

• Appearance. If a logo or a header background of the recommended specifications were uploaded in the Appearance page under System Administration, unwanted space around the images would appear. This bug has been fixed

• Board public APIs. The Procedure API endpoint documentation lacked clarity regarding the required id parameters. The "id" parameters of both the POST and GET requests have been renamed in the following way: the id parameter in the POST method has been changed to procedureName, while the id parameter in the GET method has been changed to sessionId

• Block format. A malfunction caused the "ignore number scaling" option to be ignored in Play mode for the total rows of grouped Entities, if groups were collapsed and a number scaling option was applied through the appropriate menu of the sliding toolbar. The issue has been resolved

• Bulk SQL insert. In some cases, the Bulk SQL insert Procedure step caused an error to appear and the Procedure to fail. Also, these errors were logged incorrectly. This bug has been fixed and all errors and exceptions are now logged correctly

• Capsule Snapshot (on-premises installation only). If the Board Server Configuration did not use disk C as the default “Board path”, an error would occur when creating a Capsule snapshot

• Charts. The template applied using a Select would override the template applied through a “Refer to” function. The bug has been fixed

• Collaboration. A malfunction that caused notifications of already read discussion threads to reappear when logging out and into the Platform again has been fixed

• Cube visibility. Cube visibility rules that deny Read/Write access to Rolap (RDB) Cubes were not applied correctly when those Cubes were used in a Layout. The issue has been fixed

• Cubes. A bug that prevented creating an additional Cube version other than its Structure has been fixed

• Cubes. Reordering Cube Groups from the user view page in the Cubes section caused an unexpected error. This bug has been fixed

• Cubes. When a BLOB Cube was moved to a different Group or was renamed, the Cube was cleared. This bug has been fixed

• Cubes. If a Cube with multiple versions that was included in a Transporter package was moved to a different position in the user view page, the Cube was cleared. This bug has been fixed

• Cubes. The action “Reload all Cubes” failed when loading more than 1000 Cubes. This bug has been fixed

• Database security. In the case of Screens with Layout selections, users associated with Roles with security selections on Unbalanced Entities experienced longer Screen loading times than users associated with Roles without security selections on Unbalanced Entities. The performance issue has been resolved

• Data entry. When a new value was entered in a Down Total or a Row Total cell of an algorithm block with a reverse algorithm, the number displayed in the preview of those cells was wrong. This bug has been fixed

• Data entry. In a Layout with a Screen Selection that has an unbalanced Entity set by Row, performing a Data entry on a Cube cell, resulted in the value not being saved and propagated as expected along the Relationship tree. This bug has been fixed

• Data entry. Data entry on a vertically aligned Data View would not save the correct value. This bug has been fixed

• Dataflow. Running a Procedure with the Dataflow step with the time function Yearly Cumulated Value applied to the source Cube and a Select applied to the Fiscal Year Entity would unexpectedly change the selection. This bug has been fixed

• Dataflow. In some cases, if a Dataflow had a source Layout with multiple Cubes, and a Rule was applied to one of them, no values were written if the Rule was associated with an Entity that had a large number of members. The issue has been fixed

• Dataflow. The Forecast and Trend time functions are now also supported by Dataflow Layouts

• Dataflow. If only one value of a Block in a Dataflow Layout was considered and said Block was used in the expression, in some cases, the target Cube was not cleared before writing new values. This bug has been fixed

• Dataflow. The JOIN function, not to be confused with the JOIN algorithm, is not considered by the Dataflow anymore, and if you use the JOIN function in a Dataflow, it will be ignored because Board automatically defines which algorithms to use in calculations. If you have existing Dataflows using the JOIN function, no further action is required. For example, a formula with the following syntax JOIN(a+b) is now considered as simply a+b

• Dataflow. In some cases, if the function IN() was selected while mapping a Cube from MDB to RDB, the Dataflow wouldn’t write any values. This bug has been fixed

• Data entry. If the Save/Undo mode was disabled in a Data View and a Procedure was set to trigger after updated values were saved, the Procedure would not trigger when a user pasted values in a cell range that contained some locked cells. This bug has been resolved

• Data entry. If a Layout was configured with a hidden Block, an unhidden Block with Data entry enabled, and at least an Entity set By Column, clearing sequential cells on the same row with the DEL key would also clear adjacent unselected cells. This bug has been fixed

• Data entry. In some cases, data entry on Cubes with an unbalanced hierarchy in their Structure did not work as expected when the "Sort" and "Keep top" features were applied to that Cube in the "Filters" settings of the associated Layout. The issue has been resolved

• Data entry. In some cases, data entry actions on a Data View with two Cubes failed when saving, if a Pager was set on a Entity that was present in the Structure of one Cube, but not in the Structure of the other one. The bug has been fixed

• Data entry. In some cases, values saved through data entry actions on Cubes with an unbalanced hierarchy in their Structure were not allocated correctly when the "Cube cells locked by" Block setting was applied to said Cubes and the "Filter on value" feature in the "Filters" settings was applied to filter the displayed levels of the unbalanced hierarchy. The issue has been resolved

• Data entry. In some cases, continuously scrolling horizontally in a Data View containing a large number of columns and then entering data in some cells using the ">" or "<" shortcuts without saving the entered data, caused a visualization error where the Data View would not display some of the data in the cells where the Data entry was performed until the pending data entry actions were saved. The bug has been fixed

• Data model. In the case of Entities' descriptions localized through the Entity localization feature, after adding a new language to the Language Entity the corresponding descriptions were not updated until the Engine was restarted or the Data model was unloaded and reloaded. The bug has been fixed

• Data model.  When creating a Transporter package, if Rules were unselected in the configuration of the package or if the target Data model didn’t have any existing rules, running the package afterward would result in a failure. This bug has been fixed

• Data model. If the Data reader was called in parallel on the same file from two different users at the same time, an error would occur. This bug has been fixed

• Data model. When a 128-bit Cube was configured to have access conditions from the Cube visibility section of the Database security page, the Layouts that contained said Cube would fail. This bug has been fixed

• Data model. When a custom Time Entity was created as a child of Fiscal Year and a change was made to the Time Range, the Fiscal Year was duplicated. This bug has been fixed.

• Data model Authorization/Security filter. When an Entity with a Security filter set by the “Select Entity based on Cube” is set by Row in a Layout and a Layout selection of that Entity’s parent is applied using “TO” instead of “APPLY”, the Security filter rule was ignored. This bug has been fixed 

  If you are currently using the Security filter “Select Entity based on Cube” feature released in 12.6, we recommend to update to the new version.

• Data model package (previously known as "Transporter package"). Packages containing replicated Entities related to other Entities in a Relationship tree are now working as expected and can be run in the target Platform with no issues

• Data model package (previously known as "Transporter package"). In some cases, deleting and creating Cubes and Entities in the target Data model before running a Data model package caused a misalignment between the source and target Data models. Because of this, during the executions of the Data model package in the target environment, errors occurred indicating that the transported Cubes or Entities had already been created or deleted. This was due to an issue caused by the way Cubes and Entities from the source Data model were identified and compared to the ones in the target Data model by the Board Engine. The issue has been resolved by changing and enhancing the identification and comparison method. Therefore, users impacted by this issue must recreate the Data model packages from a 2023 Summer release Platform

• Data model package (previously known as "Transporter package"). In some cases, deleting and creating Cubes and Entities in the target Data model before running a Data model package caused misalignments in the "Mapping" section between Data readers in the source and target Data models after running the Data model package. This was due to an issue caused by the way Cubes and Entities from the source environment were identified and compared to the ones in the target environment by the Board Engine. The issue has been resolved by changing and enhancing the identification and comparison method. Therefore, users impacted by this issue must recreate the Data model packages from a 2023 Summer release Platform

• Data model package (previously known as "Transporter package"). In some cases, an error was displayed when trying to access the Entity section of a target Data model after running a Data model package. This occurred because of misalignments between the source and target Data models, which were caused by deleting and creating Entities in the target Data model before running the Data model package. Users impacted by this issue must recreate the Data model packages from a 2023 Summer release Platform

• Data model package (previously known as "Transporter package"). The execution of Data model packages failed in cases where they contained "Delete" actions on an Entity followed by "Delete" actions on a Cube dimensioned by that same Entity, since the "Delete" action on the Cube failed to locate the Cube because it was automatically deleted when the Entity in its structure was deleted. The issue has been resolved

• Data model package (previously known as "Transporter package"). A malfunction that caused misalignments in the Entity groups of the target Data model even if no Entities were transferred from the source Data model has been fixed. In some cases, this issue unexpectedly changed the groups of Entities in the target Data model, while in other cases some Entities did not appear in the Layout editor

• Data model package (previously known as "Transporter package"). When transporting only a portion of Cubes from one Data model to another, the sparsity code (SparsityIdx) of transported Cubes was misaligned between the source and target Data models. The issue has been resolved. For more details, see this page

• Data model package (previously known as "Transporter package"). A malfunction that caused row templates to be transported even though their associated Entities were deleted during the execution of the Data model package has been fixed

• Data model package (previously known as "Transporter package"). An error occurred when running a Data model package which contained a Rule that referenced non-existing Entity members. The bug has been fixed

• Data model package (previously known as "Transporter package"). A malfunction that caused unexpectedly high usage of RAM during the execution of a Data model package has been resolved

• Data model Restore. In some cases, where several backups of a Data model exist, of which one was renamed without a timestamp, the system would always restore the version without a timestamp regardless of the user’s choice. This bug has been fixed.
  If the restore process is initiated in the Admin Portal, the system will always restore the latest backup with a timestamp even if the user manually selects a backup without a timestamp. This is a known limitation

• Data Picker. A Data Picker formula referencing a Cube with no data in it returned 0 even though it was configured to add an integer value to Cube value (Example: =@[V354]+14). The bug has been fixed

• Data reader. A malfunction caused Data readers to partially import data from a text file in the case of a Data reader configuration with both "Add new item" and "Replace" reading actions on two different Cubes that already contained values. The malfunction has been fixed

• Data reader. In some cases, a performance issue caused the execution of Data readers to fail when they were configured to write values in a large number of Cubes. The issue has been resolved

• Data reader. In some cases, the "Fast Track" feature created the correct Entities and Cubes on an existing or new Data model but did not load any data in those Cubes. The bug has been fixed

• Data reader. Custom environment variables now work as expected, even when used in the path to the folder containing the source file to be read

• Data reader. In some cases, a Data reader involving a great number of Cubes caused an unexpected usage of RAM. The issue has been resolved

• Data reader. When a Data source of a Google Cloud storage (ISS) or an Azure Blob storage (on-prem) was configured, a Data reader pointing to these sources could not be configured and displayed the message “Could not load file or assembly”. This bug has been fixed

• Data sources. When attempting to add an alias for a Data source in the "Data pipeline" page of the Administration Portal and entering a 'Connection String' value longer than 32 characters, errors would occur when selecting that alias in the Data sources page. The issue has been fixed and a new behavior has been put in place: the "Connection String" value now accepts a maximum limit of 32 characters

• Data View. In some cases, entering data in cells of a single column of a Data View containing a large number of rows, and continuously scrolling down at the same time without saving the entered data, caused a graphical error where the Data View displayed data in cells where no Data entry was performed until the Data entry was saved. The bug has been fixed

• Data View. In some cases, Data Views did not display the values of Cubes with an unbalanced hierarchy in their Structure and a selection applied on it, unless enabling the "Disable unbalanced hierarchy" option under the "RULES" Block settings. The bug has been fixed

• Data View. In some cases, Data View Objects displayed an error when the associated Layout was configured with a Data Block whose access was limited by the Cube Visibility feature, and the "Sort" and "Keep top" features were enabled on that same Block. The bug has been fixed

• Data View. Copying and pasting multiple successive cells from a column to another column did not work as expected: if the last cell was empty, it was not pasted into the target column. The bug has been fixed

• Data View. Templates created with Symbols applied before and after values by Row were not ignored when the “Ignore format template” option was enabled. This bug has been fixed

• Data View. Entity members whose code was a number were not displayed correctly when added as a Block in a vertically aligned Layout. This bug has been fixed

• Data View. When the Lock and Spread function was applied on a Data View associated with a vertically aligned Layout, the “Total”, “Average”, “Minimum”, and “Max” values displayed in the right panel of the Lock and Spread configuration window for cells containing decimal numbers were rounded to whole numbers. This bug has been fixed

• Data View. When a Data View with multiple columns of different widths was navigated by scrolling horizontally, the last column was not displayed. This bug has been fixed

• Data View. When a Screen Selection made with the “exclude” option and one or more of the excluded members were deleted from the Entities section of the corresponding Data model, a log line in the Diagnostic log set at a verbosity level “Warning” was created every time the Screen was opened. This event is no longer tracked in the Diagnostic log set at “Warning” but at “Debug”

• Data View. The code and description for all Time Entities would only show the description in display mode. This bug has been fixed

• Data View. When an algorithm with a reverse rule was applied to a Data View and Data entry was performed on the Grand Total, the on-the-fly calculations were not correct until saved. This bug has been fixed

• Data View. When a Data View is populated with a text Cube, has Data entry enabled, and the “differentiate zero and null” option is enabled, text that the user deleted would still appear after saving the Screen. This bug has been fixed

• Dynawrite. Entering curly brackets in a data-entry enabled Dynawrite caused all entered text to disappear when saving. The bug has been fixed

• Entities. In some case, changing relationships between Entities that had replicas caused multiple errors or the loss of data. The issue has been resolved

• Entities. In some cases, when a standalone Entity was cleared from the Entities section of a Data model, all selections that included that standalone Entity did not work as expected. This bug has been fixed.

• Excel Add-in. In an Excel Workbook, duplicating a Data View with a "Data View cells locked by" rule allowed users to enter data in cells that should have been locked, as per the rule's configuration. The bug has been fixed

• Excel Add-in. When a Screen Selection was applied to a Layout with Flatten mode enabled in Excel Add-in, an error message occurred and the selection was not applied. This bug has been fixed

• Export Data to xlsx. Exporting data from a Data View using the “Export data to xlsx” feature with the “Use Excel formulas for calculated blocks, column totals and row totals” option selected didn’t work as expected when the Layout had 2 Entities set By Column. This bug has been fixed

• Format. A malfunction made it impossible to delete the last Entity template created in the Format section of a Data model. This also prevented creating snapshots of the Data model. The bug has been fixed

• Gantt Chart. If Gantt Chart bar colors were changed in the ACTIVITY BARS OPTIONS menu from the default Monochrome to another type, after saving and switching to Play mode, the color change did not take effect. This bug has been fixed

• Impact Analysis. When opening a Cube from a Data model, the Impact Analysis was trigged. This bug has been fixed. Now, the Impact Analysis will only be triggered when the ANALYSIS tab is opened.

• Layout. The distinct count aggregation function works as expected, even when applied to different versions of a Cube other than its Structure in the same Layout

• Layout. In some cases, applying a ranged time function (i.e. Yearly Cumulated Value) on a Cube whose values are partially restricted by a Cube visibility rule, caused the Time Function calculation to still consider all values of the Cube. The issue has been resolved

• Layout. Refreshing a Layout with a Rollup Entity set by row in the most nested position caused the associated Data View to go into error. The issue has been fixed

• Layout. Layouts with Rules and filtering settings applied caused wrong coordinates to be written to the Data Entry log when data entry actions were performed. The bug has been fixed and the Data Entry log has been enhanced with new information (see the Other enhancements page for more details)

• Layout. If a Layout had a Block with both a "Refer To" option and a "Detail By" option configured on a standalone Entity, and a subsequent algorithm Block referencing that Block, the Layout execution failed. The bug has been fixed

• Layout. If a Layout had Blocks with both a "Detail By" option and data entry enabled, algorithm Blocks referencing them, and the Row totals disabled, saving entered data caused an error (exception) The bug has been fixed

• Layout. If a Layout had an Unbalanced Entity with many levels set by Column and totals were enabled, the Layout execution took much longer than before. This bug has been fixed

• Layout. In Layouts with an Unbalanced Entity set by Column and multiple Entities set by Row, data aggregation did not occur as expected.  This bug has been fixed

• Layout. The execution of Layouts containing a high number of cells was unexpectedly slower than previous versions. This has been fixed

• Layout. When a filter was set on one Cube and Data entry on another and the Screen with the associated Data View was opened in 2 different browser tabs, then, if the user filtered out a cell from one tab and then went back to the other tab without refreshing the page to perform Data entry on the same cell, the newly entered data was saved in the wrong cell. This bug has been fixed

• Layout. If a Layout with Unbalanced Entity had the “Refer to” option enabled and an Entity not in said Layout was deleted, after unloading and reloading the Data model, the Down Total value was not calculated correctly. This bug has been fixed

• Layout. A Layout having more than one Entity set by Row and two algorithms, one that displayed the same value of a Cube block and the other one that performed a formula where the Cube block is divided by itself and also had a filter applied, produced an incorrect Total value. This bug has been fixed

• Layout Configuration. When a Data View was configured with two or more Entities in the axis by Column and Data entry was enabled with the “reverse algorithm” setting, the value entered into a cell would appear in another column to the left before saving. This bug has been fixed

• Logs. When opening an HBMP log file of at least 2 GB in size from the web client, BENG crashed due to RAM consumption. This bug has been fixed. The log is limited to the last 1 million latest records

• Name field. When special characters were used in names for Labels, Cubes, Procedures, and Entities, the whole field would disappear. This bug has been fixed

• Power Bi add-in. In some cases, when signing in using SSO, an error would appear, preventing the user from signing in. The bug has been fixed

• Presentations. When a Presentation was shared to all users, CPU consumption and time elapsed was unusually high and, in some cases, could lock the system. This bug has been fixed

• Procedures. Copying a Procedure step group and pasting it right above the copied group caused Layout configurations to be erased from Dataflow Procedure steps, and selections to be cleared in selection Procedure steps. The bug has been fixed

• Procedures. A malfunction that caused the R Calculation Procedure step to partially populate the configured Cubes has been fixed

• Procedures. Excluding half of the members of an Entity in a "Selection" Procedure step caused the Procedure step to show "None" in the active selection, and the Select window did not show any members if an interactive selection was performed on the same Entity after the "Selection" Procedure step. The bug has been fixed

• Procedures. The %computername% environment variable now works as expected when used in a "Reload all trees" Procedure step

• Procedures. In an "Extract entity" step, entering a path with folders that did not exist caused the Procedure to fail. The bug has been fixed: the system now creates all missing folders referenced in the "Path/File" field

• Procedures. In some cases, Capsule Procedures caused random "Cannot save procedure" error messages to appear when saving them. The bug has been fixed

• Procedures. A defect that prevented the modification of a Data model Procedure when it included a Temporary Cube with a deleted Entity in its Structure has been fixed

• Procedures. If the path in an Extract Tree Procedure step configuration didn’t exist, the Procedure which contained it would fail instead of creating the necessary folders. This bug has been fixed

• Procedures. There was an error where Layout configurations were lost when editing a dataflow step, replacing the target Cube with a temporary Cube. This issue has now been fixed

• Public APIs. In some cases, if a Data model was renamed, starting one of its Procedures via an API request would not successfully execute. The bug has been fixed

• Rest API. If a REST API step configuration used a data picker for a query parameter value on a temporary Cube, an error would occur. This bug has been fixed.

• Rest API. When a REST API Call was made and the response had a body in gzip format, this step could not read it properly in the output configuration dialog. This bug has been fixed and compressed body types are now supported: to do so, we have added two new default headers (Accept and Accept-Encoding) to notify the remote server. If a custom value is entered, those headers will notify the remote server with said custom value. If no value is entered, those headers will not be added to the call. If an unsupported custom value is entered, an error message will be displayed in the output configuration dialog.

• Roles. Security selections configured through the "Select Entity based on Cube" feature were not saved if they were added from the Roles configuration panel. The bug has been fixed

• Rules. In some cases, when exporting and re-importing Rules created at different times, newer Rules did not work as expected. The issue has been resolved and new user notification messages have been implemented

• Rules. In some cases, when exporting and re-importing Rules created at different times, newer Rules did not work as expected. The issue has been resolved and new user notification messages have been implemented. Also, the data entry log has been enriched in order to be able to track errors regarding Rules: these new log lines include the name of the Rule that caused the error and specific information about the error 
  Example: 
  2023-07-06 16:42:10.545 +02:00  WARN  BOARD\admin      1ee9314f-6491-4ab4-8a93-484e0555f5aa  DTB_OpenByName  Rule        {"Exceptions":[{"Name":"A", "Formula":["Error in formula [a2] +10 - 01] - [a5]: Wrong codes delimiter balancing"]}]}    Anomalies in rules found

• Rules. The Data picker within a Rule was not working as expected. This bug has been fixed.

• Rules. When a Rule with “column offset” (dot annotations) was applied to a Data Block and Entities were sorted by “None”, the values were sometimes displayed incorrectly. The issue was appearing if the order in which Entity members were loaded was different from the default display. This bug has been fixed.

• Screen Object. When a button was created with the Action Type “Go to previous Screen”, the action did not always go to the previous Screen. This bug has been fixed.

• Screen Objects. In cases where the sliding toolbar was disabled, it still appeared in Play Mode when performing some specific actions. The issue has been resolved and a new behavior has been put in place: if disabled, the sliding toolbar will never appear in Play Mode 

• Screen visualization. In some cases, a malfunction caused the scrollbar of horizontal selectors to cover the list of displayed members on Firefox browser. The visualization issue has been fixed

• Screen visualization. A visualization issue prevented users from scrolling back to the first column after scrolling horizontally and having performed data entry actions in cells of other columns of a Data view. The issue has been fixed

• Select. In some cases, performing a selection through an "Interactive selection" Capsule Procedure step caused the Select window to exclude all other Entities from the active selection and to apply an incorrect selection. This bug has been fixed

• Select window. The hierarchy of Relationships is now displayed as expected in the Select window: each level of a Relationship tree is displayed right after its child level

• Smart Import. When using the Smart import Object, certain string values were recognized as scientific notation values. This has been resolved by providing users the ability to set a specific data type for each unmapped column. 
  The new COLUMN TYPE menu has been added to the Smart Import Object contextual right panel in Design mode for unmapped columns. Once opened, select the column name from the “Column” dropdown menu and then, from the “Data Type” dropdown menu below, select the preferential data type: numeric, text, or date. The default option is “not set”, and in that case, Board will automatically interpret the data type based on the values entered in the unmapped columns

• Smart Import. In some cases, copying and pasting content from an Excel file to a Smart Import Object and proceeding with the import caused some row templates to disappear. The bug has been fixed

• Smart Import. In the mapping configuration, the "Fixed value" option used more than once with the same value for multiple Cubes using caused a fatal error during the data entry process in Play mode. The bug has been fixed

• Smart Import. You can now paste numbers formatted in scientific notation or values from cells showing ##### from Excel. Numbers are saved in their extended format (i.e. 6.99E+08 is saved as 698.775.465)

• Smart Import. SEARCH and FIND did not work in Smart Import as expected. This bug has been fixed

• Smart Import Object. In Design mode, when a user with a Power User license clicked on the Suggested Values menu in the right panel, the user was logged out. This bug has been fixed.

• SQL Data Reader. When Entity members had a line break at the end of their code, the said code would only be saved correctly upon its creation. Running the same Data Reader again would not write any values. This bug has been fixed

• Subscription Hub. When an administrator updated password policy options in the General Settings area, the new options were not displayed correctly and were ignored when users changed their password. This bug has been fixed 
  Subscription Hub. A malfunction in the Subscription Hub configurations caused users to be logged out after some time or to experience difficulties when trying to log into the Subscription Hub. The issue has been fixed

• Subscription Hub. When adding a user to the Approvers list in the Enrollment page, the "Apply" button did not work as expected. The issue has been resolved

• Subscription Hub. The limit of the "Link expiration time (hours)" option under the "User registration emails" menu in the "General settings" section has been extended to 240 hours and now works as expected

• Subscription Hub. A malfunction that caused an erroneous license assignment when importing users from a CSV file has been fixed

• Subscription Hub. In some cases, when exporting users to a CSV file from the Users section, the account status column field (called "DISABLED" in the CSV file) of disabled users was changed automatically from "TRUE" to "FALSE" in the CSV file. The issue has been fixed

• Subscription Hub. A bug allowed users without a Office Add-in license (i.e. the "Office" checkbox in the User profile panel was unchecked) to successfully log into Board from the Excel Add-in. The bug has been fixed

• Subscription Hub. If the Subscription Hub was upgraded to a version that supported Board's Collaboration services without activating them, a notification error would appear when opening Capsules where instances of the previous Chat feature were present. This bug has been fixed

• Subscription Hub. When exporting users in bulk using a .csv file, user attributes did not update correctly. This bug has been resolved

• Subscription Hub. When more than 50 instances attempted to connect simultaneously, the Subscription Hub connection would fail. This issue has been resolved

• Substitution formulas. The "Show Message" Procedure step displayed incorrect Entity member values, which were selected through an "Interactive selection" Procedure step and displayed through Substitution formulas, in the case where a Pager was configured in the same Screen where the Procedure was run. The bug has been fixed

• Substitution formulas. Specific substitution formulas (@datelong, @dateextendend, @datemonth, @dateyear, @dateday, @timelong, @time12, @time24) were not working as expected. This bug has been fixed.

• System and Performance. If the Chrome or Edge browser setting for inactivating tabs after a period of time is enabled and a user accessed the same Board environment in multiple browser tabs of which the inactive tabs were put to sleep after the specific period of time, resuming one of the inactive tabs would log out the user automatically. This bug has been fixed

• User permissions. A restricted user permission allowed access when it shouldn't have. This bug has been fixed

• VDB function. The VDB function is now working as expected in Algorithm Blocks (both in the Layout and in the Dataflow Layout) and in Nexel spreadsheets

• Waterfall. Text translated into other languages using the Localization feature is now displayed as expected, depending on the user's language setting

• Waterfall. In some cases, deviation values displayed in the chart where impossible to read because the space between the two chosen values was too small. The visualization issue has been resolved and a new behavior has been put in place: if the deviation value does not fit in the space between the two arrows, the system will show the value before or after them, depending on the space available in the chart

• Word add-in. If a Board-created Layout that contained an Entity with a Row template applied was opened in a Word document through Word add-in, that Row template would no longer be applied. This bug has been fixed