Build: 23.08.31502. Release date: 31st August 2023.
Highlights
The August 2023 version contains a number of new features, changes and bugfixes. Among the most noteworthy of these, are:
- Images in crosstabs. You can now insert images - e.g., product images, organization member portraits etc. - in your crosstabs along with the other dimension member information. See the full description here: Images in Crosstabs
- Performance improvements. The ways to query data and render objects are now using the "Stateless" technology, meaning that these operations now run in parallel. Obtaining and rendering data - especially for dashboards with many data objects - is improved dramatically.
- Set client language from OpenID script. It is now possible to control the document language in the Anywhere and Windows clients from the OpenID script. See the full description here: OpenID Language support
- TARGIT Mobile 2023. The TARGIT 2023 August version requires a new TARGIT Mobile app. The TARGIT Mobile 2023 comes with new graphics and can be downloaded the respective iOS and Android app stores.
New features
Server and Client
| Server | Open-ID functionality with option to control client language in Anywhere |
It is now possible to control the document language in the Anywhere and Windows clients from the OpenID script by setting the user language. Detail article: OpenID Language support |
| Server | Better support for OpenID Connect |
OpenID users need to be able to schedule jobs without necessarily having to log in on a regular basis - this has been achieved by extending the login script options. Please refer to the TARGIT Community for further information on this. Detail article: OpenID: User info lookup and user token for Scheduled job |
| Server | TARGIT Gateway Client - Performance improvements | Performance has been improved when connecting to on-premise data through the TARGIT Gateway Client when running on a TARGIT Cloud solution. To benefit from this improvement the new TARGIT Gateway Client (build 547) must be downloaded from the TARGIT Control Panel and installed. |
| Server | Logging of response times in milliseconds in auxiliary database |
To get the durations in milliseconds, the TARGIT_LOG table needs to have a REQMSECS column. It will automatically create that one if the TARGIT_LOG table does not exist in the auxiliary database. Detail article: Logging in milliseconds |
| Server | Log in event log when data model is changed | When the data model is changed, the TARGIT Server logs this in the event log. |
| Server | Support for OKTA (OpenID) identity provider | |
| Windows client | Loading limit in crosstab |
To improve general performance when loading larger crosstabs, a limit has been introduced. When loading crosstabs and the number of rows exceed 10.000, the number of rows rendered in the crosstab is limited to 1.000. When having more than 100 columns, the number of rendered columns is limited to 50. To load the full results, you can scroll to the bottom of the crosstab and click the button. Sorting will still be performed on the full data set. Detail article: Crosstab Load Limit |
| Windows client | Images in crosstab |
The opportunity to show images per row in a crosstab based on dimension members and pull an image based on criteria. Detail article: Images in Crosstabs |
InMemory Database
| inmemorydb | If you have a grouping query, with a limit and no metrics, it will try and apply the limit to limit the processing of rows. | |
| inmemorydb | New setting for targitdb.ini If you add autodelete_database_files_on_restart=true to targitdb.ini | It will delete all Targit data files in the data directory, when you restart the tiServer This is designed for people using tiServer as a cacheServer, and to have an easy way to force their targit databases to be rebuilt on restart. Be careful with this feature, because it CAN DELETE YOUR TargitDB FILEs AUTOMATICALLY when you start the tiServer. |
| inmemorydb | New Feature that analyzes joined tables, and if there is a 1-Many Join, to create a simpler table, by pushing parts of the related where clause. | Following Contraints need to apply 1. Joined table need to have < 1,000,000 rows, so that it will applied to Dimension Tables. 2. It must have at least 5 times less rows than the parent table 3. Rows/Values in this column must be at least 2. Two new targitdb.ini settings to manage this feature. 1. filtered_subtables_debug = true will turn extra debug information for this feature 2. filtered_subtables_rowcutoff defines a row cut_off where this feature will be enabled. Set this 0 or 1 to disable this feature. |
| inmemorydb | To make tiServer/tiImport more robust, we are now applying some basic Sanity checks, when reading in columns from DISK | 1. Column and Table Name can now be at most 1024 chars 2. There is a hard limit of 32768 columns per table. 3. We check against negative numbers when reading no of columns and row numbers 4. We check to make sure the row number is the same for all columns in a table. 5. We check to make sure the number of values is <= no of rows. |
| inmemorydb | In tiServer, we have refactored the code to make better use of using in Dot Net to better | |
| inmemorydb | Added support for the characters ` to be used for embedded sql inside tiImport |
InMemory Data Drivers
| Driver: Dynamics AX7 | Verified data driver is compatible with latest release of Microsoft Dynamics 365 for Finance and Operations (10.0.35) | |
| Online Data Provider: Denmark Statistics | A new Online Data Provider for Statistics Denmark is available | See details how to set it up on the portal |
| Online Data Provider: Google Analytics | Google Analytics Online Data Provider for GA4 API |
Google analytics is deprecating their Universal Analytics version (July 2023) and switching to Google Analytics 4 version. Therefore you will have to set up the new Google Analytics driver - please see the TARGIT Community for further information. Detail article: Google Analytics version 4 |
| Online Data Provider: HubSpot | A new Online Data Provider for HubSpot is available |
Detail article: HubSpot |
InMemory ETL Studio
| ETL Studio | New "Detect Numeric Values" setting for JSON data source | This setting allows driver to detect numeric values from string values |
| Scheduler Management | Improved log of events per script is now available in Scheduler Management | It is possible to see the output of a script run in Scheduler Management by browsing and reading the contents of a .log file, created after script run. Additional .events file will be created that contains enhanced log of events that occur during script execution. |
| Scheduler Service | Installation improvements | It is no longer necessary to install Scheduler Service before Data Drivers to be sure that drivers are known in Assemblies in Scheduler Management. |
Data Discovery
| Plugin: JSON | New "Detect Numeric Values" setting for JSON data source | An optional connection string parameter is added to JSON plugin tryDetectNumericValues=true|false (default false) that will try to parse numeric values (integers, doubles) from strings. |
| Plugin: TARGIT Analysis | Authentication token information for TARGIT Analysis data source | Authentication tokens are tied to TARGIT Analysis data sources to request data on behalf of that user. It was not possible to see which user's token is being used on data source's edit form. |
Changes
Server and Client
| Server | Deprecation of Signed Login | Because of technical constraints and the fact that OpenID is used by more and more customers, this version ends support of signed login. Further information on setting up OpenID can be found in the TARGIT Community. |
| Server | ROLAP: simplify >, <, >= and <= criterias | If there is a criteria using <, >, <= or >=, the ROLAP engine will always do something like this (example Time > 2020 Q1): (YEAR > 2020) OR (YEAR = 2020 AND QUARTER > 2020Q1) Regardless of the status of the quarter level. The query gets increasingly complicated the more levels there are in the hierarchy. If the hierarchy is a natural hierarchy - or the quarter level is defined as unique, you can skip the year part and just write: QUARTER > 2020Q1 If a sort column is involved, you should not do it because even if the hierarchy is configured as a natural hierachy, you cannot be 100% sure that the sort column is unique as well. |
| Windows client | Deprecate 'Ignore all changes' setting | The criteria option, 'Ignore all changes', in the right-click menu should be removed. Technically the set of criteria saved when enabling this setting should be converted to local criteria instead. This conversion happens at runtime and is saved in the document when the document is saved. |
| Windows client | Ad-hoc user dimensions by "magic drop" is deprecated | The functionality of creating user dimensions by dragging a dimension in a dashboard onto another dimension is deprecated. |
| Windows client | Deprecate automatic trend arrows in gauge objects | When creating gauges based only on a measure, a trend arrow may in specific cases be added to the gauge. In addition, if KPI values have been defined in the back-end (Analysis Services), these may be added to the object as well. This functionality will be removed because the designer should be able to control what goes into the object. This change does not reduce what the designer can do with a gauge. Only the automatic configuration with extra trend gauge and automatic time dimension selection etc. is removed. |
| Windows client | Scheduled classic reports are no longer displayed in schedule list | With the upcoming release, scheduled classic reports are no longer displayed in the list of scheduled jobs, hence it is no longer possible to convert schedules of classic reports to the new report layouts. Please make sure all schedules points to the new report layouts before upgrading to this release. |
| Windows client | Deprecation of auto expand/collapse of other objects when selecting in time slider |
InMemory Data Drivers
| Data Drivers | Upgrade to .NET 7 | An upgrade to .NET 7 allows further improvements and performance boost of Scheduler Service and Data Discovery. |
| Driver: OData | Deprecate support of OData V1-V3 versions | OData V1-V3 has been superseded by OData V4, which is an industry standard |
| Online Data Provider: Dynamics BC | Batch size parameter is added | Batch size parameter allows more precise control over the amount of data being retracted from Dynamics BC. Typically best performance is given by batches of 10,000-100,000 items. |
InMemory ETL Studio
| ETL Studio | Upgrade to .NET 7 | An upgrade to .NET 7 allows further improvements and performance boost of Scheduler Service. |
| Profiler | Upgrade to .NET 7 | An upgrade to .NET 7 allows further improvements and performance boost of Profiler. |
| Scheduler Management | Upgrade to .NET 7 | An upgrade to .NET 7 allows further improvements and performance boost of Scheduler Management. |
| Scheduler Service | Upgrade to .NET 7 | An upgrade to .NET 7 allows further improvements and performance boost of Scheduler Service. It may cause .Net Framework drivers to stop working. If that happens, please, update your drivers to corresponding .NET 7 version. |
Data Discovery
| Data Discovery | Default schedule of a data source is changed from midnight to it's creation time | This is done to evenly distribute data source reloads throughout the day. This change does not affect existing data sources and schedules. |
| Data Discovery | Upgrade to .NET 7 | An upgrade to .NET 7 allows further improvements and performance boost of Data Discovery. |
| Data Discovery Back-end | Error messages are improved when cube cannot be created | |
| Data Discovery Back-end | Improved RAM consumption | This change improves RAM consumption, especially during Data Discovery's initialization phase |
| Data Discovery Front-end | Options to modify data source's metadata is removed from Preview data source window | It was possible to adjust metadata types of columns using Preview data source window. This functionality is now removed. If you need to adjust metadata options you can use a Column Types window. |
| Plugin: OData | Deprecate support of OData V1-V3 versions | OData V1-V3 has been superseded by OData V4, which is an industry standard |
| Plugin: TARGIT Analysis | Not possible to change ANTserver address for TARGIT Analysis data sources | Previously a data source stored information about the TARGIT Server's location and used it. In the current release this information is no longer stored in data source, and is obrained during processing of this data source. This is especially helpful when TARGIT Server is being migrated and changes it's address. |
Anywhere
| Anywhere | Print of drillthrough no longer supported |
Bugfixes
Server and Client
| Management | Error moving tabs in data model | When moving tabs in data model editor, it does not crash anymore. |
| Management | Query from Data Modeler in Preview Table gets a blank result | |
| Management | Cannot Delete a Private Dimension | |
| Management | Error using login in parametric roles | |
| Server | ROLAP: Ignore unrelated dimensions + disable multidimensional subtotals can cancel out eachother | |
| Server | TARGIT server will go into maintenance mode when running a Schedule job on the 28th | A job scheduled to run on the 28th of a month specifically causes the TARGIT Server to enter maintenance mode. |
| Server | When logging in via an identity provider, the session expires after 2 hours | After 2 hours, user is met by an "access denied" when trying to continue, and has to close down the client and start again. |
| Server | Login using refresh tokens does not work with Windows Security | Data Discovery uses a refresh token to log on to the TARGIT Server when refreshing a data source from a .xview file. If the server is configured to use Windows Security, it did not work. |
| Server | Scheduled jobs do not respect forced criteria | Forced criteria do not apply to scheduled jobs potentially causing a risk of more data being displayed than intended. |
| Server | Member cache does not take cube security into account | If a SSAS cube has roles defined that restricts certain dimension members depending on the user - and using impersonation to connect to that database (i.e. Effective User), there is a member cache that does not take the user into account, so a hierarchy member discovered by a user that has access to this member will also be seen by a user with less access, causing the TARGIT Server to believe that the user has access to the member, but eventually failing at a later point - due to the user not having access to the hierarchy member. |
| Server | Cannot edit shared user dimension used in a crosstab notification agent | If using insights, adding notification agent to a user dimension will block editing the user dimension. |
| Server | TARGIT Server hanging in cloud | Occasionally the TARGIT Server "Hangs" and it is not possible to logon to either client. |
| Server | Scheduled job failed: Non-interactive login script failed | When authenticating through OpenID an issue would cause scheduled jobs to fail if the owner of the jobs has been logged in since last update. |
| Server | "Too many axes in answer" error in criteria requests through the data gateway | When connecting to on-prem data sources through the data gateway from TARGIT Cloud and empty members are fetched in the criteria bar, the dimension is empty and the error "Too many axes in answer" is logged. |
| Server | Analysis Services: when reading the response from Analysis Services, the TARGIT Server can enter an infinite loop if the response is incorrect in some way | |
| Server | Connection between TARGIT Server and the Data Gateway is sometimes lost | |
| Server | MDX connections: Drillthrough only returns one row | |
| Server | MDX connections: Drillthrough does not respect forced criteria | |
| Server | Invalid SQL generated then ragged hierarchies are enabled | |
| Server | Errors when sending mail during scheduled jobs are not being logged correctly in the job's log | |
| Server | Identity provider login should support query parameters on the authorization endpoint URL | |
| Windows client | "Designed for" not default enabled on new Dashboard | "Designed For" is now default set when creating a new Dashboard layout |
| Windows client | "Next run" column missing in Scheduled Jobs overview | "Next run" column have been added to the Scheduled Jobs overview |
| Windows client | When creating a new analyze the "Bookmark" button is grayed out, but the option is still in the right-click menu | . |
| Windows client | No confirmation dialog when deleting shared user dimension | A confirmation does not appear when deleting a shared user dimension that is not yet used in a dashboard thereby introducing the risk of unintentional deleting shared user dimensions. |
| Windows client | Advanced calculations causes error on Map charts with Empty Result | Adding Measure/Column in Advanced calculations will cause error for Map charts with Empty Result = Hide Object setting |
| Windows client | Calculation returns wrong result | Adding x:x in the dimension does not return the correct result for the last column |
| Windows client | Auto-Filter in User Dimensions doesn't work | Auto-filter has no impact on shown dimensions members when creating a user dimension |
| Windows client | Scheduled jobs - listing changes after editing a job | Editing a scheduled job, sends the job to the end of the list - that means that the sorting of the jobs changes. |
| Windows client | Initial criteria only affects objects for one cube | Having initial criteria from more than one cube, only the first cube was affected. |
| Windows client | Cannot login to clients when user is member of many Active Directory groups | If a user is member of many Active Directory groups (100+), an error is displayed when trying to login on both the Windows and Anywhere clients. |
| Windows client | Tabular: “Missing member” in global criteria | Improved the criteria request measures decoration editor to have simple/advanced modes, where simple is how the decoration behaved previously. You need to disable the troublesome measures in the decorations editor. |
| Windows client | When mapping a member to a point decimals aren't accepted | In an area map when mapping members to a point, decimals are now accepted, which is consistent with the definition of map edges. |
| Windows client | Error when opening scheduled jobs dialog | In some cases when opening the scheduled jobs dialog an error is displayed: "One or more errors occurred. (Object reference not set to an instance of an object.)". We found this to be related to jobs being set up on a schedule according to cube processing. |
| Windows client | Issue with Menu and text box templates | Problem was that anywhere was not able to handle menu files when the document links were in the template of the menu items. |
| Windows client | TARGIT crashes in shared user dimensions dialog | TARGIT no longer crashes when pressing OK button several times fast after each other. |
| Windows client | Scheduled jobs - Log and Latest result are misaligned if a job runs for a long time (> 5 minutes) | The list of scheduled jobs is now also checking for running job. |
| Windows client | Scheduled job without a schedule makes the joblist fail | The scheduled jobs list cannot be opened if one of the jobs in the list is missing a schedule. |
| Windows client | "Apply to all cells in row" option in color agent dialog has been renamed | To make the effect more clear, the option has been renamed to: "Apply to all measure and calculation cells in row". |
| Windows client | Scheduled job fails due to insufficient data | When a scheduled batch job fails with the error "Scheduled job failed - Insufficient data to build the chart", the job stops and later members in batch are not executed. |
| Windows client | Last modified and cubes not showing correct result | When filtering on eg. last modified date or cubes, the folders/files shown was not correct. |
| Windows client | TARGIT crash when first cube has errors | When having a cube with known errors (missing table or join) in list of cubes in TARGIT Management and you open TARGIT client -> New -> Empty dashboard, then the client crashes |
| Windows client | Color Agents on user dimension member shows xmlui error if member name is changed | When opening an agent based on a user dimension member, where member name has been changed an xmlui error is displayed. |
| Windows client | Delivery column is incorrect for 'Published' jobs | With Published jobs the delivery tabs in the Schedule Jobs windows, shows the wrong icon (mail) and wrong message "Mail to" |
| Windows client | Shared Dimensions: using custom values and allow multiselect doens't work | You can not multiselect when using custom values for userdimension |
| Windows client | TARGIT Client crashes when opening the Properties Dialogue for a TEXT BOX object and MENU object. | |
| Windows client | TARGIT analysis datasource fails with 500 Internal Server error when an object is unavailable | |
| Windows client | Scheduled Job: Search for mail recipients not possible anymore | |
| Windows client | TARGIT crashes when double clicking on 'Source' in Scheduled Jobs list | |
| Windows client | When creating or editing local and global criteria is User Dimension not translated to German. Userdimension = Benutzer Dimension | |
| Windows client | Scheduled job - deleting a job that is running, crashes TARGIT | |
| Windows client | Scheduled job and log does not display same date/time format | |
| Windows client | Enabling the Ragged Hierarchy Depth Column causes reports with Global Calculation to fail. | |
| Windows client | No Error Message when All Licences used | |
| Windows client | TARGIT InMemory behaves different when a user dimension is introduced | |
| Windows client | Email address for error notification is not saved in scheduled job list | |
| Windows client | Roles: Can't see existing and newly added members | |
| Windows client | Measure Selector Incorrect Formatting | |
| Windows client | Document descriptions disappear in translation | |
| Windows client | Iteration table is still in file after deleting page with iteration | |
| Windows client | Labels: Deleting Lookup value causes access violation | |
| Windows client | Crosstab looses theme if data source has changed (e.g. measure renamed) | |
| Windows client | Top list does not work in combination with user dimensions | |
| Windows client | Initial criteria on same dimension in two roles only displays data for one of the members |
InMemory Database
| inmemorydb | Fixed an IO Problem in reading TargitDB data. | If the server connection was terminated, while reading in the Values part of a column It would cause tiImport to hang, when the lazyload issue was triggered on the server. |
| inmemorydb | Fixed problem so Dot Net 7 files are copied automatically | |
| inmemorydb | Fixed a potential issue with CompileTable in InMemoryTable. | This was causing threads to wait 1-2 hours to timeout, while waiting for compileTable to finish There is a new Disposable BOole flag, that gets resets, wrapping the main table compilation, so if an exception is thrown during compilation, It should cause any waiting threads, to stop waiting. |
| inmemorydb | Fixed a separate issue with lazyloading where the row count check on the column was being applied incorrectly. | |
| inmemorydb | If a column is being lazyloaded, there is a chance the background garbage cleaner, can unload it during the load process, corrupting the state, especially the larger the column | |
| inmemorydb | Fix for case statements with more than 23 entries taking exponentially longer to run. | |
| inmemorydb | Correlated queries now work better, when NOCACHE mode is turned on. | Previously, if an inner query contained a correlated subquery, and nocache mode was on, then required terms of the correlated query could be removed by the UNUSED column algorithm causing a problem. |
| inmemorydb | Fixed a problem with Remote Streamable INMEMORYDB Data sources in tiImport | |
| inmemorydb | Performance improvements when you have non aggregative queries that have 1 or more tables and a whereclause. | 35-50% reduction in execution time, when you have many columns. Timing improvement only applies to columns, that Are explicitly database fields. Internally it just track tuples of the row numbers, instead of the lookup values for each column. Temporary memory usage is reduced as well. |
| inmemorydb | The error, "Could not find column columnName in list of tables" | When displayed for complicated queries was confusing to debug, especially in big nested queries. This error message will now contain a message, for any tables that may be missing in that part of the query from the database. |
| inmemorydb | Security enhancement to remove Autorization=Bearer and Authorization=Basic from the SQL Statement displayed on screen during tiImport | This fix does not apply to the explicit error message sent back by the database itself though. |
| inmemorydb | Now before writing a column, there is an expliciy check to see if the column is unloaded, and at least generate an exception. | |
| inmemorydb | Before writing a table, there is a final check to see if the columns are loaded, and an Error message generated if they are not. | This could happen if they are unloaded, despite being loaded earlier, and could be triggered, if the unload setting is too low |
InMemory Data Drivers
| Data Drivers | Do not wait for telemetry responses | During data driver's execution requests to telemetry service can be executed - notifying it of errors or notifying it that the driver is being used. In 2022 Spring release a change to the drivers was made to avoid loosing telemetry data and to wait for response from telemetry whenever a request to it was executed. This leads to affected performance of a driver and additional time being consumed in cases when there is a heavy driver usage and lots of queries being executed. |
| Driver: DB2 | DB2 driver does not work | Fixed issues that lead to driver not being able to query data from a DB2 database |
| Driver: Excel | Querying data by column name does not work in Excel driver | |
| Driver: Excel | "Handle empty rows and 0 values" setting is not applied | This bug caused rows with empty and 0 values be treated as data and appear in result table |
| Driver: FileList | Spaces around "=" sign in FileList data source conditions causes condition to be ignored | For example in queries like this: [FileList].{SELECT * FROM [C:\Folder] WHERE filename = '.csv' AND directory = 'full'} |
| Driver: JSON | RAM consumption improvement during Online data provider execution | Out-of-memory exceptions could occur during provider's execution when big data sets were retracted. |
| Driver: JSON | Integer and double column types are not properly detected | This bug caused JSON driver to ignore integer or double column types and interpret it as string columns. An optional connection string parameter is added to JSON driver tryDetectNumericValues=true|false (default false) that will try to parse numeric values (integers, doubles) from strings. |
| JS Connector | "Use display names as column names" option is added to SharePoint Designer online data provider | This option allows switching between internal and display names of fields in SharePoint that will be used as column names in resulted table. |
| Online Data Provider: Dynamics BC | Numeric data types are not properly detected on Dynamics BC online data provider | |
| Online Data Provider: Dynamics BC | Fixed error 401 (Unauthorized) for scripts that run more than an hour | An authentication token obtained during script execution lives for 1 hour by default, which could cause 401 errors if script execution exceeds this time limit. |
| Online Data Provider: Dynamics CRM | Provider might not retract all the possible data | |
| Online Data Provider: Jobindsats.dk | Jobindsats provider throws error if no dynamic period is specified | |
| Online Data Provider: SharePoint | SharePoint Online data provider only retracts first 100 rows | |
| Online Data Providers | Metadata options are not passed to underlying JSON driver | This could affect data types during data reloads |
InMemory ETL Studio
| Data Drivers | Do not wait for telemetry responses | During data driver's execution requests to telemetry service can be executed - notifying it of errors or notifying it that the driver is being used. In 2022 Spring release a change to the drivers was made to avoid loosing telemetry data and to wait for response from telemetry whenever a request to it was executed. This leads to affected performance of a driver and additional time being consumed in cases when there is a heavy driver usage and lots of queries being executed. |
| Driver: DB2 | DB2 driver does not work | Fixed issues that lead to driver not being able to query data from a DB2 database |
| Driver: Excel | Querying data by column name does not work in Excel driver | |
| Driver: Excel | "Handle empty rows and 0 values" setting is not applied | This bug caused rows with empty and 0 values be treated as data and appear in result table |
| Driver: FileList | Spaces around "=" sign in FileList data source conditions causes condition to be ignored | For example in queries like this: [FileList].{SELECT * FROM [C:\Folder] WHERE filename = '.csv' AND directory = 'full'} |
| Driver: JSON | RAM consumption improvement during Online data provider execution | Out-of-memory exceptions could occur during provider's execution when big data sets were retracted. |
| Driver: JSON | Integer and double column types are not properly detected | This bug caused JSON driver to ignore integer or double column types and interpret it as string columns. An optional connection string parameter is added to JSON driver tryDetectNumericValues=true|false (default false) that will try to parse numeric values (integers, doubles) from strings. |
| JS Connector | "Use display names as column names" option is added to SharePoint Designer online data provider | This option allows switching between internal and display names of fields in SharePoint that will be used as column names in resulted table. |
| Online Data Provider: Dynamics BC | Numeric data types are not properly detected on Dynamics BC online data provider | |
| Online Data Provider: Dynamics BC | Fixed error 401 (Unauthorized) for scripts that run more than an hour | An authentication token obtained during script execution lives for 1 hour by default, which could cause 401 errors if script execution exceeds this time limit. |
| Online Data Provider: Dynamics CRM | Provider might not retract all the possible data | |
| Online Data Provider: Jobindsats.dk | Jobindsats provider throws error if no dynamic period is specified | |
| Online Data Provider: SharePoint | SharePoint Online data provider only retracts first 100 rows | |
| Online Data Providers | Metadata options are not passed to underlying JSON driver | This could affect data types during data reloads |
Data Discovery
| Data Discovery | Cubes can randomly break after restart of Data Discovery | This bug occurs if Data Discovery needs to reload lots of cubes during it's initialization phase |
| Data Discovery | Cube breaks after being queried and then reloaded | |
| Data Discovery | Unable to update cube due to data source's metadata changed | This bug was introduced in 2022 Winter release and is caused by the following conditions: 1) You have data sources or certain type: Excel, CSV, JSON, TARGIT Analysis 2) Out-of-process creation of data sources setting is enabled (this setting is turned on by default in 2022 Winter release) 3) Metadata types for these data sources was manually adjusted from the UI via Column Types or Preview window 4) A cube contains one or more of such data sources |
| Data Discovery | DataDiscovery notification settings reset after restart | |
| Data Discovery | Data discovery is not starting after restart until UI is opened for the first time | Fixed the issue by enabling "Preload enabled" web site setting on IIS for data service's web site |
| Data Discovery | Renamed cube will exist with old and new name in Source Data of TARGIT Windows client | |
| Data Discovery | Cube remains in 'Restore pending' state after restart | This bug is caused when a cube contains a data source that points to a local file and this file was removed during the time Data Discovery was shutdown. |
| Data Discovery Back-end | Stop data discovery if errors occur during initialization phase | It could happen that data discovery experiences an error during it's initialization phase and still tries to operate, however that could lead to various unrelated errors. |
| Data Discovery Back-end | First query to the cube fails with error and causes Data Discovery to hang unresponsively | |
| Data Discovery Back-end | JobindsatsDK datasource formats are unavailable after reload | Fixed issues with column names being incorrectly generated for parent data source when 'Enable data source generation out of process' setting is enabled |
| Data Discovery Back-end | Cube's targitdb file remains on disk after cube was renamed | |
| Data Discovery Back-end | Cannot create a cube with data source named "Dates" that contain date time field value | |
| Data Discovery Back-end | Data Source Format remains in 'Restore pending' state if parent data source is broken | |
| Data Discovery Front-end | Dates on UI are always shown in US format | Dates on UI will be shown according to culture of user's TARGIT Windows client |
| Data Discovery Front-end | Subcube designer's delete button does not properly work | |
| Data Discovery Front-end | Data Source Format is invisible on UI if parent data source is broken | |
| Data Discovery Front-end | Unable to call context menu on data sources | It was not possible to call context menu of a data source after opening and closing it's Column Types window. |
| Data Discovery Front-end | Multiple UI improvements and fixes | |
| Plugin: CSV | Incorrect error message when loading CSV data source | A misleading error was shown for files containing malformed CSV values. It will now display exact line number and place where the malforming is. |
| Plugin: DB2 | DB2 plugin does not work | Fixed issues that lead to plugin not being able to query data from a DB2 database |
| Plugin: Denmark Statistics | Minor fix to data source's form | |
| Plugin: ECB | Minor fix to data source's form | "From start date to today" option was always visible. |
| Plugin: Excel | Querying data by column name does not work in Excel plugin | |
| Plugin: Excel | "Handle empty rows and 0 values" setting is not applied | This bug caused rows with empty and 0 values be treated as data and appear in result table |
| Plugin: MySQL | Multiple improvements and fixes | |
| Plugin: PostgreSQL | Plugin's edit form automatically overwrites SQL query | |
| Plugin: Python | Unable to save data source due to validation error | It was not possible to create data source if metadata option was set to "Manual" and any column name was specified. |
| Plugin: TARGIT Analysis | TARGIT Analysis data source mixes thousand/decimal separators on reload | Data source could be successfully created, data types could be properly detected, however after a data source gets reloaded it messes up decimal separators leading to incorrect data |
| Plugin: TARGIT Analysis | Endless loader is shown on data source's edit form | Edit form was showing endless loader if target analysis was missing on the server. |
Anywhere
| Anywhere | Usage data from Anywhere is not logged to Insights | |
| Anywhere | Criteria containing & does not work when used in Generic embedding | |
| Anywhere | 'Unable to render chart' error when loading dashbord | In some cases objects wrongfully display the error 'Unable to render chart'. |
| Anywhere | Export to Excel in Anywhere includes unused objects | When exporting from anywhere to excel all unused objects were also exported. |
| Anywhere | “Repeat dimensions” does not work for exports to Excel | |
| Anywhere | Open Previous document (Last drill state) trigger behaves wrong in Anywhere |
Comments
Please sign in to leave a comment.