Skip to main content

Aperture Data Studio release notes

Release 1.1.4

Jul 2, 2018

This release includes a number of high priority bug fixes and a few minor new features.

New features

  • Added a no delimiter option to the configuration of files.
  • Added a new option allowing administrator users to download log files via the browser. To do this, click on your username in the top menu and select 'Download log files'.
  • Updated the versions of Experian Match (2.9.0), Standardize and included the latest default USA match rules and blocking keys.

Bug fixes

  • Users will no longer see an error when attempting to load large (over 10 million row) tables from SQL Server using the default driver.
  • ‘GC overhead limit reached’ error messages no longer appear when joining two files with a very large number of unique values.
  • Selecting files with similar names will now always select the correct file.
  • Filters containing invert logic in the multi-compare function no longer filter out all rows.
  • It's now possible to delete additional administrator users.
  • Datatype-specific functions won't output an error message when a null value is supplied as the input.
  • The mapping dialog will appear correctly when a workflow is imported where the source file is not present.
  • Improved the way that the source mapping dialog handles long names.
  • Certain functions (e.g. compare) will no longer standardize the supplied input strings.
  • You can now duplicate more than one column from the explorer menu of a Transform step.
  • Snapshots taken after renaming a workflow will now have the correct workflow name and be accessible (snapshots taken before the name change will retain their original workflow name).
  • ‘DatabaseException’ stack traces are no longer thrown when removing files via the UI.
  • Improved the performance of using snapshots as lookup tables.
  • The edit column dialog will no longer replace the tagging dialog in preview.
  • Batch all datasets license keys will now be accepted in the Data Studio 'Update License' dialog.
  • The correct error message from data-type specific functions will now be displayed for each row (rather than duplicating the first message every time).
  • Retrieving the base license on server start-up will no longer throw an OutOfMemory error when there is less than 2 GB available.
  • Dragging one function on top of another in the expression editor will not throw a NullPointer exception.
  • Fixed the ‘Extract Date/Time’ function to apply the correct time zone offset.
  • Viewing the filter on a Split step of an imported workflow before viewing the data will not throw a NullPointer exception anymore.
  • Changing the source file in a workflow and not viewing the data will now correctly update the created snapshots on execution.
  • Reporting on a workflow with no source file defined will now show a meaningful error message.
  • The Validate emails step no longer returns a false negative for certain domains.
  • The Find duplicates step will now correctly trigger a matching job to be run when placed after a Split step.
  • Fixed a refresh issue with the selected columns validation behavior on the Find duplicates step.
  • Removed the unused scale and precision options from columns in preview and configure. 

Known issues

Database
Chinese characters or numeric values as column headers in source data are standardized incorrectly on load.

Profiling
The 'Documented type' column is empty in Profile view.

Grouping

  • Grouping columns which contain large numbers of unique values may incorrectly return 0 rows (this occurred in testing between 500k to 1M unique values with default memory settings).
  • Re-arranging columns in the side menu may result in unexpected results. 

Transform step
The 'Download as csv' option doesn't reflect columns being moved/re-ordered unless the grid is re-opened.

Validate and Join steps
These known issues stem from the fact that both steps store their rules/key columns in the previous workflow step.

A workaround for all of these issues is to add a Transform step in between the data source and the Validate/Join step.

  • Using a Validation step immediately after a Join, Splice or Union step may only show the columns originating from the first of the two inputs in choosers.
  • Validate and Join steps will not save rules or join columns when used immediately after the scripting (JS / R / Python) step.
  • Validation rules can be lost when the following steps precede the Validation step: Union, Multi-View, Splice, Validate emails, Chart, Use snapshot and custom steps.
  • Deleting a workflow step immediately preceding a Validation step will cause the validation rules to be lost.
  • A Join step applied immediately after any of the following steps will cause the key columns to be lost, resulting in an invalid join: Find duplicates, Script, Use Snapshot, Union, Multi-View, Splice, Validate emails and Chart.
  • Using a Join step after a step with multiple outputs will only allow you to store one set of key columns. If you wish to perform two different joins on the two outputs, you will have to insert a Transform step in between this step and the Join step(s).

Lookups

  • Attempting to edit a column that is already being used in a lookup causes the edit to fail without showing an error (the only indication of this is an exception in the log).
  • When performing a lookup transformation in a workflow on data which has been joined, the lookup values returned only refer to the original source file rather than the joined rows. This will show as incorrect or missing results for Lookup list, Lookup min/max and Lookup first/last transformation functions.

Workflow Designer

  • Undo and redo actions exhibit unexpected behavior in some cases.
  • JDBC column mappings for exports can't be deleted after saving. To make changes, you will have to re-save the Export step and start again.
  • The configuration changes (e.g. data tags) to source files in workflows aren't updated when changed in Data Explorer. To pick up the new configuration, they will have to be deleted and then replaced.
  • The Take Snapshot step doesn't ghost out 'Show data' when the input becomes invalid. Clicking on it will display a blank view.
  • Installed dataplus items that aren't licensed will still appear in the list of ‘Additional data’ on the Validate addresses step. Selecting unlicensed dataplus items will return empty columns.

 

Character encoding 

  • The 'Download as .CSV' option doesn't export multi-byte characters.
  • The Export step will always default to the Windows-1252 character set. Trying to export other character sets will show an unmappable character exception. You can fix this by changing the character set in the advanced settings of the Export step.

Snapshots
Renaming a workflow after having taken snapshots won't assign the existing snapshots to the new workflow name.

Scripting
Multiple R script steps in a workflow will result in unexpected results.

User management
It's currently not possible to create a user with a same name as a previously deleted user.

UI

  • Setting the browser zoom to 110% may cause the bottom bar to be cut-off.
  • Reordering columns after using Fix First Columns results in column data appearing in incorrect locations.
  • Users will occasionally be automatically logged out when configuring files.

     

SQL Server
Connecting to multiple SQL Server instances on the same server causes the instance name to be ignored. 

Date/time/time zone
In some cases, changing the locale settings on a file won't change how dates are interpreted. 

Data Explorer
When changing the delimiter in Preview and Configure, the column counts and data types are not updated automatically. 

Release 1.1.3

May 18, 2018

This release includes a number of high priority bug fixes. Download Data Studio 1.1.3. Download ODBC drivers 1.1.3.

Bug fixes

  • The ODBC drivers installer now creates correct registry settings for the 32-bit driver.
  • We've made several fixes related to SQL Server:  
    • You can now preview and load data from other schemas, not just the connected user’s default schema.  
    • You can now use Microsoft’s latest SQL Server driver (mssql-jdbc-6.2.2.jre8.jar) to create custom connections.  
    • You can now load views from SQL Server. 
  • The JDBC exports will now log a single representative SQL statement for exports when debugging is enabled.  
  • Applying an invalid Data Studio license is now handled correctly and doesn’t prevent further use of the product.  
  • Defining a filter on a Split step after the Validate Addresses step will no longer throw a NullPointer exception.  
  • Clicking on column choosers for files with over 100 columns will no longer throw an IndexOutOfBounds exception.  
  • Importing a workflow from a previous version will no longer throw a NullPointer exception.

Known issues

Database
Chinese characters or numeric values as column headers in source data are standardized incorrectly on load.

Profiling
The 'Documented type' column is empty in Profile view.

Grouping

  • Grouping columns which contain large numbers of unique values may incorrectly return 0 rows (this occurred in testing between 500k to 1M unique values with default memory settings).
  • Re-arranging columns in the side menu may result in unexpected results.

 

Transform step
The 'Download as csv' option doesn't reflect columns being moved/re-ordered unless the grid is re-opened.

Find duplicates step
The Find duplicates step immediately after a Split step will not execute when the workflow is executed.

The workaround is to add a Transform step in between these two steps.

Validate and Join steps
These known issues stem from the fact that both steps store their rules/key columns in the previous workflow step.

A workaround for all of these issues is to add a Transform step in between the data source and the Validate/Join step.

  • Using a Validation step immediately after a Join, Splice or Union step may only show the columns originating from the first of the two inputs in choosers.
  • Validate and Join steps will not save rules or join columns when used immediately after the scripting (JS / R / Python) step.
  • Validation rules can be lost when the following steps precede the Validation step: Union, Multi-View, Splice, Validate emails, Chart, Use snapshot and custom steps.
  • Deleting a workflow step immediately preceding a Validation step will cause the validation rules to be lost.
  • A Join step applied immediately after any of the following steps will cause the key columns to be lost, resulting in an invalid join: Find duplicates, Script, Use Snapshot, Union, Multi-View, Splice, Validate emails and Chart.
  • Using a Join step after a step with multiple outputs will only allow you to store one set of key columns. If you wish to perform two different joins on the two outputs, you will have to insert a Transform step in between this step and the Join step(s).

Lookups

  • Attempting to edit a column that is already being used in a lookup causes the edit to fail without showing an error (the only indication of this is an exception in the log).
  • When performing a lookup transformation in a workflow on data which has been joined, the lookup values returned only refer to the original source file rather than the joined rows. This will show as incorrect or missing results for Lookup list, Lookup min/max and Lookup first/last transformation functions.

Workflow Designer
Undo and redo actions exhibit unexpected behaviour in some cases.

Character encoding 

  • The 'Download as .CSV' option doesn't export multi-byte characters.
  • The Export step will always default to the Windows-1252 character set. Trying to export other character sets will show an unmappable character exception. You can fix this by changing the character set in the advanced settings of the Export step.

Snapshots
Renaming a workflow after having taken snapshots in the workflow causes the snapshots to be created in the wrong location and to be inaccessible.

Scripting
Multiple R script steps in a workflow will result in unexpected results.

User management
It's currently not possible to create a user with a same name as a previously deleted user.

UI

  • Setting the browser zoom to 110% may cause the bottom bar to be cut-off.
  • When mapping sources after having imported a workflow without the source file present, the mapping dialog does not handle long source file names or data source names well.
  • Reordering columns after using Fix First Columns results in column data appearing in incorrect locations.
  • Users will occasionally be automatically logged out when configuring files.
  • In some instances the Edit column dialog replaces the Tag column one when tagging columns.

     


SQL Server

  • Users will see an error when attempting to load large (over 10 million row) tables from SQL Server using the default driver.
  • Connecting to multiple SQL Server instances on the same server causes the instance name to be ignored. 

 

Date/time/time zone

  • In some cases, the Extract Date/Time functions won't apply the correct time zone offset.
  • In some cases, changing the locale settings on a file won't change how dates are interpreted.

 

Data Explorer
When changing the delimiter in Preview and Configure, the column counts and data types are not updated automatically. There's no option for ‘no delimiter'.

Release 1.1.2

May 4, 2018

This release includes a number of high priority bug fixes as well as several new features. Download Data Studio 1.1.2Download ODBC drivers 1.1.2.

New features

  • A new 'Current limits' column has been added to the My Aperture Data Studio licenses page showing the server's current licensed limits.
  • A new option in the Analyze trends step now allows you to display categories evenly along the X-axis when displaying results in a chart.
  • A new 'Clear saved results' option in the Find duplicates step now allows you to clear results of previous step execution.
  • JDBC data source connections can now be debugged using the 'Debug connection' option in the Create/edit data source dialogs.
  • SDK example steps have been updated to showcase the new functionality introduced in recent releases.
  • SDK now supports an option to allow the selection of multiple columns.
  • Improvements of the license request process to cover regional business operations.
  • Updated to the latest Standardize knowledge bases to v4.0.1 for the following countries and territories:
    • AUS (Australia)
    • BRA (Brazil)
    • CAN (Canada)
    • FLK (Falkland Islands)
    • GBR (Great Britain)
    • GGY (Guernsey)
    • GIB (Gibralter)
    • IDN (Indonesia)
    • IMN (Isle Of Mann)
    • JEY (Jersey)
    • MYS (Malaysia)
    • NZL (New Zealand)
    • PRI (Puerto Rico)
    • SGP (Singapore)
    • THA (Thailand)
    • USA (United States of America)

Bug fixes

  • The Group step menu is now updated when an aggregate column is added.
  • Changes are now saved when editing hidden column details on a Join step.
  • Update for a MongoDB data source now doesn’t throw a null pointer error when pre-check is turned on.
  • MongoDB now reconnects after a server restart.
  • Drilling down from an integer value in a unique values list in Profile now returns results.
  • Cached data is now deleted when the source file is removed via the UI.
  • Hiding columns or quick filtering in any step connected to the 'Failing rows' output of a Split step no longer causes incorrect data to be returned.
  • Using the ‘Replace’ action to replace a source file in a workflow will now update the file to the intended one.
  • 'Distinct' button on an aggregate column is now not reset in the UI when the aggregate is edited.
  • Duplicate files now don’t appear in 'My files' when uploading a file on new database creation, prior to a restart.
  • On first login as LDAP users are not asked to change their password anymore.
  • Amazon S3 files can now be loaded from a sub-directory.
  • Removing a file from a multi-sheet Excel spreadsheet and viewing other sheets now doesn't cause an error.
  • The 'Hash code' function now returns results.

Known issues

Database
Chinese characters or numeric values as column headers in source data are standardized incorrectly on load.

Profiling
The 'Documented type' column is empty in Profile view.

Grouping

  • Grouping columns which contain large numbers of unique values may incorrectly return 0 rows (this occurred in testing between 500k to 1M unique values with default memory settings).
  • Re-arranging columns in the side menu may result in unexpected results.

 

Transform step
The 'Download as csv' option doesn't reflect columns being moved/re-ordered unless the grid is re-opened.

Find duplicates step
The Find duplicates step immediately after a Split step will not execute when the workflow is executed.

The workaround is to add a Transform step in between these two steps.

Validate and Join steps
These known issues stem from the fact that both steps store their rules/key columns in the previous workflow step.

A workaround for all of these issues is to add a Transform step in between the data source and the Validate/Join step.

  • Using a Validation step immediately after a Join, Splice or Union step may only show the columns originating from the first of the two inputs in choosers.
  • Validate and Join steps will not save rules or join columns when used immediately after the scripting (JS / R / Python) step.
  • Validation rules can be lost when the following steps precede the Validation step: Union, Multi-View, Splice, Validate emails, Chart, Use snapshot and custom steps.
  • Deleting a workflow step immediately preceding a Validation step will cause the validation rules to be lost.
  • A Join step applied immediately after any of the following steps will cause the key columns to be lost, resulting in an invalid join: Find duplicates, Script, Use Snapshot, Union, Multi-View, Splice, Validate emails and Chart.
  • Using a Join step after a step with multiple outputs will only allow you to store one set of key columns. If you wish to perform two different joins on the two outputs, you will have to insert a Transform step in between this step and the Join step(s).

Lookups

  • Attempting to edit a column that is already being used in a lookup causes the edit to fail without showing an error (the only indication of this is an exception in the log).
  • When performing a lookup transformation in a workflow on data which has been joined, the lookup values returned only refer to the original source file rather than the joined rows. This will show as incorrect or missing results for Lookup list, Lookup min/max and Lookup first/last transformation functions.

Workflow Designer
Undo and redo actions exhibit unexpected behaviour in some cases.

Character encoding 

  • The 'Download as .CSV' option doesn't export multi-byte characters.
  • The Export step will always default to the Windows-1252 character set. Trying to export other character sets will show an unmappable character exception. You can fix this by changing the character set in the advanced settings of the Export step.

Snapshots
Renaming a workflow after having taken snapshots in the workflow causes the snapshots to be created in the wrong location and to be inaccessible.

Scripting
Multiple R script steps in a workflow will result in unexpected results.

User management

  • Creating a new user from the ‘Create a new user’ button under teams (Configuration > Teams > [Team Name] > Create a new user) will not bring up the ‘Create a password’ dialog after creating the user. This means that users made here can never be logged in. Note that this does not affect users created under the ‘Users’ in Configuration.
  • It's currently not possible to create a user with a same name as a previously deleted user.

UI

  • Setting the browser zoom to 110% may cause the bottom bar to be cut-off.
  • When mapping sources after having imported a workflow without the source file present, the mapping dialog does not handle long source file names or data source names well.
  • Reordering columns after using Fix First Columns results in column data appearing in incorrect locations.
  • Users will occasionally be automatically logged out when configuring files.


SQL Server

  • It's currently not possible to create a custom connection using the SQL driver from Microsoft (mssql-jdbc-6.2.2.jre8.jar).
  • Users will see an error when attempting to load large (over 10 million row) tables from SQL Server using the default driver.
  • It's possible to preview 'views' from SQL Server but they can't be loaded.
  • It's currently not possible to preview or load data from SQL Server that's in any schema other than the user's default schema. 

Date/time/time zone

  • In some cases, the Extract Date/Time functions won't apply the correct time zone offset.
  • In some cases, changing the locale settings on a file won't change how dates are interpreted.

 

ODBC
When installing on 64-bit machines, the 32-bit ODBC driver's registry settings are not updating correctly. This means that the 32-bit driver can't be used and you'll see the 'Driver's SQLAllocHandle on SQL_HANDLE_ENV failed' error. The workaround is to use the drivers from a previous release. Download ODBC drivers 1.1.1.

Release 1.1.1

Apr 12, 2018

This release includes a number of high priority bug fixes. Download Data Studio 1.1.1. Download ODBC drivers 1.1.1.

Bug fixes

  • Overall Datatype and Dominant Datatype in profile are now correct for decimal columns
  • Find duplicates step no longer throws a Null Pointer error when choosing columns to analyse after manually connecting to source
  • Exception no longer returned when using non-alphanumeric values in email validation input
  • Previews from Redshift and PostgreSQL databases now significantly faster
  • Grouping after a join is no longer slowed down by incorrect compression cache block for grouping index, giving much improved performance
  • Preview of multi-sheet .xlsx files no longer displaying blank data
  • Validation and splice column choosers now correctly show second input's columns
  • Deleting a user now deletes all their uploaded files (in My Files)
  • A Snapshot of a Profile now displays datatype names correctly, rather than as integers
  • Distribution column is now populated when drilling down to formats for a column in Profile
  • Multiple licence keys/codes  can now be added at the same time and we no longer log out the user after update
  • ComponentSGF Batch layout template has been fixed
  • The installer has been updated to install .net Framework 4.6.2 which the latest version of Standardise now targets

Known issues

Database

  • Cached data is not deleted when the source file is removed via the UI
  • Chinese characters or numeric values as column headers in source data are standardised incorrectly on load

Profiling
Drilling down from an integer value in a unique values list in Profile returns no rows

Grouping
Grouping columns which contain large numbers of unique values may incorrectly return 0 rows (this occurred in testing between 500k to 1M unique values with default memory settings).

Validate and Join steps
These known issues stem from the fact that both steps store their rules/key columns in the previous workflow step.

A workaround for all of these issues is to add a Transform step in between the data source and the Validate/Join step.

  • Using a Validation step immediately after a Join, Splice or Union step may only show the columns originating from the first of the two inputs in choosers.
  • Validate and Join steps will not save rules or join columns when used immediately after the scripting (JS / R / Python) step.
  • Validation rules can be lost when the following steps precede the Validation step: Union, Multi-View, Splice, Validate emails, Chart, Use snapshot and custom steps.
  • Deleting a workflow step immediately preceding a Validation step will cause the validation rules to be lost.
  • A Join step applied immediately after any of the following steps will cause the key columns to be lost, resulting in an invalid join: Find duplicates, Script, Use Snapshot, Union, Multi-View, Splice, Validate emails and Chart.
  • Using a Join step after a step with multiple outputs will only allow you to store one set of key columns. If you wish to perform two different joins on the two outputs, you will have to insert a Transform step in between this step and the Join step(s).

Lookups

  • Attempting to edit a column that is already being used in a lookup causes the edit to fail without showing an error (the only indication of this is an exception in the log).
  • When performing a lookup transformation in a workflow on data which has has been joined, the lookup values returned only refer to the original source file rather than the joined rows. This will show as incorrect or missing results for Lookup list, Lookup min/max and Lookup first/last transformation functions.

Split step
Hiding columns or quick filtering in any step connected to the 'Failing rows' output of a Split step causes incorrect data to be returned.

Workflow Designer
Undo and redo actions exhibit unexpected behaviour in some cases.

Snapshots
Renaming a workflow after having taken snapshots in the workflow causes the snapshots to be created in the wrong location and to be inaccessible. 

User management

Creating a new user from the ‘Create a new user’ button under teams (Configuration > Teams > [Team Name] > Create a new user) will not bring up the ‘Create a password’ dialog after creating the user. This means that users made here can never be logged in. Note that this does not affect users created under the ‘Users’ in Configuration.

UI

  • Setting the browser zoom to 110% may cause the bottom bar to be cut-off.
  • When mapping sources after having imported a workflow without the source file present, the mapping dialog does not handle long source file names or data source names well.
  • Using the ‘Replace’ action to replace a source file in a workflow may not update the file to the intended one.
  • 'Distinct' button on an aggregate column is reset in the UI when the aggregate is edited
  • Duplicate files appear in My files when uploading a file on new database creation, prior to a restart
  • Reordering columns after using Fix First Columns results in column data appearing in incorrect locations

JDBC
Update for a MongoDB data source throws a null pointer error precheck is turned on

Release 1.1.0

Mar 29, 2018

This release includes new CDM steps, bringing data matching as well as address and email validation into Aperture Data Studio.

We’ve also fixed a large number of bugs, added the ability to track data quality over time, publish to ODBC clients, extended the functionality in the SDK and much more. 

New features

  • Loading and profiling performance improvements
  • Separate loading and profiling stages
  • A new Profile step added: allows you to profile data in workflows
  • Workflow execution now automatically re-loads the data sources used in the workflow 
  • Multiple workflow executions can now run in parallel 
  • A new Find duplicates step added: allows you to find potential duplicates in data (powered by Experian Match. This release also includes the latest sample rules and blocking keys for Great Britain and Australia.)
  • A new Validate addresses step added: allows you to clean and enrich postal address data (powered by Experian Batch v7.50). This release supports the following data sets: APR, AUG, AUS, CAN, DEU, DNK, FRA, FRP, IRL, GBR, GBR DataPlus, LPG, LUX, NLD, NZL, SGF, SGP, USA, USA DPV.
  • A new Validate emails step added: allows you to validate email address formats or domains
  • A new Analyze trends step added: allows you to view data changes over time
  • Three new steps added for snapshots:
    • Take snapshot - saves a versioned copy of your data
    • Use latest snapshot - get the latest version of the snapshot
    • Use snapshot range - get the combined range of several snapshot versions
  • Option to publish snapshots to ODBC clients 
  • Various improvements to the custom step SDK
  • A new Python script step added
  • Improved password management
  • New licensing model 

Bug fixes

  • Profile now reports uniqueness of values after standardization has been applied
  • Min and max values are now correct for date columns
  • Count and Grouping count now gives the option to include/exclude nulls
  • Aggregate Count is now correct for Null values
  • The Replace First function now correctly escapes input
  • You can now update via JDBC when a key column is defined in the mapping
  • Auditing now tracks the REST API actions
  • Workflow Modified auditing setting now works as expected
  • You can now use the concatenate function to add a plus sign at the start of a value which is a numerical digit
  • Lookup results are now returned when lookup column has been transformed in a previous workflow step
  • The chart in Chart step is created correctly when a null value appears in the data label column
  • The validation thresholds no longer change to 50% when opening the validation dialog after making a rule and viewing results

Known issues

Profiling

  • Format statistics for columns containing dates may be incorrect (as they may be standardized / parsed before profiling)
  • Distribution column is blank when you drill down to formats for a column in profile
  • Overall Datatype and Dominant Datatype will never display decimal
  • A snapshot of a profile displays datatypes as integers instead of translating to the string values

Grouping
Grouping columns which contain large numbers of unique values may incorrectly return 0 rows (this occurred in testing between 500k to 1M unique values with default memory settings).

Find Duplicates step
A NullPointerException occurs when choosing columns to analyze after manually connecting the step to a data source.

Validate Emails step
A class cast exception occurs if you have any non-alphanumeric value in your email validation column. 

Validate and Join steps
These known issues stem from the fact that both steps store their rules/key columns in the previous workflow step.

A workaround for all of these issues is to add a Transform step in between the data source and the Validate/Join step.

  • Using a Validation step immediately after a Join, Splice or Union step may only show the columns originating from the first of the two inputs in choosers.
  • Validate and Join steps will not save rules or join columns when used immediately after the scripting (JS / R / Python) step.
  • Validation rules can be lost when the following steps precede the Validation step: Union, Multi-View, Splice, Validate emails, Chart, Use snapshot and custom steps.
  • Deleting a workflow step immediately preceding a Validation step will cause the validation rules to be lost.
  • A Join step applied immediately after any of the following steps will cause the key columns to be lost, resulting in an invalid join: Find duplicates, Script, Use Snapshot, Union, Multi-View, Splice, Validate emails and Chart.
  • Using a Join step after a step with multiple outputs will only allow you to store one set of key columns. If you wish to perform two different joins on the two outputs, you will have to insert a Transform step in between this step and the Join step(s).

Lookups
Attempting to edit a column that is already being used in a lookup causes the edit to fail without showing an error (the only indication of this is an exception in the log).

Snapshots
Renaming a workflow after having taken snapshots in the workflow causes the snapshots to be created in the wrong location and to be inaccessible. 

User management

  • Creating a new user from the ‘Create a new user’ button under teams (Configuration > Teams > [Team Name] > Create a new user) will not bring up the ‘Create a password’ dialog after creating the user. This means that users made here can never be logged in. This does not affect users created under the ‘Users’ in Configuration.
  • Deleting a user does not delete their uploaded files from the server’s file system. 

Licensing

  • Cannot apply multiple license keys simultaneously.
  •  User is logged out after adding license keys/update codes in the pre-release phase.

UI

  • Setting the browser zoom to 110% may cause the bottom bar to be cut-off.
  • When mapping sources after having imported a workflow without the source file present, the mapping dialog does not handle long source file names or data source names well.
  • Using the ‘Replace’ action to replace a source file in a workflow may not update the file to the intended one.    

JDBC

  • When previewing large tables (above ~5 million rows), preview rows are returned very slowly for some DBMSs (PostgreSQL, Redshift) due to an unnecessary row count query.
  • JDBC Preview Row Count has a maximum value of 1000, but this isn’t clear.

Release 1.0.4

Feb 26, 2018

This release includes various improvements and bug fixes.

New features

  • PSV data files (.psv) are now supported
  • The default administrator password has been reset to administrator
  • Password management has been improved (password policy, reset at first login, lockout)

Bug fixes

  • Validation results when a filter is applied now show correct results
  • Moving or copying the database doesn’t cause a license error
  • Transforming the result of a join doesn’t cause columns to be unmapped
  • Group step doesn’t cause columns to be hidden
  • Correct results are now returned after transforming the join column
  • Join output name is now updated after the refresh
  • You will now be prevented from uploading unsupported XML and JSON files
  • Adding more than one column from the transformation menu now works as expected
  • You will now be prompted to set a password on first log in, if enabled

Release 1.0.3

Feb 6, 2018

This release includes various improvements and bug fixes.

New features

  • The SAS data files (.sas7dbat) are now supported
  • Charts (excluding pie charts) can now display up to 1,000 separate data points (previously 100) 
  • Made several SDK enhancements:
    • The SDK now has its own exception handler 'SDKException'
    • The SDK GitHub project is now also included in the Aperture installation
  • Made several security enhancements:
    • The default administrator login credentials have been made more secure. It is now Pg994_8FQ2U%VM++
    • The account lockout policy has been implemented to guard against brute force password guessing attacks

Bug fixes

  • Further improvements to the responsiveness when configuring/editing validation rules
  • You are now able to create transformation columns that use the UK National Insurance Number or ISBN business constants
  • Circular references can no longer be created when transforming columns 
  • Files are now not duplicated in the 'My files' folder
  • A lookup table can now be selected when transforming from the side menu
  • The Group step no longer causes columns to be hidden
  • The default settings for Redshift JDBC connections have been improved
  • Extracted integers can now be parsed using the 'convert to integer' function
  • Improved Swagger documentation for REST API
  • The Hive JDBC driver has been updated to version 6.0.0.000057 (F000095.U000043)

Release 1.0.2

Jan 8, 2018

This release includes various improvements and bug fixes.

New features

  • You can now load Excel and character delimited files directly from HDFS (using Hadoop API version 3.0.0)
  • Added several enhancements to the SDK/custom step creation. You can now:
    • drag and drop new custom steps into Aperture Data Studio without restarting the server
    • hide custom steps from view
    • get a row of values from an input in one call
    • use a sample step that illustrates multi-threading 
  • A 'Row count' function has been added allowing you to return the row count for the current view
  • You can now group by, filter and create expressions from aggregate columns
  • The installer package has been upgraded to Java JDK 8 update 151

Bug fixes

  • You can now rename columns in preview
  • Changing a country in 'Format phone number' function now works as expected
  • Executing workflows that contain Validation steps now takes less time
  • Correct values are now returned for 'Matches expression' when renaming/duplicating columns
  • Validation rules can now be turned off
  • Increased responsiveness when configuring/editing validation rules
  • You can now group an aggregate column by using another Group step
  • Auditing now tracks REST API actions

Release 1.0.1

Dec 5, 2017

This release includes product name change as well as various improvements and bug fixes.

New features

  • Changed the product name from DataX to Aperture Data Studio
  • Introduced a new 'Duplicate' option allowing you to create copies of saved workflows
  • Made several improvements to the workflow execution behaviour:
    • each Export/Script (R) step now displays an individual progress report
    • the execution of Script (R) steps now runs in parallel to other Export/Script (R) steps
    • all information on Export/Script (R) steps is now displayed in one dialog
    • more detailed information is now provided if a failure occurs
    • you can now download the exported files from the job completion dialog
  • Implemented REST API endpoints to delete, import and export workflows
  • Minified the distributed js/html/css files as part of the build
  • Improved the ‘UK Postcode’ business constant to cover more UK postcode areas

Bug fixes

  • Aggregate functions no longer return blank values when used on some groupings
  • Fixed a repository corruption when the server is unexpectedly shut down
  • The profile drilldown now handles Unicode values (e.g. accented characters) better
  • The 'Remove noise' function does no longer convert results to upper case
  • The 'Replace' function with a null search value now behaves as expected
  • The default value of the ‘End of line precedence’ server setting is now 'Off'
  • Editing a script used in a workflow will now behave as expected
  • The Split step will no longer lose its connections when an upstream connection is removed
  • The grid and configure menu will now always show the same number of columns visible
  • Workflow description tooltips will no longer get ‘stuck’ on screen
  • Other application stability improvements and minor bug fixes

Release 1.0.0

Oct 31, 2017

This is the first release of Aperture Data Studio.

Main features

  • Browser-based application
  • Intuitive user interface
  • Support for local files, Amazon S3, Azure and JDBC data connections 
  • Fast discovery and profiling of data
  • Interactive and re-usable workflows
  • Flexible data validation
  • Powerful data transformations
  • Out of the box data visualisation via graphs and charts
  • Custom workflow step creation using an SDK
  • REST API with Swagger documentation