Product ideas

859 Ideas

bas.hopstakenTimeXtender Xpert

Copy data selection rules to other tablesSUBMITTED

Add feature to copy / drag and drop data selection rules to other tables.

Forum|Forum|6 years ago

bas.hopstakenTimeXtender Xpert

Create measure library for semantic modelsSUBMITTED

It would be great to have some sort of measure library for the semantic models. At the moment the measures for the semantic models are part of the semantic model. It is also not possible to copy a measure to another semantic model. It would be great to have a central place to add all measures for the semantic model. And just point to the library / individual measures from the semantic model to include them in the model. This way you can change a measure in the library and it will be changed in all models the measure is used.

Forum|Forum|6 years ago

scuthbertsonStarter

Automated, regular cleanup of execution logsSUBMITTED

It would be great to have the ability to configure TX to automatically clean execution logs older than N days/weeks, on a rolling basis. For example, a housekeeping process that is run once per day, at a time the user chooses: user choice of time would important because in my organization, we actually shut down our TX virtual machine (on Azure) for 8 hours each night. Another potentially-useful behavior configuration would be whether Successful logs should be treated differently to Failed logs. We might want to keep information about failures for longer than successes. The benefit of this automation would be to prevent TX users having to follow these steps: https://support.timextender.com/hc/en-us/articles/115004530803-Size-issues-with-your-repository-Clean-up-the-logs-and-get-more-space. The comments on that article already indicate support for this request. This is related to: https://support.timextender.com/hc/en-us/community/posts/360005494483-Repository-Cleanup, but I think it is a different request because that issue is not discussing automation, but an improved user-interface.

Forum|Forum|6 years ago

WimContributor

Send an execution package to the Execution QueueSUBMITTED

It is currently not possible to send an execution package from the execution tab to the queue. Would be handy to do this as well. E.g. run a stage full load execution package in the background while you work on your DW or Semantic Layer stuff.

Forum|Forum|6 years ago

david.zebrowitzParticipant

Errors and WarningsSUBMITTED

If there are rows in the Warning and Error tabs after execution, there really needs to be a notification. I've run into situations personally where I have forgotten to check these tabs and then struggled for quite a while trying to figure out why data wasn't being loaded into my table. I have several other clients who have asked about notifications as well. This should be implemented to support ad-hoc executions in the GUI as well as scheduled or queued execution through an execution package. 1. In the case of ad-hoc / manual execution of objects in the GUI, it would be wonderful if the number of error and warning rows could be displayed in the execution dialog window. OR, make the tabs show a different color, display the count of error or warning rows. 2. For scheduled or queued execution, email notifications need options available to include error and warning counts, possibly grouped by tablename. I would want to have a separate email sent out so I explicitly know if errors occurred during the last load without having to open up each success email. Obviously when using TimeXtender for a while, one gets used to checking the Error and Warning tabs on a regular basis, but this is not intuitive for new users. Thanks!!

Forum|Forum|6 years ago

Prefix all fields of a tableSUBMITTED

We prefer to add a prefix on every field of a table. This because when pushing data to Qlik, the fieldnames must be unique.So on every field in a table we add (in most times) a logical prefix.When having a table with a lot of columns this can be a lot of work to change it for every field.A option to prefix every field of the table for once can be very useful in this case.

Forum|Forum|6 years ago

WimContributor

Incremental load Custom Tables in stageSUBMITTED

Currently it's not possible to set a custom table to incremental loading. It would be handy to have this option when you need to bring several tables together in a custom table and the source tables are capable of incremental loading. Some ERP systems separate Invoice and Order lines based on the type of document. E.g. in Navision you need to breng Invoice Lines and Credit Memo lines together to get "all the lines". Some ERP systems go even further and separate depending on the line type e.g. items, accessories, labour, vehicles etc... If you need to have these tables brought together in a custom table so that you can only build your logic once instead of x times, you are currently forced to always full load this custom table, which can take a long time. If however you could put a custom table in incremental mode and use the DW_TimeStamp of your "source" stage table as the data selection timestamp, this would remove a lot of processing time. You can compare it to loading your DW incrementally based on the PK of the source and using the DW_TimeStamp as date field, but here you would do it in the stage itself.

Forum|Forum|6 years ago

rory.smithTimeXtender Xpert

Change types of SCD Is Current and SCD Is Tombstone to smaller typesSUBMITTED

Currently the fields added by history functionality to indicate status are of type bigint. While this may have less of an impact for relatively small dimensions, it will add up for larger tables. Using a bit would mean one byte per record and fit with the boolean nature of the field, using a bigint for both takes 16 bytes per record.

Forum|Forum|6 years ago

oscar.beckeVisitor

Ability to use custom field in semantic model for relationships (semantic layer)SUBMITTED

For versions up to 20.1.4. it is not possible to use a custom field (calculated column) in the semantic layer as a field participating in a relationship between tables. This is however supported in analysis service. This could help in development scenarios, where the model can be extended with calculated columns that participate in a relationships.

Forum|Forum|6 years ago

steven.koppenolParticipant

Impact analysis at table levelSUBMITTED

Often when doing an impact analysis, we are in fact trying to see where the table is used in any way, doesn't matter which fields. We try to mimic this by doing multiple impact analyses from key fields but this is quite cumbersome. It would be really useful if we could do an impact analysis at object level, which I imagine would result in a summarized view of datamovements (table -> table -> table) with besides the arrows or in the table boxes the fields that are affected.

Forum|Forum|6 years ago

WimContributor

Add Perspectives to "Full Load Objects" in an execution packageSUBMITTED

You can create execution packages based on Project Perspectives, but you cannot add the Perspective you wish to execute to the full load objects tab in the Execution Package... This kinda defeats the whole purpose of being able to execute a specific perspective. Please add the option to force a full load on the objects under a perspective.

Forum|Forum|6 years ago

Søren SørensenContributor

Non-root ADLS Gen2 TX containerSUBMITTED

I've got a client with a large number of ADLS Gen2 storage accounts ... and a quite rigid structure for each account. Effectively they need to specify a subfolder for the TX Container. Currently, ODX Server only accepts a root directory as TX Container but for this particular client, it won't work unless we can specify a subfolder in the format "<rootfolder>/<subfolder>/txcontainer" (to follow Joseph's nomenclature).

Forum|Forum|6 years ago

david.zebrowitzParticipant

Environment Properties GUI ImprovementsSUBMITTED

When we are deploying Discovery Hub over and over again, one of the most painful aspects is setting up the Environment Properties. I think there are a couple things which could greatly improve that experience: 1. Ability to clone settings from one environment to the next for a Global Database. If I get the Global Database setup for Dev, it can make things much faster to clone those Dev settings to QA and then tweak the info which needs to change! Right now I usually copy and paste one field at a time. 2. Ability to clone a Global Database to a new one... this would be particularly useful when there are multiple SSAS Tabular Models. Right now I need to manually create a Global Database for each model (7 of them) and manually set all of the settings for those in each environment (21 settings boxes ahhh!) Thank you, David

Forum|Forum|6 years ago

bas.hopstakenTimeXtender Xpert

Automatically deploy database schemaSUBMITTED

When a database schema is added, tables are attached to this schema and you want to deploy the table. You get an error saying the database schema does not exist yet. It would be useful to automatically deploy this database schema it case it does not exists yet.

Forum|Forum|6 years ago

hugo.winkelhorstParticipant

Dynamic work itemsSUBMITTED

It would be nice if you could create a "dynamic" work item in which all tables/views you add or amend are automatically included as long as you have it selected/active. Kind of the same way the dynamic perspective are working right now. At the moment you have to remind to create work items afterwords or add multiple work items with the same description one by one which can get overlooked quite easily.

Forum|Forum|6 years ago

david.zebrowitzParticipant

Powershell Script ActionsSUBMITTED

[This subject has been brought up a couple times, but the original post is from 2017. I'm reposting my recent comment here because I feel like nobody will see the old 2017 post because this site only shows the most recent by default. Original post: https://support.timextender.com/hc/en-us/community/posts/115018657763-Ability-to-include-powershell-script-actions-in-TX] We and several of our clients would benefit greatly by having the ability to execute a PowerShell script from within TimeXtender as Script Actions. Here are a couple scenarios: 1. Scaling Azure SQL DB instances with SQL is problematic. I can use a SQL statement to scale a DB up prior to the load, but we are unable to create a loop and wait for the DB to finish scaling. The problem is that we can wait.... but the session disconnects when the larger DB is ready and Azure SQL switches the connection. The SQL Stored Proc is never able to determine if the database scaled successfully. A Powershell script could be called in order scale the DB as an alternative and since it's session will not be disconnected during the resizing process, it would allow for a scale, wait, confirm, and continue process. This could be implemented as an external executable for execution packages, but that would require multiple execution packages to properly sequence the scale up, load, scale down process. If Powershell was available as a type of Script Action which could be called at the table level, it would also satisfy the requirement of a client of ours as well. (See #2) 2. My client has written C# processes to extract data from another system. Ideally, these extracts needs to be a part of the Discovery Hub load so the data in refreshed in a timely manner without trying coordinate many different schedules. Their C# process to pump data into custom tables may be rather unique to the user community except for a mention here: https://support.timextender.com/hc/en-us/community/posts/360033987671-Possibilty-to-execute-scripts-like-python-powelshell-as-a-custom-script-execution?input_string=Powershell%20Script%20Actions However, this is something else which Powershell could help with instead of creating many clumsy SSIS Packages and External Executables. Custom SSIS Packages really are not an option for them since there would be so many to create. SSIS *Could* be an option IF the external executable SSIS Package calls allowed parameters to be passed in. Then a more generic SSIS package could be created to execute a process for which the name is passed. This however doesn't help in an Azure scenario where SSIS isn't available.

Forum|Forum|6 years ago

rory.smithTimeXtender Xpert

Be able to load data from an Azure Data LakeSUBMITTED

We have a team using Azure Data Bricks to do processing and modelling on all sorts of sensors in office buildings, and they will store output in Azure Data Lake. We want to be able to ingest this data (at least the output) and make it available in DH. Currently this is not natively possible in TX and would require something like Azure Data Factory to move files to a place accessible from TX. Could this be added?

Forum|Forum|6 years ago

andriContributor

Visualize semantic models and it's relationsSUBMITTED

Being able to visualize a semantic model and its relations can helpful when modeling.

Forum|Forum|6 years ago

david.zebrowitzParticipant

Script Actions for Semantic ModelsSUBMITTED

Having the ability to execute script actions for a Semantic Model (specifically SSAS) would give developers the ability to use additional features of SSAS which aren't currently exposed in Discovery Hub. This could help with implementing dynamic security roles among others. I equate this with the ability to execute XMLA scripts for cubes which allows us to change SSAS properties not exposed in the UI.

Forum|Forum|6 years ago

soeren.schubertVisitor

Support for calculated tables in tabular models (semantic layer)SUBMITTED

Sometimes we need calculated tables in tabular models. It would be very helpful if these could be defined in Discovery Hub.

Forum|Forum|6 years ago

emily.wynkoopContributor

Transformation option setting defaultsSUBMITTED

Right now, the default setting for transformations is "Custom." The overhwelming majority of transforamations we do are "Fixed," and it would be nice to be able to set this as the default type when adding a transformation. In addition, when you add transformations to one field and select another, the transformation type snaps back to Custom no matter what. This adds several additional clicks if you are adding many transformations that aren't custom. If the option for a default transformation type can't be added, then at least maintaining the current transformation type when changing fields would be really useful. The use case here is that we are adding fixed values to handle cases when DSK or lookup surrogate foreign keys are empty. When repeating this process for dozens of fields across multiple tables, selecting and re-selecting the transformation type begins to feel very laborious.

Forum|Forum|6 years ago

WimContributor

LEFT KEEP support for Qlik adapterSUBMITTED

LEFT KEEP makes sure only the dimension values actually present in the fact table are kept in the dimension table for a Qlik app. It would be good to have this supported in the modeller. See also https://community.qlik.com/thread/39177

Forum|Forum|6 years ago

WimContributor

Semantic Model :: Copy EndpointsSUBMITTED

Would be handy to drag & drop an endpoint from one semantic model to another to copy it, instead of having to manually create it again. Or have an option of creating an endpoint in multiple models at once. Or a global endpoint.

Forum|Forum|6 years ago

bas.hopstakenTimeXtender Xpert

Add execution queu to tab in new GUISUBMITTED

Add a feature to 'pin' the execution que in a tab.

Forum|Forum|6 years ago

rory.smithTimeXtender Xpert

Display item sent to execution queue instead of only Default PackageSUBMITTED

When you drag a table to the execution queue, you can see what table is being executed in the queue. However when you use the checkbox after right-click and Execute, you only see "Execute Execution Package Execution Queue Default" which becomes rather confusing if I have more than a few queued and want to know which one failed. Could the name of the thing being sent to the queue be displayed?

Forum|Forum|6 years ago

Filter by idea status

Filter by product