Thinkwise Platform

Universal GUI Editable Grid Performance IssuesInstalled software:Software Version Universal GUI 2025.1.14.0 Indicium 2025.1.14.0  Problem StatementThe Universal GUI seems to be less performant in various cases, when compared to the same functionality in (older) WebGUI / WindowsGUI.Specifically, the editable grid has significant performance implications, which makes the user interface unreliable. Upon pressing the downward arrow-key 3 times in succession in the editable grid. The UGUI registers only 2 button presses.The problem is not necessarily the delay, it is the discarding user-inputs that makes the UGUI unreliable in use.Possible CausesUniversal GUI version: we are on the latest version. Indicium version: we are on the latest version. Resource capacity: we host UGUI & Indicium on a server dedicated to this environment. Data quantity: this test was performed on an editable grid with 6 columns and 7 rows, without expression fields, nor triggers.This list does not spark

Hello I have below Supplier  data model How can i prevent entering a duplicate value?In the portal, it is designed to add, edit delete but it allows to enter duplicate data I'm not sure how unique constraint works here but  I added an Indexes unique constraint but it doesn't work. Please advice Below is the values on index column, unique is also tickedThanks in advance

What could be causing a bad gateway message with an Indicium docker container?I've already verified that all containers (including traefik) are on the same Docker network (the default network made by docker compose).This is the line from the traefik log:502 Bad Gateway error="dial tcp 172.18.0.6:8080: connect: connection refused"According to “docker network inspect” 172.18.0.6 is the correct internal IP address for the Indicium container.This weirdly doesn't happen with the Universal container as it serves the login page just fine.Each container uses its own subdomain as the external host.This is my docker-compose.yaml:---services: proxy: image: traefik restart: unless-stopped ports: - 80:80 command: - "--providers.docker=true" - "--providers.docker.exposedbydefault=false" - "--entrypoints.web.address=:80" - "--accesslog=true" - "--log.level=DEBUG" - "--log.filePath=/logs/traefik.log" volumes: - "/var/run/docker.sock:/var/run/dock

Hello everyone,In our application we have a few custom components. Those custom components are integrated well. But there is one point where the integration of these components doesn't work as well: on the topic of the themes (dark mode and light mode). We now have to manually toggle the components to match the theme of the software, which is weird, why would the user need to toggle a component from light to dark mode or from dark to light mode when the software should do that automatically?Is there a way to know in what mode the software is? It would help to automatically switch the theme to the theme the application is already on, otherwise the overly bright component is highly distracting.So in short: is there a way to know whether the software is in dark or light mode, so we can change the components accordingly? 

Hi. We have a remote desktop server for our users. And they sometimes need to do an update/upgrade of a client application (over VPN). it's just update of app db and sync IAM. Is there an option to do this without admin rights, which I prefer not to give out. 

What is the logic in universal if a spinner (loading) icon is shown or not. On some screens it’s shown while the data is loaded. On others it is not. This makes it very confusing because you are not sure if the data is still loading or if there is no data. Is there a subject setting that influences whether this icon is shown?Not shown: other subject the icon shows:  

When we copy text from Word into our app (Universal GUI), all kind of meta data from Word is copied. This has no additional value, but fills the DB unnecessarily (and result in other issues).  How can we copy text from Word into our app, without losing the formatting and not copying the meta data from Word?

I try to make a full height code editor as detail tab.Is there any trick not get 2 parallel scroll bars? I tried is to set the height of the ‘Code editor‘ field to 1000. By doing this the most of the time the scroll bar of the text box is not visible, but as side effect the cursor jumps to the bottom of the field when switching to edit mode.See below. 

Hi,As I understood a document is previewed when it is the first column found in the table. Although in most situations this would suffice, I wonder if it is possible to have this be the first visible column in the table (i.e. based on the subject and layout procedure)?We already have all our documents (i.e. renditions) in a separate table, but within a document/rendition record we provide the possiblity for different storage options: cloud (Azure) or on premises (fileshare) which are based on different domains referencing their respective store definition. Based on the configuration of the system we want to show the cloud column or the share column. This can even be specified on the rendition type of the document (active documents in de cloud, archived documents on a share or vice versa). In any case, only one of those columns is visible at any time.Any thoughts in how this could be realized?

Hallo, Is there a way that the below prefilter buttons can customise the label instead of showing 1 char? example instead of showing 2, i can display 2022?These are the prefilter groups  

Hallo, I would like to check, the chart on my cube view, under the row has a number 2 which is I'm not clear where it comes from or how it derive. Please find below screenshots The volume comes from a Company (only 1 company)39 volume for 2 weeks and each weeks (start of the week) which are 30/06/2025 & 07/07/2025  has 22 and 17 volume.   What does day =2 means?  also how can i remove the 2 if i only require to show the week start date instead  on my Software factory, this is the setup on the cube  The Dimensions setup is below  Thank you in advance.  

“Our philosophy is that Thinkstore solutions are a foundation—something you can build upon and tailor to your specific needs.” ​@Remco Kort  (From Black Box to Clear Logs | Thinkwise Community)How can we build upon Thinkstore solutions?When linking a base model to a work model, everything is read-only in the work model (which I understand). However I would have expected that using the task ‘Merge base models into work model’ would solve this, but this seems not the case. When using this task, ‘generated_by_control_proc_id’ is still filled with ‘Copied from base model’ leaving all objects from the base model still read-only in my work model. Am I missing something or is this a bug?  

We have an application (in this particular case, still on 2022.2) with several application scheduled system flows, of which some flows run very frequently :system flow 1 ‘generate’ - generates new data   (interval 5 seconds)system flow 2 ‘print’ - executes print jobs - ZPL to Zebra  (interval 1 second)Handling flow 2 ‘print’ may take some time to execute. This should not result in delayed execution of flow 1.Does Indicium handle the different system flows parallel or sequential?Is this configurable for parallel execution, and how? 

I'd like to be able to run multiple Docker containers with Universal GUI so each one can have its own default meta server and application. Using the documented Docker compose method with Traefik I've tried to add a pathprefix for Universal GUI just like with Indicium but it doesn't work.Can Universal be served under a URL with the format https://host/pathprefix and if so what's a working config?Follow-up of  

Hi all,I made some cubes for quick access to aggregate data. The client wants to export the results of the cube.The build-in export function exports the table/view associated to the cube.Is it possible to export the results of the cube (e.g. the aggregates)? As context, the build-in export function exports this view:  But I want to export this aggregate cube: Hope you can help with this. Kind regards,Rick

I have a process flow where I use a web connector and I want to use a parameter in the end point query string. So I have set this up in the web connector but from the manuals I can't figure out how to fill this parameter from a functionality or something like that.Can you help me? 

Hello everyone,We are currently trying to send emails via the MS Graph API. This has worked very well so far. Now it is specifically about adding file attachments. Some of the attachments are stored in Azure and therefore have to be written to the database as a byte array before sending. The reason for this is the fact that we have to keep emails and attachments together and of course the byte array has to be converted into a Base64 for sending.So how do we get the files individually from Azure into our database as varbinary(max)? Does anyone have any ideas or experience?Translated with DeepL.com (free version)

3 rows selected in this gridGUI versie2025.1.12.1.0When selecting multiple rows in dark mode, it’s almost invisible which rows are selected. We also tried changing the colors in “Themes”, but without any good results. Is there any other possibility to change the color palette for more visibility of the selected rows in a grid?   

After the last Universal GUI updates (we switched from 2024 version 3 to 2025 1.14), reports (like PDFs) are opened embedded within the application. We would prefer these documents to open directly in the browser (e.g., Microsoft Edge), like they did previously.Does anyone know how to configure this behavior — for example via path_spec, task action, or another setting — so that the report opens in a new browser tab instead of inside the GUI?

I would expect imports work within the same versions of 2025.1. Is there a reason why a different sub-version blocks this task? And is there an easy tool to upgrade a model without upgrading the whole environment where the model is exported from? ************************************* Thinkwise Software Error Message *************************************Datum: 19-6-2025 10:54:22GUI: 2025.1.13.0.NET Runtime: 4.0.30319.42000Database: IAM_SFLanguage: NLSoftware Factory version of the imported model does not match the Software Factory version. The import cannot be continued.Imported model SF Version: 2025.1Current SF version: 2025.1.14SQL:task_import_model

In Windows it is possible to do a sum on a “checkbox”, where the domain is INT   In the Universal GUI, this doesn't work Would be very nice if this option will also work in Universal GUI….ef_evnt_event_staff_planning?$apply=aggregate(count_using_dinner%20with%20sum%20as%20count_using_dinner_sum)Results in 500 error :-(.Is there a way to make that work?

 When POSTing, the following works:"profile_picture": { "FileName": "name", "File": "file_data" }So setting a profile picture works. Now I’d like to remove/clear the profile picture of a user. I have tried the following JSON bodies in a PATCH, but none seem to work:{"profile_picture": { "FileName": "", "File": "" }}{"profile_picture": { "FileName": null, "File": null }}{"profile_picture": null}{"profile_picture": ""}{"profile_picture": { "FileName": null }}{"profile_picture": { "FileName": "" }}{"profile_picture": { "File": "" }}{"profile_picture": { "File": null }}{"profile_picture_data": null}How can I clear the columns ‘profile_picture’ & ‘profile_picture_data’ in the usr table in the IAM using a PATCH API call?

Since some time we don’t see the application icon anymore in the Windows taskbar for the Windows GUI.This is for both the SF as our end-user applications.What we see is: This is already since a longer time, and we didn’t notice it straight away, so don’t know if it's related to some Thinkwise update or Windows update.Also changing the icon in the IAM Global Settings doesn't solve the issue….Any solution for this ?(we are working on the latest WIndows GUI and latest 2024.2 patch)

I am working with a grid where auto-edit is applied. In the documentation (https://docs.thinkwisesoftware.com/docs/sf/process_flows_actions#add-row) it says the ‘add row’ process flow action should not be triggered, but it does when the row is added by pressing enter when the cursor is at the last column of the grid. Is this a bug? I would like the process action to be triggered when the row is added regardless, but I believe that should not happen as is explained in the documentation. What is the intended behavior here? And is there another way to trigger a process flow action when a row is added on an auto-edit grid?

I'm looking at public API calls..   can I require a select or set it so that API call with select criteria returns no data?