Questions & Conversations

I have set my first steps on customizing the stylesheet because there are really nice examples to improve the Universal GUI. Thanks Harm for your post on the cardlists.So I tried to read all about the custom css and I created a custom CSS and placed it in the universal folder. I just copied an example from the css blog of Jasper and copied some css code into the custom.css.It simply looks like this now, just to do a test and see how it works. However the css is not applied if I put in the universal folder. What do I miss. I restarted the application pool and the application as well on the webserver and in the browser but that did not help.As said I’m brand new to css so perhaps I miss some basics… Hope someone can help me out  

Hello!We are trying to install our app for a client on premise and until now everything seemed to work good. The strange part is that now for every task when the user is choosing to cancel the "An unknown error occurred, please contact your administrator” pops up. There is no process flow involved for that task, the deployment was good and execution of the task without cancel work fine.We checked the live error log, error log but nothing is there and on windows gui is not happening. When checking the network/console there we can see that the call failed with status 405 Method not allowed.Any ideas how we can solve this or what is causing it? We saw this post  but is not very helpful since the answer is another question and it feels strange to be from permissions because executing the tasks normally without canceling them works just fine, but maybe we skip something that you can guide us. 

After pulling and running Indicium and Universal containers in Docker on an on-premise Ubuntu host both are running but Universal isn’t reachable via the browser. Browsing to http://hostname/ shows the running Indicium instance (log-in also works) but when going to http://hostname/alias our end-product doesn’t start and I get a blank page instead.These are the docker run commands I used to pull and install the containers: Indiciumdocker run \    -p 8080:8080 \    -e SQL_USERNAME='***' \    -e SQL_PASSWORD='***' \    -e CUSTOM_JSON='{"Logging":{"LogLevel":{"Default":"Information","System":"Information","Microsoft":"Warning","Indicium":"Information","Duende.IdentityServer":"Warning"},"ApplicationInsights":{"LogLevel":{"Default":"Information","System":"Information","Microsoft":"Warning","Indicium":"Information","Duende.IdentityServer":"Warning"}},"ErrorLog":{"PathFormat":"Indicium-{Date}.txt","LogLevel":{"Default":"Error"}}},"MetaSourceConnection":{"Server":"HOST1","Database":"IAM"},"Agen

We have a Linux server with a couple of Docker containers running Indicium and Universal. We plan to move systemflows from IIS to this new server but for now we want to have the option to run them in both environments if needed. All relevant files and folders for the systemflows are reachable from the Indicium container but because file paths on Linux are different from those on Windows the correct path needs to be determined from the systemflow.An idea to solve this problem is to include two file read connectors that try to read a simple check file using two different absolute paths. The two possible absolute paths would look something like this:Windows: \\share\checkfile.txtLinux: /share/checkfile.txtThe correct path depends on which of the two connectors can successfully access the text file.Is this the best way to reach the goal?

What am I doing wrong?I want the pre-filters after the normal filters, and actually in the overflow. So I defined this in the default action bar And in screen type I setup an action bar (the pre-filter bar is also necessary, otherwise no pre-filters are visible, which is unexpected to me, as these should appear based on action bar. For the tasks, those are visible based on action bar, so these seem to work differently)Here also, filter & quick-filter before pre-filtersBut the result is, and that is not what I want As default is overflow, I have not defined custom display type Am I doing something wrong or is it not working properly? Also this is very strange.   

Extended properties | Thinkwise DocumentationThe entry CachePath has a neat little entry:A custom path where the Windows GUI stores a cache of the application definition to increase the startup time. My best bet is that this is something you would not like to set, because, well, it increases your time waiting on starting up the app right? Now to my probable problem: We’ve got it set to %temp%\awesomeness_app That location is not restricted for reading, writing etc… Can it be that for some time now this is not working anymore, because it (tries to) write its cache in the same directory as where the executable is stored? 

I try to implement a zip and download functionality that reads out a folder on azure and zips the files inside that folder.There are the following steps : Decision that either selects the next folder to zip or ends the process when there is no folder left to zip. This decision sets a process variable with the name of the folder. (not shown in the picture) This folder name is used as input for the zip process action. One of the output fields of the zip action is the file_data, which is assigned to a process variable that has the var_binary(max) datatype. This var_binary variable is the used as input for the download action that also takes the filename of the eventual file that is downloaded. The download itself seems to be working as in that a ‘save as’ form is shown with the correct filename that can then be saved on the local machine.However, the file that is created is empty. The zip functionality itself does not give any errors (status code 0 is returned) and the file_data output pa

Is there standard functionality that allows you to create an xml file of selected data?I have a function that returns an xml variable.This variable contains all the text that I need.Is there an easy way to put that in an xml file and give it back to the user so that he/he can store it locally??   

In the works of creating a whole flow to process users and its user groups in the application to communicate to IAM, I am configuring the Web connections. In setting up all endpoints, I’ve added several GET actions to obtain values from lookups which I later use in the actual POST (ex. Create user).The question now, is there also a way to directly use a Lookup value in the POST instead of needing to find its “id” in intermediate steps?For example I wish to obtain the “write_back_usr_pref_type_id” by just supplying the name of the configuration.

I just had an issue with the scheduled system flow that was perfectly handled with a TCP call by ​@Angelica van der Beek . I thought it would be worthwhile to share it in the community, in case someone else encounters the same issue.  At our customer we had a systemflow scheduled every 5 seconds, in order to treat incoming calls. The system flow did not work, wh in the production environment while in Acceptance it did. We did verify all settings according to documentation: - Configuring pool usershttps://docs.thinkwisesoftware.com/docs/deployment/indicium#configuring-pool-users - Create the web applicationhttps://docs.thinkwisesoftware.com/docs/deployment/indicium#create-the-web-application - Restarting the applicationhttps://docs.thinkwisesoftware.com/docs/deployment/iis#restarting-the-application - application Initializationhttps://docs.thinkwisesoftware.com/docs/deployment/iis#application-initialization

Hello!I know in the latest update there should be an option to limit branch access which will be very useful for main branch. I was looking for it, but I think I miss where it is, can you help me?

Hello, We have a very tight security policy within our company. This means we cannot download any files on servers in our DTA environment without explicit consent from our security department.I was discussing with the security department how we could and should download upgrades from TCP and upgrade our environments.This department is now interested in any information about the security of TCP in general and how we could know for sure that the files we download from this portal are safe? For example, they want information about:Security monitoring/checks TCP EncryptionAny information is welcome! Thanks in advance.Best regards,Nick Janssen

Hello,I am looking for some guidance on how to implement a view which is based on the API results of another internal application.To give you some context, we have a Document Management System (DMS) which offers an API layer. Based on a selected record in Universal, a query parameter is filled and the corresponsing documents are returned in the response. I want to show these filenames in a grid as a detail of the selected record.I tried the following:In a tree menu, add a task instead of a table/view This task is a start task of a process flow: Calls the API endpoint and writes the result in a table Opens a document (a view which is based on the most recent call to the endpoint) This method works for my basic example, with a slightly longer loading time than a normal view (because of the API call).However, we would like to implement this solution in the menu where it belongs. This menu has the following structure:Menu structure​​​​​In Tab 1, a record is selected. Tab 2 is a detail of

Hi, we currently have some procedures on our project database which we’ve manually added to the database instead of through the Software Factory. We have several reasons for doing this.I thought the Software Factory wouldn’t drop objects in schema’s other than dbo. I think I read some other question which mentioned objects in other schema’s wouldn’t be dropped but I can’t find it now. It’s also mentioned in this documentation:  https://docs.thinkwisesoftware.com/blog/2023.2#deployment---only-dropping-objects-in-dbo-schema Our manual procedures exist in a different schema but the Software Factory still tries to drop them and this causes errors.This is part of the code in execute source code which tries to drop our procedures: /* Drop all procedures */declare @procedure_name varchar(255)declare drop_objects cursor fast_forward forselect quotename(s.name)from sysobjects swhere s.type = 'P' and s.name not like 'def$_%' escape '$' -- Do not drop defaults. and s.name not like 'lay$_%' e

Hello,I'm currently trying to embed another internal application into our Thinkwise application via an Iframe.To some extent I have this working, but I'm having some issues with specific functionalities. The other application is a document management system (DMS). Therefore it should be possible to download/print files from this iframe.When attempting this, the following error message pops up in the DMS:Error message in embedded Document Management SystemI understand the security implications of CORS, but in this situation we trust the source. I tried playing with multiple configurations in the web.config and appsettings.json file:Added some properties in the sandbox attribute of the iframe: sandbox="allow-forms allow-scripts allow-same-origin allow-popups allow-popups-to-escape-sandboxAdd Header Content-Security-Policy: <add name="Content-Security-Policy" value="frame-src 'self' data:<dotiga_url>;" />Add Header Access-Control-Allow-Origin Is there any possibility to allow

I am upgrading our SF to version 2025.1 (from version 2023.1) and i am running some test upgrades to see how smoothly this will go and what the changes are. We are running on SQL Server 2019.I noticed that when i do an iam synchronisation of a “new” branch to an IAM that this executes faster than before 😀. But when i change some roles and do an IAM synchronisation of an existing branch in the target IAM that the synchronisation process takes forever.  In the example above i stopped the process because it took to long. In Management Studio i see it is working on this query. The role_tab_variant_look_up table has more than 750k lines.Also in general i occasionally experience that merge statements sometimes take very long.Anyone else ever experienced the same problems and fixed it.

I have created a table with system versioning turned on. When I try to create an index on the _history table, and Include a column, a create script is generated with an empty include argument: create nonclustered index "some_name"on "some_table_history"( "some_table_id", "some_other_column") include ( )Naturally, this gives an error on execution.Doing the same on the base table gives a correct create script:create nonclustered index "some_name"on "some_table"( "some_table_id", "some_other_column") include ( "some_third_column" )Why is an included column not included in the create script?

We consider using TSF_dotNET.exe instead of TSF_CS.exe from a network share for starting the Windows GUI in a terminal server setup. This way we want to avoid the growth of the App Data folder. Are there any disadvantages to this idea?

Hi,As we develop more and more system flows, it would be handy to have a tool where we can see all of the logging steps of the running system flows for debug purposes.I was checking in IAM and could not see any logs there except the running flows start and end date.Is something I can look for somewhere else? Thanks in advance,Paul

Hiya!We’re interested in using the Thinkwise Insights demo application in our own environment for two main purposes: To gain insight into how new platform features are applied in a real-world model. To provide a solid demo for (potential) customers ajustable to the customer.  Therefore, my question is: Is it possible to download the complete Insights model somewhere?  Best regards,Ruben

At the moment it is not possible to use Model Insights in combination with a form or grid that’s not in edit mode. Even though the table/ row/ column/ cell information is actually available Model Insights doenst trigger (because you cant trigger eventlisteners on read only fields I suppose). Only rarely it works on read columns.I’ve checked if the information is available by using webpage inspector and check components (and parents). I’ve added a gif for reference. 

Hi, before I had a subroutine with a parameter called Units. This would accept value "Units”:["kg”,"cm”].Due to new features I moved this to a process flow, with a web connector. This new process flow now throws an error stating;An invalid value was supplied for process variable 'units'. Only primitive values are allowed.Is it correct that this does not work? I suspect this to be incorrect..oh ohThanks, B

Hi,I have a small question. Have developed a functionality to retrieve user's notebooks from Microsoft (OneNote). The onenote itself has two URL in the response body, the one_note_client_url and the one_note_web_url.Have built a view witht he data I need and for the web_url i ve made a column with the domain type as html with something like this "cast('<a href="'+notebook.one_note_client_url+'">' + notebook.display_name + '</a>' as nvarchar(max))”. This works as intended and I can open the notebook in a different tab in browser.The issue I face is whent I want to use the one_note client url wich in technical terms is the same url except that it contains the word "onenote:” at the beggining of the url. I have tried to make it with 2 different domain types, URL and HTML (same as the web_url) and in both cases I get the action blocked by the SF Is something I can do in order for the link to be inserted as it is in browser so the OneNote app would start or it is blocked by the

I tried to import an Excel file using the standard import feature. The excel file contained approximately 8000 rows. The functionality (lay, def etc) is minimal and still it took around 30 minutes to import the entire Excel file. We are using platform version 2025.1.This is not the performance we want/need. It comes down to a little over 4 seconds per row.What can I do to increase the performance?

Hello,Based on this post we have found https://community.thinkwisesoftware.com/questions-conversations-78/open-a-universal-gui-document-from-an-externa-application-4200?tid=4200&fid=78, we made a deeplink process flow that simply executes a task, then opens a document filtered on a record with a given id.  We tried triggering this process flow externally with an API call inside a Preview component (iframe), and while it is successful, nothing happens. What could the issue be - are we missing something from the POST API call (we include the mandatory necessary id in the request body) or is this due to the component not communicating with the parent page? Best regards,Valentin