Open ideas have been reviewed by our Product Management and are open for commenting and voting.
With the advent of Data Virtualization it seems like a good time to improve JSON support in FME!It would be great if there was a simple way to convert a feature to JSON format so that all attributes on the feature become elements in the JSON object and all lists items are converted to arrays in the JSON object.The result should be output to an attribute.I know that it’s possible to do this currently using the JSONTemplater but it requires manually building the template and it would be nice to avoid that. Just do whatever cleanup of attributes is required using the AttributeManager, then have FME convert to JSON without requiring extra input. The JSONTemplater can still be used for more complex cases.(I know that it’s also possible to use the JSON writer to create a simple file from a feature but this does not support lists/arrays and requires writing out to the filesystem.)
Currently, FME Flow supports database connections via JDBC, but authentication typically relies on static credentials such as username and password. In modern cloud environments like Azure, using Managed Identity is a best practice for secure, secretless authentication.This idea proposes adding support for Azure Managed Identity including System Assigned and User Assigned identities for JDBC connections to databases such as Azure SQL Database, Azure Synapse, and other services that support token-based authentication.Why is this important?Eliminate hardcoded credentials: Reduces security risks and simplifies credential management. Cloud-native security: Aligns with Azure best practices and enterprise compliance requirements. Automatic token refresh: Improves reliability and reduces operational overhead. Flexibility: Support for both System Assigned and User Assigned identities allows organizations to choose the right approach for their architecture.How it could work:FME Flow detects it is running in Azure and uses the assigned Managed Identity. When establishing a JDBC connection, FME Flow retrieves an OAuth 2.0 access token from the Azure Instance Metadata Service (IMDS). The token is injected into the JDBC connection string or passed as a driver property (depending on the database/JDBC driver). Configuration options in FME Flow to select: Authentication mode: Managed Identity Identity type: System Assigned or specify a User Assigned identity (via client ID) Use Case Example: An organization runs FME Flow on an Azure VM with a User Assigned Managed Identity that has access to multiple Azure SQL Databases. Instead of storing credentials in FME Flow, the system automatically retrieves tokens for secure, passwordless connections. Security NecessityStoring static credentials in FME Flow or any application introduces significant risks:Credentials can be leaked through logs, backups, or misconfigurations. Manual rotation of passwords is error-prone and often neglected. Compliance frameworks (ISO, SOC, GDPR) increasingly require secretless authentication methods.By leveraging Azure Managed Identity, FME Flow can achieve zero-secret architecture, reducing attack surfaces and aligning with modern security standards.
You can download subfolders of /Logs but not the entire /Logs folder.
We would like to request support for Azure DevOps as a remote Git provider in the Version Control functionality of FME Flow. One of our clients recently upgraded from FME Flow 2024.2.1 to FME Flow 2025.1.2. In the previous version, they were successfully using Azure DevOps Git repositories to manage workspace versioning. After the upgrade, they are no longer able to push changes to their remote repository. The UI reports: “There was a problem communicating with the REST API.”And the backend logs show HTTP 500 errors when attempting to push.According to the documentation, only GitHub.com is officially supported. Azure DevOps is not listed, although it previously worked without issue. This limitation significantly impacts their ability to maintain version history and collaborate effectively. Could you please consider:Adding official support for Azure DevOps Git repositories in FME Flow Version Control. Providing documentation or configuration guidance for Azure DevOps integration. Ensuring compatibility with common enterprise Git platforms beyond GitHub.com.This feature would be highly valuable for organizations using Microsoft and would align FME Flow with broader enterprise DevOps practices. Please let us know if this request will be considered for a future release and if so, in which upcoming release. Thank you for your support! Kind regards,Joëlle Jansen-SoepenbergFME Consultant
Hi,When you select a date using the calendar in FME Flow UI the calendar persists until you click away. This is a bit counterintuitive and annoying for users . Our users would prefer it disappears on date click. date is chosen and calendar is still there. Thanks,Marc
The new DV functionality seems amazing, I haven’t tried it out yet, but I understand it as one has to provide workspaces for every api endpoint one is creating.Wouldn’t be great if one could easily, in the FME Flow interface, create CRUD (Create, Read, Update, Delete) functionality? This is all functionality that is available in reader/writers, one just have to specify required attributes, primary keys etc.
FME Flow APP with dynamicParameterData is great!However, I have installed the FME flow APP on my PC, and I would prefer to open the installed APP directly instead of a new web page. The progressive web application (PWA)) APP gives a better integration / look and feel.
It would be nice to have FME Flow have file extension restrictions as to what types of files can be uploaded to the resource folders. The ability to restrict types of extensions, especially executable, would be a major safety addition to Flow. If this could be added by user/groups level, that would allow the sys admins to restrict or allow file types based on their groups. For example, some users only allowed to upload CSV files, others more privileged users can upload spatial datasets, others can upload anything except executable. This also will have the benefit of keeping the resources clean and not be a dumping ground for less experienced users not knowing what the intended purpose of that resource is.
Currently we are only able to use standard connection to FME Flow within the FMEFlowJobSubmitter transformer. Since we migrated to SAML we cannot use FMEFlowJobSubmitter anymore. We are using very complex workaround to be able to run another workbench on conditions after another one.
On FME Server flag out Workbenches that are Part of an(one or more) AutomationWould be more comfortable in case of editing the Workbench or in organizing between Repositorys.
When discarding an Automation, I’m having to locate the workflows, using Workspaces which were linked to it.Is it, therefore, possible to have an option to delete the “linked workflows” when deleting an Automation?
I must not be alone in being driven crazy by the lack of the ability to center the run button in a Flow Workspace App. Everything else is center aligned and then the run button is out there on the left and it looks just awful!!
Hi,in the record information you can nicely see data types (uint8, buffer, date, time, varchar) for all available attributes when selecting an individual feature: However, when accessing these information in Python we only get the basic types: Would it be possible to also get the full list of data types in Python?In my current use case, I’m supposed to convert all date attributes to another format. Since it’s a generic workflow, I don’t know in advance which attributes are containing dates. When looking at the schema feature I can see that the data types have been recognized correctly by FME. There is just no way easy way to get that information for each feature.Kind regards,Dennis
Hi there, it could be a good idea to have a ‘zoom to’ button on a hidden connection, that takes you to the other end (start or end) of a hidden connection. Sometimes I instantly want to see where the other end of the hidden connection is without having to manually scroll and find it
Hi everyone,wouldn’t it be cool to have a list of favorite Transformers in the Transformer Gallery? Say, one has a number of Transformers that one uses for most of their workspaces. Instead of typing them in to Quick Add each time, it might be more convenient to have a list in the Transformer Gallery for those few Transfomers. It might also be the case that one does not remember the name of a often used Transformer. With a “list of favorites” it might be easier to add often-used Transfomers to Workspaces. But that would mean that there needs to be the possibility to mark a Transformer as a “favorite”.[Edit:Doesn’t have to be in the Transformer Gallery, a selection of favorite Transformers could also be useful after clicking on “Transformer”]Let me see your feedback on the topic. Is it something others would like too or is it just me?Cheers :)
Hi there, I’ve built workspaces for clients that I need to go over and show them. It is very helpful if I add annotations and bookmarks so the client can follow along with me. However, in a workspace a lot of secondary or unimportant transformers are used (things that are not useful for a client to see to understand the logic of a workspace - think things like attribute removers, csmapreprojectors, list exploders, area and gap overlap cleaners, etc.) It would be nice to have a tickbox inside a transformer to mark the transformer as important to the workspace. So then there could be a toggle to select that could, say, highlight all the important transformers (a different colour, a glow etc.), or it could generate a separate, collaborative workspace that you can use to go over with clients, that just show the important transformers that have been tagged, this can act as a sort of ‘pseudocode’ for FME Cheers,
Custom Transformers to automatically output a Rejected port which returns all Rejected ports from within the transformer
Instead of clicking the green magnifying glass, imagine "hovering" over a port to see a live sample of the data (schema and first 5 rows) in a pop-up, like in the picture below:
The SpatialRelator has the option to output the cardinality of the intersections. It would be really useful to also have the optional functionality to output the actual geometry of the intersection.
May be terribly niche, but I’ve just come across a situation where having a ‘counter’ that assigns sequential letters in alphabetical order starting at C through to J would have been useful. If it’s of use to anyone else, maybe the Counter could count letters...or gets a cousin called the Alphabetiser?
In Jobs; Completed I would like an extra filter option: Succesful jobs with errors. Could you add that one?
I have a need to send Kafka header information as part of a producer workflow. Currently header information is only available to receive as part of a consumer. Note that headers for sent messages are not available.
Allows the user to be able to gracefully handle failures
In our standard FME Flow training course we have one excercise where Workspace A does some checks on user parameters and if they are succesful it runs Workspace B through an FMEFlowJobSubmitter.One of our trainees accidentally set up the FMEFlowJobSubmitter to run Workspace A instead and when he published it to Flow he made a recursive loop: every time Workspace A ran, it queued Workspace A again. I was busy helping somebody else at that time so when I got around to his question I noticed it had ran 17000+ jobs in a matter of minutes. These were low-impact jobs, so it wasn’t an immediate problem, but it’s safe to say that this has the potential of completely overwhelming an FME Flow setup.After talking to Safe we decided to post it as an idea here to see if others have come across this as well or have opinions about it.The big question would be:Should FME be able to detect these recursion risks?And in a broader sense: have you had this happen to you? Or something similar that you think FME should be able to detect and warn you about?
Having user-configurable syntax highlighting colors to FME’s syntax highlighting (Python, SQL, Expression Editor), similar to how Visual Studio Code handles token-based color customization would be a nice addition.I find some of the default colors for JSON syntax highlighting to be difficult to read, so it would be great if there were some customizations for this.Text Editorhttps://code.visualstudio.com/api/language-extensions/syntax-highlight-guide
In-depth courses for all FME skill levels.
Explore our extensive collection of how-to guides and tutorials.
Log an issue with us. Our support team is ready to help.
