Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
FME Flow currently has columns in the Jobs Completed screen like the workspace, source, date/time, engine, etc.What happens for bulk dynamic data replication is you need to run the same fmw multiple times with differing parameters. e.g. a SQL Server to SHP Downloader that runs on a single table at a time.Whilst yes you COULD run the workspace only once (reading hundreds of thousands of features), this not only adversely and unnecessarily affects server resources but means a single error affects all data. Instead, it’s best to run the workspace multiple times (once per SQL Server table) via say a FMEFlowJobSubmitter.In FME Flow this means you get the Jobs screen full of hundreds of the same workspace with a differing date. There’s no way to tell at a glance what parameters caused the 10 out of 200 jobs of the same workspace to fail.I’d like to see a parameter exposed in the Jobs screen so that I can see at a glance that Table ABC and XYZ failed. Not that Workspace123 failed 10 times with unknown parameters. Yes, I’m aware you can click into each job manually and that yes we could setup a log-reader workspace to do this. Or that we could setup a postprocessing task. I have setup such before - but it’d be nice out-of-the-box in the Jobs screen.This could be a standard parameter or a FME Flow parameter that can be set by the author (e.g. linked to a Published Parameter).Thanks.
Hi there,In this article (Configure user attribute mapping with Azure AD SAML Provider – FME Support Center), a group claim can be setup to pass an AD group name which aligns to a role in FME Flow. As a group claim could be based on a search criterion and many AD groups could be returned which is a common method of Enterprise group membership in ArcGIS Enterprise software. I was wondering if this could be used to grant several roles to a user?Looking at controlling access to repositories, who can view workspace and who can run a job, etc via AD group assignment.Thanks
If you have ArcGIS Pro installed on your machine then you have access to the arcpy.geocoding module with its Locator class, which can be serverless. This puts geocoding in the hands of anyone with (say) a file-based locator.
Optional input port for some transformers is a good feature introduced in FME 2025.2 :-)I think it would be better if the input port of these transformers could also be optional.DateTimeStamper ParameterFetcer HTMLExtractor HTMLToXHTMLConverterAre there any others?
using a shift or control (any key combination with left click-drag) to “turn off” autosnap to transformer ports, connection lines, etc while moving one or more transformers into an area where there may be one or more connection lines right where you want to place the new transformer.
Please add Import from Feature Cache (Similar to AttributeFilter) where you can populate the right side or the left side from values. It would also be great to have some multi row copy/paste (drag drop) to be able to easily adjust the mappings withing the mapping editor.
Hi,Right now it’s impossible to dynamically push a different Headers to the HTTPCaller. You need to already set all the Header names prior to using it.That will be awesome to be able to use, like a Json to push the Headers you want to use, like the Body section.
Currently there is not a connection, transformer or writer that connects FME directly to Azure Key Vault or other key vaults. This would be needed to access sensitive information such as Client IDs and Secrets for Web Connections (to Portal, ACC etc), stored on Key Vault, that we could then incorporate directly into our workbenches. As a company, security is becoming ever more important and this connection would greatly help us to manage sensitive data in line with corporate requirements. We recognise that the existing web connections have decent encryption in place, however there is a general push to have a more standardised approach to storing ids/secrets at a company level.
Hello all,I regularly perform a merge (eg. FeatureMerger) where I am only interested in generating a list on the requestor from the supplier and don't want any attributes merged other than the list.In this case I usually set the Accumulation Mode to "Prefix Supplier" with a prefix of something like "_ignore_me" and then optionally use a BulkAttributeRemover to remove these merged attributes.It would be nice to have an Accumulation Mode option like "Do not Merge" or "Only use List" such that only the list defined under "Generate List" is added the requestor and no other attributes are mergedThis would also be useful in the other transformers that merge/accumulate attributes.
If the information product you want to manage is a hosted feature layer view then at present you need to manually adjust its properties outside your ETL workspace, or tackle the job using the ArcGIS REST API via HTTP.Feature service views offer ways to leverage feature services other than simply as they were created.Here is an example of using HTTP calls to automate one property - swapping the source service for the primary layer..In addition to basic create views, add and remove layers, and delete views functionality for views you might want to adjust name, description, tags, visibility, extent, editing access, definition query, field visibility or layer order.The definition-adjusting REST calls (addToDefinition, deleteFromDefinition, updateDefinition are tricky to implement.Who would like this capability?If you have other requirements comment in this idea.
The FME Sharepoint reader and writer support a boolean "attachment" field, but can't read or write attachments themselves. It would be great if FME added support for attachments so that they can be downloaded or uploaded via the Sharepoint reader/writer.
There are times where I would like to store more complex data as a parameter, I have resorted to json or comma-separated text (multi-line text parameter). This is mostly private parameters that specify data source, configuration or possible some initial data. In nature this data is tabular, most of the time.Maintaining json/text in a parameter dialog today is cludgy Getting the data into features requires multiple transformersNow, wouldn’t it be nice with a table parameter type? Using a parameterfetcher could possible read the table into a list, that one can just explode/expose. Maintaining the data in a dialog set for a table would also be easier… !
With libraries like DuckDB in your Python environment, as is the case where you either have ArcGIS Pro or Enterprise on your FME machine, or you’re using ArcGIS Data Interoperability, then it is easy to retrieve potentially millions of features from an S3-API compliant object store or other web source in seconds, but it’s a bottleneck to send them on into the workspace one by one.I would like to see performance like reading CSV files brought to PythonCreator.There is a possibly related existing idea Introduce a Python Dataframe Creator/Transformer but I don’t want to confound using dataframes with this idea, which is fundamentally about performance.Dataframes might be how this idea is implemented, but my guess is that would be a heavy lift for Safe, another way might be a way to output an aggregate feature.
When we create a Web or FME Server connection in a workbench it automatically gets prefixed with myname. And then when published to Flow we get the same…. is it possible to create these connections with a more generic name, I’m not connecting as “robert.mcpherson” and it doesn’t seem to matter but this is an enterprise system and I will be for ever named in the system. I guess it doesn’t matter but it doesn’t seem a very “enterprise” thing
It would be really useful to have the option to install FME Flow with no pre-installed workspaces, no inactive automations/schedules, no topics/publishers/subscribers, and no AR App. Just the bare minimum: the Admin user and the default roles (fmesuperuser/fmeadmin/fmeauthor/fmeuser).All the other utilities and example content could instead be made available through FME Hub for download as Flow projects when needed.On a related note, I also question the usefulness of the default “Temp” resource connection, since it isn’t required by the system and isn’t automatically cleaned up. For new users especially, its presence can be confusing.
To my surprise, there is no option to add headers to an HTTP request in Automations. We need this option. In the meantime, I'll go back to Workbench :(
As an FME Flow User, I want to be able to redirect jobs from one queue to another so that it may run on a different engine while another job is occupying the originally intended engine. At this time, the only relative alternative is to cancel the job and resubmit using advanced drop down and selecting a job queue. The problem with this solution is that this interrupts an automation’s natural flow.
Data Virtualization is a great tool to share functionalities and data translation results standardized. One great thing is the automated creation of an interactive Swagger UI documentation, which enables also non FME users, as one part of the target group of Data Virtualization, the accessibility to translations. Comparing the Flow Rest API with an automated Data Virtualization, it would be great to have the opportunitiy to authenticate with a token. Missing field for token in Data VirtualizationToken field in Flows REST API (v4)
Currently the JSONValidator only validates syntax, whereas the XMLValidator allows for Syntax only or Syntax and Schema.The Schema can defined as per http://json-schema.org/documentation.html
It is really handy to set multiple presets in 2019 and this makes life a lot easier but i'm missing the ability to share these settings/presets with my colleagues. I would really love to see an export/import button or even a linked button so when changing some of the settings in the given preset it also changes everywhere else.
I need a method for ensuring workflows that include a single workspace being run multiple times triggered by the output of an automations writer are run consecutively not in parallel.In a similar way to how the Workspace Runner in FME Form ‘Wait for job to complete’ option works.I don’t want to simply assign these to a queue with a single engine as I want the first job to pick the first available engine.
[A Message from Product Management at Safe Software] This idea has been refined from its initial posting. The original description is preserved below for reference.The built-in “admin” user account in FME Flow lacks support for modern multi-factor authentication (MFA), including 2FA or passkey-based login. This poses a security risk, especially in environments where organizations rely on local accounts as a fallback when identity providers like Microsoft Entra ID (formerly Azure AD) are unavailable.This idea proposes adding support for MFA/2FA/Passkeys to FME Flow’s local system accounts. This is critical for scenarios where SAML, LDAP, or Entra ID integrations fail or are temporarily offline, leaving only the local admin account as an option for access. Dear All,I would like to see FME Flow provide native built-in feature to allow users to run FME Schedule or default Dashboards -→ JobHistoryStatisticsGathering.fmw use Microsoft Entra SSO (FME) SAML-based admin accounts.We purposefully disable local “admin” users, since FME Flow doesn't provide 2FA/MFA option for local users but set it up to use Microsoft Entra SSO (SAML) with mandatory 2FA.Your support is appreciated :)Cheers,Reno
It would be nice to format the font (i.e. increase size, bold, etc.) in FME Flow Automation annotations. I understand that font size is somewhat controlled via browser settings and/or monitor resolution, etc. However, it would be nice to be able set bold for text elements or underline headers, etc. If it could mimic the functionality in FME Form, great, if not, a few font/text formatting options for annotation would be helpful.
The usage of deployment parameters greatly simplifies switching between test and production environments.When workspaces are used in automations, which reference deployment parameters in transformers, it is not possible to get the information from the corresponding “Run a Workspace” action whether deployment parameters are used and what values they have. The workspaces must first be downloaded and opened in FME Form to see, if deployment parameters are referenced in FME elements.Is it possible to use a symbol in the action to indicate that deployment parameters are being used and to include an overview of the parameters with values in the “Parameters” tab ins the action details (for example like in my sketch)?Thank you very much.
I would propose a switch in the backend to show or hide the geometry definition input field.My users are all over the place, some of them understand that they have to set a point on the map and are confused thereafter when a whole GeoJSON syntax shows up instead of coordinates they expect, others think they should write a street address in the input field (mistaking the input field for a geolocator), etc…I can leave a message for them but hey, how many will read it...
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.
