Open ideas have been reviewed by our Product Management and are open for commenting and voting.
SummaryRequest for a new FME transformer (or suite of transformers) that automates the migration of GIS content from ArcGIS Enterprise to ArcGIS Online, including support for feature layers, web maps, web apps, dashboards, Survey123 forms, folders, groups, and sharing settings. Problem StatementMany organizations are transitioning from on-premises ArcGIS Enterprise environments to ArcGIS Online. Currently, this migration process is manual or requires extensive custom scripting using Python and REST APIs. This is time-consuming, error-prone, and not scalable. Key Features🔐 Authentication Support: Secure token-based access for both platforms. 📁 Folder & Group Replication: Maintain organizational structure and access control. 🗂️ Feature Layer Migration: Transfer schema, data, symbology, and relationships. 🗺️ Web Map & App Migration: Rebuild Web Maps and Apps by referencing migrated layers and configurations. 📊 Dashboard Support: Recreate dashboards with updated data sources. 📝 Survey123 Form Transfer: Migrate forms and associated feature services. 🏷️ Metadata & Sharing: Preserve titles, descriptions, tags, thumbnails, and sharing settings. ⚙️ Parameterization: Allow users to define what to migrate and how (e.g., filters, renaming rules).
I would like to see a built-in option in the HTTPCaller transformer to format the response body for development purposes. I know that the HTTPCaller in 2024.2 has the “HTTP Test Request Preview” that shows the response body formatted, but it would also be nice to have the attribute “_response_body” formatted instead of having to use an additional transformers “JSONFormatter” or “XMLFormatter”. Example of HTTP Test Request Preview There could possibly be a toggle to format the response body, or to keep it in raw form. Instead of using the JSONFormatter or XMLFormatter, I would like to have the response body formatted directly in the HTTPCaller.
ProblemFME is case-sensitive with attribute names. When users upload files (CSV, Excel, etc.) with inconsistent casing (Name, name, NAME), it causes: Schema mapping failures (e.g., in SchemaMapper or Joiner). Unexpected behavior when casing isn’t controlled. Extra complexity requiring custom Python for basic cleanup. Why It Matters Real-world data often has inconsistent field name casing. Makes reusable workflows brittle and error-prone. Slows down schema inspection and integration with metadata or APIs. Suggested Solution Add a reader or transformer option to treat attributes case-insensitively. Let users define a canonical case format (e.g., lowercase or PascalCase). Add case-insensitive matching toggles in SchemaMapper and AttributeRenamer. Impact Simplifies file ingestion, mapping, and validation. Reduces Python scripting for common cleanup tasks. Makes FME workflows more robust and user-friendly.
I'm dreaming...Validate and repair geometries automatically.Optimize tolerance parameters automatically.Create a digital map from photo etc. automatically.Generate workspaces automatically!etc.
When opening a collapsed bookmark, transformers and other objects shift out of the way to avoid overlapping, but I'd like to see everything shift back when re-collapsing the bookmark.
Picking up on questions such ashttps://knowledge.safe.com/questions/4786/null-sorter.htmland comparing the FME Sorter to Excel, which sorts empty cells to the bottom (for a numeric sort)...I'd like to suggest adding options to the Sorter or creating a "PedanticSorter" transformer with options allowing the user to sort certain types of values to the end, including nulls, empty strings, non-numeric etc. There could be a drop down box with "sort to end" and "sort to beginning" for each value type. For non-numeric this would apply when you choose to sort numerically but some of the values are not numeric.A further option could be a "rejected" output port. Then the drop down list for each value type would include "Output via rejected port" as a third option.
The FME Server documentation provides multi-step instructions for configuring SSL (HTTPS). The steps include: the need to copy certificates to specific locations, running command line tool, & editing config files. The process is prone to issues for a variety of reasons. Often the skills necessary to resolve issues are beyond the typical FME Server administrator and requires the recruitment of IT Staff and possibly support from Safe Support.Is it possible to streamline this configuration?Could it be done as an option during the installation? (HTTP|HTTPS)Could it be done through the Web UI or some wizard after FME Server is installed?Could there be a way to test the configuration and determine where an issue exists?Is it possible to make HTTPS the default?By reducing the human interaction for this type of configuration there would be fewer issues and a better experience for FME Server administrators.HTTPS is becoming a very common configuration tweak after FME Server is installed.
This would eliminate the need to add an Attribute Creator after every port if I want to use these test results for things further in the process like "Groupby" Functions etc.
https://knowledge.safe.com/questions/114142/translation-log-thousand-separator.html
FME FLOW Version Control could become an awesome feature for FME but currently it has some basic issues that urgently need to be addressed. Few of our biggest challenges are, The ability to establish integration with GIT via FME Flow should be straightforward. Currently we are struggling with even this basic connectivity. The basic Fetch and Push to GIT features are not working. Not even one example works even though manual Command line GIT Connectivity is successful from our end. a. Detailed documentation on FME Flow-GIT hub connectivity with videos/examples along with troubleshooting and whitelisting on Networking including Zscaler should be available. b. Main branch has to be renamed to Master. Such hardcoding should be removed and process should be more flexible. 2. Inability to create branching or work with branches via Version Control. Can only connect to Main branch. Multiple branches cannot connect or integrate. CI/CD cannot be completed without this. How will developers promote and get reviewed on their individual work without this feature? Note - Ideally FME Flow Version Control is more preferred than REST API as Version Control is an integral part of FME Flow and users like to use this integral feature more often than others. FME should try to focus and enhance this integral/inbuilt feature over other difficult time-consuming solutions.
In FME Workbench 2024 and earlier, you could type “MSSQL” in the canvas to easy access the different Microsoft SQL readers and writers. But from FME Workbench 2025 this functionality is removed. Probable due to the Format Consolidation that was made in FME 2025.Could you please bring back MSSQL as alias for the Microsoft SQL readers/writers? With the current change FME has been less user friendly.
My idea is I would like to be able to select a line or other geometry in AR and it would display attributes about the line or other Feature Classes.Thanks,David
It would be good to replace the attribute value mapper with a ‘bulk attribute value mapper’ tool that could do the following: 1. Instead of importing data and it being static - it would be good for the tool to reference a db table/file dynamically so as to always pull in the latest data. 2. Instead of only handling one attribute at a time, it would be good to update multiple attributes at once from multiple lookup locations. Examples Source AttributeDatabase/FileFeatureTypeLookup Code AttributeLookup Value AttributeMATERIALOracle GIS_DBLU_MATERIALLCODELVALUEBEDDINGBeddingLookup.xlsxSheet 1CodeValue In this example the Material AC would be value mapped to Asbestos Cement and the bedding type SND would be value mapped SandI understand that this can be done at the moment using a series of feature mergers with process duplicates enables, but this is something that comes up quite frequently as we have multiple datasources that only store the codes and regularly need to be updates to have the description field.
Allow the selection of an attribute in the adjacent features functionality of AttributeCreator. Currently in a situation where I need to calculate rolling averages across an attribute and the number of prior features to be considered varies with each run. The number is stored in an attribute and ideally I would be able to select that attribute here:
It should be possible to set the AGOL feature service name within the AGOL Writer using an attribute.
This would be a transformer that would simply unexposed all attributes without having to do it manually like in AttributeExposer or PythonCaller. Not sure why this is not an option and why they have made it so hard to do things dynamically with new attributes when the schema changes.
I know this is a specific one, but hear me out.Using @CurrentAttribute in the attribute manager is extremely helpful and can save a lot of time. Its especially useful when you need to make the same change to multiple attributes.Rather than modifying each value to account for the field name differences, I can use @CurrentAttribute and copy/paste the value calculation across all attributes that need the same change. I would love to do the same in the test clause left value when using the conditional value option from attribute manager. Currently, it doesn’t seem to work as expected.
I like to have the option to disable (and enable of course) parameters.For example specific user parameters. So not all user parameters at once, but very specific user parameters in the User Parameter management window.
The Safe Software Product Team is looking into the possibility of developing a Connector for 12d Synergy. Right now, FME Form ships with a 12d Model Reader and Writer, which enables users to work with their .12dxml design files. This Reader and Writer, however, does not work with the 12d Synergy system which means that users are missing out on the storage, metadata, and permissions capabilities that 12d Synergy provides. Either that, or users are having to create an API connection and workaround workflows which can create extra overhead. We are looking for users who:Would love to see a 12d Synergy Connector: please upvote this Idea! Would be interested in being contacted to participate in early testing and providing feedback: please comment on this Idea!
It would be nice if FME Server supported the modern, standardized OpenID Connect authentication technology to enhance the security experience of logging into FME Server.
The Date time conversion takes only single input type (FME, ISO or dd/mm/yyyy etc.) It does not allow or take multiple input types. Many times, the input data (Excel files, CSV files) comes with different string for date type field resulting the date time conversion rejection. To address the issue with the date type we have created a custom transformer. The custom transformer uses the FME “Date Time conversion” transformer multiple times with different input type for rejected part. If the FME “Date Time conversion” gets enhancement to support multiple input date types the workspace will be simplified and will improve the performance also.
Just for clarification - this is not about fme workspace parameters, rather about SQL named parameters.Named parameters provides better security against SQL injection, improved query plan caching, ensures proper type handling among other things. Another advantage is that it makes it easier to just copy/paste the sql-code from a sql editor (like ADS, Dbeaver etc) where one would be provided with prompts to fill in the parameters.It would be amazing if one just name a sql paramter the same as an attribute in the feature, and it would be picked up that way… !
hi there,I am working on a translation for some heavy 3d data. This creates lag when I want to check on some feature attributes in the visual preview window as FME has to load the geometry in the graphical preview window.I would like to turn this off so I can just see the table and not the geometry.Thanks,Marc
The SQLExecutor / SQLCreator should support bind variables rather than string substitution in SQL statements for the following two reasons:Security, preventing injection attacksPerformance, see http://www.akadia.com/services/ora_bind_variables.htmlI suggest making bind or string substitution an option within the SQLExecutor.
An Export Action Item completes the backup FeatureServer item type workflow into a local computer or another cloud storage system like S3 or OneDrive.The portal UI for this Action Item is shown below:The backup workflow would be a combination of Export, Download, and Delete action items. Adding a List Action item in the beginning would complete a batch export and download.In addition the Export Action item can be used to download formats of the FeatureServer for use in other clients beside ESRI applications.The best part to the export API (on FGDB), it includes the attachments, if there’s any.Reference: https://developers.arcgis.com/rest/users-groups-and-items/export-item/ Current available solution is the Hub transformer from Abley. Its a great tool but it’s limited to FGDB. Quite the obvious after all since its called ArcGIS_AP_FILEGDBExport.
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.
