Shape the future of FME with your ideas
Exciting Updates to the Ideas Page! Learn more about the recent changes we’ve made to bring more transparency, ease of use, and a clearer view of how your ideas are progressing.
Open ideas have been reviewed by our Customer Success team and are open for commenting and voting.
There are certain situations where it would be nice to have a Group-By on the matcher and a corresponding Complete Groups When...
The issue with FME vs Shapefile datatypes has been around for nearly a decade. See It would be really nice if there was a parameter to use explicit numeric types on the reader. I should not end up with a different schema if I read a shapefile and write it dynamically, going straight from reader to writer.
I would like there to be a level of access to job logs which allows a user to See any jobs run by any user that are within a repository they have access to Download any job logs for jobs that are in a repository that they have access toSimilar to what is described hereGrant read only access to a user based on a role | Community
Hello,The URLs of host FME Flow services used to be customable. Now only the option "Change All Hosts" is available. While simplicity and user experience is important, let’s not lose the flexibility FME Flow has offered for many years. Networks and architecture become more and more complex and dynamic in organizations.
I have multiple FME Flow instances, it’d be helpful to be able to import/export deployment parameters from/to the deployment parameter store on FME Flow. At the moment I can create via Form and Create/Remove via web interface. A nice import/Export button here with the option to overwrite would be great: Thanks!
Replacement pairs are currently specified as a space separated list which can get quite hard to reade.g.Being able to input pairs as pairs would make it far easier to work out what should be replaced by what
It would be helpful to be able to execute external programs and scripts from within an automation, without having to use a workspace and accompanying engine.An example use case would be starting long-running batch jobs as part of an automation, without having to block an engine while the external job is running.In short, something like the SystemCaller but as a native automation action.
I’d like to request native support in FME for reading B3DM files, which are a core component of the OGC 3D Tiles standard. B3DM is widely used in web-based GIS applications, especially with CesiumJS and other 3D visualization platforms.Currently, FME can write 3D Tiles but cannot read them. For workflows such as converting 3D Tiles to OBJ, IFC, or CityGML, a native B3DM Reader would be extremely valuable.Direct import of B3DM would greatly enhance FME’s capabilities in 3D GIS and BIM workflows and improve interoperability with open standards. I hope this feature will be considered in a future release. Greetings Michael!
Some ArcGIS Server FeatureServices serve Relationship Classes, similar to Relationship Classes in FileGDBs.These don’t show up in the Esri ArcGIS Server Feature Service Reader, however. It would be very convenient if FME would support these Relationship Classes, as some FeatureServices don’t have foreign keys in their tables. In such cases the Relationship Class is the only way to join tables. Relationship Classes are listed in [ArcGIS Server URL]/FeatureServer/relationships. For example: ttps://basisregistraties.arcgisonline.nl/arcgis/rest/services/BAG/BAGv3/FeatureServer/relationships
Problem:FME apparently only looks in the MDSYS.USER_SDO_GEOM_METADATA table for metadata on Oracle spatial data. When it can't find this, it currently falls back to assuming data is in two dimensions. This is not ideal given the rapid increase in use of 3/4D data.FME advice currently (not updated since v2015.1) is to manually create the metadata in the USER_SDO_GEOM_METADATA table. With the release of Oracle 23ai, this is something that's no longer advised because Oracle now creates the metadata itself. This also introduces further problems when the tabe owner updates their metadata, but the user copy then won't be updated.This appears to be an assumption by FME that assumes that the database connection user is then automatically the schema/table ownerProposal:Any data using the Oracle Spatial reader should continue to first search MDSYS.USER_SDO_GEOM_METADATA and then progress to MDSYS.ALL_SDO_GEOM_METADATA should there be no valid metadata This would support this user-schema database setup, but additionally maintain any user-defined metadata where this is required.
***Note from Migration:*** Original Title was: An "Exclude/Ignore Selected Models" option within the DGN (v8) Reader parameters The option to "Read Only Selected Models" is a great feature that really speeds up the time it takes me to process DGN elements through a work bench. However its a bit limiting as we have some drafting workflows where we have automated processes that dynamically name models. If the parameter to "Read Only Selected Models" could have a similar option to Exclude or Ignore Selected Models it would be quite beneficial to us. Maybe a checkbox "Not" or even a conditional option that would common string parsing options such as StartsWith, Contains, EndsWith with wildcards.
Reading DGN files (IGDS) can be very time consuming. Sometimes it is only necessary to read the attributes without having the geometries in FME. The idea is to introduce a parameter in the reader to not return FME geometries (return FMENull geometry) to speed up reading. This would be similar to the "Read Geometries as: Non-Spatial" option in the Revit reader.
The documentation for the IGDS (DGN) reader states that "Design files consist of a header, followed by a series of elements. The header contains global information including the transformation equation from design units to user coordinates, as well as the dimension of the elements in the file" But apparently it is not possible to read the header information with FME. Obviously reading scale and coordinate system information is important if you are trying to automate conversion of CAD to GIS so this change would be very useful.
I would like to see a conditional value option for 'Cell Name' in the DGN-styler.A conditional value is already possible in the color and lines option. However not with Cell Name. This would be useful; if how have multiple points with a different symbol. Now you can define one symbol through the DGN-Styler and the other symbols you can define with the condition-value in another transformer (for instance the AttributeManager) by changing the attribute igds_cell_name.
As well as point and graphic cells, Microstation has parametric cells, in which the exact geometry of the cell depends on user input. More info at https://docs.bentley.com/LiveContent/web/MicroStation%20Help-v19/en/GUID-0D6C4E6A-2E0E-4004-B464-4B58440FA893.htmlWould be great to support this in the future! (FMEENGINE-78421)
I would like to be able to use the source DGN as the definition of the cells when writing out to a new DGN. For example, changing the coordinate system. I want cells defined exactly as they are in the input file to be used for the output after the transformation. I mostly have no access to the original cell file
Currently, you can only set the font of a DGN text by using the font number. You can use a DGNStyler or set the igds_font attribute for this.With FME, you cannot read a mapping between the font number and the font name from the seed file. Moreover, this mapping can vary for every seed file. So there is no easy way to set the font by the font name. My current workaround is to create a new seed file in MicroStation and add some texts, and for the text string value I type the font name. When FME reads this DGN, we can create the mapping between fme_text_string and igds_font. For the actual seed file, I remove the text objects again.Having the possibility to set the font directly by the font name would be much more convenient.
Please add CRS EPSG 20047 BCSG2020 to the list of available CRS in ReprojectorEPSG:20047 GDA2020 / BCSG2020 -- Spatial Reference
When selecting output format JSON, datetime attributes should be default be written in ISO 8601 format. There should be no need to format each datetime attribute manually before sending to a writer.Not only is ISO 8601 the de-facto standard for JSON, but it is THE international standard for exchanging serialized dates between systems.In addition, when reading JSON, FeatureReader should detect ISO formatted dates and automatically transform them into datetime data type.
hi there,In Automations you can select multiple items and edit their tags together. This is not possible with Schedules.I would like Schedules to have the same tag management options as automations.Thanks,Marc
Being able to store the HTML, JasvaScript, CSS, or other static content in FME Flow Resources folders and serve them via Data Virtualisation Manual Endpoints directly (without a workspace that feed the Data Virtualisation Writer attributes) would be more straightforward and engine saving way to provide HTML pages, so the Workspace Endpoints could deal with the dynamic contents.OK, it needs the ability to set the content_type header as well.
Hi,Wouldn’t it be great if FME could automatically generate a link to your API docs, just like the FME Flow API does? This would make the APIs you create almost self-documenting and ensure compliance with the Dutch national API guidelines.
Hi Safe team!I think it would be handy to create some separation of job logs, between Data Virtualization and ‘standard’ FME jobs.Depending on what you create with Data Virtualization, the volume of jobs which trigger could be substantial - and I think users will have very different needs for those two categories.As an administrator, it will be very challenging to find particular ‘standard’ job(s) amongst all the DV job logs.You may also want to configure the settings quite differently for the age of logs to keep. It’s much more likely you’ll want to keep logs which trigger as a batch job weekly/monthly for a longer period of time (i.e 3 cycles so you have a comparison between jobs). Whereas with the DV job logs, they’re probably a bit more disposable, and not something you’re likely to be looking back on a few weeks/months ago for a single API request.
It would be great if the email of the user running an app could be delivered as an automation app parameter so that the person who pushed the button could easily be emailed without additional workflow or having to enter via a parameter which is open to error.
In certain cases, a model has been revised or changed, while you have the FME AR app open. It would be great to be able to refresh/reload the file (same filename etc but with the latest changes), without having to send/download/upload. E.g. produce files directly to the cloud, and prompt the user that a newer version of your file has been detected (if you always get the files from e.g. Google Drive or similar).
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.
