FME Form

Hi everyone,I’m working with the JSONUpdater transformer in FME and need to add a key called layerDefinition to my JSON. The goal is to have:"layerDefinition": {}as an actual empty object, not:"layerDefinition": "{}" // stringor"layerDefinition": nullHere’s what I tried:Value Type = Plain Text, Value = {} → Result is "{}" (string). Value Type = JSON/XQuery, Value = {} → Result is null. Value Type = JSON/XQuery, Value = json:object() or map {} → The key doesn’t appear at all in the output.My JSON path is correct (json["operationalLayers"][0]), and the update type is Insert into Object.Question:Is there a way in FME to insert an empty JSON object using JSONUpdater without resorting to a string or null? Or is the only workaround to use AttributeCreator + JSONFragmenter or JSONTemplater?Any advice or examples would be greatly appreciated!

HiIs it possible to connect fme to a Microsoft Fabric Lakehouse? And read data?I use fme 2022.2 

A caseA column code contains both unique and identical values ​​(e.g., A, A, B, C, C, C, D, E, E).A second column names contains different names.I want to group a column, but the names associated with it need to be merged.How can I easily solve this in FME? Thanks in advanced!

I have a column containing values ​​(contains 1 or more sorted names) that are both unique and duplicate. Now I want to assign a unique number to each value. This means that the values ​​that are the same should also have the same unique number. How can I handle this within FME? Thanks in advance!

I am trying to connect to a SFTP using the FTP caller. I have followed the instructions in https://support.safe.com/hc/en-us/articles/25407554321805-Using-a-Private-SSH-key-with-FTPCaller-to-Authenticate-an-SFTP-Connection-for-FME-Form-and-FME-Flow-on-Windows-Systems and I can connect to the server using other methods (winscp, filezilla) so the new key is working. In the log (with debug enabled) I get the following:SystemCaller: Executing 'start "" "C:\Program Files\PuTTY\pageant.exe" --encrypted C:\PPK\PrivateKey.ppk'FTPCaller (FTPFactory): Using the 'SChannel' TLS library for HTTPS/FTPS connectionsFTPCaller (FTPFactory): FTP info:   Trying ??.??.??.??:22...FTPCaller (FTPFactory): FTP info: Connected to ??.??.??.?? (??.??.??.??) port 22FTPCaller (FTPFactory): FTP info: Failure establishing ssh session: -5, Unable to exchange encryption keysFTPCaller (FTPFactory): FTP info: closing connection #0FTPCaller (FTPFactory): FTP/FTPS/SFTP transfer error: Failed initializationSystemCaller_2: E

Hi All, I have the following set up where / separate's attributes - where Value X and Y can exists multiple times within a month linked to the Item ID.Place / SubPlace / itemID / Year / Month / YYYYMMDD / ValueX / ValueY /  SUMXPlace / SUMXSubplace / SUMYPlace / SUMYSubplace /What i would like to do is after i have the sum per month per place and subplace, is to see the sum again of these sums on a quarterly basis - ie. Jan-Mar = Q1, April-June = Q2.So the final table would be something along the lines of this - to still allow filtering per place/subplace/quarterPlace / SubPlace / Quater / QSUMXPlace / QSUMXSubplace / QSUMYPlace / QSUMYSubplace /Any ideas or help would be greatly appreciated!/Emma  

I’m using FME 2024 without any issues related to fanout. However, after installing the 2025 version, I’m facing a problem where the fanout is not working as expected. This seems to be a software bug.I am using a Creator to write shapefiles with folder fanout. Based on an attribute value, the data should be written into separate folders. But currently, it is writing only to one folder ("Low"), while the other two folders ("Medium" and "High") remain empty. Attached test1 fmw file.The Shapefile Writer is skipping features and showing the following warning message:“Shapefile writer: Ignoring feature with unexpected feature type 'High_priority_data'... Last line repeated 19 times ...”In FME 2024, this workflow works perfectly and writes to all the folders as expected. I believe others may also be experiencing the same issue in the 2025 version.

Recently upgraded a dev environment to FME Flow 2025.1.1We have some workspaces that use FileCopyWriters to move files.These workspaces still run on FME 2025 but now throw out a warning message for each fileThe move request details the correct move, but then there is a failed to delete message that appears to be saying it can’t delete from the destination not the source. The files are in fact deleted from the source, so it’s actually all working as it should be and as it did previously with the exception of these warning messages. I note that the FileCopyWriter now has a delete option so wonder if this is related somehow. Can anyone confirm they see the same behaviour in 2025.1.1.  The same warning error messages is seen in FME Form 2025.1

We have FME Form/Workbench installed on some of our Azure Virtual Desktop (AVD) machines, and it works fine overall. There is this recurring issue where the Connection Storage access is lost between each session. This is disruptive, and not always apparent, which wastes time and effort troubleshooting. In 2025.1.2.0, navigating to Web Connections under FME Options shows this each time:And navigating to the Workspace Defaults section yields:My connection storage is a Shared Database, which I thought was part of the problem, but I have more recently learned that a colleague of mine is using the default Personal Database connection storage and is also having the same failure to decrypt data between AVD sessions.I’ve searched around the Community and the web some, and am familiar with Changes to Connection Storage in FME Form 2024 – FME Support Center and the list of known issues, but was unable to find anything related to this. Has anyone else run into recurring connection decryption fail

Okay, hear me out 😄In december we’ll be doing a webinar about Best Practices, but because it’s a bit of a tradition to make the last webinar of the year a fun one we’re going to call it the Worst (Best) Practices webinar. So we’re looking for workspaces that are complex, undocumented, badly designed with renamed transformers.Way back in the day the training materials included the Chamber of Workspace Horrors Powerpoint, that’s the kind of stuff I’m looking for.Feel free to reach out to me privately, anonimity is guaranteed 😃(Normally we do our webinars in Dutch, but depending on the response we can also do this one in English, so if you’re interested in that also please let us know!!!)

Hello fellow FME enthousiasts, We have been running into an issue when using the new Esri ArcGIS Feature Service [Writer] (from the new Esri ArcGIS Connector v3.13.2 package) to write to a Hosted Feature Layer on ArcGIS Online. The Hosted feature layer is created by Survey 123.Previously we used the old Esri ArcGIS Online (AGOL) Feature Service [Writer], since this won't be supported anymore in 2026 we started experimenting with the new one. When we try to update the hosted Feature layer with FME using the old one, all goes well, data gets written as expected and the geometry is visible.But when using the new one from the package we get a few warnings and no error's. The process tells us it's succesfull, we see the new records appearing in AGOL and the attributes seems to be as expected. The geometry is not visible either. When we try to zoom to a record nothing happens. When i open the layer in ArcGIS Pro nothing happens either when i try to zoom to layer/or zoom to selection. It's li

Hi, I’m working on a script which copies data from an Esri arcsde geodatabase to a postgis database. I need to know if the table in the arcsde geodatabase is versioned before I run the rest of the script, and I have a pythoncaller which runs something like this (simplified here):layername = "schema.layerName"connectionFile = "path_to_connection_file"arcpy.env.workspace = connectionFiledesc = arcpy.Describe(layername)isVer = desc.isversionedfeature.setAttribute("versioned",isVer) self.pyoutput(feature)Now, I don’t want to type in the path to the connection file manually, but rather obtain it by some function which can collect the data from the list of database connections I entered in the settings. Is this possible? I’ve looked into the FME Web Services API reference for such a function without any luck. Best regards,Jacques 👷🏼 

One of the greatest new features of the IFC4x3 schema is that we can write 2D lines to IFC, which was not supported in IFC2x3.Why does the new IFC writer support only 3D surfaces and solids and not 2D line geometries? 

It seems that in the iOS18 update Apple made changes to the HEIC file container that means the FME Feature Reader using HEIF (High Efficiency Image File Format) now fails to read the HEIC files.We are currently using FME 2022.2.Is there a later update that will support iOS18 images? Is there an alternative way to read in these images successfully? 

I’ve been trying to load some .nwds files, and they’re taking days to fully open. The files themselves aren’t very large—around 180 MB—but some of the geometry has extremely large coordinate offsets. For example, certain objects are located around X=10,000 and Y=10,000, while others are positioned near X=1,200,000 and Y=1,200,000. Could these large differences in coordinate magnitude be contributing to the slow load times?

Goodmorning! I want to write a dataset with the postgis writer to GeoServer, but the result gives errors.  Process: A reader of municipal boundaries, and a reader with fire hydrants, the fire hydrants gets clipped on the boundaries layer, and then i want to write it to Geoserver. The postGIS writer says its succesfull, but when i inspect the layer in QGIS, there is no data, and i get the error: there is no item in geometry_colums. What have i done wrong in the postGIS writer? Can anyone help me?  Thanks in advance! 

HiI draw some 3d lines from point cloud but sometimes I snapped the wrong point so now I need to check if all my vertex are correct. So in the picture I need to detect the vertex encircled in yellow any suggestion ? thx Francesco 

I'm trying to setup the children and parents list attribute for export to CityJSON. Therefore I thought to use the listbuilder, but I get a different result than it should be. This example is how it should be: And this one is my result. It looks like there is an extra layer in it.How can I solve this?

Hi, I have a report i make where from the data i get out the following (where / separate's attributes)Field / SubField / YYYYMMDD / ValueX / ValueYFor the last date of the Subfield, i would need a line where Value X and Value Y are zero - the end date is specific to the Subfield. Is there a way to get it to reach - for Subfield - find last date - create +1M from YYYYMMDD with X and Y equal to zeroI know how i would do it in a excel macro, but i dont know how to replicate this in FME /Emma

Hey all, Currently having an issue using the ArcGIS Connector when using FME Flow (or Form without Feature Caching / Counting on), receiving the following error below. Full log posted below the error message. The outcome I am looking to produce is to update a CSV file hosted on an Enterprise Platform. This does run with form with feature caching on. Any help would be much appreciated.Thanks. Ash135 2025-11-14 17:25:19 | Message Type: fme::internal::_v0::py::Exception 136 2025-11-14 17:25:19 | Python Exception <TypeError>: Call to pyoutput() requires an output_tag argument when multiple output ports exist 137 2025-11-14 17:25:19 | Error encountered while calling method `input' 138 2025-11-14 17:25:19 | ArcGISOnlineConnector_2_CALLER (PythonFactory): PythonFactory failed to process feature 2025-11-14 17:24:59| 0.0| 0.0|INFORM|Current FME version: FME 2024.2.1.0 (20241202 - Build 24801 - WIN64)2025-11-14 17:24:59| 0.0| 0.0|INFORM|System Time: 20251114172459 UTC2025

Hi Everyone!I’m currently working on a workflow using a dwg file (water system) to convert it into a geodatabase. I was wondering if anyone has any tips on the transformers and workflow I could use to successfully do this? Right now I have my dwg reader as individual types, leading into an attribute filter, and geometry filter (3 different ones- 1 for water, drainage, and sanitary). These 3 end up writing into the geodatabase. I would like to have feature datasets separating water, drainage, and sanitary written in the geodatabase. I understand this is a big undertaking, but any tips (or perhaps links to guides) would be extremely appreciated! Thank you!

Hi, I have a list of 100+ WMS services our company is using in a plugin for QGIS. I’d like to publish these to our enterprise portal, and I thought the ArcGISOnlineConnector seemed like the right choice, due to it haven the setting:Action: UploadURL : WMS URL https://***.net/geoserver/dai/wms?service=WMS&request=GetCapabilities&version=1.3.0 URL Item Type: WMS ?  My error is :ERROR |Message Type: fme::internal::_v0::py::ExceptionERROR |Python Exception <TypeError>: expected str, bytes or os.PathLike object, not NoneTypeERROR |Error encountered while calling method `input'FATAL |ArcGISOnlineConnector_CALLER (PythonFactory): PythonFactory failed to process feature What have I misunderstood ? Or is there a better option ? Thanks in advance.  

Good morning.I’ve been testing lately a FME Flow 2024 version.When writing “Date” field types using the ArcGIS Online Feature Writer, the following warning masked as an error is thrown:D:\code\fme\source\engine\plugins\python\fme-arcgisonline\src\arcgisonlinefeatures\arcgisonlinefeatures_rest_writer_helper.py:557: DeprecationWarning: datetime.datetime.utcfromtimestamp() is deprecated and scheduled for removal in a future version. Use timezone-aware objects to represent datetimes in UTC: datetime.datetime.fromtimestamp(timestamp, datetime.UTC).The FME 2025 Change Log shows this entry:It only mentions the Reader. Would it be possible to know if that deprecation was solved for the Writer, too?Thanks!

Hello, I am attempting to copy an ArcGIS Server Feature Service from our Production portal to our Staging portal as a test migration using ArcGIS Data Interoperability (FME). The migration runs successfully, and the layer is created on the destination portal with all point features appearing correctly. However, every attribute field in the destination portal is empty for the layer.For this workflow, I used: Reader: Esri ArcGIS Server Feature Service Writer: Esri ArcGIS Portal Feature Service The strange part is that when I inspect the data immediately before the writer, all attribute values are present and correct in the Visual Preview. But once the writer publishes the layer to the portal, the attributes do not appear, only the geometry is transferred.Could you please help identify what might be causing the attribute values to be lost during the migration?

I have a job that downloads 7 text files from an FTP directory, and for the last 3 days, 6/7 files have downloaded but 1 always fails (the same one each time) with the error: Number of redirects hit maximum amountI am aware of the post below, but as no similar setting exists on the FTP caller, I’m lost. I can upload fine, and manually downloading the files by logging into the FTP site, or using a WinSCP script also works. The file that failed today was empty (0MB), but so are some of the other files it downloads, and yesterdays file had records and still failed. Empty files are not uncommon for this job, and it has never presented an issue before. This has started occurring since upgrading to 2025.1.3 from 2022, and occurs on both FME Flow and Form. This error has so far not appeared in any other workspaces that use FTP callers, even for multiple file downloads at once. I’m leaning toward this being an issue with the file, rather than the FTP, but I cannot see a problem with the file t