FME Form

Hi FME folks,I'm trying to reproject these coordinates: SRID=3857;POINT(530385.593659309 6910697.223221438) into a EPSG:28992 point.I notice the result FME gives me differs a lot from other tools / methods - the difference is about 12 cm.My method:1: Scraping JSON: "geometrie": "SRID=3857;POINT(530385.593659309 6910697.223221438)"2: CoordinateSystemSetter: EPSG:38572: Use regex to extract Lat / Lon.3: Create point using VertexCreator4: Reprojector: Source: EPSG:3857 | Destination: EPSG:28992Results FME:POINT (112820.96207905837 513493.7654737713) Other tools used:Internal geocoding tool: POINT (112820.87254041698 513493.83730794006) ChatGPT conversion: POINT (112820.87254041698 513493.83730794006) Google Gemini conversion: POINT (112820.87254042 513493.83730794) Grok conversion: POINT (112820.87254042 513493.83730794)

I’m seeing some weird behaviour on FME Flow (2024.2.2)I’m trying to debug a process on FME Flow which isn’t working properly. I wanted to see what was going on so I threw down a logger, republished, reran.To my surprise the feature never made it to the logger...now I know what you’re thinking. “oh you must be doing something wrong”. Well maybe.But the workspace is so simple. Creator>AttributeManager>Logger>File Copy Writer Here’s a snippet of the log:2025-08-22 12:19:45| 0.0| 0.0|INFORM|Creating writer for format: 2025-08-22 12:19:45| 0.0| 0.0|INFORM|Creating reader for format: 2025-08-22 12:19:45| 0.0| 0.0|INFORM|MULTI_READER(MULTI_READER): Will fail with first member reader failure2025-08-22 12:19:45| 0.0| 0.0|INFORM|Usin

I have a texture embedded in a binary file. I’m trying to extract the textures and apply them to the geometry I’m loading. I figured the easiest way to do this is to read the textures in with the FMEUniversalReader. I quickly found out that I need to pass the input file as a string (filename). Ideally I’d pass this as just a BytesIO object as the library I’m working with already loads the data as a PIL object.At this point I don’r really see a way around writing out the textures to disk and then reading them again with FME. Any other ideas?

HiI have an attribute value derived from an Excel sheet. The input attribute name is “Vægt”. FME data type is number(20,2). An example value for a feature is “2.95”. I’m trying to retrieve this value using the xml templater using <Cell><Data Type="Number">{fme:get-attribute("Vægt")}</Data></Cell>   But the output value in the xml is 2.9500000000000002 Could anyone explain why this is happening and suggest a fix? I have tried attribute rounding and changing the input Data Type but nothing seems to

I am using a python caller to read paths for files from my Input folder. Files in the input folder are ordered in a ClassName and AssetName folders. There is a variable number of AssetName folders under each class name folder. I am transforming data and using a schema mapper to read transformed schema A series of shapefiles is written with a dynamic writer.How can I copy the input ClassName and Asset Name folder structure when writing the output ? I tried using shp dataset fanout but I can’t get the configuration right.

I'm doing a relatively simple task using a RasterCellValueReplacer, im filtering a few rasters by a defined filter parameter.However, if the filter value is typed into the transformer it works fine, if i try and pass the filter value by either user parameter or attribute the result is completely different and incorrect.What's a bit more of a concern, when i use an attribute or a parameter, the process seems to be making its own filter value out of thin air, the raster passes through as if filtered randomly   

I am trying to use the ChangeDectector to compare datasets before loading. First set is AGOL data, new set is SDE data. Every features is considered “changed” because of unexposed attributes that I don’t want to compare. When I have it set to “compare all except”, I am excluding attributes like “last_edit_user” and “last_edit_date”. There are only 4 to exclude and they are the same 4 for all the datasets so I am able to duplicate the transformer. But all records are exiting the “Updated” port because of these unexposed attributes: geodb_type, multi_reader_type, geodb_oid, etc. I dont want to compare these and I also don’t want to expose them first becuse the AGO layers don’t have some of them so I would have to create them and then exclude them. Alternatively, I could use “Only compare these attributes”, but because each set of data to compare have different attributes, I would have to fix the ChangeDetector transformer for each set. Is there a way to discard attribute

Hello, The goal is to convert a large amount of DWG CAD files to SHP files.Each DWG file is located in a specific folder and subfolder corresponding to it’s location and level. Each DWG name has a prefix component indicating the plan’s specific location and a basename indicating it’s feature type.  The prefix component is identical in SHPs and DWGs The basemap depends on a combination of layer(s) and geometry type. There is also a conversion matrix identifying which combination of DWG basename, layer and geometry type result in which SHP basename.Schema mapper & scanner are used to convert schemas based on basename only.I can output the corresponding shapefiles as Prefix_Basename.SHP in one global output folder.However when I

Guys, I need some help with my XML. In some parts of my document, characters like "&gt" and "&lt" appear. Specifically inside the <data_dictionary></data_dictionary> tag, where I perform an update using the xmlUpdater transformer. What could be causing this?

Good day, I have some line segments that I aggregate based on an attribute in common. After the aggregation I would like to extract the coordinates of the first and the last vertex of the aggregated line (in green). I am using the coordinate extractor at coordinate index -1 and 0. The coordinate values I get as the result are not the first and last vertex of the combined line. I think the problem is that the new line is a multi curve. How can I extract these coordinates (in green) ?

Hello everyone,I am using an XML template like the one below to generate fields dynamically in FME Workbench:<field> <name>cod_imovel</name> <description>{fme:get-attribute("cod_imovel")}</description> </field> <field> <name>tema</name> <description>{fme:get-attribute("tema")}</description> </field> <field> <name>dat_atuali</name> <description>{fme:get-attribute("dat_atuali")}</description> </field> My question is: if a new attribute appears in my source data (for example, atrr_1), is FME able to automatically add a corresponding field like this in the template:&

Hello,I’m using a Python Caller in an FME workspace. In the python script I’m trying to read in a gpkg file using the code below: #import geopandas as gpdimport pandas as pdimport osdef FeatureProcessor(feature):        path_gpkg = feature.getAttribute('path_gpkg')    print(path_gpkg)    print("file exists - ", os.path.exists(path_gpkg))    gdf = gpd.read_file(path_gpkg)However, I get an error saying:Python Exception <DataSourceError>: '/xxx/abc.gpkg' does not exist in the file system, and is not recognized as a supported dataset name.It is strange because the line ‘print("file exists - ", os.path.exists(path_gpkg))’ returns True.Has anyone come across anything similar?Thanks :)

Hello,I currently have some workbenches which access APIs and hence have secrets within them, which is fine when this is secured on FME Flow. I also want to set up version control, and have all of my workbenches uploaded to remote repo, GitHub. TLDR: How can I secure my API secrets on GitHub?So far I’ve looked into:encrypting my workbenches  https://support.safe.com/hc/en-us/articles/25407644298637-Password-Protection-for-FME-Workspaces this seems OK, but doesnt truly secure the api secrets, it does add a level of security but it could be better Storing my API keys as published parameters on FME Flow this may be hard to manage, especially when trying to run/debug/design any API workflows

New to FME, have a client using FME and want to read and traverse through a MicroStation DGN file on a number of client machines. What’s the best way to achieve this?Does a license allow me to distribute a DLL for use on their clients?Is there a batch interface I can use?Steve @ Hexagon 

Hi all, I’m new to FME and would like to know if there is a way to generate a v5 UUID in FME 2024? From what I have read I can’t see a method for this. Thanks.

We are upgrading from 2019.2.1.0 to 2023.1 ​I have many AGOL feature services with attachments that we read and then write to a local database for archiving.​For example, I have a feature service that has about 50,000 point features. Only 4 of them have attachments (500K jpegs). I have it set to read in batches of 1,000.​In 2019, if I read this feature service with Attachments = Yes, it takes about 1 minute to read everything. If I open the same workspace in 2023 and run it without updating the reader, it times out every time. Sometimes it doesn't even log that it reads the first 1000 before timing out in about 7-10 minutes. Other times it has read for over an hour but still times out. If I disable attachment reading, it reads quickly in 2023.  If I upgrade the reader to 2023, the reading still times out when attachment reading is enabled. The same is true for FME server. On our old

Prior to Desktop Version 2022. I could read an AGOL service with attachments relatively fast. Now with 2022, it doesn't seem to work at all. I can view the source from the workbench ok, I can read the features in if I disable read attachments. But it doesn't seem to work with read attachments set to Yes any more. The log gets stuck at 'Updating Access Token for..." Updating the token doesn't help either. Has anyone else had this issue? Thanks

Hi everyone,I’m having some trouble with the XMLQueryExtractor. I need to extract from the xml_fragment field the values inside the <name> and <description> tags.Currently, the output from FeatureJoiner is in the format: Theme | ID | xml_fragment What I need is for the extraction to produce the following table: Theme | ID | xml_fragment | name | description

I feel like I must be missing something obvious but so far have not been able to resolve.  I have published a Web Layer from Arc Pro to our organizational AGOL account.  This Web Layer feature service contains three sub-layers.  I’d like to update only one of the sub layers using FME.  This works fine if I publish each layer as its own service, but if published as a single service with multiple sub-layers, I can’t figure out how to target only a single sub-layer. Publishing these as a single multi-layer service has the advantage of keeping them ordered together on AGOL.When I open FeatureWriter to set parameters, I can select the overall feature service, but not an individual sub-layer.  I suppose I can run it without specifically identifying which of the sub layers I am targeting but am a little hesitant as I’m not sure how it would know which of the sub-layers I’m attempting to update.  Here is screen shot of my FeatureWriter interface below.  Road Rock is the featur

i am trying separate stacked point data in to individual points, can suggest in FME or ArcMap 

Hi all I have a list like somelist{0}.val = 'apple'somelist{1}.val = 'donut'somelist{2}.val = 'aardvark'somelist{3}.val = 'apple'somelist{4}.val = 'aardvark'somelist{5}.val = 'apple'  the list histogrammer transforms it into _histogram{0}.value = 'apple'_histogram{0}.count = 3_histogram{1}.value = 'aardvark'_histogram{1}.count = 2_histogram{2}.value = 'donut'_histogram{2}.count = 1  now I want this information as attribute pairs like attribute_name    attribute_valueapple                   3aardvark              2donut                   1 any Ideas how to do this in FME without fancy scripting or looping are much appreciated! 

I’m using FeatureWriter to write data using dynamic schemas to multiple worksheets of an Excel file. In the particular case where I detected the issue, the data meant for the last worksheet contain just one row and value in just one of the columns. This value is ‘No’.As you can see, the native_data_type of the respective schema attribute for this value is ‘string’.In the result excel file, however, the ‘No’ value is changed to FALSE. I can’t figure out why this happens.The Writer writes other ‘No’ values to other worksheets and all of them appear to written correctly as ‘No’.

Dear FME Community,I’m using a dynamic writer in my workspace and have run into an issue with newly added attributes.In my workflow, I added a transformer to perform an analysis and wrote the result to a new attribute field. However, this new attribute is not appearing in the output file.I understand that the dynamic writer uses the schema from the reader, and since this new attribute doesn’t exist in the original schema, it seems to be excluded from the final result.How can I ensure that this new attribute is included in the output when using a dynamic writer?Is there a recommended way to update or override the schema dynamically?Thanks in advance for your help!

Hi all,I am fairly new to FME. What I wanted to implement I was able to figure out with the community here or the trainings at the Safe academy. But now I struggle as I do not understand, where my mistake is.Goal: I need to write features with a dynamic schema. So, I added via Readers a “Reader as Resource”. It is an Excel file from where I want to use a specific sheet. There a added my columns which shall be written to my target source. Writing a FFS file works as expected by using the following configuration.One important note: a sheet in excel has a limitation for the name. As I need to create a dynamic workbench which can read different source files and handle a different amount of dynamic columns per source file, I want to name it like the feature type - but shortened to what Excel can handle.Dynamic Schema De

Dear FME Community,I’m trying to cut lines with segment delimited by start and end point and all works with my method and with the transformer Snipper, but it doesn’t works with line who have auto intersection like a loop.With a “normal” line I’ve this result :But with a line who have a loop I’ve this result :The result should be this :I know how to isolate lines wi