FME Form

I have a large postgres database with over 100 layers. I would like to know how, and if, the indvidual layers relate by a shared attribute. For instance: a river will have an attribute with the county-name it flows through. However, the county will not have an attribute with the river name (but wil have an attribute with it own name). Both will have a completely different geometry obvioulsy. I am also not interested in creating that attribute (using spatial relator). I just want to know which features share which attributes. Since the dataset is so large I can not use the matcher since I will not know beforehand which attribute might match. How would I best go ahead to try and find these relations?

Are there possibilities to do an automatic texturing with oblique aerial photos in fme? Ther eare some software like CityGrid or RhinoCity who can do that. They only need some parameter regarding orientation of cameras. We have an production workflow for CityGML buildings in fme and we would like to texturize within the workspace or by implementing existing functions or something else. Does anyone know if this is possible by using FME?

Can somebody please confirm that I'm understanding this correctly:https://knowledge.safe.com/articles/23883/working-with-geodatabase-subtypes.htmlLimitationsAt the time of writing, FME will not allow you to associate different domains based on a particular subtype. For instance, you would not be able to set range domains of 0-50,000; 50,000-100,000; 100,000-250,000 and apply them to a subtype definition containing the values of Small, Medium, and Large. In order to achieve this, you would need to create the domain:subtype relationship in ArcGIS.In our database (ArcSDE 10.3.1 w/ Oracle 12c) we have a feature class which has a subtype field and within each subtype there is defined a different domain for a "subclass" field. I want to be able to resolve the subtypes and domains (both in and out of the subtypes) but the above paragraph makes me think this is not possible for domains defined within a subtype. I have had no success in my testing as of yet - only coded values are

Hiwindows 10, FME desktop 2020.2.5 I installed scipy using pip"fme.exe python -m pip install numpy --target C:\\Users\\<user>\\Documents\\FME\\Plugins\\Python\\python37 " I check numpy version and it says 1.16.5 using pip list in commando prompt. But when I run the script in FME desktop it saysERROR |C:\\Users\\<my username here>\\AppData\\Roaming\\Python\\Python37\\site-packages\\scipy\\__init__.py:149: UserWarning: A NumPy version >=1.16.5 and <1.23.0 is required for this version of SciPy (detected version 1.15.1 UserWarning) But the only version of numpy that i have installed is 1.16.5.  Any suggestions on how I can fix this problem? 

I have a workspace that has two SQL Executors connecting to an Oracle database (the same database connection). In between there is processing that happens. It appears that FME is keeping the oracle connection open from the first SQL Executor for the entire duration of the processing, and using the same connection when it runs the 2nd SQL Executor. This is causing a network timeout error to occur. To prevent this I have tried unchecking the Persistent Connection option in the 2 SQL Executors, but still get the same behaviour.  As a test, I have logged the oracle session ID from the first SQL Executor to the log file, and in the 2nd SQL Executor I have logged the session ID and session duration. Both session IDs are the same and the session duration is the entire running time of the workbench. See attached screenshot for workbench, the decelerator is to simulate processing time.  So to summarize, the Persistent Connection check box doesn't do anything in FME Desktop or Server.

I'm creating DEM rasters (1m cell size) from Lidar using RasterDEMGenerator. Then Hillshades from the DEM rasters using RasterHillshader.I'm getting hillshades that look like stacked contour layers.I believe the hillshades "should" be smooth surfaces (especially Lidar surfaces) so I'm not sure what I'm doing wrong.I've attached my model and sample output.

My Data moved from Network drive A to Network drive B.Is there a way to mass change the Source Paths from the Readers in multiple Workbenches at once or do i need to open every Workbench and need to change it manually?

Hİ,I have one las file and all point cloud data is in one class. How can I classify this las data and export only ground data from it? 

We'd like to publish the /fmerest/v3/ API in an API Manager.It's well documented and structured here:https://docs.safe.com/fme/html/FME_REST/apidoc/v3/index.html#I was wondering if there's also a OpenAPI definition of the services.

Hello, i have an e57 file with size over 100g. It contains 50 pointclouds and 300 raster. So i wan to reduce the point cloud size .I've 2 questions​ I'm not sure if my workflow for this is correct. Cloud someone give me advice？In transformer "Writer" exits much options, i'm not sure the one i've choosed enough to export and save the minified file Thanks in advance​

I have a file in format of .e57, it contains point cloud and raster.I plan to firstly reduce the number of points and export again as e57 that also contians raster and point clouds.​After processing the point cloud data is there 2 options in transform Writer: Raster, Point cloud. ​Which one should i choose? Or can i just export the raster and point cloud separately(like the picture below), and then open these 2 data file in FME Inspector at the same time and save as format .e57?​Thanks in advance.

i converted my shape file to tiff which is basically a road layer of an area. That shp contain a name attribute which shows the road names. I want that name to display on tiff image. Pls guide me  

Can i Convert my shape file to tiff image with geo location ? Pls guide me . 

I have created a workspace with the help of Mr. Takashi(a person from this forum) that reads the shapefile ,creates buildings then these buildings are separated in storeys and then the storeys are separated in roof,ground and wall surfaces. All this work is attached.I want to find a way to write this information in CityGML (Lod 2).In conclusion i want a workspace that gives me a CityGML writer with the buildings.

Hello everyone, So I have a workspace, at the end I will output the HTML result by using HTML transformer. For my html report, I would like to have my writer naming: input file name + _level For instance, if the input file is: abc.geojson, the output file is: abc_level.html I tried as the figure below, yet it is not working. Could someone please give me guide on how to make it works. Thank you and sorry for soooooooo many questions recently. Zach

I can set up a zip file extractor with a user parameter but how do I then connect it to the reader to have it extract the shapefile from the zip file?

Hello,  I am a basic user of FME and have created a workbench merging many tables. I need to add a further table, but before I use the FeatureMerger I need to specifiy a WHERE clause on the reader. I have tried doing thison the reader with no joy. I have also tried with the FeatureReader and no joy. I cannot find any help documentation in the reader to help me with syntax. Is there a simple transformer I can use?  I need a very simple clause to pass only certain attributes from one of the fields into the next transformer.  Can someone assist please? Harriet

in the data inspector , and it looks like its just those two columns in the screenshot. is m new to fme and I could just be missing something many thanks in advance

Hello, I noticed something weird with my reader.The primary key from a Geodatabase_SDE table is interpreted as int type on my 2021 version but is correctly interpreted as a double type on my 2019 version. Reader's properties on 2021 version :Reader's properties on 2019 version :Allowing reader feature type editing won't change a thing. All the id values are set to 0 through the inspector.Also, double fields type from the same dataset are well interpreted. Both versions. Thanks for having a look.Regards.

Hi there guys (and girls), I've been trying to combine several 3D-geometries (arcs, polygons, lines) into 1 big polygon.  What I have now are seperate parts (a window, a floor, a wall and so on), but what I actually want is a clean building that is only clickable as a whole.  I already managed to combine the different geometries into a geometry collection with the Aggregator Transformer, but I was not able to write this geometry to the destination file (OBJ).  Does anyone have any idea how I could solve this problem? I am looking forward to hear from someone else from this planet soon,Claire

Hi,When exporting to Datasmith all the meshes origin is located in 0,0,0. This courses some problems for us in Unreal, especially when we are working in large worlds.Is there a way to make the change in FME so that the meshes have a pivot point in the center point instead? (There was once a time that adding the CsmapReprojector from same to same coordinatsystem worked, but not in FME 2021)

How can I create a generic writer where the user can choose between DXF and DWG amongst other formats? The problem is that DXF and DWG use the same output format, but with the normal writer the format changes when you change the file extension.

i have adobe geospatial pdf and it has lots of points that connects together with different lines ,i would like to extract the lines coordinates and also the length and extract also line number for each line . FME 2020.2 Thanks in advance

Hi, I try to request data from an api by a POST-request with httpcaller. Every post works fine when I set the parameter geodata in the body manually like this:{"geodata": ["FCA","FF"],..."}These are identifiers of features, which I want insert dynamically by an other transformer (Attributekeeper). The attributekeeper delivers one field called BSTs with this content: "MLLF","DPU","WWM" {"geodata": [@Value(BSTs)],..."} As soon as I activate the connection between attributekeeper and httpcaller, the api sends a "403" not allowed. I suppose, that the httpcaller sends the data of the transformer attributekeeper within the POST, what is not allowed by the api. But strange is, that I send allready the authentification-token to the httpcaller to use it in the header. This works... Any ideas are welcome ;-)Best regards, Christoph