Transformers

I have a very large dataset that I usually split into 3+ pieces when I export to shapefile, then zip them up together. I'd like to use the shapefile writer to do both without setting feature thresholds (hi, FeatureCounter!) and using an external compression utility. However, when I leave the Split Dataset... checked and check the Create Compressed Shapefile option, all that's in the SHZ is the last portion of the split data set (FILENAME_3.shp, etc.). Can I use both? If so, what trick am I missing? Shapefile haterz to the back - it's sadly what The People want.

I'm trying to make the difference between two DEM in an overlap area. Already how do I create the overlap area of the two DEM (raster) in FME. And how do I calculate the difference in pixel value height between raster 1 and raster 2? thanks you!

Hi FME users,  I am looiking for good geometry examples of any corrupted or errors that can / cannot be cleaned with the Geometry validator 2013. I found the following list: https://support.safe.com/hc/en-us/articles/25407709561741-Invalid-OGC-Geometry-Examples which is a perfect start but i would like to be able to hilghlight the following issues at least: http://docs.safe.com/fme/html/FME_Transformers/Content/Transformers/geometryvalidator_issues.htm  Does anyone has a complete collection of geometry sample errors?  Thank you.  JP.

Hi, I have 2 sets of features (terrain and buildings) that derives from Sketchup source file in coord WGS84 and are read via FeatureReader in to the workspace. When I combine the two features via FeatureHolder and look at visual 2D preview they both have the correct geo-location, . But if I switch over to visual 3D preview the building features seem to get an offset? If I add a Reprojector transformer (having the source coord read from feature WGS84, and destination coord to SweRef 99-18-00) they both line up in both 2D and 3D preview. Any ideas why this is happening? Please see attached image. Thanks. 

Hello, I am trying to extrude a set of building footprint layer. The building footprint layer is in a 2D shapefile format. Since the dataset doesn't have building height and base height attributes I tried to extract the elevation and building height from point cloud data using Surface draper. I attached the workaround here:  I run the Extruder transformer, but the 3D visualization of the building does not represent how a solid 3D building should look like. The top of the building is not flat and when I click on a feature there are lines inside the building model (seems like it generated multiple edges??). How could I generate a building model that looks somewhat like the Lod1solid building model? What are the correct steps?

I'm trying to extract building footprints from GEOTIFF raster images. I have raster Orthophotos (Aerial imagery), Digital Terrain Model (DTM/DEM) and a Digital Surface model (DSM), but I do not have the building footprints. My end goal is to use the DEM + DSM to extract the height of the buildings and extrude the building footprint to 3D models. I found this article: https://smartcarto.com/gis/detect-buildings-in-satellite-imagery-using-machine-learning-fme-vs-arcgis-pro/, but it doesn't really go into detail, and since I also have DEM and DSM, not just orthophotos I thought I migh thave other possibilties. I've extracted building footprints from Point Cloud date before, but this time I only have a

I have a shapefile that just contains a couple of attributes. Its attribute values need to be mapped to a GML that needs to be of a structure provided by a XSD file. The XSD asks for more attributes than the shapefile provided, but I want the GML still to contain all the attributes, despite that a lot of values are not provided by the shapefile. Which steps do I need to take to convert the shapefile in to this XSD-conform GML?

Hello, I want to create a raster for each different source_point_id (number of flighline) in a single las file.Do you know how to extract the source point id from the las file in order to generate a raster on the points of the chosen source_id point? thanks you for your help!

I have over 10 list attributes that I need to explode. Setting-up 10 ListExploder in a row seems tedious. Is there another way ?

I have a workspace using several HTTPCallers. These make heavy get calls, that tend to have random errors in FME 2022.0 but only when concurrency is activated (>1). Once it is disabled (=1), it works fine. We have the same errors on version 2021.2.1, but there is no option to enable/disable concurrency. Is there an option somewhere ? We need to use this version as it is the version of our server.Thank you.

Hello, We use an FME Web app to upload a single file to a workspace for subsequent processing. Is there a way to submit multiple files to the workspace at the same time without having to submit them individually ?

I have flatten xml file, and some data are missed. For example <p>Свободная ёмкость — <b>627 млн.м<sup>3</sup></b>        <span class="up">11</span></p> it should be 627 млн.м, but it skipped and only values 3 and 11 are in place. In xml ex1 i have highlighted items which are missed in xml example.png. Also attached workbench.

Hi, we have 3D polygons inside the database. Now we wanna convert them into Multipatch. Oracle support this kind of geometry.I did it by using FaceReplacer.Inside the 3D polygons we have already some kind of splitter elements (area<0.0xxx). Is there a way like Simplifier or Dissolver to unify these features? That we can eliminate these elements during the process to create Multipatch features.Furthermore it seems that after multipatch generation there are some gaps between the elements:  If the 3D polygon is closed uand after using face replacer it should be closed too right? 

Hi, I need an easy way to split a string at a desired lengt, but I want to avoid splitting words. There once was a WordWrapper but it's not working anymore. Any suggestions?

Hello everyone, I have a dataset contaning several polygons, with sometimes hundreds of vertices and one containing lines, each polygon is paired with a line by a unique ID. I am looking to build a workspace that reads the coordinates of each polygon vertex, projects them along the corresponding line and then calculates the two projections futher down each side of the line or its extension and returns their coordinates. I have used VertexCounter and CoordinateExtractor or CoordinateConcatenator to get the number of vertices and their coordinates and I have a workspace that can calculate the projection of a point along a line, but I am having trouble connecting the two through a Custom Transformer Loop to cycle through all vertices. Any help would be appreciated. Feel free to suggest any improvements to the overall thinking process as well.

I am running into a issue with OffsetCurveGenerator. There are some instances where it drops nodes from the original linework and breaks the output line into an aggregate, I assume this is beacuse it folds on itself.The uppermost grey line below is the original. The green lines are generated offsets at various distances. Note that the one at the bottom has lost a vertice towards the corner. The linecount coming out of the transformer is correct, but if I run a deaggregator, the total number of lines has increased. As the red line illustrates it has split at the corner. This is not an issue in most occurrences, but I want to run some Snippers on the start and end vertex downstream and the Snipper treats the lines as if they where deaggregated. I tried working around this by first running a Deaggerator followed by a LineCombiner. But if I do the lines sometimes flip their original direction (depending on the order in which it was rebuilt), which results in the end vertex

We have tried using FME but it always generates an error saying that it "Could not open File Geodatabase dataset". Below is the log error: FME API version of module 'FILEGDB' matches current internal version (3.8 20210302) Geodatabase Error (-2147024893): The system cannot find the path specified. Could not open File Geodatabase dataset `C:\\Users\\guillermo.rendon\\OneDrive - SYSTEMWARE INC\\Documents\\PCS\\Database Tables\\SYSAPP1CLONE copy of PCS database\\SWIS.GDB' for reading FILEGDB reader: A fatal error has occurred. Check the logfile above for details Failed to obtain any schemas from reader 'FILEGDB' from 1 datasets. This may be due to invalid datasets or format accessibility issues due to licensing, dependencies, or module loading. See logfile for more information FME Session Duration: 0.3 seconds. (CPU: 0.0s user, 0.2s system)END - ProcessID: 23880, peak process memory usage: 44604 kB, current process memory usage: 44120

I have set up a very simple reader/writer combination that reads an ArcGIS raster from a file geodatabase (File Geodb Raster Dataset) and writes the raster to an ArcGIS Enterprise SDE SQLServer database (ArcSDE Geodb Raster Dataset). This operation fails repeatedly with the message:  “Failed to write raster data tile when writing to raster dataset 'r1_max'. The error number from ArcObjects is: '-2147155566'” I am able to successfully write the raster to a different file geodatabase, and am able to successfully write vector layers to the SDE SQL Server Dbase, so my database connection file is correct. This same error has been referenced in other postings, but I have not seen a solution that resolves the issue. Any help is greatly appreciated.Software versioningOS: Windows Server 2019SQL Server Enterprise: 15.0.2080.9FME Workbench:  v2020.2.1ArcMap:  v10.8.2

I have a string in a attribute. At the end of the string there are brackets containing one or more words. I would like the contents, including the brackets to be put into a new attribute. Ex:Attribute = "cat mouse (cheese crackers)"Desired new attribute = "(cheese crackers)"NOTE: sometimes there is one, two or more words with in the brackets, and I want all of them in the new field. Using a RegEx (\\(([\\w]+( [\\w]*)*)\\)$) I can select that portion of the string, but I don't know how to put that portion into a new attribute.  Currently I have an AttributeCreator with Attribute Value = @FindRegEx(@Value(Attribute),"\\(([\\w]+( [\\w]*)*)\\)$")but the result is not the string, but numbers.  Thank you in advance for your time.Julia

Hi All,I'm attempting to take a table that has 600K+ rows that have an individual value for a type of utility per location in it, join/merge the up to 8 values into one row to be able to join to another feature class at a later date based on the location unique id. I've tried aggregator, feature merger and feature joiner. Nothing is giving me the desired results. I've used the aggregator to get the unique location id, then used feature merger or joiner to attach the additional columns to the one location id. Tried to create the list, still not getting desired results. I get only one filled in. I've used testers to put yes in the named utility column based on the code type, and then where the values remained missing/null, I used the NullAttributeMapper to change them all to N in order to concatenate the data and get all columns accounted for, still missing some of the values.It's likely something that I'm overlooking, but have been working

I am reading in a CSV file. I need to reference the first cell (Column MA and row 1) and extract the date and time from this cell, and use these two values to populate two new columns - Date and Time, and add these two new columns to my output along with existing columns such as MA, 1, 2, 3 from current CSV files. I like to get hep on how to reference and extra the cell value, and then add the two columns through AttributeManager or AttributeCreator. Thanks in advance for your help.

From this Q&A;: Geometry Template, problem getting list attributes when using variable for indicesIn fact, I've never used Geometry Template in the XMLTemplater, but I noticed that the GUI of the transformer won't recognize a Geometry Template when testing a solution for the question above. This configuration works as expected, but the property button is still red and the tooltip reports that "GEOM" does not exist.

I have GeoJason data from USGS site which contains Epoch time such as 1650016756409 or 1649948565590. I tried to read them in either %s$ or %Es% using DateTimeConverter, all failed and cannot convert. I am using FME desktop 2021.2. Please help.

Hi,I have a dendritic river network (lines only) which I would like to use to generate catchments upstream of certain points. I have run the NetworkTopologyCalculator which generates entire river catchments (by assigning all joined branches a network_id which I can then group. But I would like to specify certain segments for which I would like to generate an upstream network. Is this possible?I would like to do this en-masse and have multiple catchments generated, defined by the origin segment id.Thanks

Hi everyone,  I'm still quite new to FME so I'm hoping this is not a dumb question! I am trying to write out basemaps in 3D perspective for a client. The goal is to get a more realistic image by avoiding the orthographic view. However, whenever I try to output the basemap, I get the orthographic view. In 3D view the Data Inspector does give a nice perspective view. Whenever I switch to 2D, the orthographic view is visible. I feel like using the ImageRasterizer transformer (or any other transformer I tried) will force the data into 2D and removes the perspective.  To showcase the desired output perspective versus the current output perspective, I've attached two images: the perspective output (2nd image) is what I'd like! They're just of a rectangular box that has been rotated in 3D.