Transformers

I use a Sorter to sort on some attributes.When using the Dissolver with a GroupBy attribute and the Accumulation Mode ‘Use Attributes From One Feature’ after the Sorter I assumed the attributes are taken from the first feature of the sorted group. (Basically the same behavior as for a combination of Sorter/Matcher or Sorter/DuplicateFilter).However the Dissolver seems to work differently. The considered feature seems to be random (maybe a geometry condition drives which feature “wins”?).So the question is, how can I use the Dissolver so that always the first feature of my sorted group “wins”?I have an ugly mitigation flow with Tester, Duplicate Filter, AttributeKeeper, FeatureJoiner which is not really smart but cumbersome.Any ideas?Klaus (Version: FME Forms 2024.1.0 WIN64)

My team has a workspace that has ran thousands of times before that has started failing in two different ways. The first failure does not actually trigger FME to fail, it stalls out on the ContourGenerator and does not move. Memory space, RAM, CPU usage, and GPU usage are all fine, none of them spike or max out. In fact RAM usage is quite low. Other times it will fail, at the same spot and give this error:An error occurred. FME will attempt to obtain more information on the error, but this may cause the translation to be terminatedAn error occurred. Here is the information that FME was able to find on the errorget_next_capacity, allocator's max size reachedI found this post from four years ago suggesting to just contact the support team. Any ideas before I do so, or is that still the recommended route to take?Ve

I have a dataset, representing a road network, in a Geographic Lat/Long coordinate reference system. The area covers three UTM zones.I want to calculate the length of roads / polylines with metres as the unit of measurement.What transformer should I use?

I use the AIXM 5.1 Reader with the default settings to read all AirspaceTimeSlice. Some of them are missing geometry. For example, I have a feature which uses xlink:href  to reference the geometry of a CTR. The CTR in itself has an UNION and consists of two CTR's which are also referenced with xlink:href. How can I get FME to get this geometry. It somehow doesn't seem to follow up on the xlink:href. I use FME Desktop 2019.2.3.1

I have a layer that contains 2 point line features i.e. start and end only, and I want to convert the start and end to points but have a count that keeps the parts in order. For example the start of the first line has count of 1 and the end of that line a count of 2, then the next line, start point would be 3 and end point would be 4. Each line has an ID so was thinking to try and group the points by line ID and then apply the count but can’t get the grouping to work as expected.   

Using FME Form 2024.1.2 and QGIS 3.34.4 with Python 3.9 the following script:import fmeimport fmeobjectsimport sysfrom qgis.core import QgsApplication, QgsVectorLayer, QgsProcessingFeedbackclass FeatureProcessor(object): """Template Class Interface: When using this class, make sure its name is set as the value of the 'Class to Process Features' transformer parameter. """ def __init__(self): """Base constructor for class members.""" # Initialize QGIS Application QgsApplication.setPrefixPath("F:/QGIS 3.34.4/apps/qgis-ltr", True) self.qgs = QgsApplication([], False) self.qgs.initQgis() # Append the path where processing plugin can be found sys.path.append('F:/QGIS 3.34.4/apps/qgis-ltr/python/plugins/processing') from processing.core.Processing impo

I am writing a PyQGIS script (Python version 3.9) within FME Workbench 2024.1.2 Python caller transformer which has to run some processing tasks from the QGIS (LTR 3.34.4) processing framework:import fmeimport fmeobjectsimport sysfrom qgis.core import QgsApplication, QgsVectorLayer, QgsProcessingFeedbackclass FeatureProcessor(object): """Template Class Interface: When using this class, make sure its name is set as the value of the 'Class to Process Features' transformer parameter. """ def __init__(self): """Base constructor for class members.""" # Initialize QGIS Application QgsApplication.setPrefixPath("F:/QGIS 3.34.4/apps/qgis-ltr", True) self.qgs = QgsApplication([], False) self.qgs.initQgis() # Append the path where processing plugin can be found sys.path.appe

Hello all ! I need to create CSV where value will end  with a semicolon. I can do it using AttributeManager > @SetValue(SUP); but in that case it would add “ “ between attribute because it is converted to varchar. I need it without those “ “ but can’t manage to do it.Example below : 1st is my output and second is what i want. Only difference is those “ “.PS : Title is correctly formattedCODE_AL;NUM_AL;LIBELLELOL;LIBELLEC;SECTSTAT;ANCIENNE_C;NOUVELLE_C;CODE_POSTAL;PRO;REGION;PAYS;POP;SUP10;10;GEROMPONT Avenue des Déportés 74;GEROMPONT ;25122C11-;Gérompont;Ramillies;1367;Brabant Wallon;Région Wallonne;BELGIQUE;82;"0,099368;"CODE_AL;NUM_AL;LIBELLELOL;LIBELLEC;SECTSTAT;ANCIENNE_C;NOUVELLE_C;CODE_POSTAL;PRO;REGION;PAYS;POP;SUP10;10;GEROMPONT Avenue des Déportés 74;GEROMPONT ;25122C11-;Gérompont;Ramillies;1367;Brabant Wallon;Région Wallonne;BELGIQUE;82;0,099368;

I would like to find for each green point the closest orange building limited by the parcel border the green points are situated in. See the attached image below. The neighbourfinder should do the the trick but should not search in adjacent parcels for closeby buildings. Any suggestions? Ps. This is a snapshot of my data. I have about 700 green points and their parcels and buildings. 

I have a dataset with spikes, and I want to flag records with spikes for manual inspection.I am aware of the SpikeRemover transformer. Is there a method of just detecting spikes? Not removing them but just flagging them? I would prefer to do a manual fix.

Wondering if FME can read QR-codes from images. Input is folder with images with QR-codes and some other content. FME should determine where to save those files based on the QR-code so just be able to read the code within workspace. Has anyone tried doing something like this? 

I have been unable to find a up to date compilation of the AI transformers and connections available in FME, is there such a thing?Specifically, I want to do some experiments with reverse image lookups.

I have a featurereader after a featurewriter which is writing 4 feature types. I have configured the dataset parameter as _dataset as per the output of the reader. What do i need as the feature types to read? I have tried _feature_types{0}.name _feature_types{1}.name _feature_types{2}.name _feature_types{3}.name and _feature_types{*}.name but neither work. _feature_types{0}.name by itself does work on the first of the 4 in that same parameter.

Hello everyone,I am using the StringPairReplacer to find and replace short codes with their corresponding fulltext from a string that may contain multiple codes. It looks something like thisA1, A2(A3), A4 → Fulltext1, Fulltext2(Fulltext3), Fulltext4with the Replacement Pairs: A1 Fulltext1 A2 Fulltext2 A3 Fulltext4 A4 Fulltext4 This works fine, however the lookup table containing codes-fulltext translations may change in the future and I would like to implement a dynamic solution which doesnt rely on changing the workspace everytime something changes in the translation. I managed to create the correct text for the replacement pairs field using the SQL Creator from my translation database but am having trouble inserting this string into the appropriate field as it only accepts text and user parameter input. I am not too familiar with user parameters and cant seem to work out a solution to insert the dynamically created string into the StringPa

Hello,I would like to use FME to search for partial name matches (essentially a fuzzy string comparison) between two columns in different datasets.The challenge I'm facing is that it’s not a straightforward ID-based comparison. I want to pick a name from a row in one dataset and search for the closest match in a column of names in another dataset.Does anyone know a way to achieve this in FME, perhaps without relying on IDs?Thank you!

I have lines and their vertices should be written to XML (using XSD-driven XML-Writer).I extract the vertices coordinates into lists using the “coordinate extractor” .Now I need to find a way to output those coordinate pairs to xml, so that the output looks like this:Is there a way using the “list concatenator”?

Good day,I have a featurewriter and I want to give the dataset the same name as the display value (and not the value itself) of a parameter for which I used a choice configuration. How can I do it ?

I am trying to find a functionality within FME that allows me to replace substrings based on a lookup table. Since I have several hundred mappings, chaining string replacers together isn’t really an option.Basically something like an AttributeValueMapper that looks at substrings instead of the whole value only or a string replacer that accepts multiple string/replacement inputs.Is there such a transformer or any workaround to do this?Thank you! 

Hello,i’ve recently worked with a raster from a NetCDF file and noticed, that loading the raster in FME gives a different raster than loading it in QGIS or ArcGIS-Pro (see Screenshots).Data can be found here: https://opendata.dwd.de/climate_environment/CDC/grids_germany/daily/hyras_de/precipitation/ My question is why? Has anyone experienced this behavior before?    

I am using SystemCaller Transformer to unzip the folders to particular directory using this command""C:\Program Files\7-Zip\7z" x "@Value(source_directory)" -o"@Value(_unzip_directory)" -aoa" This process usually runs at night as subprocess within a FME Workspace. All Zipped folders are at same location. My concern is it skipped to extract one of the zipped folder. To check if there is any issue with the zipped folder I checked by running the FME process Manually and it worked fine(extracted all the zipped folder).I also checked by running the same through task scheduler from same user with which the process runs at night, and again it worked fine and was able to unzip all the folders. What could be the cause of FME skipping it? Any help would be highly appreciated.

Good day,I have the following parameters in my script:AVE_AVENUEBLVD_BOULEVARDCTR_CENTREPL_PLACERD_ROADEach parameter is set up as a choice, when running the script you have the option to chose YES or NO for each parameter.Later in the script I want to retrieve the parameters for which YES has been chosen. I want to create a new attribute whose value is a list of the parameter names (keep only the letters before the underscore). Let’s say AVE_AVENUE, BLVD_BOULEVARD and CTR_CENTRE are set to YES ; PL_PLACE and RD_ROAD are set to NO. A new attribute called “address” is created and its value should be: ‘AVE’, ‘BLVD’, ‘CTR’. How can I do so ? Thanks for your help.

I’m fairly new to FME and need help to get started on a workbench. I have a text file i need to convert into a readable format. Each line has information about a specific device and should be parsed into separate columns. The blank spaces also need to be removed. What transformer should I use? Any help is appreciated!

Hello,so far we are transforming IFC models to geodatabase feature class, which works fine so far. Unfortunately this IFC export step seems one step to many in the overall workflow and I was asked to change the workflow, so that revit project files can be used directly. As fme does have a revit reader I thought it would be possible, but unfortunately despite trying every imaginable parameter setting for the reader it was not possible to read more than one feature that did not contain some kind of meta data...but the “survey point” instead. The fme documentation also only focuses on how to export from revit to ifc and after wards use it in FME.Is it therefore impossible to transform from .rvt files to a multipatch geometry feature class in Esri geodatabases? Or am I missing something?

I am using the Schema Scanner with a dynamic writer in my workspace and it is awesome except that it often incorrectly identifies some of my integer fields as date/time fields. If I turn off the date/time detection then my actual date/time fields get written as strings or integers. Right now I am running it with date/time detection on and then I put an attribute manager on the schema and find the ones it has incorrectly identified and manually adjust them but I feel this kind of takes away from the dynamic nature of the schema scanner. Is there a better way to do this? I wish I could tell it to ignore certain fields or specify the fields that it should look for date/time in but again that is taking away from the dynamic nature of the bench. 

HiI am able download files with the ProjectWiseWSGConnector.Now, I want to download files in a flat set, but this does not work for me.If I treat the flat set as a file and try to download it. The translation is successful, but nothing is download.And I can't list the content, because in list mode the flat set isn' visible, since it is treated as a file by the transformer.Any help would be appreciated.