FME Form

Hi, what is the best way to avoid loading in a complete huge singel Parquet file (58 million records). My memory keeps running out. I have the option to split the data into smaller pieces with a Tester transformer selecting the data on date and then saving it to further process, and have done that with a child-parent workspace, and a tester + published parameter for the dates. But it still seems to go wrong due to memory problems. The process takes very long to save by date, especially loading in 58 million records before running the data through the tester. I would love tips on splitting up the data before loading in the complete 58 million records. or any other tips for dealing with Parquet files.  

I get the following error when I feed XML as an attribute from the XML updater. Reading in an XML file is no issue though.This FME edition does not recognize the `' function. Please ensure that the current platform supports this function, the function name is spelled correctly, and that you have installed all required plug-ins 

Hi all, I have created a Choice user parameter and am importing the contents of the choice list (or dropdown) from a spreadsheet.The Import appears to decide itself how it wants to sort the contents of the list. Is there a way this sort can be turned off?I have already sorted the contents of the spreadsheet into the order I want, but the import kind of ruins this order, and then I am forced to manually reorder with the up and down arrows. This isn't too bad with a very small list, but is pretty frustrating with a much larger list (which I have). Plus I know I am going to have to refresh the contents of this list semi-regularly in the future.Does anyone know of a way around this? thankskeith 

I am trying to extract the 1st and 4th word but it just keeps grabbing the first 4 words, I've tried without the ^ and it doesn't work

Hello , i am new with working with geospatial pdf files.I have Problem that geospatial pdf has no real coordinates .My pdf consist of lots of building ,i am asking if I provide one building with real coordinates manually Will be possible to change coordinates of all other building to real coordinates automatically .what should I do to change coordinates of pdf to real coordinates ,any ideas ?Thanks in advance .FME 2021

I have situation wherein I need to add fields to all the feature classes within a SDE. I cannot seem to figure out how to loop through the SDE, and operate on one feature class at a time, though. My first thought was to generate a list of all the feature classes in the SDE using two FeatureReaders, but then I was not sure how I would iterate through that list.  Any assistance would be greatly appreciated. Thanks!

So apparently, FME does look for fields that might be a good candidate for creating a point geometry in both Excel and CSV readers (see https://community.safe.com/s/question/0D54Q000080hfENSAY/excel-format-of-attribute). This means that if in an Excel or CSV file, there is a row with column names such as X and Y, FME will automatically try to create a point with the contents of these cells. In the top-rated answer of the above link, it is mentioned that for the CSV reader, this will only occur if the option "Scan for types" is activated. This option does seem to be missing for the Excel reader, though, or am I overlooking something? I have the problem that I am using a FeatureReader to dynamically read from an Excel file (the specific file

This is not a real question, it's more an assesment. In FME 2022, it seams that every query run with SQL Executor/Creator is run as a prepared statement (see here for more information on PostgreSQL https://www.postgresql.org/docs/current/sql-prepare.html). The problem is that only certain query can be prepared, not every one...Some of my queries can not work anymore like every "CREATE" queries that can not be prepared into PostgreSQL. Is there a way to disable this behaviour ?  For me it's a critical problem because some times we need the queries to be run as we write them, not like the database want to redefine them.

We have one floating licence and it is now stuck on an old laptop and won't release it. Can I force it to release either from the device or from the licence server?

With a postgres writer it appears that if Bulk Insert is set to Yes, if you attempt to write a string value of NULL to a varchar column that has a Not Null constraint it is interpreted as a genuine null and results in a fatal errorError was 'ERROR:  null value in column "value" violates not-null constraintIf Bulk Insert is set to No, the value is written as a string as expected Is this a known issue? This is in FME(R) 2021.2.6.0 (20220727 - Build 21821) so may have already been addressed.

Hi, I'm running into an issue that I don't understand the solution to.Using the geometries attached here, I can not clip the geometries using a tolerance of 0.001 .It does not give an error, it just gets stuck.If I dissolve the clippers before clipping, it does work. So I'm guessing there's something 'wrong' with the geometries (there is a strange pointy shape in the polygon) and the chosen tolerance, but I don't know why it gets stuck. I'm guessing the pointy-bit is too small and falls under the tolerance? I'm using fme 2022.1.2. Workspace attached and data attached.

Hi, see the attached, I'm reading MVT Tileset and want to fan out the output file GDB dataset on zoom level and fanout the feature types too on fme_feature_type, I get an error saying the schema doesn't have the correct element, but it does! All help appreciated. The fmw is self contained, it downloads MVT tilesets (not too big).The error at write time is:ERROR |Esri Geodatabase Writer: Feature Type Definition 'bl' was requested, but not defined. Please visit https://www.safe.com/support

Hi,I have a process where I am using a DEM to apply values to vertices of polylines using vertexextractor, point on rasterextractor etc (see below). But I end up with gaps where the z value has been attributed as -32767. These have occurred where the vertices are just outside of the clipped DEM - is there anything i can do to interpolate?

Hi all,​I have a lengthy workbench running. I can see the temporary files being written to the relevant 'wb_run...' folder specified in the translation log. The workbench has just finished writing outputs of an AreaOnAreaOverlayer, and those outputs are now being used as the input to a Dissolver, which is in progress at the moment. I can see a 29GB file named 'FME_OBJC_DissolverInputAreas...'. Is this a temporary hold of the AreaOnAreaOverlayer output features? If so, is there any way to use this temporary file and access/export the features in it? In case it is relevant - I have feature caching turned off. Thanks!😀

I am looking for some suggestions on how I could create a workspace that would detect no intermediate junction between linear objects that share connectivity between them and are of a different asset group and/or asset type. an example would be a HV conductor connected to a LV conductor without a point feature in between.All suggestions welcome.thank you :)

GeoCode transformation working very slow its take whole day to get the result just only 28K records is there any way to run faster

Hello ,How to open file has extension pfd or DGS format from power Factory DIGSilent ? The only exporting format from power factory is PFD or DGS ,There is no license or other option in power factory to export other formats .is there a method to open PFD or DGS file in FME and export the data from it ?FME 2021

Hello I'm creating a bespoke solution whereby the bottom of the page shows the Workspace App. The FME App will basically allow staff to submit comments. I've used an existing App for testing purposes which required the use of SSO (Azure). The Workspace App URL is simply enclosed within an iframe tag. <div class="embed-responsive embed-responsive-16by9">  <iframe class="embed-responsive-item" src="https://fme.mole-valley.gov.uk/fmeserver/apps/SubmitComments" allowfullscreen></iframe></div> The page shows both contents so all good.When I login as usual, it loops back to the log in page within the iframe.The Workspace App works as usual on its own. So I decided to create another Workspace App which does not require authentication. U

We update hundreds of shapefiles quarterly. The data we use comes from many sources, but we transform all the data of a single type into a standardized "Master" schema for that type (for example we get Parcel data from 70 counties, each having their own schema, and we translate and transfer that data into a standardized "MasterParcel" file). Using the above example we currently use 70 different workspaces containing 70 different instances of AttributeManager. To hopefully make this process much more efficient, we have developed is a database table containing rows like "ProviderSourceTable", "SourceAttribName", "SourceAttribType", "SourceAttribLength",...."TargetLayer", "TargetAttribName","TargetAttribType".... basically all the data AttributeManager needs to do the job for each data provider/data type combination. The only problem is we cannot figure out how to use this data to dynamically popu

Hi, With an http caller I execute get requests. This is done in a loop to enable pagination. Sometimes the server that receives and handles the request has trouble handling it and returns an error message. Is it possible to put a decelerator after the rejected port of the httpcaller and then redirect it to the inport of the loop so that the last get request is executed again and the workspace continues? 

I have 10000 GML files that I have read in as a single merged feature type. The task is to convert them into FBX files, maximum of 300 MB per FBX files. How do I accomplish this? I have tried modulocounter but doesnt seem to work.Also, I have tried fanout, and am trying to see if I should run a script to separate those files into folders first, prior to FME. Any ideas?

I'm querying a table with geometries (PostGIS) and the resulting geometry column when run through a GeomtryReplacer just ends up with an empty geometry. The encoding is Hex Encoded OGC Well Known Binary. If I copy the string (with copy paste) into an attribute creator then the geometry gets replaced.  The only differnce that I can see is that the string attribute from the SQLExcecutor is encoded as utf-16LE and the attributeCreator string is in UTF-8. It's kind of strange because another geometry column in the same database comes in as a utf-8 string.  It's an envelope. This is the string/geometry in question:01060000A0E86400000100000001030000800100000005000000333333335C021C41000000C052BF54410000000000A05C403333333354021C41000000C052BF54410000000000A05C403333333354021C41000000C052BF5441CDCCCCCCCCEC5C40333333335C021C41000000C052BF5441CDCCCCCCCCEC5C40333333335C021C41000000C052BF54410000000000A05C40I

I have a fairly straightforward workspace writing out one AGOL feature class with a large schema (American Community Survey data). After running the writer the translation log shows 26 features successfully written, however, the resulting AGOL feature class only has one feature.  I am stumped as to why this might be happening as I've worked with AGOL writers pretty frequently and never encountered anything like this. Any advice?

When attempting to load in a VRT Raster, created with GDALBUILDVRT that is in Int8 Format, FME says it is invalid. I am licensed to use the VRT format. I am using FME 2021. I am using the correct VRT Raster format when loading it into my workspace. It seems rather strange that FME would not support VRTs in Int8 formats. I am using GDAL 3.7.1-1 to create my Raster VRT's. Either using the commandline tool directly or manually creating VRT's in QGIS results in the same error.  GEOTIFF writer: Setting feature type level option 'geotiff_compression_method' to value 'deflate' for feature type 'changedetect'GEOTIFF writer: Setting feature type level option 'geotiff_jpeg_compression_level' to value '25' for feature type 'changedetect'GEOTIFF writer: Setting feature type level option 'geotiff_deflate_compression_level' to value '6' for feature type &#3

When using the AzureEventHubsCOnnector to receive in stream mode, the documentation states "Stream: The connector continues to receive messages indefinitely. The translation needs to be terminated in order to stop reception." Am i missing something here? how do you send a termination to it? In python this is straightforward...