FME Form

I am using the aggregator transformer to take multi part polygons across NI, grouped by Local Government District, so when the data is displayed and selected on a map the records in one district will be selected as a multi part. I want to keep all unique identifiers and addresses for the records, as well as other attributes. I have currently got all the attributes to remain by using the assemble hierarchy section in the transformer using LGD as parent id and unique id as the id, but the group by doesn't seem to be working as the records are being selected as individuals polygons, not multiparts.

How do I set a xml element to xsi:nil=true in a XSD Driven XML Writer?I tried to set with "name.nil" xml_boolean. No luck. I changed the type from xml_boolean to xml_char. No luck.I created a feature_type for <name> element and set the attribute via a "nil" attribute. No luck. The output shall look like this: <aixm:name xsi:nil="true" nilReason="inapplicable"/>nilReason works without problems. 

Dear all ,  It's a hollow PIPE , I want to extract XYZ for each a) centre point (c1)b) face centre point (v1&v2) + face INNER vertex point (vf1&vf2)c) radius (e.g. distance from v2 to vf2)d) total length (e.g distance from v1 to v2)  thank you for kind support

I have different gdbs at different level in a folder.In each gdb there is the same dataset. I want to merge (usually with a junction) the similar datasets from the gdbs (all dataset has the same structure). But I not sure of what reader should I use:I use a reader with feature Type Name as "any schema".And Reader "PATH\\**\\*.gdb" Then I put a FeatureReader , but I cannot get attributes from all the dataset. Any Help will be appreciate. 

I am running checks on data. Any fails in checks, they will be written to a tab in the spreadsheet. If there are no fails, the tab won't be created. In the case of failures, is there a way for me to print a message to say "Checks ok" into a new tab?

Hello everyone,I am having random timeouts on FMEServer when i am trying to connect to a PostgreSQL database (using FeatureReader/FeatureWriter) or when i am trying to realize a simple get/post with HTTPCaller. The timeouts seems to happen randomly, mostly when there are multiple simultaneous jobs.The HTTPCaller timeout is partially solved by activating the option "Retry Failed Requests", but sometimes the maximum value of "Maximum Retry Attemps" (10) is not enough (i.e. the timeout is still there after 10 retries). For the FeatureReader/FeatureWriter there is no retry option available.A similar problem was found here

Hello, I tried to make a workbench that write ifc from our data base (building).When i try to read the final file, FME cannot because of syntax error for the ISO10303-21 format. Each building has one ".." value that cause the problem.#35=IFCBUILDING('O6Hu3fDbRB2OEJH5wcfNLw',#23,'51_102','','',#34,$,'51_102',..,$,$,$); If i change the ".." by a $ (that mean 'Null' according to BuildingSmart) the file is fine and FME is ok to read it. Does someone know what parameter or attribut I missed in my workbench that cause the ".." ? Thanks ! 

Hello All,British National Grid Long/Lat (OSGB 36) [EPSG: 4277] seems to be missing from FME's suite of coordinate systems - or am I mistaken? I have a table with Longitude and Latitude data in this projection and I want to create points from them, and then convert them to British National Grid (Meters) (OSGB 36) [EPSG: 27700]. I can try to use the Reprojector transformer, but I can't see EPSG: 4277 anywhere.Is there anything which I could use instead? Or is there another solution such as creating a custom coordinate system?Many thanks,Jane

Dear All, I have the following setting : I've created a User parameter of type "choice" where the user can tick which attributes he wants to keep and send to the Writer (column1, column2, column3, etc.).  With the MultiChoiceParamSplitter followed by a ListExploder, I can get a table where the attribute "_choice_param_values" holds the attribute names the user has selected (e.g. column1, column3), but how can I use it then in the AttributeKeeper (or any other transformer that would do the job) ?  I've tried to pass the table as input with "_choice_param_values" as Attributes to Keep but it does not work at all...  Or maybe, from the outset, I didn't tackle the issue from the right angle and you have suggestions about how I could I achieve what I want in a totally different way around ? Any ideas are welcome, as I've run out... NB : I've seen the (seem

how can I extract the Shapefile from this link via FMEhttps://tgftp.nws.noaa.gov/SL.us008001/DF.sha/DC.cap/DS.WWA/current_warnings.tar.gz   

I have a batch file (.bat) that I need to run. What Transformers would I need to use to get this to run using a workspace in FME Desktop?

I have two SQL Server spatial tables (one containing current records and one containing archived records) that I need to merge and read the features ordered by their version number in order to do onward processing in FME.  I have been trying to use a SQLCreator to read these records into FME in the correct order using SQL similar to the below: Select a.* from(        SELECT   [fid]   ,[toid_version]   ,[SHAPE] FROM [dbo].[a_boundaryline]         union all         SELECT    [fid]   ,[toid_version]   ,[SHAPE]                       FROM [dbo].[boundaryline] ) as aorder by a.fid, a.toid_versionThis SQL works in SQL Server management studio without issue. However, when I use it in a SQLCreator (or SQLExecutor) I get the following warning(s) in the log and the geometry is not read (i.e. features have no geometry): 

I am trying to set up my anaconda environments in FME, using the custom interpeter function. As interpeter I selected the .dll file (C:\\(...)\\anaconda2\\envs\\gdal\\python27.dll). This succeeds in recognizing Python and having a basic setup. However, my packages are not included. Trying to import geopandas, which works fine in my Anaconda editor, results in the following error in FME:Python Exception : No module named geopandas 2018-12-04 12:50:20| 0.5| 0.0|ERROR |Error executing string `importfmeimportfmeobjectsprinthelloPieterimportossys#sys.path.appendrC:UsersfvzfAppDataLocalContinuumanaconda2envsgdalpython.exeimportgeopandasasgpd#fromrasterstatsimportzonal_statsprinthelloPieter#TemplateFunctioninterface:#Whenusingthisfunctionmakesureitsnameissetasthevalueof#theClassorFunctiontoProcessFeaturestransformerparameterdefprocessFeaturefeature:printhelloPieter#TemplateClassInterface:#Whenusingthisclassmakesureitsnameissetasthevalueof#theClassorFunctiontoProcessFeaturestransform

I am new to FME Server.  I have a workspace that I published to FME Server. It reads from an Excel spreadsheet, and writes to a File Geodatabase and Access Database.  The data for the reader and writers reside on separate servers, they are not loaded into FME Server.  Can someone take a look at the screenshot and let me know if the Published Parameters are correct? I seem to recall that for the spreadsheet (see red arrow in the screenshot), does there need to be a parameter added to replace the UNC path? What you see in the screenshot is how the workspace was published into FME Server from FME Desktop.  Do I need to add the UNC paths shown in the screenshot as Resources? Can someone explain or point me to a link that explains this process? Can I keep the UNC paths for the File Geodatabase and Database Path or do I need to replace these with an FME Parameter as listed here:

Please can you tell me how i do this?

Hi! I have encountered some difficulties as I am trying to check whether certain microstation files contains empty levels. I've studied the feature-reader for Microstation datasets, but I cannot find a way to read empty levels but only levels that has content within. My hypothesis is that the feature reader outputs all the levels that is associated with a feature, and hence, misses the empty layers.For Autocad files, I had the same issue initially. However that was solved by using a Schema (Any Format) - reader on the Autocad-file dataset and hence reading all the layers as fme-feature-types. Any ideas?

Hi guys, I am still learning FME and have made a few workspaces which are working well for me however I am doing a merge with a SQL non spatial and a mapinfo tab file and am trying to make an end result of the poly's having no fill or patterns, just an outline and being transparent. I have tried many things, the closest I have gotten is creating the attribute mapinfo brush transparent and setting to True, and if I create a pattern with Mapinfo styler the transparency works - obviously with the pattern though, I have tried doing an attribute expose and remove for the mapinfo_brush_background and foreground. Sorry I have been trying to do this for a couple of hours and tried all sorts of combos but just can't get it.  Thanks in advance Garry

I need to do a transformation, like pivot table, using Python, and for that I would need to install libraries, like "pandas"...  How can I install custom libraries to FME Python ? I created a environment with Anaconda, installed pandas there, but now, how to use that environment in my PythonCaller?

I am trying to mosaic 228 ECW tiles into one uniform image using RasterMosaicer transformer. The individual tiles are quite large as the 228 individual tiles together are 15GB. I am finding the process using RasterMosaicer extremely slow and is taking close to 24 hrs to complete using FME desktop. Just wondering what the best method is to mosaic these tiles and have a resulting image equal to the size of the individual input tiles, 15GB...or less.

I'm having trouble removing buffer segments correctly when generating a buffer around a line dataset with roads.Input data (road lines with road name and city): When buffering this, I end up with:This method has a lot of overlap that i'm not able to process properly. Is there a way to buffer it like this or process the _overlap >= 2 in a way that this is the result:I tried the areaonareaoverlay and correct the overlap afterwards but did not get the desired result.

I have 3 workspaces; they need to be run like this:​1st to be run is workspaceA, 2nd to be run is workspaceB, 3rd to be run is workspaceC.​I know in Scheduler I can set each to run for example: ​workspaceA runs at 905p, workspaceB runs at at 910p, workspaceC runs at 915p.​But, this approach poses a risk that if workspaceA fails during its run, workspaceB and workspaceC will still run and that could corrupt my data. It seems more prudent that if, for example, workspaceA fails to run that workspacesB and C will not run.​What is the best way to deal with this situation using FME Server? ​Should I create a new workspace using FMEJobSubmitter Transformer or add FMEJobSubmitter Transformer to the workspaceA, B, and C?​​

When using a FeatureReader to read CSV file, one can manually select in the "Parameters" section which attributes we want to read from the file. Very short question : does something equivalent exist for reading only a selection of attributes in a reader that loads "Esri ArcGIS Portal Feature Service" data ? Thanks in advance

Dear All, I am facing the following very weird issue. I want to test for several attributes if, for each of them, there are some features that contain a specific value.To do that, I have created a Custom Transformer that takes as input 2 user parameters (defined exactly in the same way, published and required) : 1) colname = the name of the attribute ; 2) progname = the value to test if the feature is equals to. When I go back in the Main, right-click on the custom transformer > Properties, in the prompt I can see well the 2 user parameters (colname, progname). But when I click on "Run to this" custom transformer, the prompt only asks for 1 of the 2 parameters (always 'progname', even if I switch the order of the parameters in the parameter manager of the custom transformer). In the Translation log, I see that the following command line is run :fme.exe myscript.fmwThis results in translation failure with the error

I'm trying to write data to a csv. I have a uniqe identifier that has two leading zeros. When writing to a csv file the zeros are removed. When viewing the feature information window the attribute has a string value. When doing the same in the written csv it says 64 unsigned integer even though I have set it to string in the writer parameters. What am I doing wrong? Making sure FME writes it as a string will keep the leading zeros right?

"" Error connecting to PostgreSQL database(host=, port=, dbname=, user='postgres', password=): 'could not connect to server: Connection timed out (0x0000274C/10060) ""  How do I get FMe server to connect to my PostgreSQL db stable every time ?