Best Answers

I would like some quick instruction on how to automate the start of an FMW process. The current FMW uses a series of files that it converts based on date (i.e. the date is in the filename). I need a Python script to be able to load those files into the FMW project once a month and then run the conversion process.  I realize that this would be best done using FME Server...but what I'm trying to do is build a "proof-of-concept" for management here (they will see and understand the pitfalls of automating a desktop process to run indefinitely).Thank you

So I'm usually fairly aggressive in large workspaces in keeping the in-workspace Schema tidy and as small as possible after running data through various Transformers that all like to add their own default Attributes and Lists etc.This trimming of the Attributes and Lists at mid-points through the Workspace is under the belief the passing features with unnecessary attributes and lists that are no longer required through the Workspace would cause overall significant performance penalties. This makes something like AttributeKeeper a very common Transformer in my large or data heavy projects.However, I find that using AttributeKeeper, AttributeManager and AttributeRemover all have their own overheads to run, and you could slow your Workspace down by using them too often (well so I believe anyway!) The question is: When and how often should you use these in large projects/workspaces? Is there a best practice in FME for getting a balance on these?

So, I hope I have this right. As far as I can tell, the default behaviour for FeatureWriter is to create any field indices during the CREATE TABLE or ALTER TABLE stage by issuing commands like CREATE INDEX before any features get written.However, this would be against normal mass data append default convention? Writing bulk data against indexed tables can be very slow and in most cases best performance is to create the indices after the data is written. For instance, my FGDB FeatureWriter takes 2 hours to write my current project with the relevant fields set for indexing because every single feature written triggers a resorting of the indices, versus 15 minutes if I leave the field indexing options off.The best workaround I can come up with at the moment is to instead do post writing index creation by manually writing out the the "CREATE INDEX..." statements in the "SQL to run after Write" parameter, but can anyone think of other workarounds or why indices would gen

Hey FME'ers, We have a problem with the TextStroker. It seems like it ignores the rotation of our text features. In attachment an example of our problem.  Note that the textstroker doesn't change the fme_rotation but still the text is shifted 180° around a point different than that of the insertion point.It seems like the problem started since the update to Version_5. With version_4 there is only a slight shift in position (see attachment). I guess we could set back our textstrokers to version 4 to avoid this problem.Our FME version: 2019.1.3.0 (20191007 - Build 19642 - WIN64)we use python 3.6+  Thanks in advance!Kind Regards,Louis   

From the attribute above containing newlines, I'm trying to extract each line of coordinates into a list using a StringSearcher with regex: \\n.+°.+°.+\\nI don't want lines of text to be included in the list that are not coordinates, so the first line starting with "Buoy" would not be captured. It should also be noted that the lines that are not coordinates may contain a '°' symbol, such as "Range line at 186°". Because of this, my methodology consisted of matching any set of characters containing 2 '°' symbols that fall between two newlines.My example regex seems to close, but it is also capturing the first 2 numbers of the next string. How would I need to format my regex to avoid this?

Looking to do histogram matching. From research, the preference is to be in a color space with intensity, therefore HSV or YCbCr. I've tried RGB matching but on the darker sets of satellite images, it does not match well when its 'brightness' increases.   Is there something I can do/use in FME to convert from RGB to HSV ? Can I do this in python? Read in a feature and output a HSV image?

Hello - sorry if this is a very basic question but I'm struggling to convert a JSON response that is received from an API in to tabular data.The response is sourced as per the following:I've included two objects (apologies if 'object' is not the correct term) in the following .txt:I would like to generate tabular data from this response - i.e. based on the attached file I would like a new row for each 'object' (denoted by the "OrganisationID" value) with a selection of the values within this 'object' as attributes on the row.I've read the JSON tutorials and have tried various combinations of JSONFlattener, JSONExtractor and JSONFragmenter but with no joy.I would be extremely grateful for any advice that anyone can offer. Thanks in advance, Matt 

Is there a way to set headers for the HTTP call? Ironically I am trying to call the FME Server API, so I need to set, for example,  Authorization: token fmetoken=YourTokenHere 

hello , i want to design with sql statement on fme .i have coordinate system on a city and i want to know which cables are over lapped the border . if the cables are over lapped the border then change the colour of this cable .please if someone can tell me how can i do it .i will be thankful for him .Thanks

Hi there, I am a newbie in FME. So please accept my apology if my question sounds stupid. I want to calculate the year of the particulars date on FME. For example, I want to calculate the year of 31st March of the current year and next year. Similarly, I want to calculate the year of 1st April for the current year, and the previous year. These calculate years along with dates would be stored in columns.  Looking forward for your guidance.

I'm trying to demonstrate how FME (eventually FME Server) can automate a process for us. I have all of my NMEA data in one folder (they are stored by day in a given month; hence, there can be 700-800 NMEA files in each folder), and I can get the reader to read them all manually...but I would like to know if it's possible to get FMEWorkspaceRunner to automatically read and load all of the files in a folder that will then be converted.The FMEWorkspaceRunner variable is set to: runner = fmeobjects.FMEWorkspaceRunner()I tried to be sneaky and use parameters['SourceDataset_NMEA'] ='\\\\2018ais\\c$\\rawnema\\Feb2019\\*.nmea', but it did not work using runner.runWithParameters(workspace, parameters).I can (of course) do it manually from Python by using runner.promptRun(workspace), but this then defeats the purpose of my Python script.Is there a way to get the FMEWorkspace Runner to automatically read/load all of the files in a folder that will then be converted?When l

I was trying to use FeatureReader to read an online zipped shapefile and it failed with the messageShape Reader: Failed to open dataset 'https://ecos.fws.gov/docs/species/shapefiles/usfws_I02T_I01_Cavaticovelia_aaa_current_range.zip' which is neither an existing file nor directoryHere is one of the link I use. https://ecos.fws.gov/docs/species/shapefiles/usfws_I02T_I01_Cavaticovelia_aaa_current_range.zipBelow are the parameters I set for the FeatureReader.The regular Reader worked just fine for the link  Can anyone help me troubleshoot this issue?Thank you.

I have created a workbench to request an ID with a fake XML. This works for one ID.  I've now added in an xls and an attribute manager to attempt to edit the XML that requests the IDs to request multiple ID's. This will not work, it keeps rejecting it. My XML is set to:Rejection:Please help me get this working.

Hello,I have a point dataset, and the attributes within can be summarised as per below:Point_IDArea_IDCategory1 AAZ2AAZ3AAX4BBZ5BBX6BBX I want to extract one point per Area_ID (I don't mid which), but it must assigned the the category attribute with the highest count within that Area_ID.Therefore, form the table above for Area_ID 'A' I would want to extract points 1 or 2, and for AREA_ID 'B' points 5 or 6. Any ideas? Thanks,RB

Dear FME'ers, I have a question about the conflict resolution of SQLExecutors and FeatureReaders.I found the issue when I was reading data from a PostgreSQL database. Some of the initiator's attribute names where the same as in the result of my SQL. Therefore I used the option 'conflict resolution'. This because I supposed that the data of the initiator is then overwritten by the data of the SQL script (for the attributes with conflicting names).After that I tried to rename the output attributes with the help of a BulkAttributeRenamer but got an error. This is not the case when I set the conflict resolution to 'Use Initiator'. There is also no problem when I use a regular AttributeRenamer If I'm obliged to remove all conflicting attributes then what is the use of the conflict resolution? Or is this a problem related to the BulkAttributeRenamer? In attachment an example of my question with some extra info. I was able to recreate the issue by reading an Excel

Hello all,I am trying to automate the output of GIS data from PostGIS to Autocad DWG. The base data is in geographic coordinates (lat / long) and I have been able to successfully create the output DWG in GDA2020 UTM Zone 52, 53 or 54.The problem is that the end users need the data in ground coordinates not grid but I am not sure how to define a custom projection to either take it from geographic to ground or from geographic to grid to ground.I have been able to find the formula for the scale factor but this is something that will change for each extract of the data. Does anyone have any advice on the best way to create a new projection on the fly using the newly defined scale factor?

So, yes I've read about how to get FME 64 bit to run on a machine with 32 bit ArcGIS Desktop installed, but for ease of deployment on different user PCs that will only have the standard ArcGIS configuration, I've opted for 2 workspaces:One to run an extraction of source data with an ArcSDE Reader executing in FME 32 bit. This is due to the dependencies of FME on the bitness of the ArcGIS installOne to transform it by executing in FME 64 bit. The data model is very large and RAM intensive so needs to run in FME 64 bit.FME 2018.1The best solution I can think of it to BAT file this by calling the workspaces against the individual 32 bit and 64 bit FME executable file paths, because I don't think a conventional running them in a Parent workspace with 2 WorkspaceRunners can work, since you can't specify the bitness of the WorkspaceRunners and they would all have to execute as 32 bit or 64 bit inheritted from the Parent workspace mode, but not a mixture of both.But, intereste

I recently upgraded my FME Server and now when I run jobs using the data download service the zip file returns an unwanted file hierarchy with additional folders named after each Writer in my workspace (e.g. XLSXW2_1 and GEODATABASE_FILE_1), how can I remove these folders?

After I found out what was causing the crash (2 days and 50+ eliminated transformers later) I wanted to share.I had a couple of workspaces that ran fine in version 2018.0 that stopped working in 2019.2It came down to this - the Sampler transformer - set to "Process When Group Changes" and the sample type of "First N Features" and disabled will crash FME.exe on both Win7 and Win Server 2012 R2.The screen cap below contains a Creator, disabled Attribute Creator for a dummy Attribute, the Sampler and a closing disabled Inspector. Surprisingly, the Sampler never receives any features - yet seems to cause the crash.I hope this saves someone else some hair pulling.ThanksMike B

Hi there, I am trying to filter (using tester) the data on the basis of dates (i.e. month and Year). I have converted the dates into desired format using DateTimeFormat function. When I run the tester, it does not give the right result. In the passed data tab I could see the dates row that should have been in the failed data tab. I have attached the screen shot of my test parameter. Could anyone guide where am I making the mistake?

I have the following SQL being used by a SQL Executor; FME_SQL_DELIMITER ;DROP TABLE dbo.[PREMISE_CONTRACT_ROLLBACK];SELECT * INTO dbo.[PREMISE_CONTRACT_ROLLBACK] FROM dbo.[PREMISE_CONTRACT]; It runs successfully using the "Run" option when I'm accessing the transformer itself. However, if I just add a Creator > SQL Executor > Logger and run the workbench, the following error occurs;Microsoft SQL Server Non-Spatial Reader: Query failed, possibly due to a malformed statement.  Query Text `SELECT * INTO #fme_tempAdoDataTypeTable_spatialReader FROM ( SELECT * INTO dbo.[PREMISE_CONTRACT_ROLLBACK] FROM dbo.[PREMISE_CONTRACT] ) AS customQuery WHERE 1=0'. Provider error `(-2147217900) Incorrect syntax near the keyword 'INTO'.'Microsoft SQL Server Non-Spatial Reader: Getting Custom SQL schema failed. Make sure all columns are named (any ambiguous columns, e.g. GEOM.STBuffer(1), must have an AS clause). If there is an ORDER BY clause, either remove it or add a TOP clauseIt looks like FME 

I have a number of DXF input files feeding a workspace that writes to a SQL Server File Geodatabase. These DXF files are sometimes there, sometimes not. The fme workspace fails with a Fatal Error when the Input Reader fails to find the dxf file during "version check". As failure occurs before there is opportunity to incorporate a transformer, I can't use FileExistanceChecker or other file exists processes. Is there a workflow I can use to ensure the translation continues if the input file is missing?

Dear learned friends, I am attempting to automate the cleansing of a polygon dataset that contains a large number of overlaps. I have managed to identify where instances or overlap occur by using the AreaGapAndOverlapCleaner transformer and then overlaying the output from this against the original data set.I also have a point geometry dataset that contains an attribute linking it to a parent polygon. Many of these points reside within the identified overlapping areas.I now need to 'cleanse' the overlaps within the polygon dataset, but need to apply some intelligence to the process so that the assets within the overlap instance are retained within the correct parent polygon.Hopefully the simplified attached picture will explain this - the red and the blue polygons are overlapping, and the point assets (black dots) within this overlap contain an attribute stating that they relate to the red polygon.Thanks,RB

I tried installing the safe.robj-detect-pkg in to FME desktop 2019.2 build 19801 and I got the following errorRasterObjectDetectionModelTrainer.fmx ignored because it was created with a newer version of FME WorkbenchI downloaded and installed 2019.2.1 build 19813 and the same error occurredTransformer safe.robj-detect-pkg.RasterObjectDetectionModelTrainer version 1 in file C:\\Users\\apadilla\\AppData\\Roaming\\Safe Software\\FME\\Packages\\19813-win64\\transformers\\safe.robj-detect-pkg\\RasterObjectDetectionModelTrainer.fmx ignored because it was created with a newer version of FME WorkbenchTransformer safe.robj-detect-pkg.RasterObjectDetectorSampleGenerator version 1 in file C:\\Users\\apadilla\\AppData\\Roaming\\Safe Software\\FME\\Packages\\19813-win64\\transformers\\safe.robj-detect-pkg\\RasterObjectDetectorSampleGenerator.fmx ignored because it was created with a newer version of FME WorkbenchTransformer safe.robj-detect-pkg.RasterObjectDetectorSamplePreparer version 1 in file C

Hello Team,While installing FME Server 2019, System was asking to uninstall existing setup of 2015. Is it okay to uninstall it? Is there any further consequences with this action?Thanks in Advance