Transformers

Hi, I have a new source with poor metadata. I would like to create a quick overview of the dataset by producing a matrix, where I have each attribute field and all unique values within that field listed. With aggregator or duplicate filter I can easily do one. But I have close to 100 fields, so there must be a more clever way... Thanks for any suggestions!

Hi, I just yeasterday started to play with htmlreportgenerator and html as format is new to me also. After playing one day with it, I managed to make with small effort make point in map and make some charts out. Now was thinking if there is some "easy" way to make those together? I have about 300 points in map what show their label when they are clicked. How much html_content I would need to modify to make charts pop out instead of label? All of these points have their own charts. To make it simple.. 300 points and 300 charts. How to make them together? :) Aki

I have a custom gsd file provided by our client to go from NAD83(CSRS) UTM zone 18N (or EPGS 2959) to UTM83-18 (or EPSG 26918). Using the data read with these projections or datums, and trying to relate the two spatially within FME, produces no overlap between the features. Whereas, there is only suppose to be a 1 meter shift between them. When I try to edit the FME desktop options to point the transformation screen to this gsd file, it gives me an error "failed to override grid...contact SAFE."

Hello, i have a big network(_network_ID = 1) as displayed in the left section of the picture below. The right section shows some lines(_network_ID = 2,3,4,5,6,7) which are not included to the network yet, but are networks of their own. I would like to snap each neighboring line with each other and then snap it to the big network, which should be easy using a Snapper Transformer. Now to the problem. Some of the distances between those lines require a snapping tolerance bigger than the distance between their own end points. So some lines snap to their own end points. How can i avoid that? Many thanks in advance, Andreas

I am using Salesforce reader to read files from data loader - is this the correct approach? I am getting the following error when I use the custom domain (https) with username and password (with or without security token): "Failed to obtain any schemas from reader 'SALESFORCE' 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" My purpose is to read files from salesforce with lat/long and write to SQL server (enterprise geodatabase). Any suggestions?

Is it possible with FME Server to create a shortest direct URL? (like in Google Maps, when we share a map, we have this option) I've pasted my URL in Notepad and it seems that I have 500 characters. This is not a beauty into an email and it is not so nice to copy paste for anyone. And for a newbie, all the information into the URL isn't very useful. Thanks.

Hello, i use an AnchoredSnapper to snap lines(candidates) to points(anchors). The snapping type is set to 'End Point Snapping'. The problem is that some lines have both end points in the required snapping tolerance, so both end points get snapped. I cant reduce the snapping tolerance. So is there a way to avoid this? Many thanks in advance, Andreas

My xml code (contained in an attribute of my incoming features) looks like this: <?xml version="1.0" encoding="UTF-16"?><document authority="Forstamt Kanton Thurgau" authority_url="" category="main" cycle="Öffentliche Auflage" doctype="decree" enactment_date="2007-10-31" federal_level="Kanton" id="3039" instance="Gemeinderat" number="" subtype="Waeldi" title="AWB Detailplan Nr. 1" type="Statische Waldgrenze"><file category="main" href="/api/attachments/9989" title="AWB Entscheid-Ersatz.pdf"></file></document> How am I to extract the XML attributes (e.g. 'authority', 'authority_url', 'category', 'cycle', etc.) into seperate FME attributes of my features. In other words: I want to extract 'Forstamt Kanton Thurgau' into anattribute called 'authority', 'Öffentliche Auflage' into anattribute called 'cycle', and so on.

Hi, I have a dataset I want to check matches with another one. In particular I need to make sure that if the business area names match between the 2 datasets that the sub-business area listed is one named under that business area. For example if in the 1st spreadsheet; Business area = Environment, Sub-business area = River or Forest Business area = Air, Sub-business area = Quality or Pollution. Then in the 2nd spreadsheet, if the Business area is Environment then the sub-business area should only be River or Forest. If is it Quality then I need to know this is incorrect. Also is there are way of doing this without having to list out all the sub business areas (e.g. if you used a tester with environment = river, forest,etc). Thanks in advance and also happy to clarify if what im asking isn't clear.

We use a Directory and File Pathnames to extract filenames and path_created_date and store the value for all new files in a database. When we move to summer-time and winter time, it seems that FME gets 1 hour different and adds all files again. So I have checked the following, to see if the timezone settings were off: FME Server: own custom server - Western europe standard time - UTC+1 Database: MS SQL - Not sure File server: - Western europe standard time - UTC+1This is really annoying - and it would be great with a universal solution. Any tips or someone had a similar issue?

In a control file (.fme) in which I chose the source file and the destination directory, I would like the log file to be in the same directory as the destination directory. How can I do ? Thank you

@takashi As you have probably worked out I'm fairly new to more than push button FME. This should be really simple and probably is but I want to expose a numeric raster band value and create an attribute from it so that when I covert the raster to polygons I have the value as an attribute of the polygons. I have tried using attribute creator but can't work out the syntax. This leads on to another probably basic and related question: I want to replace values in a raster based on ranges of numbers - eg 0-12 becomes 12 and more than 12 but less than 20 becomes 20, etc. I can string together multiple instances of RasterCellReplacer to approximate this but there must be a more efficient way using RasterExpressionEvaluator for example. Again my syntax to do this is lacking. Any resources for a self-taught FME'r on that front that start at the basics?

I'm now working with localization of FME Descktop 2017.1. It's also a good opportunity to take a closer look at what's new. I would post something I noticed through the job. FME 2017.1 build 175391. ListBuilder The "Input Ordered" parameter seems to be shown at a wrong location. 2. NeighborFinder It's great that the new parameter "Take Measure/Z From Candidate" has been added. However, I'm not sure if it's reasonable that "Yes" is set to the parameter by default. How do you think about this? possibly to be continued...

Hello. I am using FME 2017.1 The issue I am trying to resolve is stopping the ArcGIS Transformations window from popping up when I add a shapefile that was created by FME. The FME shapefile is in the correct projection (same as the dataframe) but the FME PRJ file is slightly different than the ArcGIS PRJ file. I am assuming that this is what is causing the Transformations window to pop up. So, my question is: how do I create a PRJ file from FME that is the same as what ESRI expects so that the Transformation window doesn't pop up in ArcGIS? Here are the details of what I have done: Right now, the projection I am using from ESRI's ArcGIS is not in FME. The FME projection that is the same as what I am using is 10TM115-83. I created a Custom Projection by created a .fme file and putting it into the following directory (C:\\Users\\<<USERNAME>>\\Documents\\FME\\CoordinateSystems) as suggested on this form: http://docs.safe.com/fme/html/FME_Desktop_Documentation/FME_Workbench/Workbench/coord

Hi, I have a point Shapefile that I would like to recreate in raster, specifically, ESRI Ascii Grid format. However I am having issues in that when I use the 'NumericRasterizer', some of the resultant cells in the raster are up to 0.25m away from the original location of the points. Is there any way that I can set the rasterizer (or use a different transformer) to force the resultant cells to be overlaid exactly on the original points (ie: points will lay in the centre of the cell)? Thanks

I am tryintg rasterize one dwg. I am using this scheme that can be seen on screenshot but unfortunately I am getting blank result. Does anybody have an idea what could be the problem?

Hello, this time I do not come up with a problem but with a request for ideas/opinions etc. The scenario: static geographic locations, where samples are taken on a regular and irregular basis are hold in different database environments such as Oracle Spatial and Smallworld Data Stores. The measuring values are hold in an Oracle non-geographic environment. There is an ID that connects both "worlds" (yeeepie!). Probably there will also be non-static sample places (ad hoc samples) in the future whose positions are not yet stored anywhere. Here, the question how the data will be georeferenced, has to be answered first. For this scenario I exclude these sample positions.The idea now is to join both worlds (geographic and non-geographic data) and visualize it in a map environment. Visualize in this context means:- overlay the sample positions with any WMS (own and third party) - delimit the sample positions by location and / or attribute(s) - select which type of, e.g. biological parameter,

Whats the minimum FME Desktop / Server version to use with ArcGIS 10.6?

Hi all, Upon creating rgb values it appears that the lueb and green channel are reversed. When I create the values with an expression evaluator and set them on objects (linear), the gradient is correct. In the raster however the blue and green are reversed. (swapping green and blue channel values corrects this) Is my observation correct? Is this intended?

I have two input files, and I want to check whether or not the address in one input file (comprised of a street attribute, number attribute, postal code attribute etc.) matches the adress in another input file (also comprised of more or less the same columns). It's a string/text object, so I don't need it to be an exact match. I've already tried feature merger, but it seems as if feature merger only accepts exact matches. I could be mistaken of course.

I have line feature which I have cut down from the administrative boundary and I need the result which have only the administrative Areas outside of line Buffer Areas. I used Generalizer,Bufferer, dissolver then Clippers("Clippers First" to the Clipper Type parameter in the Clipper). Dissolve areas are in Clipper port and the administrative boundary areas in the Clippee port, then shows this error and all dissolve area comes on the rejected port of Clipper and output is written only the administrative boundary. ClipperVector(ClippingFactory) Extra clipping feature encountered and ignored ClipperSolid(SolidIntersectionFactory) No clipper supplied for group,all clipees will be treated as OUTSIDE have anyone any suggestions?

Hello everyone, I need the [B] (Green) areas, the administrative Areas outside of road Buffer Areas. Be precise, I need an area which have no Road. Therefore, I used: Bufferer -> Generalizer -> Clipper ("Clippers First" to the Clipper Type parameter in the Clipper)-> Buffer areas are in Clipper port and the administrative boundary areas in the Clippee port and output is Esri Geodatabase (File Geodb API). Unfortunately, the Clippers is too slow. After 14 hours, it finished only (clipping group 16985 / 33079) I have total 60001975 buffer road. How can I improve the speed of the Clipper and get my desirable output. Thanks

Hi, I have a simple txt with many coordinates like (no header): 554174.746999408, 5598075.08612377, 554175.302999408, 558090.53212377, 554177.963999408, 5598129.63312377, 554182.011999408, 5598160.71912377, 554182.176999408, 5598181.49312377, 554180.422999408, 5598212.50312377, 554172.863999408, 5598249.02912377, 554167.975999407, I want to validate that geometry. I tried using different readers (txt, csv) and transformers like stringreplacer. I found this question https://knowledge.safe.com/questions/39344/use-of-the-geometryreplacer.html but could not modify my workbench with success. Any hints on that?

Hi, My linked custom transformer ('ColorAndReproject') is used in many places in my workspaces and I am regularly working on it. This is why it's not embeded. I have installed it with double-clicking the *.fmx file. It's stored in a "Shared FME Folder". The workspace does exactly what is expected if I run it in the workbench or if I run it with a *.bat file. Sadly it does not work when the scheduler kicks-in during the night and tries to run the script (*.bat). The scheduler is a special Windows user with some admin rights. I get this error log : The custom transformer 'ColorAndReproject' referenced by this workspace could not be found. The custom transformer must be installed before the workspace can be runProgram Terminating There are other workspaces run this way with no problems. But none have a linked custom transformer. Any Ideas ? Thanks for your help ! edit : it seems the real problem is that each user executing a script with linked custom transformer should have FME configured

I think this could be either a bug, or an issue with my understanding. image below, why would the exact same feature class, put through the exact same geometry validator yield different results between readers Top scenario in image is with Geodatabase File Reader Bottom in image, is with File GDB reader