Authoring

I need something like this: if the value is NULL or non-existent, then name field as "UNKNOW", else where bring the values assigned to it.How can I tell FME to separate the NULL and empty cells from the numbered ones?For example:feature 1 has attribute ‘someattr’ == blank, then attribute ‘someoutput’ = UNKNOWfeature 2 has attribute ‘someattr’ with a value, then attribute ‘someoutput’ = the value from ‘someattr’ attribute

I have a date attribute in the format %Y%m%d%H%M%S. I want to use everything in 2016, so how can I create a wildcard similar to 2016% ?

Hi I’m currently using the workspace ‘Write Microsoft Word (.docx) with FME’ to produce a report which highlights errors with in the data.What I would like to know is how using this workspace can I create a single word document with multiple tables? Each table will be showing different attribute errors depending on its output from a single workspace.ThanksAndy

Hi Friends,The problem is distance marker's preserve behavior. I have 14 lakhs features. But distance marker waits till last feature to come. after 9 lakh fme gives memory issue and comes out. I dont know why Distance marker to wait. because this is not group function. it can send the each features points as soon as it completes each feature. and write immediately. if the distance marker will preserve the incoming features, then definitely memory issue will come. how can i do that ?

I am trying to create a workspace that when run will pull the current date into python (either as start up script or python caller) and then pass that date out as a published parameter in FME. I have the following in python:import datetime today = datetime.datetime.now() #print str(today) print today.strftime("%Y%m%d")This gives me the date (for today) as 20161103I want to pass that out as a Published or Private Parameter [Start_Date] to be read in a transformer later in the process.I will also need to be able to calculate a date two weeks away and pass that as [End_Date].

I have a work space setup with multiple reader types. One reader will read values from an HTML table to populate attributes within the work space. On rare occasions when reading the HTML table, there will be an empty value from which I currently have the work space set to terminate with a message stating that a certain value was not found.Is there a way to prompt the user for that certain missing value if not found in the HTML table, instead of just terminating the session like I have it set to now? I know user parameters have to be set before the work space is run, just wanted an idea of how to combat this situation.FME Desktop 2017 Beta 

Hey all, I have a point dataset with over a million points. However, I am trying to aggregate them based on their flight number. The thing is some of the points have the same flight number in different counties, so I dont want to use a hullaccumulator because that would take in to account empty areas or areas with other flights present. I have tried transformers like the neighborhood aggregator, I am just unsure how to go about this.I just dont want ArcMap to have to draw over a million points every time I pan to a different area. Any ideas? I have attached an image of what I mean. thanks in advance!

I am new to FME and I love it but I'm stuck on something that I thought should be easy. I want to convert DGN or DWG to Shapefile. I want to produce shapefiles for Point, Line and Polygon. Most importantly I want to retain what we would call attributes in GIS. This is information such as Entity, Level, Layer, Color, Linetype, Elevation per feature. These are visible in ArcGIS if you view as CAD but when I convert using FME workbench these are not carried through. If I look at the CAD in FME inspector I see many attributes but not the ones I need that I can see within ArcGIS. Can anyone help?

I have a set of features in a table that in my example is classed into two types by area - Big and Small. Each feature has a numeric value for the area. In my hypothetical example I need to decide on a sensible number to use to make the decision between small and big area types. I therefore want to choose a value for the area and then check to see how many features below that are correctly classified as Small, how many are correctly classified as Big, and how many are incorrectly classified. I want to be able to loop through a set of values for the area and write out to a table how many are correctly and incorrectly classified for each iteration.I have created a workbench that does exactly what I want it to do for a single value for the area. However, when I attempt to create a loop, nothing seems to be happening inside the loop. I have used statistics calculators to get a count for each classification decision which I believe are blocking transformers and therefore FME creates a sep

I have a few hundred ortho-jpgs, but i want to be able to read only the ones that intersects or are within a polygon.The jpg's are georeferenced and the DWG too and botha re set to the same coordinate system and both show up at the right positions in the inspector when opened individually.This method worked for me using .las point clouds on the same spot, but I cant seem to manage to get the jpg's I want.Is there something wrong with the settings? Thank you in advance,Robin

Hi,I would like to create an output folder based on the input file.I created a user variable <output_folder> and I'm trying to determine the value  using a python script.Below is the error that I'm getting using this workspace INCLUDE -- failed to evaluate Python script `def ParamFunc():  import fmeobject  def output_folder(feature):  feature = FME_MacroValues['featuretypes']  string='C:\Users\Administrateur\Documents\oracle_spatial\tmp'  if feature[10:]=='roads'     output_folder=string+'\shape'  else     output_folder=string+'\mapinfo'  return str(output_folder)value = ParamFunc()macroName = 'output_folder'if value == None:  return { macroName : '' }else:  return { macroName : str(value) }'Program TerminatingTranslation FAILED.  File "<string>", line 5    feature = FME_MacroValues['featuretypes']          ^IndentationError: expected an indented blockThanks

Hi All,I have an excel file with header-1 names and header-2 names. I have used AttributeSplitter and made list for both like header-1_list and header-2_list for each record. Now I want to aggregate the values like mentioned below...If 1 record in excel containsheader-1_list contains 0,1,2 valuesheader-2_list contains A,B,C,D,E valuesOutput should be 15 records which contains 0-A,0-B,0-C,0-D,0-E,1-A,1-B,1-C,1-D,1-E,2-A,2-B,2-C,2-D,2-E as values.(I have attached sample excel file and my progress work bench to this message for better understanding)Thank you in advancePratap 

How does this get remedied?I rebooted my workstation and still get this message...How do I get going again, please?

My format attributes are exposed from the source up to my destination feature class but these format attributes are not written on my FGDB. I can tell which of these formats have values from my inspector. Should I map these format attributes with an AttributeCreator or AttributManager so that they will be written?

Hi!I want to load a Web Map Service into my workbench with the FeatureReader. It works almost perfectly, but I want to use a bbox as a geographic filter for my WMS. The problem is that you can't insert a bbox in the WMS parameters of the reader. Is there a possibility to specify a bounding box?Best Regards,Felix

Dear all. I completed the configuration of a Directory Watch subscription. I spent time trying to check with the Topic Monitoring without success (i.e. no results on the monitoring page). Searching the community here, I found an answer to a topic where an expert suggested to look the subscription log. I did and I found the service working properly. I know that the important thing is that the subscription is now working but I'm interested to know, if possible, the reason why the monitoring page seems not work properly. TIa Roberto   

I have linear spatial data that consists of types (A, B, C, ... J, K) as well as condition (1, 2, 3, 4, 5). I am tasked with creating a list of total length of each type split by condition, ie Length of A1 = 100m, Length of A2 = 30m, ... Length of K5 = 200m.Is there a way I can create a function that splits the input into condition and then measures it without grouping all the output data? Currently I am splitting the data into types, then splitting each type into condition, then performing measurements, which results in dozens of identical functions being run. It seems that this can be streamlined, but I am somewhat new to FME and I am unsure how to approach this.I'm currently exporting the resulting data into excel, if that helps.Thanks in advance!

When I use a PythonCaller (on build 16673) to set the feature type by callingfeature.setFeatureType('something')it actually doesn't seem to do anything, because when I connect a FeatureTypeFilter transformer to the PythonCaller and filter on 'something', nothing comes out of that port. I discovered that it only works if I do:feature.setAttribute(fmeobjects.kFMEFeatureTypeAttr, 'something')Where kFMEFeatureTypeAttr resolves to the string 'fme_feature_type'. But then what is the point of the setFeatureType() method? Is this method actually not implemented, perhaps?

Hi,A simple question:Is the index range in an FMEObjects FMEAggregate geometry parts list 0-based or 1-based ?The documentation doesn't say anything about this either way.It's for code in a PythonCaller.Cheers Lars I.

Hi,I'm having difficulties with the following problem:I'm having X polygons (with unique ID)(lets assume 1000 polygons) and Y points inside those polygons. First I do colorsetter so see which polygons are disconnected. So I have X polygons with an attribute _color (with 5 different possible values). Then I use spatial overlay to know which points are inside a certain polygon and in the end I'm using a statistics calculator to count the amount of points inside the polygon. As end result I have X polygons with as attributes: unique ID, _color and count of points.Now I want to split these X polygons into Z groups (public parameter, let's assume 10), but in a way that these 10 groups contain sort or less the same amount of points.Does anyone has some ideas to group them on an average? I don't mind the amount of polygons in a group. ex group 1 can contain only 1 polygon (as it contains a lot of points) while rest of the group can contain 11 polygons.ThanksVAT 

Hey guys I am trying to figure out how to clean up some data before using the google geocoder transformer. Is there anyway to drop text from my address column that is messing up the geocoder, i.e. apt or suite number?I was thinking some kind of string splitter or something, but not sure. I have attached a screengrab below, thanks!

Hi!As the topic tells, I'm wondering if it's possible to write dgn file with xattributes?I have dgn file with building information in it as xattributes. Im trying to write it through fme back to dgn file with all the same features.Just to see if its possible. And there is some projects where it might make my working bit faster.Regards,Aki

Hi, I'm trying to use a featureReader to read a spatial table, however I need to link to a non-spatial table in the same connection to complete some filtering within the where clause. However it seems that the featureReader only allows either/or rather than both types. The reason for using the featureReader is i'm told its quicker.My question is can you use the featureReader with spatial and non spatial tables (oracle) within one? if not what would be the best way to accomplish this?