Data

Hi Everyone,  I am getting an error when trying to create a TAB file or .GDB features from a certain dataset in Oracle Spatial. The error is "Error occurred trying to create area geometry". It happens at the point where the features are attempted to be written to the TAB file. It doesn't make much sense since the table was originally loaded into Oracle from a TAB file so it should be happy to translate back to that at the very least.  My thoughts were there must be something wrong with the geometry which is preventing FME from "calculating the area geometry" so I tried the steps below to fix/trouble shoot the problem.  I have put the data through a geometry validator and geometry filter however that has no effect on the error I'm getting. I ran the Oracle geometry validation function over the table then deleted all geometries it found which were not valid.  If anyone could give me some ideas as to what this error message actually means and how I can get FME

I want to read the lat/long in a gps'd jpg and create a point with a hotlink to the jpg. Does anyone know how to do this?

I'm experimenting with incorporating Excel workflows into my FME workspaces and am stuck.  My workspace reads in a raster, then I get raster properties using RasterPropertiesExtractor, then I create a few more attributes with AttributeCreator, and finally I want to write to an Excel file with each attribute name in the first column and its value in the second column.  I tried many settings under "format parameters" in the excel writer, but with no luck. The main problems I am having is that 1) nothing seems to write to the excel cells no matter what settings I have under "format parameters", and 2) the image of the raster is inserted into the excel file, which I don't know how to stop.  I'm obviously missing a few things in my first attempt...    

Hi, Having some problem using SQLExecuter and SQLCreator with a Postgre/Postgre.  Error connecting to PostgreSQL database(host='**.***.**.***', port='5432', dbname='*******', user=postgre, password='***'): 'FATAL:  password authentication failed for user "postgre"  Its not really a authentication error since both the postgre and postgis reader and writer works fine with the same credentials. I can connect to the tables in the SQL transformers but when I hit run I get "authentication failed"  Anyone have an clue?  cheers Henrik Rinne   

Hi Folks, I want to do something that should be simple but I'm stymied.  See the attached workspace and simple sample data. Due to the nature of the dataset, the single file in "DATA" gets a slightly different file name each time we receive it. So all I need to do in theory is: Readers -> "Import Feature Types" -> (Change to | delimiting) -> Press "ok" -> Select the updated "feature type" -> delete the old one from the workspace -> link the new one up. Done!  Except it doesn't work. I've done it in the attached but as you can see if you run it, it just skips those features. Despite the fact this feature type is listed under "feature types".  Even if I delete that feature type and use "Import" instead, I get the same thing.  The only way to fix it is to add an entirely new reader.  Here: https://drive.google.com/file/d/0B22cYd3gwE-6NjIxRHhWUVhOY0E/edit?usp=sharing  Can anyone tell me what I'm d

I have a workspace that writes out one multipart shapefile (many rows in the dbf).  I'm looking to instead write out separate shapefiles that would correspond to each part (i.e. one shapefile for each row in the the multipart shapefile).  Not sure if my workspace requires additional transformer(s), modifying the writer parameter settings, or would involve both transformers and writer parameter settings.  I tryed experimenting with fanout settings in the the writer, but I haven't got anything to work.   

I noticed today that the dwg writer asks for a file name in a folder (unlike other writers that simply ask for a folder).  How can I force FME to call the resulting dwg by  the "feature type name" that I define in the name parameters of the writer? It seems that when I do define a name here, it still reverts to calling the output whatever I chose when I was first forced to chose a folder and file name of the writer.  

I use FME Desktop Proffesional. A minor inconvience I have is I cannot drag files from ArcCatalog to FME workbench. So instead I often use Windows Explorer, which means that for a shp file, I have to spend an extra few seconds making sure that I choose the file name with the .shp extension.  Is it not possible to drag from ArcCat to Workbench because thuis functionality is part of what you pay for with the ESRI FME version, or is it not possible in either case? If it isn't possible in either case this would be a nice little addition to the program.

I think there was at one point but I can't find it in the list of readers.  Thanks 

Hi dears,  I've been try create a route with KMLTourBuilder. Could you help me with this?  I make a route in Google Earth and exported to kml. I add in a reader KML conected in KmlTourBuilder and writer KMZ. This correct?  Thank you for any help you can provide me.  Best regards.  Ramon Wallace.    

I've got a Feature Class that I'm reading from and SDE GDB, where it is in a Feature dataset. I update an attribute, then send it to a File GDB Writer (not the API one). This is in FME Desktop 2013.  This writes the Feature class out, but also writes it into a Feature Dataset in the destination. I would like it just at the top level. Can't figure out how.  The reader has ignore network and relationshipd set to Yes. The Writer Feature Type has teh Feature Dataset set to "<not set>".  Ideas? Thanks.

I am wondering if it is possible to output polylines that are created through seperate worlflows in a single workbench to a single dxf file? I am reading  x,y data out of a database and using FME to create polylines and write that to a DXF file. I am able to get that to work with a single polyline file but if I try to do it with two only one will appear in the output file when I would like to have two seperate layers. I'm hoping I just need to find the transfromer to do that.  Thanks,  Vince

Is it possible to use a wildcard to select the Feature Type to read when single merged feature type option is checked.  I would like to be able to choose what table to read without genereating the list of tables merged in the reader because it takes too much time.  I would like to simply type something like: *711.SHP to select all tables that finish with 711.SHP  It seems it is not possible. Any other solution?

Hi I Am new to the geopdf translation I would like to have an entire feature table of oracle spatial converted to geopdf with each record as a separate layer on the basis of an attribute .Currently fanning out by the attribute  creates new file

Greetings,   I have two shapefiles of roadway networks for a state.  One is just of state maintiained roadways (that I got from one source), and the other has about every road in the state, including the  state roads(from a different source).  Is there a way, or transformer, that will look at the two and create a shapefile using all of the data from the state maintained file, and add the data that is present in the second file, but not in the first?  Sorry if this sounds confusing, what I am trying to do is a bit odd.  I got these files from two different entities, and I have a feeling that the one that has all the info doesn't have as accurate data as just the state roads file, but they want a file containing what they know is the good state data, and the somewhat good other data.  Thanks, Mike

I'm writing out to an old SQL 2000 database, and I get the following error using the MS SQL Non-Spatial writer:  "'TYPE_ID' is not a recognized function name."  I checked the log file and before writing to the database, FME is trying to execute the following query  SELECT name FROM sys.columns WHERE object_id = OBJECT_ID('[TableName]') AND user_type_id = TYPE_ID('int') AND is_identity = 1  When I execute this query I get an error too: SELECT name FROM sys.columns  The error is:  Invalid object name 'sys.columns'.  Is there a problem with writing to such an old SQL database? 

I'm reading a geotiff and am unsure why when I'm writing to a new geotif, all my dark dark green cells, such as RGB (0,16,0), get rewritten to (NODATA, 16, NODATA).   Using the RasterBandNoDataRemover does fix the problem, but also changes the NODATA cells that I want to be NODATA into balck (0,0,0).  How can I fix this? Seems to me that this should not be a default behavior...   

Hi Folks, I have six 3D objects I want to place into my 3D PDF. The six objects between them make up two different "Things".  I'd like each of these Things to be controlled independently. Currently they share the same anchor/base, so if I try and rotate one, the entire scene rotates around a common centre.  Can this be done with 3D PDF and if so how?  Thanks! Jonathan

I would like to ask FME experts to point me any documents/code samples... for FME and IBM datastage integrations.  By searching the Web, I only fould a presentation by your boss did a few years ago. It doesn't have too much technical details.   Your help is very much appreciated.  George Shi Lead, Canada GIS team, IHS global GIS Calgary, Alberta  

I have this workspace: read LAS ->PointCloudCoercer->RasterDEMGenerator  I also have read into the workspace a raster (ECW)  In the parameters of the RasterDEMGenerator, I have defined the resolution of the output as: Output DEM X Cell Spacing: $(SourceDataset_ECW) Output DEM Y Cell Spacing: $(SourceDataset_ECW)  This fails with the error: RasterDEMGenerator(SurfaceModelFactory): Fail to process the SAMPLE_SPACING clause because it contains an invalid entry or it contains an invalid number of entries.  The ECW file is very much not corrputed, and reads fine in Data Inpector.  How should I be referencing  the properties of an already existing raster with the RasterDEMGenerator?  The overall goal here is of course to output a raster from a point cloud that has the exact resolution of an already existing raster and is also "snapped" to it.     

Hi, can anyone help me?  When I'm logged in as admin I see the options to view or download the logfiles after running a transformation in the web user interface, but not as other users. I guess it has to be some security setting I have to give for example to the role fmeuser, but I can't seem to find the right one.  Does anyone know?  Thanks, Tomas    

I have a workspace that reads from a File Geodatabase (FGDB), cleans up some attribution and then writes the updates back to the FGDB.  The problem is that I am getting ArcObject errors for some of the features.  All the features are lines.  One of the errors says that "Closed polylines are not allowed" and the other says that "The connected feature does not exist".  I am stumped by this as it is reading and writing to the same data set.  In looking at the feature(s) in an inspector the closed attribute says no.  For the other error... well how could it read something that doesn't exist.  I have successfully updated single and multiple records in this FGDB.  In comparing a feature that updated and a feature that did not update in the inspector I can see no difference.  This FGDB is a copy of a data set from our LGIM and does contain a network.  Any suggestions on how to handle this is much appreciated.

Hi everyone,  I'm trying to take some complex 3D shapes and use them to clip a lidar pointcloud. Here's an example of the shape:    I need all the points that lie inside those cylinders places into a seperate las file. The 3D shapes are stored as a series of polygons with height attributes in a shapefile, I've been using the 3DForcer, then extruding them, then using the geometry coercer and aggregator, finally then feeding them through the clipper.    The problem is the clipper seems to clip everything above and below the largest polygon, as if it were 2D. Below is the result, as you can see it's clipping points outside the yellow cylnders which is what I don't want.    Are my shapefiles not truely 3D when I'm feeding them into the clipper? Should I not be aggregating them? I'm trying to clip trees out fo a forest, so whatever I do has to be batch processed 20,000 times.  Thanks a bunch!  -EAYREY

Hi all,  I'm just trying to create a workbench which is able to write proper CityGML with CityObjectGroups, but unfortunately I can't find anything about that feature type in any documentation or post on this site. There is just one little sentence on the "What's new 2012" page that this feature type was added, so I think it should be possible to write CityObjectGroups and I just don't know how.  My solution so far is, to create an new writer feature type which is called "CityObjectGroup" and every cityobject that should be written also goes to this feature type. I guess all these GroupMember need to receive the citygml_feature_role attribute with the value "groupMember" so I already did that. But what to do next? Nothing really worked so far. Maybe someone of you already has some experiences on this topic? Any help would be appreciated. Thanks!  Jens

Hi all,  I am wondering if any one has experienced an issue with fields being populated with out making connections. The green arrow is not signifying  a connection and we are definitly not making any physical connections and we offten have intigers showing up in fields that should be charactors even if they were a connection made. Tahnks Pat