Transformers

The TESTER transformer let pass only 32bit INTs when using 'type is' Integer as operand. I need to test if the value is 64bit Integer or smaller (but no double or float).I used also AttributeValidator with the same result. 

Hi,I'm trying to programmatically assign colours based on an attribute to 3D geometry features for visualisation in Data Inspector. However, I can't work out how to do this without splitting features using a filter and then branching to loads of different ApperanceSetter transformers and setting a separate colour in each of those.I discovered there is what looks to be a very helpful custom transformer called GradedColorizer which has a few algorithms for binning an attribute value then assigning a fme_fill_color property. However, setting that fill_color property doesn't actually seem to result in my model being shaded correctly in Data Inspector (ie. still appears as a wireframe, see image below).Any ideas why this might be? If I use ApperanceSetter it works fine, but I can't programmatically set the colours using that.

Hi all,I was unable to find an answer I needed so I thought I would post.I have been using the StringReplacer and am trying to learn the regex commands to no avail.I have a column with values that are mostly numeric, some alphabetical, some blank and some number with decimals. I would like to ensure that every number that does not have a decimal gets a decimal added to it.Example of value range:000625.224512566254662512566751256.335TLL003640.123CODDMy ask is how can I use the StringReplace (or any other transformer) to take the numeric values here. And insert a period in between the 6th and 7th characters. Leaving 6 before the decimal and 3 after.To note. Every numeric value that has a decimal is 10 characters. 6 before the decimal, one decimal and 3 following. Every numeric value that does not have a decimal is 9 characters. I want to keep all Alphabetical and Null values untouched.Thank you for your help!

Hi all,  In the middel of a workbench I want to do some kind of schema check for my 100+ attributes: All values of attribute 1 are composed out of less than 255 charachters All values of attribute 2 are composed out of less than 50 charachters All values of attribute 3 are integers out of maximum 10 numbers All values of attribute 4 are decimals with maximum 10 numbers before and 3 after the decimal seperator All values of attribute 1 are dates ...  For the charachter fields I can work easily with the StringLengthCalculator and statistics calculator. But the number and date fields are a bit less easy.  For the number fieds I split them at the decimal seperator and check the length of the first and last part and with a StringSearcher I check if there are only numbers.  For the date field I split it up by day - month - year and check if the value is below 12, 31 or 9999  This all seems a bit "dumb". Is there some kind of intellegent way to check type and structure of each attri

I have this topology and would like to merge the red segment with its corresponding blue segments. (Blue segments are on both side of the red segment.) I can group all red and all blue segments but I haven’t an attribute to group the related blue and red segments. Is there a way to do this with FME? Thank you for the help.

Hi,I'd like to do an offset for polylines (left or right) with starting distance d1 and ending distance d2). is-there a simple way to do that (affiner with some explication please or another method))thanks.

Hi, sorry for something probably very basic.In my geodatabase I have a polygon featureclass with a lot of attributes. On a regular basis I receive new polygondata in the form of Shapefiles which shall be written to the existing featureclass.If the new data from the shapefiles do not overlap the existing data, they are inserted to the geodatabase (via fme_db_operation = INSERT).But if the new data overlap existing data I just want to compute a new geometry for the polygons with the Dissolver and update the geometry in the geodatabase with fme_db_operation = UPDATE.The geodatabase writer is in UPDATE writer-mode, and I specify an "Update Key Field", but still run into (the probably famous) error number from ArcObjects '-2147220985'.Could be a number of reasons probably, I have a feeling it might be that my "Update Key Field" is not an ESRI ObjectID but I am really not sure. Comments, ideas, help is much appreciated. bestPål Herman 

I have managed to create fast writer for many 3D-formats - except 3ds. I am writing a terrain model, buildings, roads, water, trees. I have tried combinations of MeshBuilder, Triangulator, GeometryCoercer, Deaggregator etc. However it seems that when writing to 3ds it still takes alot of time (6 hours, compared to 15 minutes with the approx 10 other 3d-file formats).If I create the data to an IFMEMesh before writing I assumed the writing would go fast. I am using a FeatureWriter to start the writing at once. I havent added any textures on the terrain model - so that does not slow things down.Does anyone here have some tips with respect to this particular format?@takashi

I have a line feature class (with a field SURVEY), which I want to separate into 2 feature classes. In the first one I would have those features that have more than one feature with the same value in the field SURVEY - and in the other one - those features that have a single entry for a value in the field SURVEY.I believe I could explode the original Feature class by the value in the SURVEY field, then count features in each of them, then test for value "1" and merge if required. But I don't know how to create separate features based on the value in the field.Just to give you a context: For the first one I will later create convex hulls based on field SURVEY, but for the other one - buffers (as the convex hulls wouldn't look too good for a single lines). But this is not part of the question - as I know how to do it.Please advise.

I am attempting to develop a workflow that detaches a SQL Express database, downloads and copy an updated version of the database, and reattaches the database. I have the middle section working (FTPCaller and filecopy writer), but I'm struggling with the attach and detach. I keep thinking there might be a way to do this using a startup and shutdown script, but I'm not having much luck. Has anyone had any success doing something similar? Any suggestions are greatly appreciated. Thanks!

HelloI have some GPS data from a dashcam which contains a strange coordinate system which i have never sen before, I am guessing it is decimal degrees but can not be certain. Does anyone have any ideas as to what it is and more importantly how to convert it to British National Grid.Thanks in advanceAndy

I am trying to replicate what these two ArcGIS APIs below do in FME, between a polygon and a point, and a polygon and a polygon respectively. Which FME transformers can do best this? I tried SpatialRelator for the 1st,  but it gives  a lot less records than the arcpy. The arcpy adds a polygon for each point. For example, BHP_WELL_PATH_BUFF_FC has 28 polygon, there are 91 points that fall into those polygons, so WELLS_FRAC_HITS_TEMP_FC duplicated some of the polygons to be 91 polygons, one for each polygon. The SpatialRelator only gives 28 polygons. It does not have the "JOIN_ONE_TO_MANY" ,"KEEP_COMMON", and  "KEEP_ALL" options.arcpy.SpatialJoin_analysis(BHP_WELL_PATH_BUFF_FC,THIRD_PARTY_FC,WELLS_FRAC_HITS_TEMP_FC,"JOIN_ONE_TO_MANY","KEEP_COMMON"..arcpy.SpatialJoin_analysis(WELLS_FRAC_HITS_TEMP_FC,PRODUCTION_UNIT_FC,WELLS_FRAC_HITS_PU,"JOIN_ONE_TO_MANY","KEEP_ALL"

Background: I am maping a system of marked trails in a national park. There are dedicated trails for hiking, short walks, skiing, cycling, snow mobiles, snow shoes and what have you. In short it looks like someone droped a bowl of noodles over the place. In many places there are many uses for the same trail section, and it is these overlaping sections that I am trying to clean up.  Question:I have used LineOnLineOverlayer to find the overlaping sections and have created a List over the attributes that identify the type of trail, but from there I have not found the tools to move on. What I would like to do next is to break appart the overlaping sections from their original longer lines and apply a number of sorting rules to the List. Some examples: Where hiking and skiing overlap I would like to delete all but one polyline and give this a new code. If there is a Nature trail I would like to delete everything else but that. If there is a cycling trail I would like to keep that for later

Hi,I have Yellow lines with values ( attribute Value ) and i need to assign value to quotes ( red lines ), respecting the interval Value+1:My situation:What i want ( black numbers ):Attached my files.Thanks, - Danilo

Hi! My problem is that I have 2 readers combined, en excel sheet and a path folder with subfolders and documents inside. The inspector reads them without problem. At a certain point they go to a feature merger, from where I get an error report. The error reports shows that non of the documents from the folders are present. Does anybody know why that happens? The next screenshot shows how are they connected and what happens. Those 21 unmerged requestor documents should not go to the error report because they are present. Any help?? Thanks in advance. I am not very good at this but I try...

Just wondering if there is a way to transform a revit file from London Survey Grid into British Grid within FME?

Hello!I need som help with creating a outer boundary line with some other transformer than the "Hullaccumulator".The "HullAccumulator" (concave) in combination with "DonutHoleExtractor" works great and do what I wanted it to do but it cannot run Alpha value below about 0.5 because then my computer is chrashing! (out of virtual memory) And it have a fast i7 CPU and 32 GB RAM and a SSD drive but it does not helping me much...Is there a way to do this in an exact way and not rounding corners like in the "Hullaccumulator" with for example Alpha = 1 (that is meaning 1 meter radie in the corners and that is too much).I attache 3 Shape files that has totally 9 polygons that I want to be combined to only 1 outerline.Best regards,Jonas

I have a number of space delimited XYZ files (DTM data) and need to merge them and then split them into 1km x 1km tiles. The final dataset needs to be an .xml file that Civils 3D can use and produce a surface out of it.I've used the tiler transformer but i'm running into errors during/after the tiler processAny thoughts? Am i missing something in the process?

Greetings FME Community!I have a CityGML of buildings [1] and I want to attach new textures to the roof surfaces.The Appearance should be a texture based on a specific GeoTIFF that was generated for every buildings.Every GeoTIFF has the CityGML feature unique id as name. I don’t find a way to save every texture in the FMELibray and, mostly, retrieve the right Appearance for every feature of the CityGMLd to I think this should be related to the AppearanceSetter [2] settings or SharedItemRetriever [3], but I can't figure how to use them properly. Kind regards, [1] http://donnees.ville.montreal.qc.ca/dataset/maquette-numerique-plateau-mont-royal-batiments-lod2-avec-textures[2] https://docs.safe.com/fme/2017.1/html/FME_Desktop_Documentation/FME_Transformers/Transformers/appearancesetter.htm[3] https://docs.safe.com/fme/2017.1/html/FME_Desktop_Documentation/FME_Transformers/Transformers/shareditemretriever.htm

Road Number Runoff loss of control driver side Runoff loss of control passenger side off path Head on loss of control Head on overtaking Intersection Property access Other Fatals Serious CRA_ROUTE = 1 CRA_RUM_CO IN(72,73,82,83,84,85) CRA_RUM_CO IN(70,71,80,81,86,87 ) CRA_RUM_CO IN(74, 75, 79, 88, 89) CRA_RUM_CO = 20 CRA_RUM_CO = 50 CRA_RUM_CO IN(10,11,12,13,14,15) CRA_RUM_CO = 47 CRA_RUM_CO IN(30-39, 40-46, 48-49, 51-69, 90-99) CRA_DEG_1 = 1 CRA_DEG_1 = 2

In System Cleanup on FME Server UI you can set tasks do remove files (in Source Folder) older than some amount.Does it delete all subfolders in that given Source Folder?

Thanks all for attending and thanks to Mark for taking the time to run through things with us.We want these sessions to be driven by the people attending so that you find them useful. We originally came up with a set of topics to talk about the first of which were the general best practices for loading large datasets.The other topics are below but if you have other areas you think would be useful to cover then please let us know.I will propose another time in a couple of weeks for another session and I suggest maybe covering COU as that is something i think that could be a tricky area and is something we are looking at in ALT at the moment.Change only updatesAdditions, deletions and updatesTracking changesCreating history tablesCreating copies of updated features to use in comparison (COU Preview) Flattening/expanding of structureAddresses that have a master record and then multiple sub records Normalising/de-normalising data structures Differences between FME and data interoperabilit

For a customer I had to exchange the FME 2015 UpdateDetector in a workspace with the FME 2017 version of the same transformer.I have copied the same settings from the FME 2015 transformer and use the same dataset.The performance for the workspace went from 5 minutes to 50 minutes, and a lot more (75%) updates were detected.Has anybody the same experience with this custom transformer.Just wondering what could cause changes to be detected with the new transformer that were not detected with the older version.Any suggestions are appreciated.