Data

As is stated in the title. However I need column names to be created exactly as they are, for example I have incoming attribute "OGC_ANGLE.uom" and even though I have Lower Case Attribute Names unchecked,tables in database have column names are created lowercase ("OGC_ANGLE.uom" becomes ogc_angle_uom) - this is a problem, if I read those tables and try to use xsd to create gml again, as these attributes are unknown to the xsd, therefore failing creation of the gml. I could manually rename those columns, however I would need to rename those columns every time xsd file changes, which is why I need to do this dynamically in the first place, as those xsd files are going to be updated quite often... And it seems that the dynamic PostGIS writer ignores Table Qualifier, it always creates tables in default schema, unless I specify schema in Table Name (schema.@Value(fme_feature_type))

Hi, my workspace reads WFS and gives me the warning“CS-MAP: Coordinate (approx) 3E:47N outside grid file coverage; fallback used”It seems to run fine any further, just I like to know where this message comes from, since this coordinate nowhere pops up on neither GetFeature nor GetCapabilities request. Indeed the approx coordinate does not fit within the dutch Grid, but it does not seem to come from the WFS source? My FME version is 2022.2.4

I’m converting geotagged mobile photos to dwg format. Inserting them at correct photo location.All works fine except that FME automatically converts my jpeg images to png files.This results in image files (png) which are many timmes bigger then the jpeg files. Is there a way to make FME write jpeg images instead of png with the dwg writer?

I need to incorporate a custom coordinate system definition into FME Form, but can’t find the right parameters in the online form to fulfil the definition.The following parameters define the map projection, but e.g. ITRF does not exist anywhere in the online tool.projection type: Gauss Kruger (Transverse Mercator)reference ellipsoid. GRS80datum / reference frame. ITRF2005central meridian. 11° 20’scale factor at central meridian. 1.0false northing. 0.0 meterfalse easting. 1000000.0 meter How can I create a proper fme file with these parameters?

Hi everyone,Curious if I could pick your brain on the workflow below.I would like to just write the asset_id field values from the GIS feature class into the corresponding assetID field in the API table. I’d want to write the asset_id from the GIS FC to the assetID field in the API table only if the assetID doesn’t exist in the API table. They have different field names.Curious where I would place the HttpCaller transformer with the PUT request in the workspace.The GIS EGDB reader would be at the start of the workspace exposing the relevant fields. Once I rename the asset_id field from the GIS feature class to match that of assetID for the API table, what would I do? Could I hook up the output from an AttributeManager directly to an HTTPCaller without a Creator?  

Hi everyone,I need to update the block attributes with the csv data, but I don’t know what I’m doing wrong. Can someone check my script and with this issues? I’ll really appreciate it. 

Hi, I have a JSON that I want to "flatten”, Currently, it has this structure. {"A":{"id":"1","Name":"Aa","IsTrue":"True"},"1stList":[{"Type":"Yes","Number":"100"},{"Type":"No","Number":"100"},{"Type":"Maybe","Number":"1000"},{"Type":"Unknown","Number":"11"},{"Type":"Hh"}],"2ndList":[{"X":"10","Y":"100","Z":"Abc","Q":"Def"}]}{"A":{"id":"2","Name":"Bb","IsTrue":"True"},"2ndList":[{"X":"10","Y":"11","R":"100000","Z":"Efg","Q":"Klmo"}]}{"A":{"id":"3","Name":"Cc","IsTrue":"True"},"2ndList":[{"X":"10","Y":"12","Z":"Lmno","Q":"Stuv"}]}Just reading it with the following parameter settings doesn't work:Flatten Nested JSON Values into Attributes: YES I've tried with 2 listexploders after reading, but that also doesn't give the correct result.I just want the endresult to be a flat table. How can I best achieve this in FME? 

We have some text fields that shouldn’t have any UNICODE characters.  Turns out at least one has snuck into the data.  Is there a way to Test to find UNICODE characters and then remove them?FME 2022.2.3

When exporting to shapefile from GeoServer/WFS one can specify encoding in the request, for my case I specified UTF-8.This shapefile has a cst-file that establishes the encoding (similar to cpg from ArcGIS).When I read this in Workbench/Server 2022.2 the fields have encoding cp-1252 and some garbled characters. If I change the filename to from cst to cpg the attributes come in with the correct encoding.Is this fixed in later versions, or should it be reported as a bug… ?

I saw that in this idea, writing dynamic autocad blocks was implemented in build 22098. Is there any documentation available for this yet? I understand this is still a beta build so it might not be available. The reason for this is im trying to write out FME aggregates as Autocad blocks with different colours stored as appearances for individual geometries within the aggregate and i cant find a way to preserve these when writing to Autocad.

Hello everyone, I have a small problem I am just starting on FME so I need your help. I have at my disposal a database in which there are many schemas and many tables inside the schemas. I would like to be able to retrieve a list of each last update date of each table and a list of each first creation date of each table. Here I would like not to do it table by table but all at once! Thank you for your help and excuse me for my approximate English! (For information I am on FME 2021)to be a little clearer I want to create a loop that separates me and sends me the tables one by one to sort the attributes then writes the result in an excel file the problem is not the sorting table by table but the loop and how to automate the sending of a table by a table, Thank you!

Hello,I was hoping to be able to create an FME workspace to be uploaded to FME server, that would run an SQL script (via SQL Creator) and create a CSV file that would then be uploaded directly to Azure Blob storage.I tried the 'AzureBlobStorageConnector' (using the 'File' upload option), but it appears that the file needs to be stored locally to use this option, which seems to be no good if I want to publish and automate this script via server. I also tried 'FeatureWriter', but couldn't see an option to write to a web service.Any help would be greatly appreciated!ThanksChris

Hi all , i hope you can advise on this issue. i’m using the cloud FME Flow , i uploaded my fmw file but when i run the workspace i don’t find the shape files at the Destination Esri Shapefile Folder.  any advise ?

We are moving to short-lived token authentication for ArcGIS Feature Service access.  I am able to use the FeatureReader transformer and pass the token and read all endpoints (132 in total).  However, I only want the transformer to retrieve a subset of endpoints. I attempted to create individual readers for each endpoint similar to what we had when using long tokens, but layers are now passing to the Initiator port.  I thought adding a GeometryExtractor transformer would work, but it did not.

Heya FME specialists, can you please give a thought on the next issue:Created a new table in PostGIS with FME from shape, using the PostGIS writer. (see 00_initiate.fmw) Point feature, attributes pkey and a, b, c with values. Insert, drop and create.Updated this table with the PostGIS writer. (see 01_update.fmw) Point feature, pkey, and only a, b with values. Update based on pkey, use existing.The problem is: We decided to maintain attribute c with QGIS directly in PostGIS but kept editing the geometry and attributes a and b in the shape. I deleted the attribute c in the writer assuming it wouldn't write it anymore to the database. That assumption appeared to be wrong and to prevent updating attribute c I had to use an AttributeKeeper in front of the PostGIS writer. My problem solved but curious if this is a DB writer syntax or a bug?Using FME Desktop 2016.1.0.1 on Windows 7. Uploaded workbenches in zip for reproduction.Thanks!

I am getting the following error message when I try to run my workbench.“FeatureWriter_UtilMaster (WriterFactory): MULTI_WRITER: No dataset was specified for MULTI_WRITER_DATASET or FeatureWriter_UtilMaster_13_DATASET or FILEGDB_DATASET”I cannot for the life of my figure out what the issue is. My only theory is that because it is trying to write to a geodatabase, but the data I have been given to test with doesn’t actually have any geography. The writer is Featurewriter with output format Geodatabase (Esri Geodatabase (file gdb open api)) with the dataset as one of the attributes.Does anyone have any ideas that might help, or think I might be on the money that the lack of geometry is causing a geodatabase writer to fail.

Hello, I have a point dataset which represent the noise level at 1x1 intervals on a wall.  I need to convert those point into a raster  which i need to convert into a graphic (jpg or tiff)  Any ideas how can i do that ?Thank you, David

Hi there!I want to let you know, that I'm writing a new tutorial covering XML Filter Expressions in WFS Reader:https://olivers-expression-lab.gitbook.io/filter-expressions-in-fme/ The basics are done.Please let me know, if it works for you and escpecially if not! Best regardsOliver 

Hey FME team,My coworkers and I are trying to figure out a tough polyline overlay issue. We are building a routable transport network for a city. We received 2 polyline datasets for the city’s road network. One dataset (Roadmap) contains geometry that we are licensed to modify the geometry for (as the original data were poorly connected and incomplete), and the second dataset (Carriageway) contains geometry that we are not able to use for the transport network but can use attributes for. Attributes include the number of lanes, whether the centrelines is in a urban or rural location, and the start and end measures for the carriageway feature. The link between the two polyline layers is a field called road_id, which forms a many:many relationship between the map layer and the carriageway. An example is shown below (3 Roadmap features, 5 Carriageway features):Roadmap (modified, x3)Carriageway (x5)Our issue is there is a slight offset between the modified Roadmap geometry and the Carriagew

I have Roads which are at the borders of municipalities (polygons). But for many of these road segments, the LEFT/RIGHT municipality fields values are wrong. I am looking to rectify that data. I have looked into using LeftRightSpatialCalculator and NeighborFinder transformers but they don’t seem to be helpful for identifying which municipalities are on the LEFT and RIGHT side of each segment. Please suggest, how do I accomplish this?

Hello , I got error from Python caller .i would like to get all paths from start node to end node then I have written this script but I got Ann error . could you please help me to fix this problem ?import fmeimport fmeobjectsimport networkx as nxclass FeatureProcessor(object): def __init__(self): self.graph = nx.DiGraph() self.start_end_pairs = [] def input(self, feature): start_node = feature.getAttribute('start_node') end_node = feature.getAttribute('end_node') arc_id = feature.getAttribute('arc_id') self.graph.add_edge(start_node, end_node, arc_id=arc_id) self.start_end_pairs.append((start_node, end_node)) def find_all_paths(self, start, end): return list(nx.all_simple_paths(self.graph, source=start, target=end)) def close(self): for start, end in self.start_end_pairs: all_paths = self.find_all_paths(start, end) for path in all_paths: feature = fmeobjects.FMEFeature()

Hello everyone, I’m currently working on a project where we have 3D mesh models (SLPK or 3DML) and I’m looking to convert these into point clouds. My goal is to use ArcGIS Pro with the Imagery Analyst Extension to extract and classify all tree points. We’ve had great success using the Tree Classification deep learning model with LiDAR data, and I’m curious if this approach could help us save costs by not needing to collect new LiDAR data when we already have 3D mesh models from the same year to convert 3D mesh to point cloud. Has anyone here had experience with this type of conversion? Any insights on the process, potential challenges, or tips would be greatly appreciated! Thank you in advance for your help!Best regards,Reno

Hello , I got error from Python caller .i would like to get all paths from start node to end node so I have written this script but I got an error . could you please help me to fix this problem ? import fmeimport fmeobjectsimport networkx as nx class FeatureProcessor(object):    def __init__(self):        self.graph = nx.DiGraph()        self.start_end_pairs = []     def input(self, feature):        start_node = feature.getAttribute('start_node')        end_node = feature.getAttribute('end_node')        arc_id = feature.getAttribute('arc_id')                self.graph.add_edge(start_node, end_node, arc_id=arc_id)        self.start_end_pairs.append((start_node, end_node))     def find_all_paths(self, start, end):        return list(nx.all_simple_paths(self.graph, source=start, target=end))     def close(self):        for start, end in self.start_end_pairs:            all_paths = self.find_all_paths(start, end)            for path in all_paths:                feature = fmeobjects.FMEFeat

using the Autodesk AutoCAD Map 3D Object Data reader and setting the parameter to “Group by Entity” (this is so i can get all the objects with 1 reader). I have the autocad_map_odtable{} format attribute enabled. i then have a testfilter to get >0 to separate the features with od values. do i need to enable an extended data value to get the table? How do i get the OD table values for these features?spoiler. I already have the data.i ran the workbench but looked at the wrong output. by separating the data i already have the attributes on all features within the testfilter. 

I have a set of records that read via SQLExecutor which includes a timestamp field (driver_date). This seems to be stored a “buffer” datatype. I don’t seem to be able to explode this field using the DateTimeExploder as it can’t autodetect the timestamp value as a date/time value. I’ve tried including a DateTimeConverter, but get the same result. Is there someway to convert the timestamp from “buffer” datatype to datetime?