FME Form

We have a large number of photos taken in many countries and contexts. They are from a range of people from professional photographers to amateurs, using everything from professional DSLRs to smartphones. Preparations are ongoing to load them into a new professional online media library. The online media library can read certain IPTC metadata fields. We are considering an interface where staff around the world can upload their photos and add the required metadata. To facilitate easier integration into the media library, being able to write the user input to IPTC metadata fields/tags with FME would cut out a significant manual step. All I found on the Safe site was this:https://community.safe.com/s/question/0D54Q000080hV8xSAE/can-iptc-metadata-tags-be-exposed-using-fme-

Revit files seems to have a non linear time to process. Is there anything I am doing wrong or anything I can do to speed this up. Currently a:8MB file ~ 10 minutes70MB file ~ 1 hour 10 minutes+200MB file ~ waited over 12 hours before cancelling. I have many of these files I need to process so just 1 file taking over 1 hour is way too long. What is the best way forward? I have a similar pipeline with civil3D files however they are processed fairly quickly, it seems to be just revit files take a long time.

I have a dynamic replication workspace which reads the contents of one file gdb and uses its schema to truncate/overwrite another. I've added some contour datasets to both gdbs that are static, and don't want them being included in the replication because of the time overhead they add in reading and writing. Mid-workspace workaround is the humble Tester, but not reading at all would be better. I've had a google of regular expressions to use for this, rubular rejected the solutions here ( https://stackoverflow.com/questions/2078915/a-regular-expression-to-exclude-a-word-string ), and couldn't think of how the merge filter would be used for it. I need to exclude feature classes like '%Contour%' and also those like '%0_5m' - in the latter case it can be an 'ends with' scenario as it is always at the end of the feature class name.

I am trying to understand why, when using the FME RasterResampler (cell spacing) for multiple datasets at various resolutions, the output is variable resolutions rather than just the cell spacing I have defined (10m).  For example one dataset with a native 2m resolution does not get resampled to 10m when I use the RasterResampler setting cell spacings to 10m, but instead is is resampled to 14.3561626759533. All datasets are resampled slightly differently and only a few to the requested 10m. What I want is the resample output to be the same as resampling in arcmap, would I need to use a python caller to do that?  

I am having a problem getting a HTTPCALLER do a GET from a custom API. I can successfully connect to the same API url in a web browser but get the error below from the HTTPCALLER. I am using the HTTPCALLER with similar APIs without issue. The url for this API is a little different than previous urls I have used so maybe that is an issue?Previous urls were like https://.../name/2A159Problem url is like https://.../name/2A159/Family?rtuFamily=C where the GET has to match both 2A159 and rtuFamily=C HTTPCaller_4 (HTTPFactory): HTTP/FTP transfer error: 'SSL connect error'HTTPCaller_4 (HTTPFactory): Please ensure that your network connection is properly set upHTTPCaller_4 (HTTPFactory): No proxy settings have been entered. If you require a proxy to access external URLs, please ensure the appropriate information has been entered

Hi,I have some polylines that I buffered with Bufferer in order to make some checks.Once the checks are performed, I'd like to get the original polylines back. I was wondering if I can copy the lines geometry into another geometry attribute and read them after, or maybe some other technique. Thanks in advance!  

I am working on a integration Workbench between 2 softwares using their REST API. One of the REST API request JSON body as arguments to use some GET methods. HTTPCaller seams not to allow to use body with GET. Any idea ? Thank you. Mathieu.

If a execute this workspace: Using an existing (and empty) database with this trigger on parcel table: I get this error:  There seems to be a conflict between the "prepared statement" used by FME and the trigger. Are they incompatible? Note: This project is a simplified version of another in which the database schema and the logic implemented by the trigger are considerably more complex.    

In FME desktop I have a geometry extractor which converts simple polygons to FME binary and stores them in the attribute _geometry. I need to use this geometry in another workspace. I pass the attribute _geometry to the other workspace using a workspace runner and a parameter. In the child workspace the Geometry Replacer fails. The error is INVALID_PARAMETER_GEOMETRY_SOURCE. Does anyone know if it is possible to pass geometry as text from one workspace to another or what I might be doing wrong? I've attached a simple example of the problem.

I am attempting to use the arcpy.da in the python caller transformer in FME workbench 2021.1 build-21627 W 64 bit , ArcGISpro-2.9v installed. I am wondering if there are any suggestions to getting my workbench to successfully access the arcpy.da module. I have setup the preferred python interpreter in the FME options. If you provide any help /documentation related to the Arcpy in FME it would be highly appreciated. Also getting following error while running the FME Workbench? 

Hello everyone,In my workspace, I have one point cloud feature (IFMEPointCloud). The points have x y and z components , but many of these points have unnecessary digital precision. My goal is to round all coordinates up to two decimal digits. After reading the FME documentation and checking the FME Forum, it turns out that: "CoordinateRounder" does not work with point cloud datasets;"PointCloudExpressionEvaluator" only rounds float components using ceil and floor expressions, thus resulting in integers.As a workaround, I am trying to use "PythonCaller" to go through each individual point of the IFMEPointCloud and then round their xyz components. The documentation for IFMEPointCloud is, however, limited, and I don't know how to reach the point/component levels.Any ideas? Thank you in advance!

I am trying to utilize the Anchored Snapper to snap the edge of a zoning area to the underlying parcel boundaries. The tool visually works great; however, it is utilizing the vertices of both features resulting in overlaps and slivers that are later detected in our workflow. Is there a means to have the Anchored Snapper only utilize the vertexes of the anchor? Is there a better tool option that will not change the parcel feature?

Hello.I am looking to split some of the attributes of a filed into 2 separate fields while rest stay in the main field contents. even the split one part goes to separate field and rest stays in.example I have following data:Banana 24Banana 25Banana 26Banana 27Apple 01Apple 02Apple 03Apple 04Pear PearI want it like Banana stays in original field and 26 goes to separate field.but nothing happens to pear.I tried using tester to filter them out then attribute manager with string function to replace but length is not same for all attributes how do I handle different length?Thank youNikki

Hi All, Hoping an extra set of eyes on this might solve this.I am using XMLUpdater to update geodb_metadata_string and everything works fine until I use a Conditional Value. Then I produce the error below. I have tried several different conditional statements, with and without quotes and have run it as PlainText and as XQuery but always the error. Pics from the Updater and Conditional Value below. Thanks for any ideas.Starting translation...Unbalanced quotation marks encountered on line 438 of file E:\GIS\FME\test\wb-xlate-1644023348637_43516 -- line was `FACTORY_DEF {*} XQueryFactory2    FACTORY_NAME { XMLUpdater }    INPUT DOCUMENT_FEATURE FEATURE_TYPE AttributeExposer_OUTPUT    INPUT QUERY_FEATURE FEATURE_TYPE Sheet1_XLSXR_2    FLUSH_WHEN_GROUPS_CHANGE { <Unused> }    XML_TEXT { "@EvaluateExpression(FDIV,STRING_ENCODED,<at>Value<openparen>geodb_metadata_string<closeparen>,XMLUpdater)" }    COLLECT_CONTEXT_ITEM_NAMESPACES YES    SEPARATE_DOCUMENT_FEATURES YES 

Hello,The following settings of the StringReplacer works for some feature attributes, and doesn't work for others.This is aimed to replace dots for commas (or vice-versa) in a decimal number.So, it normally works for example like 100.5 -> 100,5 But, I just used the very same transformer in a different feature, and the result is zero, for example: 100.5 -> 0 Both columns are 64-bit real. I can see in the the "cached features" in Data Inspector that a valid decimal number goes in, and out goes the zero. Any ideas?Thanks in advance

The ultimate goal here is to orient overlapping lines in the same direction so that later offsets will come out neat. The lines have been processed in multiple steps and when I apply a OrientationExtractor it says: no_orientation. Is there another way to get the orientation, other than the Topologybuilder which is quite slow and requires a few extra calculations? Also is the Orientor the only way to rewrite orientations to a line, or is it possible to copy it somehow with something like a FeatureMerger?

I'm playing around with security settings and it seems that if you password protect a custom transformer, it still spews out info in the logs from the internal guts of what's going on. Is there a way to prevent the internal guts of a custom transformer from being logged? Ie. I just want to show the input and output of the transformer. Is that possible? Seemed maybe like it was from this article https://community.safe.com/s/article/password-protection-for-fme-workspaces

I have tried reading out to a file geodatabase and then reading that out with the where clause to shapefile but would like to try something more direct

We can't seem to get it to work using any of the available transformers, nor any of the help we've searched for online. We just end up with either the data in the output table staying the same, or an error in the transformation. We are using a standard file GDB in ArcMap with a FC containing about 10 million records, and we want to replace about 200,000 of those records with updated info that is located in another FC in the GDB (based on a common field (county)).  Is there a way to do this??

Hello, I'm working with FME2020. I want to import several excel files with the same schema from one folder as different features. With a xlsx-reader (select multiple folders/files) all files are written in one feature. With a path reader I can't define a schema. Is there a way to read all the files with one reader or do I have to use one reader per file? Thanks in advance

Hi Guys, any ideas of how to solve this please (I'm a novice so take it easy please, I have a workbench that has 2 readers and 2 writers, the first part of the process reads and SDE featureclass , manipulates some attributes and writes to another SDE Featureclass - it has a run before and a run after sql for dropping indexes and constraints and recreating indexes and constraints and running a stored procedure. The second part of the process reads the updated featureclass and filters by query like value=1 then writes to a csv file. This all works however the second part does not select the correct number of records - its 6 records short. Its possible due to the before and after sql scripts taking a while to finish and commit. _ I can brake this into 2 workbenches and its all perfect but I'd like to keep them together - is it possible? Thanks Paulg 

Hello! I have a workbench in FME Desktop 2021.2 where I need to write out several XLSX files from the same dataset, with exactly 1500 rows in each, and each one named FileName_1.xlsx, FileName_2.xlsx, FileName_3.xlsx, etc. I have managed to do this using a Counter, and ExpressionEvaluator. The ExpressionEvaluator takes the attribute from Counter (_count) and creates a new attribute called "_result" using this expression: @floor(@Value(_count)/1500) + 1 So that gives me the sequential file numbers that I want to append to the file names based on 1500 rows in each one. I can do this if I include the "result" attribute in the Excel writer, and use a FanOut Expression like this: FileName_@Value(result).xlsx This produces the Excel files with the correct file names, with 1500 rows per file. However, my question is, how can I exclude this "result" attribute in the Excel files? I don't want that attribute being written out. If I remove it, there is then no

I want to do some transformation on a schema/datamodel of a GDB and write it into a new GDB (which should be basically the same schema with these adaptions). But i do not get any output if i use Format: schema (any) and feature reader. Does anyone has some ideas for me? I´m a newbee using FME.

I am in the process of developing a custom transformer that can write a coded domain list to a new file geodatabase file, using this article as inspiration: https://community.safe.com/s/article/creating-and-associating-geodatabase-domains-dynam  The hope is to create these domain values either from the existing data or a look-up table within FME. Then upload the file geodatabase with these new domains to ArcGIS Online. I have followed the guidance in the article and believe it should work. The attached transformer edits the schema feature of the relevant attribute to something like this: "attribute{7}.native_data_type (string: UTF-8): coded_domain(FloorType:char(255):2:"2":1:"1":0:"0")". But when trying to write dynamically to gdb from the Schema Feature, the domain doesn't carry through.I compared the syntax of writing a coded domain manually to gdb in FME (by editing the user attributes) and looking at the log, and the format is the same as

Hi, can FME write PDF/a format files? I don't see any options in the Geospatial PDF writer.