Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
We ported all ETL processes to FME to have a single systems for all our dataflows. We tried to maintain a Enterprise Architect documentation about all our sources and sinks and the coresponding FME Workbenches to get a big picture of our dataflows.Since everything is in flow all the time it´s hard to keep up with the external documentation since it´s always manual work. It would be nice, since all needed information is already in the FME Workbenches, and all relevant workbenches are on FME Server, if the system would be self documenting and would generate some sort of enterprise dataflow overview, with all sources and sinks and the workbench names that are feeding and manupilating the data between the systems. Since all relevant wb´s are on the server it would be nice to click on them and get to them in some sort of way.
As an FME user, I would like to see support added for insertion scale units on the Autodesk AutoCAD Map 3D Object Data writer so I can specify the units. Currently, the workaround is to use a blank template and specify the units but this is a tedious task that needs to be done each time I want to write a new dataset.It would be great if there was a parameter similar to the one mentioned in this thread: https://knowledge.safe.com/questions/1906/write-dwg-with-coordinate-system-and-units.html
Sometimes in complex multi-drawing (workspace runner) scenarios it would be handy to be able to de-aggregate blocks into their component elements within the workbench instead of the reader.
Hi,I’m trying to build a dataflow using FeatureReader to read my files.Whenever I encounter a file with an error, the log file contains this snippet: 2016-11-24 09:50:12| 994.1| 0.0|INFORM|Expected a value string for '%L1KR', found '%KM1' instead, in line '916' of input file2016-11-24 09:50:12| 994.1| 0.0|INFORM|DSFL halted, approx. line in input file: '916' 2016-11-24 09:50:12| 994.1| 0.0|ERROR |Error DSFL expected a value string from the input file2016-11-24 09:50:12| 994.1| 0.0|INFORM|Closing the DSFL readerHowever, the feature output by the “<Rejected>” port of the FeatureReader only contains a single “_reader_error” attribute with very limited content:Error DSFL expected a value string from the input fileWould it be possible to enhance this error output significantly, preferably adding more error information (e.g. the offending line number), but at least so that the error message attribute reflects the full error messages shown at the top in the log file ?CheersLars I.
Import --> Dataset option to be added to JSONFlattener, feature reader (output port --> attribute to expose). Also applicable on JSONFragmenter, etc...
Imagine, some day, a new dimension in Workbench with transformers in 3D...So we could connect transformers in every direction of the space and move everywhere in the workspace.Transformers could be cubes or octahedrons or even dodecahedrons according to the number of needed ports.I don't have enough talent to draw my vision with a 3D software, but you probably got it.Ok it was my "friday delirium". I'm not even sure if it could be practical to use.
Hey guys,When utilising MapInfo Extended writer DELETE operation, the features are flagged for deletion, however the database is not "packed" at the end, and I have to go into map info to perform this operation to permanently remove these features. This causes large file sizes if there are constant deletions and additions on a native x file.Thanks,James
Hello,I often work with DGN files, especially version V7, and many times I need to map, if element is view dependent or not... As I mentioned below, value for this attribute is filled only in version V8. How complicated is to add support of this attribute in version V7, please?Thank You so much!Luboview-element.jpg
I'd like traceability as to who modified what workspaces/topic assignments/etc. It would be nice to be able to see what is taking place in an FME Server. Especially a production instance.
From FME2018.0 it is possible to collapse bookmarks. Maybe it is an idea as well to collapse annotations into an icon. It would be nice if you could choose between two or three different icons, for example an exclamation mark, a question mark and a ligthbulb. The different icons (maybe it should be possible to manually add them) can be you used to distinguish between the type of annotation
Currently to use the batch deploy option the workspace must have a writer. If there is not one, the option is greyed out. Given that are multiple ways to create data via transformers, this restriction no longer makes sense.
https://github.com/mapbox/geobuf Geobuf is a compact binary encoding for geographic data. Geobuf provides lossless compression of GeoJSON data into protocol buffers. Advantages over using GeoJSON alone: Very compact: typically makes GeoJSON 6-8 times smaller. 2-2.5x smaller even when comparing gzipped sizes. - Very fast encoding and decoding — even faster than native JSON parse/stringify. Can accommodate any GeoJSON data, including extensions with arbitrary properties. The encoding format also potentially allows: Easy incremental parsing — get features out as you read them, without the need to build in-memory representation of the whole data. Partial reads — read only the parts you actually need, skipping the rest. Think of this as an attempt to design a simple, modern Shapefile successor that works seamlessly with GeoJSON. Unlike Mapbox Vector Tiles, it aims for lossless compression of datasets — without tiling, projecting coordinates, flattening geometries or stripping properties.
It would be great if FME could check if all destination files are free for writing, and not currently locked by another application, during startup. Especially AutoCAD oder Excel file are often affected by existing locks that cause the workspace to fail at the very end causing some considerable aount of wasted time. If a destination file is locked the workspace should stop before any reader / transformer starts to work.
With a new computer and 64GB RAM there is not much that can slow down my computer to a halt. Using SurfaceModeller, working with raster and liddart takes time, howeve it does not render my computer laggy/non-responsive when running. Even with large datasets.But when using the PointCloudSimplifier this happens. It seems to be using 115% CPU. 60% memory still available, "no disk or network activity". Nothing else is running, computer is restarted and I have tried different las-files, smaller, larger etc. If it runs for 2 minutes or 2 hours the behaviour is the same.Is it possible to tune this transformer to use a little less CPU so you can "work normally" withouth this lag?
We often build up quite complex test conditions using the Tester, TestFilter etc. It would be fantastic if there was an easy way to export this as a WHERE statement in SQL (and maybe even other syntax...) that could be used in other applications and workflows.
The ESRI SDE Reader should support circular arcs (like the File GDB Reader)
When clipping a raster it would be nice to specify how to deal with pixels that are crossed by the clipper boundary.The options could be:include pixels that are entirely within the clipperinclude pixels whose centers (and/or origins) are inside the clipperinclude pixels who intersect the clipper - not that this should not include pixels that only share an edge.As it currently stands, minor (floating point) differences in raster extents and/or resolutions can lead to differing results
The LDAP reader in FME Desktop for Windows is currently unable to read objects that contain characters from other languages (e.g. Å? and Ä?) and therefore fails to load. I work for a global organisation and the usernames in our LDAP server contain characters from many different languages, therefore the workaround to change the locale of Windows is not feasible. I believe that this issue is also preventing our instance of FME Server from connecting to active directory, which is a significant barrier to the deployment of FME Server across the organisation.I therefore suggest that Unicode (UTF-8) support is added to the LDAP reader to enable multi-national organisations to connect to LDAP using FME Desktop and FME Server.
Hello,I think it would be a good idea to have another parameter for the SQS publication: Max Number of Messages.My scenario is as follows:Multiple FME Servers are polling new messages from SQS. Each message should trigger a FME Workspace, which will eventually remove the message from the queue if successfull. Basically, a load balancer for FME Server.Current Status:If multiple messages enter the queue very fast, the first FME Server to poll for new messages will basically grab all messages. Therefore, you have to work a lot with SQS visibility timeouts and FME queued job expiry times to achieve an equal job distribution.The parameter MaxNumberOfMessages should be available in the SQS API: https://docs.aws.amazon.com/AWSSimpleQueueService/latest/APIReference/API_ReceiveMessage.html What do you think?Kind regards,Dennis
It would be great if FME Cloud came in a "Home Edition" free of charge for hobby use. Any plans for that?
Currently even if the Ignored Failed Readers is set to Yes, "If all readers in the workspace fail, FME aborts the translation, regardless of this setting."Given that we have multiple ways to read data, including the FeatureReader, the various SQL transformers, HTTPCaller etc. It would be nice to have this parameter setting observed if there is at least one Creator type transformer in the workspace, even if all the readers failed.
The option to search by user has been added to the Knowledge center.Lately this function is not working anymore.I would suggest adding "My Questions" and "My Ideas" to the selection in these lists:That would make it much easier to keep track of the questions and ideas you submitted.
It would be nice to be able to to do a few bulk actions on schedules from the schedules page. Be able to change the sharing options of the schedules in bulkBe able to execute multiple schedules in bulk. Currently you have to open each one to "Start Immediately" and then go and update the scheduled time again back to the original time.
Please add a possibility to List the Reader Sources. This would be especially helpful for workspaces that use many Readers.In this way, you can quickly get an overview of the used sources!Currently it is very cumbersome, because you have to open each item in the Navigator to see source.
In-depth courses for all FME skill levels.
Explore our extensive collection of how-to guides and tutorials.
Log an issue with us. Our support team is ready to help.
