Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
I would love if FeatureReader could read data from an attribute, not just from disk file (and the same way, FeatureWriter could write the contents to an attribute instead of a disk)Eg. binary data files (image, PDF, Excel, etc) are data URI encoded in JSON from a web service. It is straightforward to extract and decode back to binary to an attribute. However, this binary must be written to disk and read back by a FeatureReader. Writing and reading a disk is always slower than memory and it would be one less dependency (eg. folder and filename) to take care of if the FeatureReader could read from an attribute.
It currently takes 6 transformers to read JSON or CSV or whatever from HTTPCaller with FeatureReader, because the http result `_reponse_body` needs to be written to an intermediate file before FeatureReader can utilize it. This is a lot of extra work when all the information needed is already present. FeatureReader should be able to take `_response_body` (or any string for that matter) as input in addition to files and urls.
Please add a <Rejected> port to the EsriReprojector.Currently, if the reprojection fails on a feature the entire translation is terminated. This becomes an issue when bad geometry is encountered on a single feature (out of potentially millions of features). It would be much better if that 1 bad feature could be logged and the translation continue.
Many coordinate systems have bounding boxes defined but if data that falls outside of that area is transformed the results may either be wildly unusable or (as one of our customers is seeing) can cause an infinite loop. So I would suggest adding a <rejected> port to the Reprojector and other projecting transformers to optionally reject features that fall outside of the bounding box of the target coordinate system.
***Note from Migration:*** Original Title was: Enable Rule-Configuration for the "IN" operator on AttributeValidator to be set from an attribute I often need to test if an attribute's value is compliant with a comma-separate list of allowable values (enumerated types) held against another attribute. I know I can test for this in a TESTER but it would be nice to have it all in one place, and with a TESTER I still need to build the validation error message.A workaround is to use Regex operator which does support attribute substitution.
Can the attributevalidator be made to allow for composite tests....e.g. Field type is Double but null is also possible.
Hi,It would be great to be able to implement conditional validation rules.If Att1 = Value1 then Att2 must be in "A,C;F" else Att2 must be in "A,B,C"This would replace many combinations of "TestFilter and AttributeValidator". Something similar to the conditional value setter in the AttributeCreator..JP
The Logic section is missing, preventing multi tiered validation as well as inverse operations such as "Not" In and "Not" Contains Regex. This could be such an amazing tool, but currently only half the functionality is unlocked!
Currently in the AttributeValidator the Unique rule is treated as each attribute to validate is required to be unique (individual) I would like to see a configuration option added for "composite" which would treat all selected attributes as if they were concatenated. Yes this can be accomplished by simply adding a StringConcatenator and computing a new concatenated column to validate, but many other transformers that have key parameters accept multiple key fields and treat them in this composite fashion, so why not the AV.
Would be nice if the shape of the Junction could be changed from the dot to a regular Transformer. I used the AttributeRemover as Junction before they were introduced. (I still do, but this can be tricky...)Having a hard time to click / select the tiny junctions and can't get used to the different look. I know the workarounds, like a Custom Transformer, but it might help others as well.
A few suggestions for improving the visual feel and flow of workspaces.The ability to hide unconnected ports on a transformer - e.g. featureMerger has 7+ ports but often only 3 are utilised - leaving an unnecessarily tall transformer.Height adjustable ports (minimal if not required, allow for banked text, etc)Style Junctions - change size, shape and colour.Better connector alignment - difficult to get a truly horizontal connector at times.Improved alignment of transformers and bookmarks (something akin to the Visual studio alignment functions on Windows Form development would be fantastic)
Essentially builds on this idea a bit: https://community.safe.com/s/idea/0874Q000000TlqlQAC/detailThe idea is to have the option to hide/remove output ports for each transformer. The Benefit would be that these removed ports would no longer cache, but would also provide a way to further tidy up/reduce the size of a workspace.There are risks that these hidden ports would cause more trouble than they're worth because of missing features.
There is currently no way to assign a useful name to junctions in a workspace. I have workspaces with dozens of junctions, and they are all named "Junction_1", "Junction_2", etc. When collapsing bookmarks and adding custom labels to each input/output, and a junction is one or more of the inputs or outputs, it can be difficult to figure out which junction is which without repeatedly opening and collapsing the bookmarks. It is also challenging to connect to/from junctions in the Feature Type Connections window without knowing which junction is which.On a related note, being able to browse the usefully-named junctions in the Navigator would also be very helpful.
Move Workbenches between Workspaces on FME Server. I can delete and republish, but I want the ability to move workspaces between the Different Server Folders.
Add the ability to drag and drop a workspace that is in one Repository folder to another Repository folder. Currently, a user must remove the workspace complete from FME Flow and republish it. This makes it more cumbersome if a user simply needs to move workspace into a different repository.
Allow users to create a sub-folder ("sub-repository") within a Repository (folder) would allow for better organizing of workbenches, especially if a user has a lot of workbenches they are running in FME Flow.
There will be cool to specify a Connection in Emailer transformer as for example a WebConnection. With this feature the maintenance of workbenchs that use a specific connection will be easier.
Probably some of these transformers can be improved about their performance, if they had an option specifying first input (such as "Clipper First" in the Clipper) and the user could control the input order.AnchoredSnapper: Anchors FirstListBasedFeatureMerger: Suppliers FirstNeighborPairFinder: Candidates FirstSpatialRelator: Candidates FirstLineOnAreaOverlayer: Area FirstPointOnLineOverlayer: Line First
Re-opened as Idea, originally posted in Q&A as https://knowledge.safe.com/questions/32480/dlis-support.html DLIS stands for Digital Log Interchange Standardand is widely used in Petroleum industry. The standard is described here http://w3.energistics.org/rp66/v1/Toc/main.html Couldn't find any library for processing that format.
Feature caching is very powerful when authoring complex workspaces. It would be great if the VariableRetriever / VariableSetter transformers supported then feature caching mechanisms with partial runs, etc.
It would be really helpful for both average users and newcomers if you could use the new ChatGPT feature in GPTs to create a tool that assists in building workspaces, maybe even ask users questions about what they need, the format they prefer, and even provide examples cases to make workspace creation easier.I've used ChatGPT a few times, and while it's helpful, sometimes it makes up transformers. Having a dedicated GPT for this purpose could potentially resolve such issues and even help generate Python code for PythonCaller for tasks that don't fit a transformer.If something like this already exists, please let me know where to find it.
Since it is very common to retrieve GeoJSON and EsriJSON, how about combining the fragment-as-format functionality of JSONFragmenter into HTTPCaller, in the response handling?
In the Desktop User Parameters dialog there is the option to "Save As User Parameter Default Values". It would be great if this feature was also available for use in server apps so that the parameter information the user entered is retained when they run the app again.
It would be great if user parameters in automation apps could be imported from some of downstream workspaces, and updated accordingly if that workspace is changed. Similar to what we have in Workspace apps now.
Would it be an idea to have an extensive webinar summary in written text available, in say a week or so, after the webinar has been held? When looking for information one could be so much more time-efficient reading the summary, rather than having to sequentially access serveral hours of webinar talk.
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.
