Shape the future of FME with your ideas
New and Open ideas are awaiting review or have been reviewed by our Product Management team, and are open for commenting and voting.
The Directory and File Pathnames reader is useful for getting the files paths as an attribute without reading in the data itself. However it currently only takes a folder as the input parameter.There are situations where the user should be able to pick any combination of files/folders through the Select Multiple Folders/Files parameter interface and the result is one feature per file (not selection) with an attribute containing the full path to the file.It currently takes 8 transformers to achieve this result.
It would be nice is there was a ConnectedComponentCalculator that would work similar to the NetworkTopologyCalculator except with Area features that are intersecting.The features would be output with an attribute _component_id that would be shared by all the features that belong to the same graph.Bonus would be an optional attribute that contains the number of features that are part of the graph.
***Note from Migration:*** Original Title was: Run FME Job using active directory account of user (not windows service logOn user) I was chatting with a customer and they said they'd love to be able to let their users run scripts that access data and directories they are already entitled to using the Active Directory account. Currently, the jobs all run as the single account specified as the Log On user for the FME Engine windows service). That account needs permission to access ANY data that might be processed by ANY job on FME Server.I'd prefer if the jobs ran as the user that creates them.
Hello all,Since the revamped date/time implementations in FME, my use of small python-snippets has been reduced dramatically. No more 'import datetime' to do some simple operation.Now the bulk of my python-use involves the os.path library, like splitting, joining or testing for existence.There are a couple of transformers who have path-functionality, but I often need it in text-editor like situations. e.g. building the output-file-name of a writer from parts of the input parameters. Splitting and concatenating using 'text-methods' is not platform-agnostic due to the path-separator. So I put in some python-scripted-parameter.So please, give the os.path methods the same treatment as the datetime library got some years ago.Martin
So we have the good old writer with that nice feedback if all attributes are connected. But yes it's a dead end an I'm not using it that much anymore.Then we have for some year the FeatureWriter with the ability to keep flowing. It's great and I can not do without it anymore.Shoulden't we combine the strengths of both solutions and make an enhanced FeatureWriter?
Since the Inspector supports changing the drawing style of geometry it would be nice to have the FeatureColorSetter have opacity and width parameters.
I was curious if this would be possible to add a future enhancement that would allow for tabbed workspaces within FME Workbench? I think this would be really nice in order to have all of your open workspaces available to you within one instance of FME workbench.
When using the JSONExtractor the JSON Query can only be hard coded or user parameters. It would be nice if the query could be constructed using attribute values.ex. json[@Value("json_index")]["test"]
To be able to use the mouse less and keyboard more I have tried to use TAB key to switch between parameters within settings of a transformer. I notice that the order they switch between is not logical in many transformers. Below is some examples. Would be a great summer-job to go through all transformers in TAB-way and fix the order in a more logical one!.
Hi,As long as upgrading FME version is time consuming, there will be people with multiple installations on their computer (no mass upgrade for transformers etc..). Would it be possible to have some improvements to deal with this situation such as :list of recently edited workspaces with the build and not for all FME installations,a way to automatically open with the version used for the last editing if it is present,an automated naming option offered based on build number,a warning in workspacerunner if the build used to edit the launched workspace is different.etc... Thanks!
Currently, if you set an HTTPCaller method to GET, the body section of the transformer configuration window is disabled. I'd like to request this section be enabled.I have a REST API I must interact with that, for whatever reason, requires the parameters of the GET request to be submitted as json in the body. This makes FME and this REST endpoint incompatible.Thank you
It would be nice to have a type "Time" as an option of user parameter. Currently, only Date/Time is available.
Having had my fingers substantially burnt when upgrading Writers recently, I think it would be a really good idea to show a side by side comparison when doing this, as happens with Transformers.I upgraded just the Writer, not the Feature Types, and didn't notice that because some of the Writers were quite old, the options had changed. This meant they switched from the equivalent of "Truncate existing" to "Create if needed".If I had seen a side by side comparison, I would likely have spotted this and been able to correct it before running the Workspace and having duplicate data to sort out. Of course I should have done due diligence and checked the Writers first, but seeing the side by side might have made that job a little easier.
I'm working more and more with JSON data coming from an FME Server Automation webhook. Since the data isn't passing through a Reader in my workspace (it comes in via a published parameter), some JSON values may be encoded for URL (% codes for many special characters). After passing through a JSONFlattener, rather than having to chain together multiple TextDecode transformers to account for all the attributes that might be encoded, it would be simpler to be able to decode all the attributes inside an AttributeManager (or even an AttributeCreator).Unfortunately, there are no string functions currently available in AttributeManager/AttributeCopier to encode/decode attributes.It would be great to have encode/decode functions to provide this functionality!
Duplicate to Writer is very useful but some additions/improvements could be made I think. For example:When "Duplicate to Writer" is chosen, automatically include the reader as a writer to duplicate to.Option to select an existing FT on a writer and choose "Duplicate to Writer" and select same or another writerDuplicate to Writer automatically connects to the reader being duplicated - this is needed in about 5% of cases for me. An option to duplicate with no connection would save manual deletion of connection almost every time.In the data sources list, it would be useful to be able to to say "duplicate reader as writer".If you duplicate a writer FT now, 00 is added to the name. It isn't possible to change both the Writer and remove the 00 from the FT name in one step, unless one adds something to the end of the FT name.
In the hub, a Custom Transformer can be downloaded to count the number of characters in a string: https://hub.safe.com/publishers/gisinnovationsb/transformers/specificcharactercounter This involves creating a list, so it is rather slow.@takashi has some excellent ideas in: https://community.safe.com/s/question/0D54Q000080hYeESAU/count-number-of-unique-characters-into-a-stringAlso some really good ideas in: https://community.safe.com/s/question/0D54Q000080hIUkSAM/count-the-number-of-characters-and-numbers-in-a-given-stringIt might be useful though, to have a dedicated Transformer 'CharacterCounter'. I've looked for it several times, before I started implementing a solution with the ExpressionEvaluator. Especially when you can make it faster than the current solutions.
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.