Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
To be able to sort and/or filter by individual columns in the log viewer in desktop.Use case, identifying long running elements of a workbench
(Note: Not to be confused with the previously suggested idea “GeoBuf” - this is “FlatGeobuf”!)https://gdal.org/drivers/vector/flatgeobuf.htmlhttps://github.com/bjornharrtell/flatgeobuf/blob/master/README.mdExample: https://observablehq.com/@bjornharrtell/streaming-flatgeobuf
As we can zoom on a transformer by clicking on the log, it would be great to access the first call to the transformer in the log by clicking on the transformer (like the preview updates when you click on the lens ). It would save some time scrolling or searching.
There had been several suggestions to make design of FME Server automation workflows more graphical, and provide a higher level overview: https://knowledge.safe.com/idea/42491/system-architecture-design-view.html https://knowledge.safe.com/idea/18987/automation-workflows.html https://knowledge.safe.com/idea/19014/better-workflow-control-of-multi-step-workspace-jo.html I think a good addition to these ideas is the ability to display the status of each workflow step on this graphical framework, which makes this a visual monitoring / debugging interface, as well.
For a use case of this idea, please refer to this forum post.In summary, currently you can only access a single item in a list when using the adjacent features functionality. We should have the ability to access an entire list attribute, regardless of size/count, in order to copy it over to another existing or new list attribute. It would also be nice if perhaps the adjacent feature functionality is made available in other transformers such as the ListCopier, so we can copy an adjacent feature list attribute for each feature. Thanks.
Just like in some video games, having a list of all the available badges and how to get them could enhance the "gamification" of the platform. I know in my case I tend to give an extra nudge on someting if I know I'm close to get a badge or sticker :)
In short: Readers and Writers have and option to change a named database connection into an embedded one.This is our standard way of making sure that workspaces published to FME Server don’t overwrite other users' connections.And it is easy to do in FME Desktop:However, this not possible for FeatureReader, FeatureWriter, SQLCreator or SQLExecutor ! How many of your workspaces contain one or many of these transformers ? More than 50% ? 80% ? Yeah, I thought so …Consider the time savings, when in development you can add your transformers with named connections, and when publishing to FME Server you don’t have to embed the connection parameters by hand in every flaming transformer.In fact, it should be made possible in the ”Database Connections” section of Navigator - so that all transformers using that same connection would get embedded with one click !!!Read on for a case (almost) from real life ..Consider this scenario:your organization has several eager FME users, some of them fairly new, and you decide to let them publish their workspaces to FME Server. The users have learned enough FME to use saved ”Database Connections” in Desktop (because you teached them – or they learned it from an FME Webinar!).You yourself have made and deployed some critical scheduled workspaces using your own ”standardized” Database Connections like ”REPORTING”, ”CRM”, etc.After awhile you get emails from business users complaining that they haven’t received their critical reports for after sales marketing or contact lists of potential new customers.You start investigating and find out that your workspaces have crashed when trying to connect to the ”REPORTING” database. So what on earth has happened ?You finally find out that someone has actually over-written your database connection in FME Server. Someone has used the same database connection name ”REPORTING”, clicked ”Yes” when asked by FME whether existing connection with the same name should be replaced, and done the damage.But you remember that you told your FME users to replace their named database connections with embeded connection parameters in Readers and Writers when publishing their workspaces, and everyone says they have done so!So now you look at the Database Connections definition of ”REPORTING” in FME Server and find out that it connects to a completely different data warehouse (the one called HR_REPORTING). You then call the HR guy who made the latest deployment and ask him why he didn’t embed the connection parameters.He claims he did, only that he added one FeatureReader, tested with his named database connection ”REPORTING”, but could find the command to embed the connection parameters in the transformer. So he thought FME would somehow automagically take care of it, as it does for many things, and published his workspace anyway ...
It would be helpful if the fmeobjects API had better support for complex lists. Consider the list:_list{0}.id = '502' _list{0}.text = 'abc' _list{1}.id = '36' _list{1}.text = 'xyz'The following fmeobjects method currently returns None:items = feature.getAttribute('_list{}') To access child elements, you currently have to access them invidudually and then assemble them into an appropriate structure yourself.It would be very handy if the getAttribute() method could return a nested Python dictionary containing the entire list with its elements, e.g. using the sample list above:items = {0: {'id': '502', 'text': 'abc'}, 1: {'id': '36', 'text': 'xyz'}}This would get rid of a lot of boilerplate code in Python scripts dealing with FME lists, as we could access individual list items using syntax like:>>> print items[1]['text'] 'xyz'
A Geocoder for FME Server Apps would be a useful tool for locating users 'areas of interest' over a large area.This could be an addition to the geometry picker tool (new in 2020)orA stand alone map that can be added to FME server appsThis idea comes from feedback from users in the month of April 2020.The useful part of FME Server Apps is the speed you can deploy a workspace to server and make it available to users, this automates users for self-serve.based on a questionhttps://knowledge.safe.com/questions/112959/fme-server-app-geocoder.html
When one makes an error in the attribute manager it could be useful to have an undo button or support for CTRL+Z to undo. Asking for a friend.
https://www.cogeo.org/FME is mentioned but is official support coming?
***Note from Migration:*** Original Title was: Ability to Set Attribute Order in AttributeExposer: Alphabetical vs. Default Order via Import It would be nice if the AttributeExposer could honor the sort order of attributes if they are imported via a CSV or other type of source. Currently the AttributeExposer defaults to put them in alphabetical order vs. honoring the order they we were listed in the CSV. Alternatively, if the CSV contains a separate attribute for 'sort order' could the AttributeExposer transformer use that to maintain the sort order? Maybe give the users a choice: Maintain Existing Order or Alphabetical? Manually moving attributes in AttributeManager or in the Writer seems like something a transformer should be able to do.
I recently received several .numbers files with data that I want to relate to points on a map using FME. I had to ask the sender to export them from her Mac as Excel or CSV. Fortunately, she was willing and able. This is the first time that I have received .numbers files, so I don't know if this is a larger problem for the community. For the current project, I anticipate receiving more data in this format.This from the web: "You can only open NUMBERS files with Apple Numbers in macOS and iOS. You can also upload NUMBERS files to the iCloud version of Numbers in your web browser (requires an Apple ID). This is helpful for Windows users who don't have access to the Mac or iOS versions of Numbers. In the iCloud application you can view, edit, save, and download NUMBERS files."
We need an Active Directory connection with group managed service account (gMSA).We can only use gMSA (for the "search account") to connect to the AD. Right now, FME Server supports only regular accounts. So, i cannot use our AD.That's the interesting point for me. Please support the gMSA in FME Server :-)
Include an option to control what is added to the Windows Context Menu (right click on file/Workspace). With multiple versions the list can get quite large - and I never use Quick Translator. OK I could edit the registry but would prefer something less extreme.
When using FME Desktop, the "Run" button is always enabled, allowing a user to potentially run a workspace at any time, and therefore overwrite data at any time. While the "Redirect to FME Data Inspector" option is available for Writers in a workspace, that does not disable all transformers, including Executors and FeatureWriters.This idea is proposing a "Disable Run" option for the entire workspace, allowing a user to toggle on/off the ability to run a workspace. An additional functionality is to allow an author to "Lock" the "Disable Run", therefore if the workspace is opened by a colleague (or other unauthorized user) the workspace will not run unless the enabled.
Integration frameworks like Make, Power Automate & Tray support the creation of webhooks when connecting to Online hosted feature services, FME should too, including managing them (events, frequency, payload URL etc). This should come to Portal too when it gets webhooks.
It would be nice to have a panel with an overview of the existing FME Server apps in the main FME Server window. It's definitely one of the most used functionalities.
Hi,Are you planning to support IBM Notes? I actually don't understand much what this is, I've just got this question from a customer.Thanks a lot!
The current writer for JPG exports a WLD file. It would be useful to have it be an option to export JPW files.
Its would be great to be able to create world file for Jpeg2000 (.j2w) and for the ECW (.eww)
Many times, I have a workflow that reads multiple tables from one database, performs a few simple tasks to each one individually, and writes to another database. If there are 10 tables, it completes one before moving on to the next. This can add up to long processing times in the end. However, if I split that job up into 10 separate workbenches and run them concurrently on FME Server, the total processing is drastically reduced since they are run in parallel. Unfortunately, that method creates a data management nightmare. My suggestion is to allow parallel reads and writes within one workbench (when those reads/writes don't depend on the other reads/writes in the same workbench).Obviously this wouldn't apply if there are table joins or any transformers that hold features.Example: The screenshot below is a job that takes nearly 24 hours to run due to large amounts of data. However, no single table takes more than a few hours. But since FME runs in "serial" mode, all those hours are added together in the end.
This thread outlines it pretty well:https://community.safe.com/s/question/0D54Q00009QoujHSAR/best-way-to-remove-fractional-seconds-in-date-timeCurrently if you want to drop the fractional component of seconds you need to do extra steps. This is fine, however, it feels like it would be neater it the DateTimeConverter/DateTimeFormat function could handle this. Just a nice to have really.
Hi everyone.I was wondering if there ever was the idea of an "Tester" within all Transformers (such a the "group-by" is in almost every Transformer now).I couldnt find the idea yet and didnt see that this feature will be added in 2022.Is there any reason not to implementt?Not sure about increase of performance, but for overview and reduction of Transformers. Might be only makeup but i have around 60 Testers in one of my workspaces and i thought it might be a not too difficult to implement.I guess the necessity hasnt been too big yet, but i feel there will be more and more workspaces on fmeserver/fmecloud where the users can choose parameters which have different impact within the workspace.Would be glad to see this or to read if/why not to see.Cheers
What happens on FME Server 2017 and later versions (Linux) - when FME_TEMP is out of free space during translation, FME Engines are just crashing with "ResourceManager: Optimizing Memory Usage. Please wait..." message. The message is quite confusing and we had to dug many days about the cause. I propose to change message so the user know exactly what happened: "FME TEMP is out of space, please reserve more free space in temporary directory" or something similar.BTW - Shouldn't the memory to FME_TEMP be dumped in a clever way to use only what's available and not crash?
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.
