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 have been playing around a bit with Duckdb lately and its performance is impressive, even with its spatial extension. I understand that 0.8.1 is still a young version and lacks some backward capability but it would be interesting how Duckdb could play with FME.
Notify Workspace Authors on FME Form about Broadcast Messages in FME Flow when connected. I talked to a group of FME Flow authors during a recent training and all agreed that this would be a welcome feature in their organization to notify FME users also on Form about upcoming downtimes due to updates or maintenence. Broadcast messages aim to reach as many Flow users as possible, so why not include the ones on Form.
Currently it is possible to select (multiple) fields when adding fields in the DatabaseJoiner. What I am missing is the possibility to add the field(s) with a specific name.I am bound to add them with the name they have in the joined table, leading to a AttributeRenamer after almost every DatabaseJoiner.It would be much more straight forward to have the option to add the new name/alias directly in the DatabaseJoiner.
Hi guys,in fmeobjects there is a function FMEFeature.getAllAttributeNames used to retrieve all attribute names. However, it also includes all format attributes available on the feature. Of course one can easily filter out attributes by checking for specific prefixes (e.g. fme_). However, if you are writing code for generic use cases, this can get quite hard since every format provides new format attributes (e.g. kml_, csv_, xlsx_, ...). It would make life way easier if this there would be a function getAllUserAttributeNames to only retrieve user attributes. Best wishesDennis
Use Case: We treat FME Server as a shared resource across our enterprise, if one department user submits thousands of jobs with the highest priority, that user can consume all engines blocking all other users for an extended period of time. In such a scenario today we have only one option: to ask the users to cancel thousands of jobs and resubmit those jobs. and this is not efficient. We as admins should have the option to select jobs in bulk and switch the queue for pending jobs.
Can we have a reader for the ESRI Vector Tile Packages, so they can be used within the background maps in Data Inspector
Hello,There is an Automation Writer to configure the geojson to be sent as a message. This is great, especially the debug mode to see it locally. I wondered if the generated geojson could be also imported as a reader for the subsequent workspaces.In the current examples, I have only seen ParameterFetcher used, even the geometry is extracted as an attribute (and so replicated..). It is a bit painfulIs there some work in this direction already? Thanks for the great work, FME Flow UX is getting better and better!
Can you update the API call to enable us to see who ran a job? I am trying to better support my users but I would like to be able to include the user that ran the automation into an email that I receive when a job fails.
It would be nice to have the ability to put descriptions in next to attributes for what you're doing to them (ie. if you're doing some complex math equation or formatting). This way you can better see what's going on in the attribute manager.
Details of this idea are described in this post.
With the following ministerial decree, Italy formalised RDN2008 as the new official reference system.It would be appropriate to include the EPSGs referring to the RDN2008 datum natively in the software, in order to avoid Italian users having to write the definition of reference systems by hand.The list of reference systems can be found in this post
Dear FME Enthusiasts, I'd like to draw attention to a current limitation within FME regarding job statistics on Flow Streams. Unlike other FME processes, the Flow Stream lacks the capability to provide detailed job statistics due to its reliance solely on translation completion. While other FME processes exhibit CPU usage metrics, the Stream does not offer this visibility. Introducing job statistics for Flow Streams would immensely benefit the community. With access to CPU usage data specific to Stream processes, users would be empowered to optimize their workspaces, potentially reducing CPU load. Moreover, this enhancement would enable better monitoring of resource utilization. In our company, we've observed that a running Stream consumes approximately 4-5 seconds per hour. However, despite this resource consumption, monitoring tools such as the database and job history reflect zero usage, even after the Stream has terminated. Effective management of CPU resources is critical, especially for Streams, to avoid excessive CPU hours on Dynamic Engines. Providing insights into CPU usage for Streams would aid in this endeavor.I encourage the consideration of this enhancement to improve the monitoring capabilities within FME, benefiting users across various workflows. Reference: FMEFLOW-21844. Examples of jobs completed and CPU usage
Parameters are great at controlling the proces of a workspace, and is adequate for the most part. However, they're mostly static.Sometimes it would be nice to be able to first gather information from e.g. a database, and then prompt the user to select items in this information before finalizing the workflow.This could be accomplished by an inline dialog box, presenting the user with options when all features have been received by the transformer.Alternatively, if a parameter could be dynamically populated from a database view, this might accomplish the same thing.Case in question: I have a transfer workspace, that selects data from one database based on an owner ID, and writes it to another database into a specific schema. Currently I have a frontend workspace, where I manually match the two sides into a single parameter, and starts the transfer workspace with the appropriate parameters as defined by the user selection. It would be nice to read that list from a database, not hardcoding it in a workspace.Cheers.
When resolving fields it would be useful in certain instances to have the option to NOT have the _resolved field set to <null> when the original value can’t be found in the domain. In these instances it would be useful to have the original value put in that field:For range domain fields: There is no code-description look up so all values are currently set to <null> in the _resolved field. All data from the original field is lost. For coded domain fields: When the coded value isn’t found in the domain the value is set to <null> in the _resolved field but there are instances where it would be good to have the option to not lose the original value (maybe also include a warning in this instance as stated in this other idea).
As similar to the quick toolbar for readers and writers, add a button to disable or enable a transformer using the mouse. A draft idea is shown below.
I’m in the process of working on a workspace which interacts a lot with an Open API. The api has endpoints like “HOST/api/projects/list” or “HOST/api/{projectId}/apps/list”. As I have no idea what projects are available so I have to have an end user type in the project(s) they’re interested in as a text input and i can’t use the choice option. It works fine, however, for the user the experience is a little annoying because they have to copy/paste or type in the value which is just not ideal when in theory it could be coded in.Assuming that a web connection was already defined for the service it would be really nice to be able to have a user Parameter which is able to perform the query and populate the list for users to choose. Even if it needed to be scripted with python or something else that would be ok.When the end user clicks on the “Select Project(s)” button a query is run to list the projects and render them for the user to choose. Taking it a step further another parameter could be built/scripted based on the value of that result. For example another “Select App” could made a query dependant on the Project Value.It’s probably just a rewrite of other ideas out there but with what feels like Open APIs everywhere now it seems like something which could be worth doing
I think that FME Lizard should be given extra vacation hours this year as a reward for all of the overtime hours that were put in last night to get all of those posts migrated. /s
You cannot move files and folders between an FME Flow resource folder and a user-defined network resource [https://docs.safe.com/fme/html/FME-Flow/WebUI/Manage_Resource_Connections.htm]. Please add this to FME Flow functionality!
The user has interest in an OData Writer. The user explains that writing seems to be possible using the HTTP caller, but they are wondering if there are plans for an OData Writer since there isn’t one. They want to make sure they are building their processes correctly in FME.
I would like to set the Tab Color when writing to an Excel spreadsheet (without the need for a template file or Python script). Community member alc33 asked a similar question 4 years ago:
As stated here I think it would be great if there would be a way to import database connections on a Linux system using the command line interface
Not sure why this is not already implemented, but it is something that would be extremely helpful. I use view extents hundreds of times every day.
Credit to Joshua Abaygar for this idea.Could row numbers be added to the Tester transformer so that it would be easier for users to create Composite Expressions?
Some transformers, like the tester and the testfilters, gives you an simplified version of the expression you currently have in the "Composite Expression" by representing each clause in the Test Clauses table by its respective row number. This may help you in reviewing your tranformer, but this may also become hard to understand if you have a lot of clauses included because you have to manually count the rows in order to locate which clauses the numbers refer to.I suggest that we should be able to show the row numbers within the table itself.
Good afternoon,I just experienced a situation that caused my workflow to stop because the source had a value that changed the datatype of a row. For instance, the table was expecting a date, but because the date was not formatted properly the SQL Writer thought that I was trying to pass a string into a date data type.Another case were this could have been useful is when my input has orphan records from a related table. the SQL writer had an issue because the source had items that violated database relationships-- meaning there was a record that didnt exist in the parent table.These two scenarios were painful to diagnose and currently there is no work around for them with out having to build another tool from scratch.Thanks for the help
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.
