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.
This idea is for Desktop and Server versions of FME software to be able to be patched where bugs are encountered & with a simple patch process. Background ... After updating to FME Desktop & Server version 2019.1, a couple of bugs were noticed, with StatisticsCalculator, and Counter (with Excel). Due to Desktop Server versions needing to be in sync, it is not always easy or cost efficient to simply update the software, particularly on the Server, each time a bug is encountered. Workarounds have been used in relation to the above, though they are 'painful' and inefficient often needing to be fixed in later versions. Having a simpler, easier process to patch the software (the transformer) would be much preferred and have business benefits and cost efficiencies.
It's possible you forget you have a reader disabled and when you run your workspace, it's not part of the result. Would be nice if FME notifies you before you run your workspace that there are readers/transformers/writers disabled (and which ones).
Hello, here's something I've come across many times in FME. When a translation sends no records to the writer, no files will be written for that writer. This means that if I am updating something daily and one day no records are sent to the writer, the file will remain as the day before (outdated), and this could cause confusion to customers.It is the same issue expressed in this thread https://knowledge.safe.com/questions/3785/empty-mapinfo-tab-or-shape-file.htmlHowever, no satisfactory answer is provided.The idea would be to include an option (a checkbox) in the writer or in the workbench to "allow writing empty files if no records are sent to the writer".The thing is that some systems need a file to be there, even if it is empty. Otherwise, loading will fail. So things like, removing all the files from folder before running the workbench won't work for me because the system is requiring all the listed files.Probably this is something that many people has come across already, and there are many workarounds, but this option should at least be considered, as it would save a lot of time applying those workarounds.Hope that makes sense.
I love the emailer transformer- it's great, however, there is one limitation which is a little annoying which I think would be nice.Sometimes you want to attach an unknown amount of attachments to an email. At the moment you must zip them up (which is fine, but not always desired). If you want to work around this it's a little annoying an involves multiple emailers each with a different number of attachments defined .The solution would be to allow a list attribute. It's pretty minor really but would be a nice to have.
It would be nice to control a job's running expiry time by which Queue it's been assigned to. At the moment Queues can be configured to a specific Repository, Engine, and designated a Priority level. So for example:Fast Queue ( < 15 mins )30 Min Queue ( 15 mins - 30 mins )60 Min Queue ( 30 mins - 1 hr )Slow Queue (1 hr + )1) This would help when your limited to a small number of engines and have to manage a high volume of jobs.2) It eliminates processes that may hang up an engine or two and cause the FME Server Job Queued List to rapidly grow. This could be to due to external sources. i.e. Database Performance Issues, Memory Constraints, CPU Constraints, and other constraints from a data source or target source perspective...3) You wouldn't need to then set the running job expiry time parameter for each individual schedule or automation, which makes it easier for FME Admins and Authors. 4) It would be easier to designate the correct Queue for a schedule or automation. You wouldn't be mismanaging Queues and you'd be able to minimize the amount of time a job is waiting for an available engine on the Server. Related to: https://community.safe.com/s/idea/0874Q000000TkrlQAC/detail
***Note from Migration:*** Original Title was: Respecting field name length limitations and reserved keywords per feature writer type 1) Field name lengthAs the maximum length of a field name in various formats is known (databases, shapefile, etc), it would be very helpful if FME Writers validated this on attributes entered.https://desktop.arcgis.com/en/arcmap/latest/manage-data/gdbs-in-sql-server/enterprise-geodatabase-limits.htmhttps://desktop.arcgis.com/en/arcmap/latest/manage-data/gdbs-in-oracle/enterprise-geodatabase-limits.htmhttps://pro.arcgis.com/en/pro-app/help/data/geodatabases/manage-file-gdb/file-geodatabase-size-and-name-limits.htm2) DB KeywordsIn a similar fashion, there are reserved keywords which can't be used as field names in many formats - this could also be validated on writers.https://docs.microsoft.com/en-us/sql/t-sql/language-elements/reserved-keywords-transact-sql?view=sql-server-ver15Error messaging for this in the translation log could also be clearer.
The time has come. I ditched python 2.7 at my first client when upgrading server to 2019.1And now of course, some workspaces using python don't run anymore, but which among the approx. 100 workspaces do use python, and have the version set to 2.7??I went for the .fmw reader, reading all .fmw files in the repository-path. With the Transformers features, I did find all workspaces containing a PythonCaller transformer.But, for this task it would have been super helpfull If I could read all private parameters (looking for python scripted) and read all of the workspace properties (especially the python startup- / shutdown-scripts and the Python Compatibility setting).So... Add all workspace properties and the private parameters as features to the .fmw reader please, as a help for all future upgrades....Kind regards,Martin
I often find myself needing to do certain things on list members such as removing or extracting the first/last element, or extracting/removing the nth elementh, reversing the list, etc. It would be great to have a transformer that allows these types of operations so you can manipulate list data easily (spoken by someone that doesn't do Python for a living!). I have also found one of @takashi's custom transformers (ListSampler) that is sometimes useful, but doesn't have the options I use most in the Sampler such as "Last Nth Element", which would be a great workaround to drop the last element in a list by picking the unsampled output. Thanks!
FME Server Automations provide the ability to set most workspace parameters via Global Keys but date/time parameter values are not supported. There is no option to select a Global Key as a value for workspace date/time parameters and the parameter value must be set individually for each workspace in the automation.Please add support for date/time values as Global Keys in FME Server Automations.
Sieving is removing clumps of contiguous cells in a categorical raster that are less than a threshold tolerance (number of pixels). The replacement pixel value can be one of NoData, default value, majority filter, value of largest neighboring clump, depending on what algorithm is used.This is used to clean up noise in categorical rasters without contaminating the boundaries of 'good' pixels the way a simple majority filter would.GDAL already has an implementation:https://gdal.org/api/gdal_alg.html#_CPPv415GDALSieveFilter15GDALRasterBandH15GDALRasterBandH15GDALRasterBandHiiPPc16GDALProgressFuncPv as does R:https://www.rdocumentation.org/packages/bfastSpatial/versions/0.6.2/topics/areaSieve as well as various python packages:ex. https://rasterio.readthedocs.io/en/latest/api/rasterio.features.html#rasterio.features.sieve but it would be nice to have a native FME version.
Is it possible to get FME Automations more similar to FME Workbench? The GUI feels like a "completely different software" than when working within FME Workbench. For people that have used FME Workbench the FME Automations feels a bit "old and rusty" with respect to how to easily add and remove items, connect items etc. My wishlist:- Easier add new items (start writing on dashboard, same as FME Workbench)- Easier to connect items (drag and drop, click and click, same as FME Workbench)
I have a situation where a user needs to select the relevant web connection for authentication, but as there is no means to get the base url from that web connection, they still need to enter an additional URL that the web connection should be used against. This leads to cumbersome double entry and is therefore prone to mistake, when the relevant information is currently held, but inaccessible to the workspace.I believe it would be beneficial to have the URLs that is held in a Web connection accessible in the workspace. For Aoauth2 this would be the URLs stored in the Web services, while thing like FME Server wuld be the Server URL stored in the connection. I would suspect that this could be an additional transformer that could extract details from a selected webconnection and extract them out into attributes within the workspace.
The new concept of Packages is superb... but, for organisations who want to have a little more control over what is downloaded from the web, especially in this time of a need to comply with things like Cyber Essentials, allowing everyone to be able to download and locally install a package doesn't fly for lots of larger companies. Can packages be downloaded once and stored at the end of a 'shared' path, like the concept with custom transformers and database connections please. Then central IT can be given the option to control the packages in use and importantly the version deployed to end users. Thanks!
I usually use Sublime or another tool to examine the logfiles, make this easier after a run by adding a button visualized below.
At a customer site I have the need to "calculate" the name of a web connection based on a number of environment settings.So, I use an AttributeCreator to create the _FMEServerName attribute.But then the FMEServerJobsubmitter and FMEServerLogFileRetriever do not accept an attribute to define the named connection.I would like this to be changed, so the name can be taken from an attribute (not just from a published parameter or a hard-coded value).
You can add and download the FME Hub custom transformer by searching for keywords, as well as adding a general transformer, Once you have downloaded the custom transformer, FME will always use the downloaded local transformer. Even when the custom FME Hub transformer is upgraded, the user will not be notified of the upgrade.I suggest that when you add a transformer that is downloaded from the FME Hub, If the FME Hub has an updated version, the upgrade prompt is displayed in the transformer brief.
Along the lines of this idea and this idea, but I would really love FME to be able to support PBR materials as appearances.With modern 3D formats like GLTF (and 3DTiles) gaining popularity and wider use, it would be really nice to have some way to support PBR material types. I wouldn't expect the Data Inspector to render based on these, although that would be very cool. It seems like it is pretty difficult to interchange between the two types of materials so being able to support PBR materials would allow for better preservation of the original dataset.At the moment, there is no real way to properly preserve a file which carries a PBR material, I must convert these back into a standard appearance, losing information.
Writing in Excel files, Like the HYPERLINK (see https://knowledge.safe.com/comments/35259/view.html )It would be nice to define a commentary over a value fieldE.g. value -> measurement / value.comment -> remark about measurement.It would avoid too large sections between the lines for big comments when they are shown as a column.
Try and make resource browsing experience closer to what you would expect from a file browser. - Altered UI/UX to a more familiar file navigation experience for users (similar to file navigation on the computer)- File tree to allow quickly dive deep into folders- Bookmarking folders for quick access from the file tree- Quickly create new folders in directories without the need of filling several fields- Right-click to quickly copy and paste files/folders
Create a transformer that can be used to simplify complex meshes, so it would be nice if we had a transformer that could be used to create smooth meshes from simpler ones.
Being able to run Windows containers with FME Engines on them in a Kubernetes cluster would enable users to read/write to Windows only formats in a K8S cluster.
Please add 3D Oriented Bounding Cube to BoundingBoxReplacer
I'm liking the new stats calculator setup in 2020, however i think the fact that you cant rename the output attribute inside it like the previous version is a bit limiting, meaning you have to add a subsequent transformer to handle that.
The ability to be able to send and receive data streams to an ArcGIS GeoEvent Server (like what we can currently do with the KafkaConnector)
Hi, at the moment FME support the OGP P1/90 format, and this has now been replaced with the OGP P1/11 Geophysical position data exchange format. Suggest that FME implement this new format as well"Originally released in 2012 to replace the UKOOA P1/90 and P2/94 formats, Version 1.1, released in August 2018, incorporates industry feedback from the first release and provides greater flexibility – without changing the fundamental format structure. For example, the format now enables the recording of land and OBC formats and additional peripheral sensor, or QC data. It also includes enhancements such as: handling channel grouping, ‘virtual’ position objects such as mid-points, global pre-plot attributes, event-based attributes, receiver and source relation records."URL: https://www.iogp.org/bookstore/product/ogp-p111-geophysical-position-data-exchange-format-user-guide/
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.
