Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
***Note from Migration:*** Original Title was: Setting ArcGIS Pro 2.4 in the python interpreter is not available yet - fixed in 2019.2 release? Hi,We are finally switching to FME 64 bit and have new machines to test it on.We are going to install version 2019.2 and are wondering if ArcGIS Pro 2.4.2 is the best choice due to the notice on the esri compatability matrix that Setting ArcGIS Pro 2.4 in the python interpreter is not available yet?Was this issue fixed in the 2019.2 release?Or should we go with ArcGIS Pro version 2.3.3? Thanks,AnnetteStedin, Rotterdam
Kind of an addition to:https://knowledge.safe.com/content/idea/18817/pdf-page-formatter-parameters-from-attributes.html Would love to be able to see some extra options for the Page object location based on values in the attribute table. With the idea that you don't need input ports but can use a generic input port that filters the objects that need placing based on basename, feature type or some kind of group (i.e. map, legend, rasters). Adding a generic input port where you need to define an attributes for the upper left x, left y, height en width of the object. By doing so all object can be placed in the PDF page mathematically instead of drawing boxes by hand and always have check if the aspect ratio is in order.
Would be great with improved logging on the TINGenerator, right now it is not possible to know how long the operation takes. In percent or in time or in vertices. For instance a % of points used or % of vertices created (estimate) would be handy. Then you can easily spot neverending and huge TINGenerators.
FMECommunity - add all transformers as topics or add a new category called "Transformer" so you can find all topics related to a specific transformer.
Make it possible to upgrade existing Reader / Writer to a FeatureReader / Writer with the same parameters, featuretypes and attributes.
I'd like to be able to configure the text paramater to set a maximum length. The use case is quite simple. People have an option to make a remark with a certain workspace. The remark will be added to the dataset in a field. However the field has a maximum of characters allowed. I would like to set this number of characters, so I want have to make the workspace fail if people add to many characters in the Parameter.
***Note from Migration:*** Original Title was: Ability to move down and up feature types in Navigator to specify their order of reading Its possible to move up and down datasets in Navigator to specify their order of reading by FME. Why not the same ability for feature types ? Maybe because each reader has his own characteristics of reading ?
Hi,I quite often want to know how well my workbenches are running and check the status of the last few jobs. The sort of summary stats i pull out are, avg runtime, max runtime, avg feature count, max feature count, it'd be also nice to plot specific jobs on a graph for these types of stats.Historically i have always accessed this data through a database connection to the repository fme_job_history table (and it's predecessor), however i'd like to move to using REST if possible.To do that efficiently though i need a new end point which is something like:GET transformations<repository><workbench>jobsThis would return a list of all jobs running, queued, and finished etc, relating to that workbench.ideally i'd also like to be able to pass a query parameter to filter the response, maybe start_date=<ISOdatetime> and end_date=<ISOdatetime>Surely i can't be the only person doing this?
Would be very nice to send a workbench with a token using FMEJobSubmitterBye Bye the username and password !!
***Note from Migration:*** Original Title was: Integrate Desktop and Server more closely. Execute tasks on Server thru workbench gui.
XMLXqueryExploder and XMLXQueryUpdater are very powerful transformers when you use XML inside FME but they lack an easy starting point. It would be such an improvement if the "XQuery Expression window" had code samples in the left panel (similar to the XQuery Functions) providing the user with basic code snippets that would help them forward quickly.Such snippets for example could be- How do you count child-nodes - Replace content of a property- Replace content of a node based on a string in child-node- Find domains from an XML
I have around 60 scheduled FME Server jobs and I find it quite hard to keep track of what job starts at what day at what time."Start Time" and "End Time" are not very helpful since they only show a date maybe 2 years ago, when I first set up de task.Therefore it would be nice to be able to set up schedules with weekdays and time of day and show this on the schedules site as well. For example I set up three scheduled tasks:workbench 1 / Tuesday, Thursday / 20:00 / every weekworkbench 2 / Sunday / 13:00 / every second weekworkbench 3 / Mo-So / 22:00 / dailyThese informations should be visible (and preferably sortable) on the /schedules site as well.
FME ought to natively support ECW version 3. See also https://knowledge.safe.com/questions/24065/i-cant-open-a-version-3-ecw-in-fme-data-viewer-err.html@jlutherthomas @Mark2AtSafePR#54009
While the HDFS Connector and Hive Reader can be useful, an Hbase Reader and Writer are essential to our Hadoop workflow.
Adding authorization header token support to data streaming, job submitter, data download services. Possibly by aligning or incorporating them into the REST API via the transformations api. For example:POST /transformations/submit | transact | transactadata/<service>/< repository >/< workspace >Or maybe through a new Services Manager API?
Currently the tester allows you to use a comparison mode of Case Sensitive/Case Insensitive with greater than and less than operators. There may be good valid reasons why someone would want to consider 111 to be less than 20 whilst also considering 20 to be greater than 199 but I'd hazard a guess that these use cases would be far more rare than someone wanting to do a numerical comparison and having the incorrect comparison mode selected. Currently the In Range operator is only available when the comparison mode is set numeric or automatic, but greater than and less than are not restricted in the same way.Additionally, it means that if you send string characters to a greater than/less than test that is set to automatic the test is attempted when it might be preferable to produce an error.
In addition to @courtney_m's idea to replace WorkspaceRunners by FMEJobSubmitters (https://knowledge.safe.com/content/idea/33203/ability-to-replace-workspacerunners-with-fme-serve.html), I would like to suggest a hybrid Transformer handling both. It might be a technical hassle because of constraints of the two platforms, but if it existed it would save time both during development and maintenance of the Workspace.
It would helpful if the FME Server System Share could deployed to an Amazon S3 bucket. This would remove the use of a fileshare, increase user options during fault tolerant deployments as well as assist with automatic backups. For users already leveraging AWS EC2 machines as their server component host this also allows for seamless AWS deployments. If the ability to leverage Azure and Google Cloud equivalents to S3 could be implemented that would also be useful for those who want to architect cloud agnostic deployments while removing the dependency on a fileshare.
In the same vein as some tools like 'make', and the APT (aptitude / apt-get) ecosystem of Debian, I think it would be useful to have a dry-run mode for workbenches. It would have to be really simplistic, and would probably only be useful in simple workspaces (i.e., without workspacerunners, etc). I imagine in the most basic implementation a system that when activated, replaces any Writers with inspectors. The idea is to assist with gaining a more rapid understanding of how a workspace works. Many times I am presented with something made by someone else, and want to run it, but first have to spend a bit of time disabling parts of the workspace, or attaching a few inspectors, etc etc. Instead, with dry-run mode, I could safely hit F5, and be confident that no files or geodbs would be overwritten, hidden output files clobbered, and so on.
I have a hard time finding the ideas I proposed to see what happened to them.
Stanford's Temporal Tagger - SUTime - would provide the ability to recognize, extract, and normalize time expressions in textual data. It will convert next wednesday at 3pm to something like 2016-02-17T15:00 and would be an extension of the Natural Language Processing (NLP) functions that have recently been implemented in FME transforms.https://nlp.stanford.edu/software/sutime.shtml
The Bufferer transformer has the parameters Dissolve On Attributes, Dissolve On Attributes Mode and Generate Dissolve List. It would be nice if the GeographicBufferer had these parameters as well.
It would be nice if there were a parameter on the ListIndexer to remove the ListAttribute similar to the one on the GeometryReplacer.80% of the time I have an attributeRemover immediately after the ListIndexer.
If you have a large area for which you generate Aircom Enterprise Files and you make smaller subtiles to speed up calculations, a new export overwrites the existing files. At the moment, this can only be solved by exporting to different subfolders. It would be ideal if the Aircom Export could update the index file in a existing folder and add more entries corresponding to existing *bin-files.
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.
