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.
Create animation or dynamic image in data inspector window within specified area at break points when running workspace to visualize the transformation process and save as a video clip or GIF.Could be nice presentation material.
Select output ports that you would like to have the link from, click the input port of a transformer or a Writer, and Workbench would create links.Usually it's too slow to go to View > Feature Type Connections and select the sources and destination, thus I'm linking the ports one by one.It would be handy to have the functionality of linking multiple output ports to an input port straight in the UI, and not in an separate window (Feature Type Connections).
***Note from Migration:*** Original Title was: ChartGenerator to have the ability to change the orientation of the axis label text At the moment the axis label text is only horizontal. It would be great to have the ability to rotate the x-labels text to 90 degree, for example to account for axis with a large number values.
Add Option to Turn OFF Accumulation Mode in FeatureMerger i case I don't want any attribute from supplier.Problem occurs when there is just 1 element in the list of Requestor features and 2 elements in the list of Supplier features. Both lists have the same name.After FeatureMerger with Accumulation Mode: Use Requestor.The Second list element from Supplier will become the Second list element in Requestor Which is quite annoying.Final note: Switching Requestor and Supplier port can provide a workaround, but In my case it does not give me the result I want and causing more problem than I already have.My final solution is to always remove list from supplier before merging.
***Note from Migration:*** Original Title was: Enable Different 'WHERE ClAUSE' for Different Feature Types From the Same Reader(ArcSDE) Enable Different 'WHERE ClAUSE' for Different Feature Types From the Same Reader To Avoid Creating Multiple Readers.
I recently needed to resample some RGB rasters to a different size (using RowsColumns), causing the aspect ratio to change. FME then changes the raster cell spacing, so that it's no longer [1,1] but [0.8,0.65] for instance. Seems more like scaling than resampling to me, but okay... Now what if I want to output this raster to a format that only supports square cells? Depending on the size, FME resizes it again before writing, so I end up with the original raster or some weird compromise (e.g. [0.7,0.7]).This is why I propose a "Force Square Cells/Preserve Cell Size Ratio" option for the RasterResampler (when resampling using the RowsColumns option).By the way: I'm now using a SystemCaller with an IrfanView command line to achieve this. It works, but it feels a bit hacky.. ;)
I'm proposing a tool/wizard that could publish a whole bunch of workspaces and/or custom transformers to FME Server at once. This could be a separate tool, but it could also be integrated into FME Workbench. Preferably though, it would be a standalone application that also accepts a command line. This way, you can hook it up to a SVN/Git repository for instance, so whenever you commit, all the stuff you created also gets published to the desired repository on FME Server.It would be nice if this tool had a "recursive" option, so that it checks if custom transformers contain some other custom transformers and publish these "dependencies" as well. Right now, FME does warn me for missing transformers whenever I publish, but it doesn't check inside these transformers.
Logical and Performance Optimizing Concept: Similar to finding the shortest(Best) path1 Map out the logic of all the transformers. (map creation)2 Detect unnecessary logic steps. (shortest path calculation)3 Merge transformers without breaking the logic. (shortest path implementation)3.1 Merge multiple adjacent Testers into one Tester or TestFilter 3.2 Merge multiple adjacent AttributeManipulators into one AttributeManager3.3 Replace (Bufferer+Dissolver) with AreaAmalgamator or vice versa3.4 Replace PointOnLineOverlay with SpatialFilter or vice versa3.5 Replace AttributePivoter with Aggregator or vice versa3.6 ...
An additional option in the statistics calculator to count unique would be great.
If your Sql executor crashes, your workbench crashes. So is it possible to add a new output (Failed) for Sql Executor ? this new feature would be very appreciated...Or what is the best thing to catch every exception of Sql Executor ?Thank you
When setting up Notifications in FME Server, everywhere that it's necessary to enter email account credentials such as for SMTP and IMAP publications, provide a 'Send Test Email' or equivalent button to prove that you've added the account correctly before you go any further and spend ages troubleshooting only to find out you miskeyed a password. Also have an option to show password so that you can see what you've typed is correct before hiding the password again.
I have a script in FME Server that gets called by a URL with parameters for the WHERE clause. The querystring on the URL looks like this:some_example.fmw?Year=2010&Neighbourhood=Kensington-Cedar%20CottageThe WHERE clause in the database reader then looks like this:neighbourhood = '$(Neighbourhood)' AND year(event_date) = $(Year)As I understand it, there is no way to "parameterize" the WHERE clause to avoid an SQL injection attack. Since we are planning to expose our FME Server to the world, it would be great to have this ability in database readers.
Right now FME Server and FME Cloud allow the user to upload (or select from the resources) any file they want as a source dataset e.g. if I set up a workspace with a Shape reader but the user uploads a DXF file it's just going to fail with some horrible error messages. That's not very user friendly.What I'd like to see is an option to disallow uploading/selecting of unsupported file types, kind of what the Desktop version is doing at the moment. It can still be circumvented, but this would prevent a lot of the accidental cases.
Gravitysketch is a new 3D sketch tool where you can create 3D features. http://www.gravitysketch.comIt would be great to create 3D objects there and convert them into features for use in City Engine and Esri GIS formats.
***Note from Migration:*** Original Title was: Temporarily hide table view in data inspector when showing table in central area Instead of having the Table View window display "The table is being displayed in the main window's central area" when the main window is set to Table View, close it and regain the real estate and then reopen it when switching back to 2D or 3D view.
Good day FME'ers,One of the most useful features to come in FME 2016 is the improvements to the regex engine. With this improvement came a nice little quick help guide to writing regex statements. I would find it useful to have a similar quick help guide in the DateFormatter with the most frequently used codes.Thanks
Set Conflict resolution per attribute (e.g. FeatureMerger) in stead of supplier / requestor. Sometimes you only want a few to be overwritten and a few to be kept. This can be done now by using a AttributeRemover before the merging transformer, but it would be nice to manage these settings in the merge and filter transformers.
For Example,Editing the SQL in an SQLExecutor requiresopening the Parameters form (popup)opening the SQL Statement form (popup)at this point the workbench is now hidden behind 2 layers of windows which demand focusi.e. both of these forms have to be closed before getting back to the workbench, making jumping in and out of tools more fiddly. One example is having to close the SQL form to read any error messages generated in the Transaction LogI suggest the forms can instead be made tabs on the main window, allowing for a less cluttered interface experience and quick switching between the forms via the tabs.
I was recently playing with some notification topics for successful or failed jobs. So I created a Push subscription and didn't specify the subscriber_content. Guess what? FME pushed the JSON job result to the URL I specified. Great! Exactly what I need. No REST polling necessary!Then I tried the SQS subscriber without subscriber_content. In this case, nothing was sent, because FME did not replace the empty message body with a job result and SQS does not allow an empty message.This is what I mean with consistency: why would this behaviour be different for each subscriber?Furthermore, does anyone of you know of a documentation page where I can get a proper overview of all the job result variables I can use in a subscriber message template? For instance: as a workaround to the SQS problem I mentioned above, I am now sending an SQS message that only contains the job ID (so I can then get the job result through the REST API using that ID). To do this, I used the {id} variable in the message template, because I read that in some example somewhere. However, other variables (e.g. {status}) do not seem to work here.I would love to see a proper overview of all the variables we can use and for which subscribers we can use them.
***Note from Migration:*** Original Title was: FMEServerJobSubmitter/Waiter: Callback result or custom message from successful job Wouldn't it be nice if we could post something back to the calling workspace, e.g. in a _customMessage string attribute on the FMEServerJobSubmitter output, or by actually having the FMEServerJobSubmitter output the data from the "Data to Post" writer setting (set when you publish the job)?
Accessing and editing settings and code within parameters transformers is often fidely, often requiring: selection --> right click --> further selectionThis could be replaced with a standard 'edit' panal that always offers editable settings and is populated on selection of a transformer / parameter, immediately exposing relevant settings of whatever is in focus
When you bring in a dataset with multiple tables it would be nice to change the order the tables are displayed in to help clean up a workspace.
Currently the NumericRastrizer always gives a constant string "NumericRasterizer" to the resulting feature as "fme_basename", even if you set "fme_basename" to the "Group By" parameter. It would be better if the transformer could propagate "fme_basename" value of the input feature when the user set "fme_basename" to "Group By".In addition, the transformer doesn't recognize attribute names which are set to the "Group By" parameter if the names contain Japanese characters and "Input is Ordered by Group" is set to "Yes". This also should be improved.
Validate the attribute values against a variety of specifications (within ranges, of proper types, etc)
The File Copy Writer (already in FME) can be used to move or copy files.I use this writer quite a lot.It can be used to ZIP files too, if the destination is a ZIP file.I would like to be able to UNZIP files using this writer (source is zipped file, destination is directory to store unzipped 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.
