Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
When working with a linked custom transformer, I often have to temporarily embed it to properly debug it. Then I have to export it to make it linked again. However, when I do that, any version history on the transformer is lost, as the versions start over at 1. This will, of course, break any other workspaces using the linked transformer that have not yet ready to upgrade to the new version.When exporting, there needs to be an option to add a new version to the .fmx, rather than merely overwriting it.
It looks like the AttributeManager doesn't offer the ability to Expose attributes that exist on a feature as yet and then do something to them, like rename. This would extend its capability.Merged Ideas:Add "exposer" to AttributeManager by bakkermans on May 1, 2016Is it possible to add the function of the AttributeExposer to the AttributeManager?
Hello,when working with linked custom transformers I often have to temporarily embed the transformer to properly debug a workspace. However, there is no good way to revert to the originally linked transformer. Once the transformer has been embedded I either have to remove all versions of the transformer to insert a linked transformer again (which means I have to repopulate all parameter values and properly connect the transformer) or I can overwrite the existing custom transformer which replaces the transformer via the "Export as Custom Transformer" option.I think an option to transform an embedded transformer to a linked one would be nice. Does anyone have another good way to develop/debug custom transformers that are integrated in multiple workspaces?
Just googled this and I'm sure there's a way somewhere, but if you want to read just the attributes and not the geometry for features, e.g. to query/review just the attributes, a 'table only' option could be useful to reduce read times where you have a large number of features. Admittedly I'm not certain how much overhead geometry actually puts onto the reader!Seems like this might have occurred to someone earlier, so if it's old hat then that's all good!
Would be great to do away with the type (metric & repository) and combine them so then you can select both repositories AND/OR metrics as rules.
As a user of FME HubI would like to create new versions of Custom Formats and Transformers whilst keeping older versions availableso that my desire to upgrade doesn't block users with older FME versions from using my work.I have a Custom Format that I maintain in FME 2016. I'd like to make a nicer and shinier version in FME 2020. Doing this would mean anyone on a pre-2020 version of FME can no longer use the version on FME Hub. Ideally, I'd upload a new version, but keep the old one available.I don't know how FME communicates and downloads from the Hub, but if that communication involves the FME version, the Hub could return the latest supported version of a Custom item.Then on the Hub you could upload a number of different versions, or mark in the history which versions are to be sent out.Of course, I'd love to see everyone on the latest and greatest version of FMe, but in practice that just isn't the case. I could also create two listings on the FME Hub, one for the old and one for the new version, but that seems like an ugly solution.
On the item page of a hosted service you can export to fgdb and other formats.The benefit of this over an fme reader and writer workflow is the easy preservation of unique ids, relationships and attachments. This also takes no local compute power/time. Please make a way perform to the "Export To FGDB" (or other formats) found on the hosted feature layer item page within FME.There is already a method to download items from arcgis online with the use of the ArcGISOnlineConnector (https://docs.safe.com/fme/2020.1/html/FME_Desktop_Documentation/FME_Transformers/Transformers/arcgisonlineconnector.htm)If we could create the fgdb through fme, we could also download and manipulate it within a workspace. I'm doing some lengthy transformations on a million+ records. Performing these actions through the Online Readers/Writers is taking hours (90% is reading from the REST API) and I need to reduce that time.I also see this workflow being a simple daily backup tool for ArcGIS online. Thanks!
It would be nice if we could choose WebP as a Preferred Texture Format in the CityGML Writer.
It would be very useful if the feature types were written in the order in which they are arranged by the user. In this way, writing to databases with foreign key restrictions would be simplified.
a transformer to generate heatmaps as raster based on density and/or metric-based
Add the ability to expose paper space sheet names via the Schema Port within the FeatureReader when reading AutoCAD DWG files. Currently, a solution suggested by @jedi121 (https://community.safe.com/s/question/0D54Q000080hBiUSAU/how-to-read-all-paper-spaces-from-acad-file) works well for the DWG Reader but does not expose all sheet names within the FeatureReader transformer.Or can you add an option to read all DWG sheets in the AutoCAD DWG reader parameters vs. manually selecting them? Maybe add the option here: I am attempting to read all sheet names for a DWG file using parameters within a FeatureReader.
The Clipper (in FME 2019) seems to reject vertically oriented area features. It should clip them properly like it does with with other areas.Our use case is generating 3D Block models with Buildings and Terrain, we have a nice workflow working, however, the vertical wall features are getting rejected which results in holes in the model.Of course this feature would also help: https://knowledge.safe.com/content/idea/19471/clipping-operation-for-surface.html or this: https://knowledge.safe.com/content/idea/86914/meshclipper-or-3dclipper-or-anygeometryclipper.html With 3D picking up more it would be really great to see better behaviour with Clipping 3D features. #3DisTheNew2D
I use TestFilter as a means to allow only certain feature classes to be downloaded from an ArcGIS enterprise geodatabase. If I didn't have TestFilter populated with the specified feature classes, someone could conceivably guess the name of a non-public feature class and download it. I have well over a hundred feature classes listed in TestFilter. When I need to find one of the feature classes, or have added one or more feature classes, things get messy real fast when trying to locate a specific feature class. If sorting and/or searching (including wildcard) was enabled on TestFilter, my task would be much easier. Thanks!
If you want to run a batch file asynchronously it requires you to add a wrapper to execute the command with a min command. (Eg. START /MIN CMD.EXE /C C: empmybat.bat). It would be nice to have a parameter to enable an asynchronous silent mode.
The ArcGIS Earth Automation API is a simple way to visualize 3D data dynamically.https://doc.arcgis.com/en/arcgis-earth/automation-api/use-api.htmIt is still in active development, now is a good time to provide input on functionality you want. I'm picturing supporting 3D point input with camera, flight and graphics modes so that Earth users can 'subscribe' to dynamic content coming from FME.
It would be great to be able to give end-users more information about a user parameter. E.g. through an info button next to the parameter, or - like in the App Gallery - the possibility to hover over the parameter to show more information.I.e. something like this would be great!Info button:Hover over:
The SchemaMapper's filter is curently supporting '='It would be great to have operators like for the test filter: =; >; <; Begins with; Contains Regex...
Introduce a reader for Mine Sigth (.msr) files
Would be useful if the reader could support selecting multiple feature layers at once and creating associated feature types in the workspace rather than adding individual readers - current task is to archive several layers' data at the same time so would be good for that.
Present an option in customer transformers to use the existing icons that describe transformer functionality. This would provide more immediate readability for the intended function of a custom transformer.For example, a FeatureReader has the database read icon attached to the transformer:If presented the option, I would use this same symbol on a custom transformer I have that is dedicated to reading files with a loop for re-attempts at reading (for better error handling):I believe it would read a bit better on the canvas (pun intended) for users.
***Note from Migration:*** Original Title was: ChangeDetector - pass attribute from original features to updated features when not one of the selected attributes @ebygomm asked this question. I ran into the same issue, so it might be helpful to other people as well. With this idea, it might be caught by Safe's radar.In the change detector is there any way of transferring an attribute value from the original feature to the updated feature when it's not either a Key Attribute or one of the attributes listed in the matching strategye.g. Like @mark2atsafe mentions in his answer, it is very weird. You would expect a Merge Attributes section, like on other transformers. Current workaround: follow up with a FeatureMerger/Joiner transformer to copy info across.
Is it possible to have some sort of estimate feature cache file size before workspace ran? We ran into the issue where the dataset was so large (let say there's no way of chopping the dataset to make it smaller), FME ran into memory issue and it filled up the storage space from the temp Cache file. If we can see the rough estimate cache file size before we ran the workspace with cache on we could determine if we have enough disk space or if we have to change where temp file store.Thank you!
One of the things I love about FeatureReader is the ability to enable caching on first execution, so subsequent executions pull from locally cached copy of the data. I often wish I could do the same with SQLCreator, when I need to pull the results of specific ad-hoc queries (often with complex table joins). I suppose I could create a view in the database just for these ad-hoc queries, but that seems heavy-handed.
I would like to have the option to choose Azure file share as a resource from FME Cloud
It would be nice to be able to automate the Tile Layer update in ArcGIS Online, preferably without using credits. Something that could replace manual steps: export tile cache to tpk, upload tpk, import tiles from uploaded tpk to existing tile layer.
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.
