Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
Since the introduction of caching and the tools on bookmarks to "Run Just Contained" and "Between Selected" I find myself almost writing 'multiple workbenches in one' and only working with them interactively.Which has made me think it would be neat to have a way to actually Run them in a traditional FME sense.Original IdeaI've only thought about this a little bit, but maybe it's a mix of:a property that we can set on the bookmark to state its 'Run-able' a section in bookmarks where we can order run-able bookmarksShow a hierarchy of linked bookmarks that would run after one anotherInclude an option to set bookmarks to run between to handle forking in a data flow.A step that allows disabling bookmarks to prevent running un-wanted upstream sections when running betweenFrom the traditional Run prompt, add an option that allows you to set specific bookmark(s?) to run or 'All'Example"Bookmark One" runs some transformers to generate an AOI to a database"Bookmark Two" reads this AOI to download and generate more data saved locallylinked "Bookmark Processing" runs some initial processing that generates results and errors that require manual tweaks on the downloaded data"Bookmark Three" reads the downloaded in "Bookmark Two"Linked "bookmark Processing" runs again against the corrected data, generates results and errors ...Evolved ideaAs I've typed this out I'm thinking maybe rather than run-able bookmarks we create "named run-instances", which has a similar concept where bookmarks are ordered within each instance but de-couples the bookmarks themselves.Now you could create a separate tab view for each named instance which shows just that data flow and links between bookmarks that don't have feature-flows connecting them etc..'Named run-instances" is a much better idea and I've changed my draft title from "Run a workspace by bookmark"ps I've now ended up thinking about this more than just a little and I left this as a stream of conscious because I think there's value in understanding how people get to ideas (and I didn't want to rewrite everything!)
Adding support for vector images. For example: SVG file type (and output port based on vector)
The option to deactivate Desktop License with the licensing assistant through the internet when moving a license to a different machine. (For fix license)
I would really appreciate if you could use your own geocoding services with the FME GeoCoder. Here our service as an example : https://digitransit.fi/en/developers/services-and-apis/2-geocoding-api/
Then we can chain the trainer and classifier easily...
The Windows installer offers three file-location options when doing an Express installFMEServer componentsFMEHome for all data related stuffLocation for the PostgreSQL database filesThe Ubuntu 18.04 installer offers two:FME installation (both components and the FME Home, mixed as by default in the Windows installer)Location of the database files.Having the option to configure a separate FMEHome would be appreciated (without the need of tedious manual config-file editing). Also this feels a bit inconsistent across platforms.
When writing to Excel I would like an option to set a range (referencing the template file & Feature Type sheet) that will set default values and formatting for each new row FME will write.Use cases are:Sheets where you have to click a button to add 5 more rows of template/formattingSheets where the first 20 rows are formatted but we have more rows to writeHow it could look:In "Sheet Settings"property: "Row Template Range" values: A12:XX12Result:Any new row written by FME will first take the cell ranges A12:XX12 and paste that as the template before writing the feature values. It will copy in:formattingvalueslookupsetc...Possibly related tohttps://knowledge.safe.com/content/idea/23780/set-excel-template-sheet-in-feature-type-propertie.html
Many transformers have the option to save settings as My Defaults, it would be great to have that functionality on writers as well.For instance, on the jpeg2000 writer, we have standards in regards to the Compress By Percentage parameter which is not the default 75. Every single time we add the writer we need to remember to go and change it.
Add the keyboard shortcut Ctrl + Shift S in order to Save As... within FME Workbench.
In FME Desktop, when I press Ctrl-F to bring up Workspace Search, the cursor should automatically be inside of the search box. (Currently, I have to click inside of the search box.)Additionally, when the most recent search term appears, it should highlight the entire string so that I can either press Enter to redo the last search, or just start typing to enter a new search term.
***Note from Migration:*** Original Title was: JSONFragmenter/Flattener should route input feature to port if HTTP request fails You can set a URL to the File/URL parameter in the JSONFragmenter/Flattener to read a JSON document from a Web server. Currently FME stops the translation if HTTP request fails with the transformer, but I think it would be better that the input feature would be routed to the <Rejected> port so that the user could define a subsequent process if necessary.cf. the XMLFragmenter routes the input feature the <Rejected> port if HTTP request fails.
I think it'd be great if we could have an option to handle 'no-data' or write off values like '9999' for point clouds.
I think there is a PR for this somewhere but no Idea that I can find. The most common use case I see is datum transformation between ellipsoidal heights and gravity based heights. CsMapReprojector can do the job but not write the VCS metadata to ArcGIS formats.
I've been thinking for a while, it would be great to have a button in the mini-toolbar that comes above the feature types or transformers for opening datasets in the OS default application. For example when writing in Excel, you could open the written file directly from FME Workbench without having to open the containing folder and clicking on the written file.
The current version of FME (2019.0.0.0 (20190328 - Build 19238 - WIN64)) S-57 (ENC) reader only accepts data with file extension (*.000) for ENC base data files and (*.030) for Catalogue file. According to the International Hydrographic Organization (IHO), the current ENC standards had been upgraded to version 3.1 hence the file name of the most of the ENC catalogue files would be "CATALOG.031". It is suggested the FME S-57 reader shall also be upgraded to allow the change in the IHO standards. Furthermore, according to the IHO ENC standards, the data structure of the updating file (with the file extension *.001, *.002 etc) for ENC data is identical to the ENC base data. It would be a huge benefit to users to validating this S-57 ENC updating files if the FME S-57 reader allows files with extensions *.001, *.002 etc.
To speed up work, with cartographic base, you do have to draw areas on an existing cartography. Using the PDF annotations, they could be exported as well as the rest of the information. To this day you can not export the annotations created with the adobe. thank you
It would be very nice if the administrators could adjust the FME Server website so that possibly even a company logo can be placed there. We use, for example, 2 different FME Server instances. To distinguish these we have different logos placed on the page. It would therefore be very nice if there would be a wildcard on the website for such purposes.
I'd love a way to expose data type/encoding (32 bit integer, UTF-16LE, 64 bit real etc) as values. These encoded values already show up next to the Attribute Name in the Feature Information Pane. An example could be:OBJECTIDNameAgeAddress01Andy22.51234 main St, college station, tx02Joe43143 2nd st, college station txwould output as:OBJECTIDNameAgeAddress32 bit intString64 bit realstringThe reason I'm trying to do this is to help with describing sensitive tables with support groups, be it Esri or FME or other. I've been trying to make a custom transformer that'll do it but can't quite crack it. Thanks
It would be useful if the list created by Matcher would include an attribute that indicates whether the element (matched feature) is oriented in the same direction as the resulting feature, or if it's in the reverse orientation. In other words, include an attribute that indicates whether this element was matched because we selected "lenient geometry matching".
When Labeling closed polylines with "Label position" parameter set to left or right and a value for "Label offset" parameter, some label feature texts overlap the polyline or are offset more then others (visualy).I got a workaround for this by setting "Label offset" to 0 and using the Offseter transformer to Polar coordinates Mode, Distance by Conditional value: 0-90 half the label height, 90-270 the label height and 270-360 half the label height, and Polar angle set to @Value(_parallel_rotation)+90. Actually, the label height plus something to clear the linewidth.It would be nice to have this implemented in the transformer.
Would it be possible to add a refresh button to the FME server job submitter so that when a workspace is republished to FME cloud with additional parameters, you don't have to re-map every parameter in the workspace.Currently to update the transformer you must select a different workspace before going back to the one you wish to update and then re-map all of the parameters.
Right now the Microstation DGN readers doesn't support Mesh element types. When the reader encounter such a object the following warning is shown in the log: WARN |DESIGN READER: Found element of type '105', which is unrecognized or unsupported. Skipping elementWhen working with infrastructure projects Microstation is widely used and the mesh element type is as common as butter on a toast.A question about this topic was asked already in February 2014 here.(For further reference to Microstation element types, here's a good link)
In FME 2019.0 transformer presets were introduced. This is very helpful to reuse existing settings for transformers.It would also be very handy to have a possibility to have preset groups for multiple transformers. For example for certain areas, when you load the preset group it will choose the presets for all transformers that are in that group according to your settings for that specific area.
if I need to use the function of a transformer in a PythonCaller for example, please add a sample Python snippet after the help description of the transformer to quick utilize in the editor.
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.
