Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
So you need not use a cloner-transformer or search in the logfile to get the exaxt number of Features entering a transformer in the workbench
The GeometryReplacer should have a parameter to optionally set the coordinate system (preferably allowing attributes) instead of requiring a CoordinateSystemSetter
***Note from Migration:*** Original Title was: Create a breakpoint that saves all of the data up until that point, so when you run the workbench again it will continue from that point instead of starting over from the beginning.
When using the Quick Translator for running multiple workspaces it is hard so see wich workspace is open.I think it would help if a dialogs appears that shows the workspace name and/or workspace description.
When working with workspaces with large numbers of embedded custom transformers it can become difficult to navigate between them. The tabs at the top of the screen become too small to read the transformer names and there is no easy way to go back and forth between views. A web browser style "back" button would be very handy, for example, if navigating in to edit a custom transformer that is embedded in another transformer, then needing to go back to where you came from.
A transformer to create word clouds based on attribute names and/or values would be pretty coolA FOSS Python library exists here
Hi,MS/SQL is quite picky about geometry validation, and as such, a prior pass by a GeometryValidator seems to be a sensible choice at all times. Especially if one has 20+ feature connections in a "generated" workspace.How about making this an easy option on the writer itself, turning validation on/off ?This might even be a beneficial feature on all spatial database writers, maybe on all spatial writers ?CheersLars
When creating a custom transformer, and editing it in the canvas, it would be nice to right-click on the output port to add an output object (like you would to add an inspector/logger) . Currently the only way to do so is Transformers>Insert Transformer Output, or right click on the canvas.Maybe a hot key as well?
The RasterBandInterpretationCoercer allows only for hardcoded choice or user parameter for all of its parameters like Destination Interpretation Type, and Conversion Options.It would be nice if these value could be set by an attribute on the Raster. The Conversion options could also be simplified to Convert from Color, Convert from Numeric, Convert from Float to Integer, as the "to" interpretation is redundant when supplying the Destination Interpretation Type.Any features with invalid combinations (like palettes restricted to UINT) could be output the Rejected port.
https://www.actian.com/data-management/ingres-sql-rdbms/ The fastest Hybrid RDBMS on the market with Geospatial capabilities. Previous versions of Actian Ingres have worked with FME so it would be great to see this latest version certified especially since they have improved Geospatial performance.
I recognized that the XMLXQueryUpdater can be used to update not only an XML document but also the schema of the input feature.Example from Extract attributes from XML attribute:Assuming that the source XML is:<SHD> <feature datadefinitie="Sleutels" code="SLEUTELS"> <attribuut naam="gisobjectnummeer">105963</attribuut> </feature> <feature datadefinitie="IMG" code="SHDV"> <attribuut naam="type">muur</attribuut> <attribuut naam="IMGEO status">bestaand</attribuut> <attribuut naam="In onderzoek">Nee</attribuut> <attribuut naam="relatieveHoogteligging">0</attribuut> <attribuut naam="identificatie">P0027.cdffbb11ae7c4f56a96a</attribuut> <attribuut naam="tijadstipRegistratie">20170626172844000</attribuut> </feature> <feature datadefinitie="GBK_LKI" code="C"></feature> </SHD> An XMLXQueryUpdater with this small XQuery expression extracts pairs of attribute name and value from the source XML and adds the attributes to the input feature.for $a in //attribuut return fme:set-attribute($a/@naam, $a/text()) I think the XMLXQueryUpdater (XQuery expressions) would be a powerful and elegant solution in some scenarios such as the example above. However, depending on the situation, you have to add some other transformers e.g. AttributeExposer, AttributeRemover to clean up the feature.It might be nice if these improvements would be done on the XMLXQueryUpdater.Add some new XQuery functions to process feature attributes. e.g. "fme:remove-attribute" and "fme:remove-list-attribute".Add "Attributes to Expose", "Attributes to Hide", and "Lists to Hide" parameters.Make the "Result Attribute" parameter optional.Rename the output ports (QueryResults -> Output, Invalid -> <Rejected>).
In order to be able to easily share both plug-ins and custom transformers (both using the .fmx extension), it would be great to introduce a new file type, that basically is a packaged/zipped FMX with a bunch of auxiliary files added to it. This would be similar to the .fmwt file (Workspace Template), that has the ability to include source datasets with the workspace.Why?If you developed an FME plug-in using Python for example, you typically end up with a .fmx file and one or more .py files. This can't be shared on FME Hub unfortunately (as discussed here). Bundling it and adding support for it might solve that for those users who are not familiar with Git (if FME moves in that direction): the code file would simply be included in the package and when the user downloads the plug-in, it will all be installed in the FME transformer directory.An FMX package would allow us to include required resources, e.g. when the custom transformer requires some kind of lookup table (e.g. CSV).You could include images (GIF, JPG) that can be used for creating examples in the transformer description/help section. If you want to do that now, you'd have to host the image on a server (which means keeping it online "forever") and link to it in the transformer description/help.
I would like to see some sort of workbench version control on the server. We have staging and live servers and undergo testing of workbenches before making live. However, as the other systems largely use URLs the workbenches have to stick with the same names. I would love the ability to drag and drop versions of a workspace for testing that automatically keeps tally of the version date and release number as well as auto backup the last iteration of the workbench. Yes you can create separate folders but it can get complicated keeping a tally on each version of a workbench.
Add 'Global Search Envelope' that can be applied to all the readers.
Currently there is no way to get font name and symbology from DGN reader. The symbology and font is stored in the seed file. The DGN writer accepts a seed file. Please add seed file to the DGN reader as well.
Sometimes when I'm cleaning up my workbench to make it easier to read, I need to move transformers or writers near other existing transformer connectors. The green triangle auto-connect gets in the way and starts connecting transformers inadvertently. I would like to be able to hold down shift or something that would disable the green triangle.
Currently. The header is added with no option to skip it. Would like to have this option like the other XML transformers
***Note from Migration:*** Original Title was: To finetune the Questions form to avoid people form typing the questions in the title. Hello,I may be wrong but the text in the Title text/box is misleading the users; reaons why they type long questions in the title text Field/Box. Please see the following questions for more details:https://knowledge.safe.com/questions/51222/asking-question-in-the-title-isnt-the-right-way-bu.htmlCheers.Lyes
All the JSON transformers except the JSONFragmenter allow the input JSON to be an attribute, free text, or a parameter containing the json, usually called JSON Document.The JSONFragmenter requires an attribute containing the text or a parameter containing the name of that attribute, which is inconsistent with the behaviour of the other transformers
I am looking for a transformer, which will take a shapefile as input and then publish this shapefile on to ArcGIS Server or ArcGIS Online, this should be possible either in ArcGIS Desktop or ArcGIS Server, this is to remove the whole manual process involved in the publishing of such files.
Takes a long time to download and update FME desktop. For example Microstation Bentley has an app called Connection Client that keeps the user current. It would be nice to have something like that in Safe FME environment.
I would like to be able to change the default fill opacity in DI i.e. set it like a preset, or to be able to set it for more than one layer simultaneously.I spend lots of time changing the fill opacity in order to see through a layer or to find out if there is overlapping. Frankly said, I don't know when I really need 100% fill opacity ever.That said the simplest solution would be to change the default fill opacity permanently to 50%
One of the challenges that my office sees is that our GIS holds valuable asset attribution, which then needs to be entered into ETAP modeling for our engineers. These 2 systems work in completely silos from each other. There isn't a very easy way to get the data needed for the modeling databases from GIS other than to start typing it in. The ETAP model then follows a flow or order of detail of assets along a circuit. It's not spatially aware or to scale. These 'one line drawings' are referred to often by our electrical engineers and field crews. I'd love to be able to work between these 2 systems and seamlessly exchange data. https://etap.com/sectors Thank you for your consideration.
To be able to efficiently process raster I miss the option to read a raster not as a single object, but by "logical unit".For instance for these formats:MrSID / ECW: I want to "read by row" as these are row-based formats. So feature nr 1 will be all pixels on the first row of the raster (creating an image of 123923x1 pixels). Resulting in a number of features = the height of the raster.TIFF-files can be stored in 2 different ways. By tiles or by rows. For TIFF-files stored by rows the reading will be similar to MrSID/ ECW for TIFF-files stored by TILES the reading will be a bit different:o feature nr 1 will be all pixels in the first tile of the raster (creating an image of for instance 256x256 pixels). Resulting in a number of features = number of tiles in the total raster.This will significantly speed up processing of rasters and make FME able to read huge rasters without having them in RAM.
I often find I may have several FME Knowledge Center or other safe.com tabs open in my browser while I work on FME Server. Sometimes it takes me a while to find the right tab because all of these pages have the same icon. Would anyone else appreciate a distinct icon for FME Server?
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.
