Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
Setting the proxy info in FME Server currently involves using the command line, see https://docs.safe.com/fme/html/FME_Server_Documentation/Content/AdminGuide/Using_FME_Server_with_Proxy_Server.htmI suggest including these settings in the admin GUI to make them easier to maintain without needing console access to the server. These settings should override whatever proxy setting is defined in published workspaces (HTTPCaller, etc).The most important bit is to set the proxy server and enter the authentication details. Defining custom proxy maps is probably less important.
It would be neat to have a mobile app that allowed me to provide all the sensor data from my phone to a workspace run on FME Server or FME Cloud.I'd want to be able to use any FME Server, probably multiple and any workspace. In some cases I would just want to run the workspace, but in other cases I would want the output of the workspace to be provided back to the phone. For example: PDF, Powerpoint, or cool stuff like Augmented Reality or any new app that might come out and which uses data.I don't want a complicated app! Just let me run the tasks (workspaces) and see the result.
The current FME(2016.1.0.1) only allow the user to save one setting. But many times we need to switch between in many Settings, I hope I can save multiple setting, and can be set each name.
The ArcGIS Online reader and writer are quite slow to use. For every feature type, in every AGO reader and writer, FME currently loops through all accessible groups twice (once to find the feature service, once again to find the layer). When a user only has access to a few groups within his organization, it's merely an annoyance. When a user has access to a large number of groups, waiting for FME to find the correct layer can add minutes to the time it takes to run a workspace. There is a workaround where the HTTPCaller can be used to directly access a layer, much faster than the AGO readers/writers can, but this method has several other downsides. It would be great if the official AGO readers/writers could be improved to be a lot faster.
From @lefred06 at https://knowledge.safe.com/questions/96349/s57-customized-files-location.html Customized S-57 CSV files should be stored outside of the FME installation directory so they aren't lost when FME is upgraded or reinstalled.
We run 32 and 64 bit versions of FME Desktop along with Dev and Prod environments for FME Server. Some of the more complex workspaces we create involve a fair bit of manipulation of inputs, outputs and resources between these instances.I am currently in the process of writing a startup python script to identify which environment (desktop 32, desktop 64, Server Dev, Server Prod) a workspace is running on - this works. From there I can read a config.csv which has parameter names and a separate column for each environment - this also works.The next step would be to set appropriate user parameters, to circumvent the need for manual changes in the different environments, but this seems to be a sticking point as user parameters need to be predefined.Could provision be made for the dynamic creation and setting of user parameters at startup?
If you get a error on FME Desktop. there's some hints here: FME Server Troubleshooting: Resources, one thing that's not mentioned there that might also be affecting this is low disk space on the drive that holds your shared resources (which is also where the job logs end up), triggering aggressive cleanup tasks.
We have a c-tree database we are currently pulling data out of via daily file reports, but we are looking for a way to query it directly. FairCom's site has an ODBC driver download, but we don't have any developers in-house to integrate this sort of thing.
I have a workbench where I have multiple readers and each reader goes to their own DuplicateFilter. Each DuplicateFilter creates a DuplicateRemover table in the Temp folder. When the reader finishes, I wish the DuplicateRemover temp file used for that reader would be cleaned off. I've begun running into memory issues as the files I've been reading in have grown.
With the CoordinateSystemExtractor, only a description of the coordinatesystem is given. It would be great to have the EPSG code as well.For example:_coordsys : _Netherlands-RDNew-2008_0_epsg: 28992So I can use the 28992 code to extract more details from this coordinatesystem by building this URL, for example: https://epsg.io/28992.gml
***Note from Migration:*** Original Title was: Ability to turn off outputs on line/ point on area/ line overlayer to improve performance Currently on the line on area overlayer/ point on area overlayer you are required to process both outputs for the transformer but in most cases you only need one output. Could you have the ability to chose to turn off the processing for the output not required?
It would be great if column settings can be saved and re-applied when a workspace is run and data visualized in a table within FME Data Inspector. When building a workspace, I find myself constantly re-running to test and look at the output, but sometimes I want to focus on specific columns, yet have other columns accessible if needed without having to re-run the workspace. I would like an option where we can save various configurations, or at least have Inspector remember the column settings when a workspace is re-run and use the same column order and visibility instead.
It would be helpful if the COLLADA reader would support reading Geometry definitions stored in the DAE as definitions so we could use the Shared Item tools. Currently the COLLADA writer supports this - I can create a COLLADA file with a single model but unfortunately it seems when reading that these definitions are read as aggregated mesh objects. When writing the same file back out to COLLADA the size blows up because a new geometry definition is created. When working with a file with 1000 of such geometries it blows up from about 1 MB to over 250 MB...
Surface Weather and Marine Observation Markup Language (SWOB-ML) is an XML format based on the OGC Observations and Measurements standard.This format is used by Environment Canada to publish detailed and highly-frequent weather station information from both EC and third-party sources across Canada.Documentation for the format is available here:https://dd.weather.gc.ca/observations/doc/ And sample data by date and station (CVOO has good info) is available here:https://dd.weather.gc.ca/observations/swob-ml/
I love the glTF writer and we have been using it a lot in conjunction with our Cesium based applications to help bring customer data into their interactive 3DCityModels.glTF is awesome because it supports both vector (lines/points) and True 3D surfaces. One issue we're running into, however, is that I'm unable to style the vector features in the output. The fme_color attribute appears to be ignored and the GeometryColorSetter/AppearanceSetter rejects it because it is InvalidGeometry...This means we're unable to bring in colors from CAD drawings which makes it lose some of its valueI would love if the gltf Writer would recognise the fme_color attributes.
When pushing the refresh button in the Inspector, it refreshes all data sets. Sometimes, you want to refresh one data set only. The idea is to add 'Refresh Dataset' when you right-click on it. Similar where you find the possibility to 'Remove Dataset'.
In the Navigation tree, we can click on a Bookmark and the canvas is centered in its content.It would be also useful (at least for me) if in an Annotation we could create a link to a bookmark.In my case, I tend to create individual bookmarks for the SQL sentences used in Readers and Writers or for code samples of the PythonCaller or SystemCaller... The idea is adding an Annotation to the Reader/Writer/Transformer with a short sentence about what they do and a link to the bookmark where extended information can be found.
Could you make the icons for transformers that have more than one input have a visible difference to those with only one (a small number, different colour or input lines on the left).Often times I don't really know what to search for to achieve a goal, but I will know that I need two inputs not just one, so if I can visibly see the difference when searching for a transformer it will make narrowing it down much easier.
Currently on FME Cloud when you launch an instance you have no control about the virtual private cloud (VPC) it is launched into (we actually launch it into our own VPC). I am interested in soliciting feedback on whether adding support to launch into your own AWS VPC would be beneficial. We have already had quite a few users request this, as for a lot of customers their first step in moving to the cloud is the hybrid cloud. Just to propose some ideas. Allow you to launch into your own VPC in FME Cloud and then allow you to link to your VPC to ours via VPC Peering.We could allow you to launch directly into your VPC but this is likely to bring with it some limitations and additioanl configurations.
***Note from Migration:*** Original Title was: Possible solution for: "Geodatabase Error (-2147024893): The system cannot find the path specified." Not sure if this is the right space to post this but here goes:I just spent most of the morning trying to automate my data workflow using FME, .bat files and Windows Task Scheduler.The logic is:Task Scheduler runs a workbench .bat fileThe .bat file executes an FME workbenchDetailed Errors: 2019-07-09 11:52:00| 0.6| 0.0|ERROR |Geodatabase Error (-2147024893): The system cannot find the path specified.2019-07-09 11:52:00| 0.6| 0.0|ERROR |Could not open File Geodatabase dataset `geodb path' for readingMy troubleshooting:The translation is successful when running from within FMEThe translation is successful when running directly from the .bat fileThe translation is NOT successful when using Task Scheduler to run the .bat fileSolution(?)I changed the following setting in Task Scheduler:My translation is now successful using Task Scheduler.I'm writing this post to help others who are having similar issues.I would appreciate if anyone can explain the difference between the two settings... I must have misinterpreted what they actually do.
It would be great to be able to test an external database connection from within FME Cloud to ensure it is set up correctly before having to run a published job.
I would love the ability to push one update to FME server that would automatically sync with all my workbenches saving the time and hassle of opening each one up and making changes. The best way I can think of for doing this is to make the workbenches into a series of modules, which could be custom transformers or something new. If I open and change that module and then save up to the server it should copy this change to a number of workbenches on the server - or better still give me a choice. That way I can keep on top of tweaks and updates that I need to do and ensure that all the workbenches are working in the same way. For instance I could have a back ground mapping module, a PDF export module etc. All the workbenches would basically be an amlagation of several modules and same stand alone tranformers.
It seems that the FeatureColorSetter blocks features always even if you set Fixed to the Color Schema parameter.It would be better if the transformer won't block features when setting a fixed color.
I attended the Partner Summit 2019, which was fantastic! Only thing I was missing was specific, advanced training sessions to the partners. We should know new functionalities of FME very well, and of course we study by ourselves, but it would be a great opportunity to learn e.g. IFC writing from Safe experts or how to write IMDF and other indoor mapping formats. The sessions don't have to be long; just 1 to 2h would be enough. I suggest that in upcoming partner events, some of the tech sessions and roundtables could be changed to training sessions.
***Note from Migration:*** Original Title was: Option to give FME Data inspector to show results in web page as alternative to data download Sorry for the long title, but I think it would be really powerful to let users explore fme server result data online instead of downloading it. How cool would it be to have an interface like desktop's data inspector appear once the translation is complete?Thanks, Andy
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.
