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.
It would be very handy if FME (Desktop and Server) had a mechanism for storing key/value paris in an external database, similar to the database connections introduced in FME 2016.These key/value pairs could be used to store local workspace settings, similar to this post.The idea would be to define a list of key/value pairs on Desktop and then upload it as a resource to FME Server, where it would contain settings for each server instance (e.g. development, staging, production, etc).It should be very easy to retrieve these settings inside the workspace, e.g. as a type of parameter.This would enable workspaces to retrieve an external configuration and not being dependent on hardcoded settings or python scripting.
It would be great to have the option to right click "Duplicate on Writer" for Feature Types listed in the FeatureReader. It's a great option to have on a normal Reader.
As of right now the GoogleDriveConnector does not account for user rate limit set by the Google Drive API. As suggested by FME support and Google Drive API documentation would be to add batch request functionality to the GoogleDriveConnector. See here for batch requesting functionality available with the Google Drive API:https://developers.google.com/drive/v3/web/batch
This Coordinate System is China’s new-generation geodetic coordinate system, was officially put into operation. And more and more geo-softwares support it, such as ArcGIS.
It would be awesome to be able to add a new/custom coordinate system to FME from within FME Workbench, rather than having to manually add or modify a document within the file directory. Maybe there could be an "Add New" button within Tools>FME Options>Coordinate Systems, that would allow the user to define a new/custom coordinate system?
A tiny addition.My workspaces tends to quickly have many runtime parameters, and these are easiest maintained in the "Run with translation parameters" dialog. Unfortunately you either need to cancel, cancelling any changes, or press ok to save the changes _and_ run the workspace.My tiny suggestion is adding an "Apply" button to save any changed parameter values without running the workspace.Cheers
The Filecopy writer can be used to copy a folder structure and populate it with files. However sometimes the folder structure will contain empty folders that you want to copy because they will be polulated manually when you work on a project. This works on FME Workbench, but not when you move the workspace to FME Server.
I would like to suggest some enhancements for the WMS Reader:Allow the user to choose the exception type if the WMS propagates more than one via getCapabilitiesIf the WMS servers sends an exception, FME 2017.0 does currently show:<WMS> Exception detected for the GetMap operation, requested Content-Type, 'image/png', returned Content-Type, 'application/vnd.ogc.se_xml; charset=utf-8;charset=utf-8'. It would be nice if FME could parse the exception xml and show the contained error message.Bonus: Even handle the error xml if the server sends it with the wrong content type ;).Allow choosing of WMS version 1.1.1/1.3.0
@XiaomengAtSafe this one is for you.I made a throwaway comment that there should be an Entomologist Badge (aka Bug Hunter) for those who find and report a given number of bugs to Safe support, and got asked to post it as an idea.
With the addition of being able to set execution order on Creator transformers and Connections a great method of workspace control has been added. This functionality is a little bit obscure, and that's fine, you don't always need to change it. But the information about that order seems more important and should be visible to the user.I would suggest a small numbered icon that from now on would mean 'Order'. This could be shown on the corner of a Creator transformer or on the first half of a Connection to inform the user on which parts of the workspace would execute in which order. This should only be shown when there are actually multiple Creators/Connections.
When performing a merge there are often records which one would like to exclude - e.g. where a field used has a few entries with no/a dummy value.It would be a nice addition to be able to integrate optional testing conditions as a pre-requisite to merging data (without having to employ additional transformers).
Currently FME Desktop asks for elevated permissions several times during the installation. Could it ask for it only once?
In many occasions I use a FeatureMerger to verify the match of ID between two datasets.In this situation I really don't want to merge attributes at all. I use the Feature Merge Type "geometry" but then I need to restor the original geometry of the requestor (GeometryExtractor / Replacer).My life would be much easier with an option "None" which checks the match between two dataset but leave untouched both requestor and suppliers so I don't have to worry about attribute accumulation.
Hi,The AttributeManager works well for most purposes, but I've run into some problem (2016.1) when using it to handle cases of 50-100 attributes. In these cases the dialog box is essentially buggy, and one quickly loses ones overview, rendering it unstable. I have a few bug issues and a few suggestions for improvement.I've experienced the following bugs:- The whole transformer is very, very slow, taking many (e.g. 5) seconds to display itself. This also - When I'm using the "Filter" to limit the list, at any time the insertion cursor may jump from the filter entry textbox into the "Action" dropdown for any of the filtered attributes. A longer wait, or a mandatory press on Return, would be better.- Often when I'm clicking on the Action dropdown on an attribute in a filtered list, another attribute pops into the top, shifting the whole list, so I essentially end up changing a setting on a wrong attribute.- Often, when I'm updating e.g. "Remove" to "Rename", the display will not update, so I have to save and reload it.Here are some suggestions for improvement:- Color-code the lines wrt. the action chosen. This will enable a much better overview of the status of each attribute. E.g., I often mistake "Rename" for "Remove" when glancing thru the list.- Enable temporary sorting of the list using any of the 4 columns (Input Attribute, Output Attribute, Attribute Value, and Action). Again this will enhance the overview immensely.- Filter on Action may also be of great use.Hope this'll be of use, and hopefully my suggestion may be implemented soon :-)Cheers
Using the AttributeManager some users have reported that it is hard to tell the different Actions apart and colour coding or icons might help. If you agree please up-vote or if you have a better idea please comment.
The SQL transformers allow you to hardcode your SQL in the interface or select it from an attribute constructed earlier in the process. It would also be nice to be able to run SQL "from file" where the code is stored externally in a *.sql file. This could be a little like the flexibility offered in the FTPCaller "upload from a file" or "upload from an attribute".
***Note from Migration:*** Original Title was: Format support for HL7 - Personal health records, outbreak management, vaccinations There is broad consensus on exchange of health information for pharmacy, hospital records, personal health information, vaccination program delivery, and outbreak management through the Health Level 7 body of web services and data exchange formats listed at HL7.org. We are seeing more integrations of front-end CRM systems with back-end government systems supporting HL7 standards. And leveraging data stored in back-end systems for epidemiological analysis in GIS environments such as EpiInfo or statistical packages such as SPSS (or Tableau) is becoming more necessary for public health professionals.
It would be great to have a date/time rounder transformer that can round time to example nearest even hour, day, 5 minute etc. (inspired by this Q&A : https://knowledge.safe.com/questions/30353/rounding-time-to-5-minute-intervals.html )
For customers that run 'Production' and 'Disaster Recovery' environments the 3-server redundant failover licensing model is not appropriate. In this case the 3 servers must be able to see each other, so having them span the Prod and DR environments can't be done as typically these environments can't see each other. What would be great is if FME Desktop could allow you in the license assistant to point at a primary and a secondary (failover) license server so that it could failover between them automatically if the primary production license server goes down. This would mean that one license server could sit in the Prod environment and the other in DR. This is clearly supported by FlexLM as ArcFM and ESRI (desktop applications) licence servers are FlexLM based and function in exactly this manner. There are some basic details here: http://support.esri.com/technical-article/000010729 but this has been inherent functionality within the ESRI licence server for some time. I appreciate that FME Desktop is moving away from FlexLM but I'm informed that it may be far enough away to justify looking into the viability of this approach. I'd be interested to hear the thoughts of the community and how you manage FME Desktop license failover aside from the above. Thanks, Dave
I would like the AttributeManager this to work with multiple actions. For example, I would like to rename a field as well as Set Value in a single attribute action rather then doing it in two steps.
My client uses SQLExecutor (FME 2016) to call a procedure inside SQL Server.This complex procedure sends back 10 tables : A Table, B Table, C Table, ... J Table.But some tables could have no data.SQLExecutor outputs only tables (as entities) with data (ok normal behaviour, FME dont create empty entities).But we dont have an index (the cursor ?) to know which table is outputting.By example :- The procedure sends : A Table with data, B Table wih data, C Table to I Table with no data and J Table with data.- SQLExecutor outputs 3 entitites : A Table, B Table and J Table.-> But we don't know in FME that first entity maps A Table, second maps B table and third maps J Table.If SQLExecutor could generate a cursor attribute, it would be fine :Fist entity : _cursor = 0 (means A Table)Second entity : _cursor = 1 (means B Table)Third entity : _cursor = 9 (means J Table)
https://knowledge.safe.com/questions/38456/clip-las-at-feature-reader.html
Have the ability to connect to SharePoint 2010 and have more integrated functionality with SharePoint Lists and Data
It would be helpful if FME could handle changes to file destinations on the fly. This way if a destination drive starts filling up faster than expected a user could re-path the rest of the files to a new resource. This obviously would not work for single file outputs but would be useful for multiple file outputs.
Right now the Chopper doesn't keep track of which remnant was where in the original feature. E.g. if I use it to split a line into individual segments (max vertices 2) or points (max vertices 1) there's no way of telling what the original order was, other than that they're being output in that order. Fine if you manage to maintain that order, but that's not always the case.So I would like to argue for a _remnant_id attribute that keeps track of that.
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.
