Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
Add 32 and 64-bit on the ICON of all exe-files. Makes it easy for me with all versions installed (32-64bit, 2013,2014,2015) to distinguish them :) https://fmecommunity-ideas.s3-us-west-2.amazonaws.com/4036.png
It would be fantastic if you had the option, when feeding in two polygons, to have a spatial relationship of requestor contains CENTER POINT of supplier. That way you could ensure a 1:1 relationship when doing a select.
Due to the annual drift of the continents, there is a growing shift between coordinate systems with the reference Systems ITRF2000 and ETRF89. The annual shift is about 2,5 cm each year which results in a gap to WGS84 in FME, as this coordinate system seems to have a reference to the epoche of 1989. This gap is now about 75 centimeters. Comlexe formulas will have to be implemented to solve this problem, an idividual grid-shift file will not result in the desired accuracy. The integration of this annual shift for every year is important for many clients, especially airports.
See Title.
Would be nice if the Geocoder transformer have the precision result as an output field With Here API or Bing API, we can get the precision of any geocoded addresses (Address, Street, Postal Code, Country, etc.)
Really just promoting this question to an "idea". External attributes which are exposed on a custom transformer input port get applied to all input ports in that custom transformer. It would be nice if the exposed attributes could be tied to specific input ports.As was noted in the original question, if different features are coming into different input ports, then they will not all necessarily contain all of the attributes being exposed and this can lead to confusion within the custom transformer as to what is really available on a feature.Similarly for output ports. If features exiting a specific output port only contain certain attributes, then invalid attributes should not be made available to the caller.
It would be nice to have the option of selecting all or caret insert when selecting/editing transformer names.As a side note when opening the "Workspace Search", the Find box should have the focus and it would be nice if the contents of the 'Find:' are all selected (similar to what I am complaining about above).
For our projects we have a series of tests we run on FME Server to check over data. The tests are separated by utility pipe network and live in their own repositories. At the end of the project, when all networks are completed, there is a final process to run which is uploaded to all 3 repositories - when updating this workspace it is obviously necessary to upload the workspace 3 times. Hence it would be good to have the ability to shortcut to one location of a workspace, so that it can be uploaded just once but run from any number of locations.
From the reader gallery you can create a custom format reader and use it, however, from the writer gallery it seems you can create a custom format writer, but that can not be used.
If I have a user that needs to interactively do a Road-Name search then road names containing diacritics will fail the search. The workaround is to have 2 road name attributes, one for searching, and one for publishing. For my Road-Name search attribute, I need to be able to strip out the diacritics. PostgreSQL offers a "unaccent" function to do this - can we have this same capability inside FME StringReplacer or similar? Refer https://www.postgresql.org/docs/9.1/static/unaccent.html
Hello, I request this functionality as a result of the question discussed here: https://knowledge.safe.com/questions/42506/readingwriting-microstation-properties.html The information seen via File->Properties in Microstation can currently not be read from / written to DGN-Files with FME. It would be nice to have FME attributes corresponding to the various text fields found therein. Regards
Create a new interface which allows users to graphically view the entire FME Server system architecture, all the key components and their interconnections. It could be a web based Workbench like interface, but would need nodes that represent individual services, notification subscriptions, topics, publications, file and web resources etc. This way it would be a lot easier to trace the dependencies through the services and notifications.
Working with large (GB) of JPEG files - reading and writing with FME takes a long time. There is an open source implementation (beta) that uses GPU to compress/decompress JPEG if you have an NVIDIA - possible to implement this in FME? Cutting edge! https://github.com/CESNET/GPUJPEG
It would be nice to have an option on the FeatureReader to allow for case insensitivite feature types and corresponding ports, particularly using folder based formats where the feature type comes from the filename. So that FileX.shp, and filex.shp are treated as the same feature type.
I would like to see the inspector far more integrated with the FME Workspace. Use the results inside the inspector to guide and insert transformer in order to clean data. There are lots of simple but tidies data cleansing tasks that could be discovered and cleaned while you analyse and view the data. For example removing/replacing strings. Split or concatenate strings, find/replace text, change projection, date formatting, aliasing, data type conversions, etc... The inspector could give hints while users simple select text in the result or selects attributes... The inserted transformers (suggested by the inspector) would basically be inserted in the work-space after the inspector-transformer that exposed the results.Such radical changes could allow FME to come a step closer to basic users and it gives advanced users a quick tool that would improve building work spaces much faster. Even cooler would be to re-run the last result so you instantly see the effect of the added transformers...
Since GeoJSON features can be represented completely with text, it would be great to read GeoJSON features directly from attributes and parameters. This would allow, for example, supplying database spatial envelope filtering or clip geometries as a parameter to the workspace.
Within my workspace I have the need to filter several datasets at one or the other point, utilising AttributeRangeFilter. However, the range used is always the same. To avoid typos I was looking to set the range as user parameter and link to the different AttributeRangeFilter transformer.While able to create user parameter with range configuration, I couldn't find any possibility to link user parameters to the 'Range Lookup Table' of the transformer.Hence, I would propose to extend the AttributeRangeTransformer to use user parameters.
***Note from Migration:*** Original Title was: List more descriptive less abbreviated coordinate system names; preferably matching ArcGIS when imported. ie. "NAD_1983_10TM_AEP_Forest" instead of "CANA83-10TM115", so as not to be confused with “10TM115-83”
I'd like some quick functionality for when I'd like to enter a list of values (no geospatial data involved) into something like a creator where each line would be interpreted as a different feature. I can do this currently by creating a single attribute then splitting by line and exploding but some sort of scratchpad functionality would be great. It's something I'd mainly want to use when testing/creating new workspaces.e.g.
I have what I think is a fairly simple use-case, but FME doesn't currently has the tools necessary to readily implement it. So this is a transformer suggestion for a ListAttributeMapper or something to that effect.Basically I have a list like this:_list{0}.class = 1_list{1}.class = 2_list{2}.class = 3_list{3}.class = 5_list{4}.class = 2_list{5}.class = 1I need to know how many of any given class there are with the feature. ListHistogrammer does this nicely:_histo{0}.value = 1_histo{0}.count = 2_histo{1}.value = 2_histo{1}.count = 2_histo{2}.value = 3_histo{2}.count = 1_histo{3}.value = 5_histo{3}.count = 1My problem then is turning those into arbitrarily named attributes without a needlessly complex set of testers, let alone a PythonCaller.I want my output feature to have attributes like these:first_class = 2other_thing = 2third_item = 1fourth_item = <missing>final = 1A kind of ListAttributeMapper would be one way to solve this. You'd point it at _histo{}.value, and for any given value in that field, tell it into what attribute name to to put the index equivalent value of _histo{}.countHopefully that's clear
It would be really useful if the TableAdder could have a group by function where multiple tables can be created for each group of data.
Most times we do not see an entire log content in a chronological order; seeing last 100 lines, or logs created in last one minute or so would suffice. This would also save a bit of bandwidth.
Having the option to specify a WHERE Clause would be invaluable to workflows that leverage the FeatureReader transformer. Limiting the amount of features read into FME Workbench would considerably reduce processing time.
Hi,When a Workbench workspace (*.fmw) is run, it's converted into a mapping file (*.fme) on-the-fly to be run by fme.exe.I'm using "fme" files for production (called via a BAT file), while the Workbench workspaces are for testing and debugging. This is my "best practice".But if I have one workspace that runs another, I cannot run the mapping file itself, only the Workbench workspace.I think it would be a huge enhancement, if the WorkspaceRunner could activate a mapping file (*.fme) directly, removing the need for the on-the-fly conversion. A win-win wrt. performance too.CheersLars I.
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.
