Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
***Note from Migration:*** Original Title was: Also allow the operators 'Attribute has ...' for 'non attributes' (like e.g. parameters or direct fme feature functions) in test conditions It would be nice if the operator 'Attribute has ...' could also be used on 'non attributes'.It is quite tempting to use a testfilter with the condition 'input parameter', and operator 'Attribute has a value'. However, this won't work as expected as the input parameter is not an attribute. When your not aware of this it is also quite tricky to detect that a (test) statement like this doesn't work, because no error or warning is generated when you do use such a (invalid) statement.Also, it seems that I'm not the first one to have been confused that this operator is not possible for non attributes, see e.g. the following topic on the forum; linkOf course there are quite some easy workarounds to this;a) if you just create an attribute from the parameter (with e.g. a parameterfetcher) you can use the teststatement then on this newly created attribute. (Although this is the simple and obvious fix, I do feel it is a bit cumborsome as I think it shouldn't be neccessary to go into such a workaround. Also creating an attribute from your parameter will actually create an extra 'column' to your dataset, which might be quite costly on resources when you are dealing with large 'feature sets')b) I think in many cases you can also get away by using the operator 'Contains Regex'.E.g. I would think that the operator 'Attribute has a value' would be very similar to using the operator 'Contains Regex' with the regex '.+' (meaning 'one or more' (i.e. '+') of 'any single character' (i.e. '.')).
RFC 8142 outlines GeoJSON Text Sequences, also known as line-delimted GeoJSON. It would be great if FME could both read and write in this format, mainly for use in the Mapbox Tilesets API.
Hi Safe,This is now been a long standing issue and has not implemented or resolve since version 2015. I think it is driving ITS and FME developers wild
***Note from Migration:*** Original Title was: Add option to Data Inspector to set (/alter?) coordinate system of dataset if initially unknown (or wrong) Sometimes I get shapeFiles from users/clients that don't contain the .prj subfile specifying the coordinate system. In this case by default FME seems to assume that the coordinate system is equal to 'SPHERICAL MERCATOR'. However, in most of my usecases this is not correct. In this case a warning 'some features may not align with the background map' is displayed in the data inspector.However, there doesn't seem to be the option in FME data inspector to set/correct the unknown (or maybe in some usecases wrong) coordinate system. It would be nice if this could be added.Another less preferable option could be if the default coordinate system that is used in the data inspector could be altered in the settings.N.b. a quick search about this on the forum only provided this result , which seems to ask for a sort of similar implementation to be added to the data inspector (only more from a 'stability viewpoint').
It sure would it be nice to have the ability on any transformer to change the order of the outputs. For the people who hate it when the lines cross, this would ease our minds. I do realize we can hide the lines, but having a clean workspace where nothing is hidden is like a piece of art.
Add support for writing to Mapbox Vector Tile format: https://www.mapbox.com/vector-tiles/specification/
Currently, when an automations underlying workspace(s) parameters change there is no way to get the automation to refresh the parameter list without removing it from the automation and adding it again or changing the action to another workspace then back to the original workspace.
In ArcGIS 3D Analyst, you have this option when generating a TIN surface.I'm missing something like that in FME's SurfaceModeller and/or TINGenerator transformer.It would be nice if an extra parameter could be set in which you can specify the maximum edge length so you don't always end up with convex TIN's that have these huge triangles on the concave sides of the surface, like so:[image adapted from Esri]Just like 3D Analyst, if the value is omitted, it could do a best guess by using the average (weighted) distance between all vertices in the surface for instance.[edit]I actually discovered a MAXIMUM_EDGE_CALC_LENGTH parameter in the SurfaceModelFactory documentation. Just a mention though, no description. Is that implemented? I've never directly used factories before...[/edit]
Add possibilities to create multiple charts at once. There is a group by - but no option to parallell process. Workaround today is to put it in a Custom Transformer with parallell processing possibilities.
I have random points that need to form a line (straight line) with an attribute that identifies the different lines. When using the LineBuilder if the line is defined with more than 2 points, in some cases the node order is not following the logical order-> 0,1,2,3... example-> incorrect node order: 0,3,1,2) . As result of this lines are not OGC valid, containing self intersections. An option to reset the node order will generate valid lines without overlaps.
Support for a "directory watch" on OneDrive to run an FME job whenever something gets modified or is new
I always run into problems finding that one workspace - among hundreds of them - I used last time to do this or that.... Even though I consistently name my workspaces using the order number, the name of the customer and a couple of key words in the filename. It stuck me that Windows displays a preview of the contents of a file, whenever I click on a tiff, or a png, or even a csv or html file, whereas clicking an FME workspace file in Windows Explorer results in a piteously:Would it be an idea to have Windows Explorer display a miniturized overview of the complete workspace (or the last view on the canvas when saved) if a .fmw file is clicked and the preview pane is on? This for some people might recognize the overview better as the work they produced, at least better than the name of the .fmw file.
I do not know if this is possible or not, but it would be nice to have the ability for Readers and Transformers that connect to databases (SQLExecutors, Feature Readers, DatabaseJoiner), to have the option to close the connection after the data has been read out. Currently the connections stay open for the entire ETL run, even if the data has been read out, so if the ETL takes 35 minutes to run, and we have a firewall timeout of 30, the ETL will fail even though it has already read out all the data from the source, as it cannot close the source connection, as it has been terminated by the firewall at that point.
FME has recently integrated property sets support.Would be nice if values from properties that can be chosen from a drop down list were supported too.
When using a reader to a very long path and filename. Or to files within a zip-file with very long paths and filenames FME usually gives "File not found"-error in the logfile. Give a better explanation to the user when the path is very long so the user can just move the files to a folder with less letters.Or when reading a file with long name (more than 260 characters - mspoweruser.com/ntfs-260-character-windows-10/ ) try to move the file to FME-temp folder and then read it from there.Since some project-disks are long by default it is not possible always to shorten the path where the original data lies
Would be great to visualise line direction in the inspector!
The srsDimension property can be set on multiple locations in a GML. Some clients require it within the gml:posList tag. However FME places them by default on the gml:geometryType, e.g. gml:LineString. I would like to have to option to choose where FME places the srsDimension property.
When gathering a list of relevant values, it would be nice to be able to compare an attribute value directly to these values, to determine whether it's found in the list or not. Now I must create a concatenated attribute, which must be comma separated, prompting all kinds of formatting problems. Simply comparing to a list would be much cleaner and efficient. This is also relevant in all other transformers that utilize test conditions, like TestFilter et.al.
When configuring a distributed FME Server install, it seems that there are still ports opening in the high numbers, that are not controlled by setting the FME_SERVER_PORT_POOL. These are causing problems where firewall restrictions mean not everything works as expected in FME Server.It would be useful to add port_pool configuration parameter for the web app (tomcat).
An option to use multiple spatial tests (similar to the SpatialFilter) when using the FeatureReader would be useful when translating large datasets, especially if you want to read features that are disjoint or touching the initiator (e.g. lines that are disjoint or touching an area, but not contained).
Currrently in the Ifc reader there is the option to substract openings from their parent feature geometry. However if you make that choice, the geometry of IfcOpeningElement will no longer be available. If you choose not to do so, you can use the CSGBuilder to reproduce the holes, but with a risk of keeping 'Wallpaper' geometries. I would like to have the option to substract the openings, but also to keep the geometrie available. This way it is possible to rebuild the relationships in a later stage, but also to use the geometries of the IfcOpeningElement for other purposes, e.g. if a OpeningElement isn't filled.
This should be a fairly simple addition, but it is almost certain that the underlying code in InlineQuerier has an inbuilt option for creating and executing the DB in RAM. All it needs is to be exposed to the user in the Transformer Parameters.Currently InlineQuerier creates a temporary SQLite database file on disk, and executes SQL queries against this. This makes the Transformer dependent on both the available disk space and the speed of writing/reading from that disk.However the underlying SQLite engine used by the Transformer has In-memory database creation options as a standard part SQLite. There are three ways of creating different databases, but all use the same base functions and commands.I've tested all 3 methods with the similar SQLite FeatureWriter, and all 3 seem to work which seems to indicate that FME can already do these, it is just lacking a Parameter in InlineQuerier to expose the different performance optionsIf you call"sqlite.exe filepathname" or "rc = sqlite3_open(filepathname, &db)" depending on accessor being usedthen this creates the database on disk. This is what InlineQuerier does now.However if this is altered to:"sqlite.exe :memory:" or "rc = sqlite3_open(":memory:", &db)" depending on accessor being usedThis creates the database in RAM, if available.A third hybrid option is available that again uses the same command, and creates a temporary "pagefile" like file that if allocated RAM is consumed then it will mix data between RAM and on-disk:"sqlite.exe "" " or "rc = sqlite3_open("", &db)" depending on accessor being usedSQLite technical documentation here:https://www.sqlite.org/inmemorydb.html
Sq-lite Default User Parameter description should say Source and Destination
Reading Civil 3D property sets with FME. See also these topics:https://knowledge.safe.com/questions/40822/expose-autodesk-civil-3d-property-sets.htmlhttps://knowledge.safe.com/questions/34435/new-autodesk-civil-3d-2017-property-set-definition.html
I often encounter the same problem and need: I want to add text input in my workspace for various reasons. You could use standard transformers for this task, as AttributeManager or AttributeFilter, but I think a new type of transformer would be more appropriate. I give you the most common scenario:You want to filter out certain id:s (perhaps from a long list or a database) to try your methodology. The id:s could be found elsewhere and you just want to copy and paste them in your workspace and then hit Run. Let's say that you have 10 or 20 id:s you want to check.The new transformer, called TextInput or something else, would look something like a merge between a standard tranformer and an annotation. You could just click the annotation area and paste your text and it will automatically be put as the attribute fme_text_input.The only parameter for this transformer would be a checkbox if you want the textinput to be transformed in to a list and a choice which delimiter that should be used for creating this list.
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.
