Shape the future of FME with your ideas
Open ideas have been reviewed by our Product Management and are open for commenting and voting.
The online FME Server Authoring Course is good, but quite limited in terms of getting the most out of the product. For custom solutions, the API is really where its at and the Server Playground is a good starting point. However would it be a good idea to include two online FME Server courses: one for authoring and one for API development? A bit like the way the two Desktop courses are basic and Advanced, my idea is that Authoring and API Development are two possible FME Server course titles.
The internets suggest they're a GeoTIFF file, but when I force them to read into FME as a GeoTIFF it loads weirdly (looks as though it's loading multiple pyramid levels on top of one another?)
It would be helpful (especially to FME beginners) if workspaces (templates) would be provided to demonstrate how each regular transformer works. Each workspace should allow the user to test a regular transformer with various parameters setting and see the result visually as much as possible. How about starting from some most used transformer?
As mentioned in this question and discussed in the answers, it is now almost impossible to make FME export an XML document with a <![CDATA[ ]]> section a.k. literal character data.My intended use was to communicate some HTML-formatted descriptive text in a XML-response about the capabilities of a transformation. This can thus be presented to the end user of the application communicating with my service.My preferred implementation would be to add a XQuery function in the XMLTemplaterget-attributeget-xml-attributeget-cdata-attributeKind regards,Martin
It can bit a bit of a hazzle working to improve existing custom transfomer. To be able to save the custom transfomer in a new version, you have to close down the project open it up, to be able to select modify as new version, this is tricky if you have already started to modify the existing transformer and you don't want changes to the existing version. What would be nice would be an option to save as "new version", and an option in the workbench to update the custom transformer to a newer version.
It would be great if FME could monitor log files of one database, and when a transaction occurs, relay that change to another type of database in real time.
Could we please have a new transformer that enables basic vector operations [intersection, difference and union as a minimum] between two geometry columns (i.e., columns that have been created using GeometryExtractor within a processing sequence) and stores the result as a geometry column that can be accessed using the GeometryReplacer. This feature request acknowledges that the AreaonAreaOverlay does this job when configured correctly with other transformers. However this requires two datasets to be compared across all spatial combinations each time the process is run which can take a very long time for very large row counts. It is common in spatial database thinking, to have long term storage of multiple geometries that travel together in the same rows (with incremental update at times) and to do repeat assessments in different queries. The proposed transformer would in effect enable the multi-geometry operations that are possible (and extremely useful) as in the Oracle as an example: SDO_GEOM.SDO_INTERSECTION, SDO_GEOM.SDO_DIFFERENCE and SDO_GEOM.SDO_UNION operators (google these words for details)Having multiple-geometry operations would speed up very large and repeated table vector operations that compare two shapes in the same row.Thank you
Repeat shown warnings at the end of a log file. Now you have to search through the log file to find the warnings.
As data storage and transfer costs are billed in addition to the FME Cloud instance cost, it would be handy to be able to track usage over time. The current Billing Dashboard provides month-to-date metrics but it would be nice to be able to see daily transfer and storage usage (in KB/MB/GB, rather than - or in addition to - $USD). This information could be displayed in a graph or table. See this question for further details: https://knowledge.safe.com/questions/75739/fme-cloud-data-transfer-statistics-reporting.html
Supporting Netcdf4 format. Trying to work with Sentinel 5p data and it is only manageable using netcdf4 python module, R netcdf4 library or SNAP management tool. It would be nice to handle data directly in FME!!!!
In the GML3 spec, you can link to features using xlink:href /references based on GML:ID.This is currently not available in FME, which means we have to develop our own workaround external to FME to provide those details. This will enable CityGML features (for example Buildings) to use the surfaces to construct a solid, without replicating the geometries in the file.
Hi, it could be interesting working on the element lists like an AttributeManager do with attribute. Grouping the functionnalities of ListHitogrammer, ListSummer and other list transformers in one could improve fme projects like the AttributeManager do. Some other ideas on it ?
***Note from Migration:*** Original Title was: Add Flag or mark in transformers for visual identification of debugged/updated parameters Summary:I suggest the implementation of a visual mark (icon) or color usage for pinpoint/ identify transformers visually in workbench canvas that were revised/ updated by user/ workspace creator when performong a debug or are developing a certain workspace.The need:Often a user become confuse or "lost" in his own workspace when performing debugg action or re-parametrize. This is specially sensible when the workspace contains dozens, if not a couple of hundred transformers. The proposal.Could be included/ turn on a visual flag (similar to cashed inspection) or option to change transformer colour by user in order to identify a certain transformer that was already revised and is setted up as the user intends that it stays. This will allow for the user on a very easy and effective way to quickly identify which is the part of the workspace7 transformers already "worked" and the remaining ones, supporting efforts of debugging actions.The benefict.- significant gain of time and focus when restart a workspace debugg action;- visual- easy to understand what is done and what is pending when workspace goes from one user to another to continue the work.
provide better query functionality within the FME Server logs, maybe Elasticsearch
How about adding the ability to protect/encrypt zip files created in the FME Server download service to provide security for sensitive data. Security, can never have enough.
HiI need to append a lot of different file (XLS CSV SHP) and output a unique csv table but I would like to remove as first step all the fields present in my input that doesn't contain any valuemaybe an option in the Attribute manager .......thxFrancesco
Since it is possible, and nobody has posted it yet. What about a Movie Reader/Writer (MPEG, WMW, AVI, MOV...)?It can be done today with use of third party builds like ffmpeg, so why not add it to FME.
CMIS is an OASIS open standard for interfacing with content management repositories (upload and download documents, versioning, metadata) and appears to be supported by recent builds of Sharepoint, Laserfiche, OpenText and more. https://en.wikipedia.org/wiki/Content_Management_Interoperability_Services I'm not sure how feasible this is, but it looks good on paper..
Please split the enviroment variable FME_JVM_MAX_HEAP_SIZE to:FME32_JVM_MAX_HEAP_SIZE FME64_JVM_MAX_HEAP_SIZE The 32-bit version can use ca. 512m, the 64-bit version can use the full memory
Would be helpful to have a way to explicitly order MSWordStyler transformers. Currently the order is determined primarily by the order of the connectors (unless you have a Creators initiate every MSWordStyler with their order set). The simplest way to order them would be to allow the user to daisy-chain the MSWordStylers where successive MSWordStylers append to the the incoming from the previous MSWordStyler. An alternative would be to have an Order parameter in the MSWordStyler that you can set with the Word writer sorting them before it begins writing the output.
Hello Everyone, Those that use Socrata, Socrata has updated their SODA version (2.1 I believe). With this update, it allows customers to upload an ESRI Shapefile and translates it to a tabular dataset. Here is the support link describing the datatypes now supported: https://dev.socrata.com/docs/datatypes/index.html#, To have a Socrata writer that can take geometry input to the writer and translate it to any one of the spatial datatypes (listed) would be a huge boost for efficiency for the workbench and execution time and write time. Thank you
The FMEServerResourceConnector is a fantastic transformer for working with resources on FMEServer but it lacks the ability to create new folders.Even when uploading a file, the path that the file is being uploaded to must already exist, which means that it needs to either be created manually through the web interface, or by using the FME Server REST API. The probem with using the REST API (via an HTTPCaller) is that it cannot be used with a named web connection, so you need to mess around with Tokens, etc. for authorisation.It would be great if the FMEServerResourceConnector could be enhanced to either:add the capability to explicitly create a folderadd the capability to specify a path for an uploaded file, with that path being created if it doesn't already exist.Additionally, the error returned when something goes wrong with the transformer could do with some attention. The only error that seems to get reported back, regardless of what went wrong, is, "fme_rejection_code: ERROR_DURING_PROCESSING", which really doesn't help at all.
In FME 2016 the AttributeClassifier has been deprecated in favour of the AttributeValidator. I agree that the AttributeValidator is the future but unless I'm mistaken it doesn't yet carry the full volume of tests previously offered by the AttributeClassifier?Of course you can get the old AttributeClassifier from an old Workspace, but we shouldn't have to do that really.None of the other Transformers that the AttributeValidator is designed to supersede have been deprecated yet, so can the decision be reversed, for the time being at least?
Please add support for ADAC V5.
Originated from comment by @taojunabc in this Q&A.The idea is to provide a way to find Feature Type names that matches a regular expression, to add more flexibility to this transformer.
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.
