Transformers

Hi all,  I have one dataset containing a set of features with attributes like ID, Name, Date, Location. While none of the individual attributes are unique, the combination of all of them are. (Record) I have another dataset of features with one attribute containing freeform multiline text.  (Text)  Each Text feature contains ALL of the values of ONE feature of the Record dataset, but not in any order, and generally not an exact match on a line.  I need to identify which Text feature corresponds to which Record feature.  Each Record should have a zero or one match with a Text feature.  I am assuming that python and regex is the way to go, but I'm not sure as to the most efficient way to process the data.    Record FeaturesIDNameDateLocation24AAA23 MAY 2019X32AAA07 JUN 2019Y24BBB07 JUN 2019Z A sample text feature could contain something like:SEE 242926m7000'Search shelter X32500 2000 2500800 32 200AAA/ABC07 JUN 2019YThe correct record in this case is 32-AAA. 

Howdy,Are there known errors/solutions related to the FileNamePartExtractor transformer's handling of 'DateTimeFormat' functions in a file name? I am experiencing this issue in both FME 2018.1.1.2 & FME 2019.0I have a workspace that reads in information from FME Server on scheduled jobs (http://fme:8080/fmerest/v3/schedules?limit=1000&offset=0) and writes details about each scheduled Job and its destination tables to an Excel workbook.I believe this workspace was working fine a few months back, however now I am encountering an error with a FileNamePartExtractor transformer's "Full File Name Attribute" parameter. The error is triggered where a Job has a parameter for the destination file name that includes @DateTimeFormat or @DateTimeNow . For example processing a destination file name like this triggers an error:\RC Metrics_DailyUpdate_@DateTimeFormat(@DateTimeNow(),%Y%m%d).xlsxThe error message received is below:2019-06-18 12:11:24|   0.7|  0.1|ERROR |FilenamePartExtractor_2 (

I have a number of ESRI shape files that need Sqlising. Each shape file goes into a SQL Server Database table as a single row. How do I achieve this in one workflow? Doing these one by one is not practical as there are hundreds of shape files and I can easily make a mistake plus not repeatable.

Howdy,Are there known errors/solutions related to the  JSONfragmenter transformer's handling of attributes with no value?  For some reason instead of sending those records out through the Rejected port the translation ends with an error message.Here is the full warning/ error message:2019-06-18 12:43:35|   0.6|  0.0|WARN  |JSONFragmenter_14 (JSONQueryFactory): A feature was passed to the JSONQueryFactory with no value in the 'JSON_text_HTTPCaller_ws' attribute.  Features with no 'JSON_text_HTTPCaller_ws' value will be ignored2019-06-18 12:43:35|   0.6|  0.0|STATS |Storing feature(s) to FME feature store file `D:\AppData\Local\wb-cache--EeQqMH\Main_JSONFragmenter_14 -1 10 fo 1 _lt_REJECTED_gt_  0  6d87df4aecc2d633e5a7711702c99e6532a5683f.ffsupdating'2019-06-18 12:43:35|   0.6|  0.0|WARN  |JSONFragmenter_14 (JSONQueryFactory): No JSON data was found. Make sure the 'JSON Document' parameter has a valueError running translation.I am having this issue with both FME Desktop 2018.1.1.2 & 

We often get data with invalid dates in date fields. What is the best way to find invalid dates and replace the invalid dates with a common valid date. There are multiple date fields and there could be invalid dates for multiple date fields for the same record. This is what we need to accomplish:1. find the invalid dates (could be multiple fields for the same record)2. update the invalid dates with a generic valid date (for instance, all the invalid dates would become 1/1/2001)3. combine all the updated records with updated invalid dates with all the records that did not have invalid dates so that the outcome is a table with all the original records with just the updated invalid dates Thanks,David

I need to set the tiling for an orthophoto made on Zephyr. I want to conduct object based image analysis on the mosaics made in eCognition software. I am hoping to classify broad tropical marine habitat types, e.g. coral, sand, seagrass etc. Can anyone recommend how to go about deciding on the best grid resolution?… Read more  

When reprojecting a spatial Autocad DXF file with labels, I'm finding that the geometries all get reprojected correctly, but the labels are wildly wrong. I've reprojected a variety of attributes such as autocad_alignment_x and autocad_alignment_y but haven't been able to get a solution. When manually setting fme_text_size, the overall bounding box of the data does change, but the labels themselves don't seem to behave correctly. Has anybody run into similar issues in the past?

HEllo,I would like to build a line from polyygon center points. The coherent polygons have an id (ivv) on which a I want to create a single line.First I am using a centerPointReplacer to get the centre point of the polygons. After that I am using aggregator, to group the poins by (ivv). Now I want to build the lines, but it gives back rejected even if I set the connection break output to ivv or not. What should be the correct order or parameter setting to get lines from the aggregated points?

I'm new to FME, and most of the time I work with ArcGIS Pro. I have a few building polygons, and a few points containing additional information. Now, in ArcGIS Pro I can spatially join these two, creating one layer with the polygon "look" but added with the point information. This then needs to be exported to a new shapefile.How can I recreate this in FME? I now used the SpatialRelater, with a tester, but when I export 'passed', it still only contains the attributes from the input area, and the attributes from the points are not added. When I look at the passed data with the inspector, it does show an attribute table wit the combined data, but when writing this to a new file, it does not.Can someone help me out?

I have a set of polyline features and a set of point features in a ShapeFile. Each set of features has a RID field. The RID for a polyline record corresponds to an RID field for the points. As expected, one line has many points that comprise it. There are 10-20 lines in the input dataset and ~1500 points. I just want to show each unique RID value as a different color in my output KML file. The workbench goes something like this:1) Read the Lines and the Points2) Read a SQLite DB containing a table listing a RID, and a Color (both RGB 100,100,100 and KMLColor FF00987F format (http://www.zonums.com/gmaps/kml_color/)3) FEature Merger moves the attributes from the SQL Lite DB to the Shapefiles, the reprojector, 3dforcer, KMLPropertySetter then the KMLStyler where the colors for all symbology is set using the fields containing the colors merged from the SQL lite databases.4) The resulting data is written out to the KMZ. The issue is that neither the RGB or the KML Color values are applied t

I need to get the closest distance between points and lines calculated. The points and lines have matching attributes.I only need distance calculated between point A and Line A, Point B and Line B.. Is it possible?

I used the ShortestPathFinder to generate the shortest path from a start point through a set of > 1,000 intermediate points to an end point. Therefore, the "From-To" line that is used as input to the ShortestPathFinder connects all of these points/stops. Only the intermediate points on the From-To line were set to be re-ordered and not the Start and End point. The ShortestPathFinder did output a "Path" line. Now, I simply want to figure out the sequence/order of these intermediate stops on the shortest path.I have allowed for U-turns in the shortest path and there are many segments in the path that overlap one another as U-turns are allowed. Therefore, two points/stops at a very close geographic location can be at completely different orders/sequences along the shortest path from one another. I tried several approaches already: 1) Use Chopper to convert all the vertices of the ShortestPath line to points, then Counter to order the points in a 'Count' attr

Hello, I have sql query from our db (sql creator) containing records as e.g.Id; Source, CreationDate1; WTT Australian Notice 123, 201903122; German Notice 12GG3 and VTS, 201904123; VMS and Canadian Notice 53T, 201903124; Canadian 78UU, 20190312I want to use a regex stringseracher for the source attribute and assign a value based on the search result.I have an excel table that contains the regex I want to search for in the db records and the value I want to assign e.g.Regex; Country Australian (Notices|Notice); AUSTRALIA Brazilian (Notices|Notice); BRAZIL Canadian|Canada (Notices|Notice); CANADAHow can I use this table as my regex string?My output should look likeId; Source, CreationDate; Country1; WTT Australian Notice 123, 20190312; AUSTRALIA2; German Notice 12GG3 and VTS, 20190412; null3; VMS and Canadian Notice 53T, 20190312; CANADA4; Canadian 78UU, 20190312; nullThanks for your help!

Hello,I am beginner in android development. I want to upload the model from mobile device to fme server and want to download the converted file to local storage of android. But I can't find any proper documentation about how to use fme api's with android studio.Someone please help me to get this work.. Or share me trick to use REST API's in android. Thanks, 

Hi,I was wondering where to find a sample working FME script that uses NDVICalculator transformer with a satellite image data.I remember Dimitri presented the concept way back during one of those FME World Tours (2016 or 2017?) in Calgary but I'm kinda lost where to look for it in the KnowledgeBase, or if it was even posted.Could you then kindly please advise. Greatly appreciate it.Thanks, WillyWilfredo de Vera, M.Eng.Sci., P.Eng.

I am trying to clean up dissolved features but area gap cleaner doesn't clean up the gaps.what am i doing wrong?Thanks

Hi, I have a file with a 10 points where I need to combine all points to multipoints. It shall produce 100 multipoints and each object should contain 2 points(one combination) .Its more or less a simple cross join without a database. I can produce all the points with unconditional merging in the feature merger but cant figure out how to aggregate them pairwise. The Featurejoiner will do the jobb but it doesnt allow me to keep the attributes Anyone have an idea? 

Hello,I would like to merge pointcloud data spatially.To do that, I create polygons representing the hull of point cloud data using the following steps:coerce point cloud data to individiual points,Bufferer every individiual pointDissolve the buffers that overlaps/intersectsCreate a polygon using hull replacerHow can I perform such operation without using a pointcloudCoercer because it is too much time consuming. Any idea on how to improve this part of my workbench?Thanks

Hello there, It's my first time dealing with CityGML. I am working on my university 3D buildings project and I am trying to figure out how change from EPSG:2180 coordinate system to EPSG:2179. I was trying to use CsmapReprojector, yet even tho FME considers it as success, there is no geometry after the process. I was also trying easy way to simply use writer, yet it losses additional atributes.  Also, is there any tool I can use to "mosaic" CityGMLs?

I am trying to use a GMLFeatureComposer for parsing the input data to the format of one input xsd file. While doing this translation, I am getting the error: GMLFeatureComposer(GMLFactory): Unable to find matching GML feature type from the parsed schemas. Not sure if this is due to any attributes missing in my xsd file or any other issue.

I have two shapefiles, x and y. I would like to filter the polygons in x to show only those that contain/intersect/touch the polygons in y. I've tried spatial filter but it doesn't give me any results. Any ideas?

So I have an Oracle table that has a Username field, Crew_Leader field, and a Crew_Number field.In my process I am reading from a table that has a username value in a field called Editor.I've been referring to this article: https://knowledge.safe.com/articles/19634/using-the-sqlexecutor-to-do-a-join.htmland decided to try to use the sqlexecutor transformer to run a query to join the Crew_Leader and Crew_Number fields to the table. However this seems to be very slow process as it tries to read over a thousand records. Do you know any other way I can accomplish getting the 2 fields from another table and then copying the values into fields in another table?Thanks.

Hi,I'm needing to use a custom python interpreter in 2019 on MacOS in order to use the pandas library.The option to use Custom Interpreters appears to be missing in 2019.0.0.1 - Build 19246. Is this normal behaviour i.e. has this functionality been removed?  Thanks,Campbell.

Hi Guys,i know maybe this was asked before and i searched the forum for this but i was not really successful in finding....if there is the same question please llink it:After my last post i have now set up a workbench which brings me following result: i have now 744 lines with same length of 1600m e.g with always a 50m shift first line begins 0 to 1600next from 50 to 1650next from 100 to 1700 and so onin an attribute i have stored a class number (1,2 or 3)so line 1 to line 16 class1then from 17 to 21 class222 to 51 class 1 and so onbut i have of course duplicate geometry always (with this shift)so i want long segments (and no duplicate geometry any more) where the class is the same BUT: on duplicate geometry parts i want to keep the line with the higher classi have part where i have class 2 and class 3 e.g.so i want the delete the geometry parts with 2 there and keep one with 3...hope i was clearly with my description otherwise ask meGreetz and ThanxFranco 

Hi  I'm looking for a quick way (hopefully) to identify fields that will have their values truncated on writing to output. Using the StringLengthCalculator works great but only on a single field, so I would need to string lots of them together - as my data is being written to a database with multiple tables and multiple attributes per table, I'm trying to avoid this. I need something like the StatisticsCalculator which can generate summary stats for multiple fields at a time,.. but the problem being calculating the string length first as StatsCalc works on the values. And my output values will differ to input values. I expect this is a comon problem - is there something I'm missing? Any ideas & help much appreciated,  Thanks Mary