Transformers

I need to output an SQL script created with SQL Creator in a custom text file. The file is essentially a csv file but with '|' as the seperator, the header row prefixed with a 'H|' the rows of data prefixed with 'B|' and a footer containing F|<file_id>|<datetime_stamp>|<number_of_rows> creating a csv file using '|' is easy enough but I haven't found a way to do the prefixes or footer. I was looking at using XML Templater but not sure if I'm barking up the wrong tree. Any guidance would be greatly appreciated.

We're trying to run a script in FME Server 2018 that replaces a geodatabase with a new dataset using arcpy. In a few months we'll be attempting to use FME's native geodatabase writer, but for now I need to troubleshoot why the following doesn't work:import fmeimport fmeobjectsimport arcpy# Main Functiondef delme(feature): # Variables table_2b_deld = feature.getAttribute('SOURCE_ETL_TABLE') sde_conn = FME_MacroValues['SDE_Connection'] full_del_str = sde_conn + '\\' + table_2b_deld + 'X' # print full_del_str # Process try: arcpy.Delete_management(full_del_str) feature.setAttribute('deletion_result','success') except Exception, e: logger = fmeobjects.FMELogFile() logger.logException(e,2) feature.setAttribute('deletion_result','failure') feature.setAttribute('deletion_result_detail',e) passAfter that script runs, we move data in Oracle from our staging schema into our production schema, then attempt a geodatabase registration script with a virtually identical syntax.If the geodataba

I looking to calculate the intersected area between polygonsI used the Areaonareaoverlay transformer but I am getting one of the feature removed and replaced by another one,

I am converting an excel spreadsheet to an esri geodatabase feature class of points. Working well so far, except for one field of elevation values. There was no standard of input and the values are varied. For example:75 feet75 ft.75 ft75f75 MSL25 m / 75 fIf it weren't for the last example, which occurs too often to waste time updating manually, I'd just use a string replacer to extract just the numbers, but in that last example, I'd end up with 2575 instead of just 75. I'm guess that first I need to remove everything before and including the '/', and then can use a stringreplacer with regular expression '[0-9]+' . Also, I've gotten another spreadsheet of data in which "##m" occurs. Those rows would need to be converted to feet. Hoping this is solvable with transformers, as the data is not to be edited.

I have a custom transformer that caches API call results to a PostgreSQL table. It tries a DatabaseJoiner on the key, and if it finds a matching record, it passes the joined feature directly to the main Output port. Otherwise, it calls the API, passes the result to a SQLExecutor to INSERT it for joining on same key value on subsequent features passed to the transformer.Unfortunately, the DatabaseJoiner is unable to find records stored by the SQLExecutor for subsequent features during the same workspace execution. I've tried explicit COMMITs (with appropriate delimiter declaration) in the SQLExecutor (it says no transaction is in progress) and a 5-second-per-feature Decelerator after the SQLExecutor in case there is some kind of race condition between FME Workbench and PostgreSQL. No luck. I still get duplicate key violations because the failure of the DatabaseJoiner causes the same key to be sent to the API and storage attempted again via the SQLExecutor. I've even tried

Good morning,I have a question about FMEI have many readers and writers that transform files into oracle tables, and he reads everyone and does commit only after all. I would like you to conduct a commit every transaction from reader to writer

Using 2018.0 I placed a logger in-line with a translation. Max messages to log is set to 999999, but it stops logging after 115 features.Why is that - and how do I get it to log all the translation info (looking for info on the 428'th item in this case)?

Hi,Just wondering if there is any way to link the components of a BIM (Revit) model to values so the component changes a color if its value is above a range (?)Any insight is more than welcome.Many thanks.

Is there a way in FME DataInspector to make 3D solids "partly" transparent, like area filling is now ?I have quite hard times to inspect the parts of a solid when I need to always rotate it so that the marked vertex/point is in foreground.It would help greatly if it would be possible to see through the surfaces. Michael

I'm trying to run a workbench on my MacOSx 10.13 using DateTimeNow function but I get this error :AttributeCreator: Failed to load timezone data. Please visit http://www.safe.com/supportAttributeCreator: Failed to evaluate expression '@DateTimeNow()'.  Result is set to null.It runs perfectly on Windows 7.Do you know what could cause that? I've check into my options and my time zone is defined.

i have a lidar data for which i have to create a lod2 city gml model . how can it be done?

i build a 3d buildings scene in sketchup, when i convert it to 3d tiles, i found there is only one feature in b3dm, i can't divide different buildings each other, can i split 3d entity in skp file ? or any good solution to solove it ?thanks!

Thanks for your help in advance, I am creating composite content, stacking polygons together from different 'surfaces' then clipping. This works as expected but it is almost exponentially creating additional vertices. This is an issue and as a result I need to generalize the output. Unfortunately when I use the generalizer tool the output geometry reports lots of invalid outputs, mainly 2d self intersections. Does anyone have any suggestions to help.Many Thanks

I have an IFC File that was exported by Revit. FME triangulates the surfaces as can be seen in the picture. Is there any transformer or parameter option to solve this problem? @DaveAtSafe

The issue: When using the FME 2017 and 2018 NETCDF writer the coordinate system of the output file is 'unknown' even when set. Writing the same data to geoTIFF has the correct coordinate system set. So far I have tried1. Setting the crs in the reader and then writing out with no specified crs2. Setting the crs in the writer3. Using the CoordinateSystemSetter to set and then writing with no specified crs4. Using the CoordinateSystemSetter to set and then writing with a specified crs (same as what was in the crs setter)5. Setting the crs in the reader and then using a reprojector (GDA94 to WGS84) and then write with no specified crs6. Doing 1 and then writing to geotiff Results: 1,2,3,4, and 5 have a crs of 'unknown'6 has a crs of 'GDA 94'Attaching an inspector to readers with a specified crs shows the crs as 'GDA 94'Attaching an inspector after the coordinate system setter and the reprojector shows the crs as 'GDA 94'

Hi All,  I am building validation rules for 3D buildings. One of the rules I would like to test is to check that all the surfaces (walls, ground surfaces, roofs) are not crossing each other.  When I test this with dummy data, this works fine using the SpatialRelator. (Test to perform: Crosses) [I create dummy data by creating 2D-lines an extrude them (some of them with a constant extrusion height, some of them with and extrusion vector) so I get surface which intersect somewhere in 3D (and the intersection is a line) and the SpatialRelator detects my crossing objects. So this works fine.]  When I test the same with real data (and I know some of the surfaces do cross, the SpatialRelator doesn not detect the crossings.  I have the impression that the difference is the geometry type of my "surfaces". In my dummy data the Geometry is IFMECompositeSurface and contains a IFMEFace and an Appearance. (it is the extruder who changes the geometry-type from line into IFMECompositeSurfac

I am currently working on creating polygons from a gps tracks based on tracks of same attributes using hullaccumulator but my points (tracks) are scatter and producing overlapping polygons so I am looking for how I can set a distance constrain so points closer to each other can form a separate polygon from tracks further apart even though they have same attributes. Please how can I be able to set distance constrain in this scenario?

Hi Everyone,  I have been using FME for some time now and am now getting into little details. My question revolves around parsing out filenames and eliminating portions of them to create a new name. This is probably elementary to many of you.  I have a name like this:   2_038_1558_0001.ecw  In this case, the number after the last underscore "_" is always an incremented value. What I want is to always eliminate the last underscore and following number and extension and have the following result:  2_038_1558  The sequential number portion is not always 4 digits. Sometimes it is 2 or 3 or 5 (2_038_1558_0001.ecw,  2_038_1558_023.ecw, 2_038_1558_00012.ecw)  Any ideas of the which transformer to use and/or the code the place within it?  Thank you for any help .  

Hello,can someone please help me to understand this error:ERROR |Microsoft SQL Server Spatial Reader: Failed to fetch post-ADO SQL Server columns for table I can't replicate it every time I run my workspace. But when I capture that, the process ran and I got expected results.Thanks

I am new, so I appreciate your patience.I have tried to search for my answer, but so far no luck.I am attempting to compare the values from 2 datasets (no geometry at this point) with the same attribute names. I used the changedetector so we could see which attributes were non-matching and which were missing from one of the datasets (updated, inserted, deleted).The features have dozens of attributes, so when something doesn't match it can be a process to determine the exact attribute that it failed on. How can we retain/build the relationhship back to the attribute name and value to pinpoint where the failed match occurred?

Hi I'm applying change updates to OS MasterMap stored in a geodatabase. However, it is not inserting the new featuresI am using two feature mergers, one to find the deleted features and one to find the updated and new features to insert. These then go to attribute creators to create the fme_db_operations field with DELETE, INSERT and UPDATE values.My writer parameters are as follows: TOID is the unique value field. It is deleting and updating existing features but not inserting the new ones.Any help and advice would be greatly appreciated.I have attached my workbench for you to look at if that helps.Thanks,Matt

I have a spatial table (an ESRI feature class) that consists of many points. I want to divide the points into groups based on an attribute "projectCode", then create a center of mass for each group. All the centers form a new feature class. Any hint how to do this?.

I have a test2 shp file representing a 3D door, and then try to convert test2 to the fbx format file, but the fbx file appears translucent area. After query, it is found that the normal line needs to be set perpendicular to the surface. How do I set it? This file contains test2.shp and fwm project files

I've got a dataset with lots of columns and the contents of the ones I need to tackle contain many zeros (0) and lots of ones (1)... there are also lots of 'Yes' and 'No' values. I want 'selected' columns to be checked for the existence of a 0 and it should be replaced with a 'No', 1s need to be 'Yes'.The AttributeValueMapper will do this on 1 attribute, but I don't want 1 AttributeValueMapper per column.An AttributeManager allows me to set conditional values, but I don't want to have to create this conditional rule for every column, even with FME 2018 supporting copy\\paste, that's too much effort.I'm currently using 2 chained NullAttributeMappers, with one of them handling the 0 mapping to No and the other the 1 mapping to Yes.But what would you do? I'm looking for ideally a elegant single transformer solution and really an AttributeValueMapper that could work on more than just 1 column seems to be an 'Idea' I o

I am producing an XML file in a loop. I write one feature formatted as XML at a time. In the next loop execution I wish to write over what is in the file or wipe it out completely and write the current feature. Thanks.