Re: [gdal-dev] FileGDBs and FIDs of -21121

2024-08-29 Thread Jeff McKenna via gdal-dev
I also like the subtle nod to Rush's legendary album, "twenty-one twelve", inside the GDAL code ;) -jeff -- Jeff McKenna GatewayGeo: Developers of MS4W, & offering MapServer Consulting/Dev co-founder of FOSS4G http://gatewaygeo.com/ On 2024-08-27 6:38 p.m., Even Rouault via gdal-dev wrote

Re: [gdal-dev] FileGDBs and FIDs of -21121

2024-08-27 Thread Even Rouault via gdal-dev
I'm afraid you are questioning design choices dating back to the last millennium: https://github.com/OSGeo/gdal/commit/9d4076f3672b0b6b31de175fcfa677b378c0b577 :-) This could be any value really (perhaps except 0). I assume this was a nice one when looking a OGRRaw value in a debugger to actu

Re: [gdal-dev] FileGDBs and FIDs of -21121

2024-08-27 Thread Seth G via gdal-dev
Thanks Even! I got as far as setting up a test case in MapServer and starting the debugger before your fix arrived :-) As an aside, is there a reason -21121 was chosen as a special value? Seth -- web:https://geographika.net & https://mapserverstudio.net twitter: @geographika On Tue, Aug 27, 202

Re: [gdal-dev] FileGDBs and FIDs of -21121

2024-08-27 Thread Even Rouault via gdal-dev
Le 27/08/2024 à 23:01, Even Rouault via gdal-dev a écrit : oh, I've managed to reproduce. This is a bug in the OpenFileGDB driver. Fix to follow soon ==> https://github.com/OSGeo/gdal/pull/10662 -- http://www.spatialys.com My software is free, but my time generally not.

Re: [gdal-dev] FileGDBs and FIDs of -21121

2024-08-27 Thread Even Rouault via gdal-dev
oh, I've managed to reproduce. This is a bug in the OpenFileGDB driver. Fix to follow soon Le 27/08/2024 à 22:31, Seth G a écrit : Thanks Even for the details. Sorry, the FID is correctly populated (with an autogenerated FID values starting from 1). However the unique ID of the feature is exp

Re: [gdal-dev] FileGDBs and FIDs of -21121

2024-08-27 Thread Seth G via gdal-dev
Thanks Even for the details. Sorry, the FID is correctly populated (with an autogenerated FID values starting from 1). However the unique ID of the feature is exported in a separate attribute field, and it is in this field that all values are present in the FileGDB except for the record with -

Re: [gdal-dev] FileGDBs and FIDs of -21121

2024-08-27 Thread Even Rouault via gdal-dev
Seth, I'm not totally sure to understand the exact scenario. -21121 is a special value for OGRField, which is a distinct concept that the Feature ID, and normally it shouldn't prevent setting Integer64 fields to the value as this canary is set in 2 fields of OGRField. from osgeo import ogr l

[gdal-dev] FileGDBs and FIDs of -21121

2024-08-27 Thread Seth G via gdal-dev
Hi all, I'm using a GDAL within MapServer to export features as a FileGDB. Using the following OUTPUTFORMAT: OUTPUTFORMAT NAME "FileGDB" DRIVER "OGR/OpenFileGDB" MIMETYPE "application/x-ogc­filegdb" FORMATOPTION "STORAGE=filesystem" FORMATOPTION "FORM=