Re: GDAL_COMPILATION and standalone plugins
Even Rouault via gdal-dev <[email protected]>
| Newsgroups | gmane.comp.gis.gdal.devel |
|---|---|
| Message-ID | <[email protected]> |
Andrew, You can redefine safely the definition of your GDALRegisterXXXX() function in your driver code (prvided you use the same signature as in gdal_frmts.h of course) GDAL_COMPILATION is used to protect "private" GDAL functions that we don't want external users of the C++ API to be able to access. As cmake/helpers/SetupStandalonePlugin.cmake is designed for code that is in the GDAL source code tree, it is logical it sets GDAL_COMPILATION Even Le 25/03/2026 à 14:30, Andrew C Aitchison via gdal-dev a écrit : > > When should GDAL_COMPILATION be set ? > > I would like a source tree for a driver. > Ideally it would be possible to build this in tree (both > as a plugin or compiled into libgdal.so) and standalone against > an existing gdal installation (with development libraries and headers). > > I thought that GDAL_COMPILATION was a way to tell whether my driver > was being built as part of gdal, but I find that > cmake/helpers/SetupStandalonePlugin.cmake includes > cmake/helpers/GdalCompilationFlags.cmake which sets GDAL_COMPILATION > > Why do I care ? > An in tree build will declare my GDALRegister function in > gcore/gdal_frmts.h > (since it will be used in > frmts/gdalallregister.cpp > ) but an out of tree build will need to declare this itself. > I thought that testing GDAL_COMPILATION would be the way to avoid my > GDALRegister function being declared twice under some circumstances. > > Any idea what I am missing or misunderstanding ? > > Thanks, > -- http://www.spatialys.com My software is free, but my time generally not. _______________________________________________ gdal-dev mailing list [email protected] https://lists.osgeo.org/mailman/listinfo/gdal-dev