+1 Yes please!! Thoughts: - I've battled with the types so many times over the years. The pain caused by a switch will be well worth it - Q1: I'm for going to GDAL 4.0. - Q2: Having a GDAL_USE_OLD_INT_TYPES for a while seems like a good idea to expose the alias to users of the library - Q3: Yes to vsi_l_offset -> uint64_t. I'm 90% for it. For me: Yes offset has meaning, but if the type is critical to understanding a function / method interface, we should have a comment. Knowing that it's an unsigned 64 bit int is usually more important. Would be great if anyone against this change could speak up and give some examples.
Short term pain, but long term big gains in readability for people new to GDAL / stability / maintainability / ability to cross compile. On Fri, Sep 15, 2023 at 2:24 AM Even Rouault <even.roua...@spatialys.com> wrote: > Hi, > > I've submitted an initial version of RFC95: Use standard C/C++ integer > types > > https://github.com/OSGeo/gdal/pull/8399 > > It has important implications on backward compatibility of C and C++ > API. The "questions to be answer" paragraph need to be answered. > > Even > > -- > http://www.spatialys.com > My software is free, but my time generally not. > > _______________________________________________ > gdal-dev mailing list > gdal-dev@lists.osgeo.org > https://lists.osgeo.org/mailman/listinfo/gdal-dev >
_______________________________________________ gdal-dev mailing list gdal-dev@lists.osgeo.org https://lists.osgeo.org/mailman/listinfo/gdal-dev
