Hi,
I'd like to have a broader feedback of the community on
https://github.com/OSGeo/gdal/pull/14763 . Please read it in detail. In
practice, the main backwards incompatibility would be the point 1 I
mention in the PR text, that is that for current users of
gdal.Open(dsname), they would now get vector datasets opened in addition
to raster ones. While I see this is as a long term benefit, at least one
user (https://github.com/ubarsc/rios/pull/192) had to work around it
since they expected gdal.Open() to fail on vector datasets. The
workaround being to use gdal.OpenEx(dsname, gdal.OF_RASTER) that will
work before and after the PR, or if targeting only post PR state,
gdal.Open(dsname, gdal.OF_RASTER)
PR 14763 would be potential material for a GDAL 4 if that did occur, but
I don't know if it is desirable and even if it was, I don't think it
would happen in a short or medium term (mostly because the GDAL dev
community is too small to be able to gather a significant set of
backwards incompatible change in a reasonable period of time, while not
blocking features to be released at a reasonable frequency). So we're
more in a situation where we land a few backwards incompatible change
each (non bugfix) release.
Even
--
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
