On 22.10.2017 07:19, Robert Elz wrote: > Not really (aside from quoting glibc as being an example of how things > should be done is usually just funny) - apps can define as many of those > symbols as they need, the lines from featuretest.h that I quoted are > just picking the default. Just to be clear. GLIBC wasn't a model example how things should be implemented, but how they are actually implemented. So going for multiple namespaces to be specified explicitly will break software. I noted that there is a problem porting software to SmartOS as it's difficult to figure out what is the proper set of _*_SOURCE defined in order to ship function prototypes for all used features in software. This is another extreme example. The usual solution porting software to NetBSD is to disable _SOURCE_POSIX-like defines or change -std=c++11 to -std=gnu++11 to unbreak things. Getting closer to SmartOS-like solution will increase the pain porting GNU software.
Attachment:
signature.asc
Description: OpenPGP digital signature