Re: Removing the "install" targets from scripts/makefile.*
John Bowler <[email protected]> Sun, 18 Sep 2022 19:38:06 -0700
| Newsgroups | gmane.comp.graphics.png.devel |
|---|---|
| Message-ID | <CAP7U398PCk0TAQ6=BwgeCMJgpwJNjTwK+gWfmCQO=2er4h5F6w@mail.gmail.com> |
> I noticed that most of the makefiles inside the scripts/ directory did not execute the install targets correctly, and it's been like that since libpng-1.5.0beta01. Clearly, those things weren't tested in ages. Because they are third party contributions that were, mostly, maintained by those third party devs. Do you you a riscos machine (scripts/makefile.acorn) to test that on? Why is scripts/makefile.vcwin32 there? There used to be a Visual Studio project (which I tested with every release; Glenn did not have appropriate equipment, but he did test VMS, IRC.) I think I'm about to become a badly scratched record; delete them all (not VS, not configure, your choice on CMake), see who whines. In fact VS can probably go too; it most likely hasn't been tested in ages either. Bear in mind that all the makefiles, cmake and VS all rely on a prebuilt configuration. I couldn't work out how to make cmake do a local build operation, so configuration seemed impossible. I think I might have put in some support for VS, which can handle the concept of things that must be built locally versus things that are for the target (I was testing the ARM targets, even though I couldn't run the result), but all the same it's a nightmare. Simplify, simplify, simplify, then delete everything and start again. _______________________________________________ png-mng-implement mailing list png-mng-implement-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org https://lists.sourceforge.net/lists/listinfo/png-mng-implement