-
Steve Revill authored
Detail: You can now rely on the options passed in the components file to adjust the "Locale", "UserIF" and "Machine" environment variables for a component. This allows you to build components which use paths like "LocalRes:", "LocalUserIFRes:" in their Makefiles and get the correct resources out. I also fixed the install rule in the Makefile. Admin: Builds and quite thoroughly tested. As an aside, I do occasionally see data aborts during builds, but this has been seen for a couple of years and on various build machines. An example is: INSTDIR = 'ADFS::Slayer2.$.Builds.Baseline.InstallAMU: *** Internal error: abort on data transfer at &0380D278 *** (line split for clarity). I don't know if this is amu's fault or scrbuild not terminating strings correctly. This is always where things seem to fail, i.e. the Kernel outputting a string which aborts before reaching the terminator - at the byte which should be it. Maybe it is the OS_NewLine code triggers callbacks in a task window which occasionally causes some nasty abort? Version 0.37. Tagged as 'srcbuild-0_37'
Steve Revill authoredDetail: You can now rely on the options passed in the components file to adjust the "Locale", "UserIF" and "Machine" environment variables for a component. This allows you to build components which use paths like "LocalRes:", "LocalUserIFRes:" in their Makefiles and get the correct resources out. I also fixed the install rule in the Makefile. Admin: Builds and quite thoroughly tested. As an aside, I do occasionally see data aborts during builds, but this has been seen for a couple of years and on various build machines. An example is: INSTDIR = 'ADFS::Slayer2.$.Builds.Baseline.InstallAMU: *** Internal error: abort on data transfer at &0380D278 *** (line split for clarity). I don't know if this is amu's fault or scrbuild not terminating strings correctly. This is always where things seem to fail, i.e. the Kernel outputting a string which aborts before reaching the terminator - at the byte which should be it. Maybe it is the OS_NewLine code triggers callbacks in a task window which occasionally causes some nasty abort? Version 0.37. Tagged as 'srcbuild-0_37'
Loading