The test has been changed again to fail since making the test pass would
mean this bug could later silently show up inside a large GNU configure
script where it would be impossible to find. Therefore, when using pkgsrc
on NetBSD 7, it is better to switch to another shell for doing pkgsrc
work.
The comment has been updated to be more accurate.
In Makefiles, the variable values are aligned vertically. This format is
now also used in the show-all target, which makes it easier readable.
Some more variables have been marked as multi-value, and single-valued
variables ending in space are clearly marked. Without the latter, the
regression test would have a line with significant trailing whitespace.
The *_ENV and *_ARG values are typically very long, and reading them in
a single line is unnecessarily difficult. Therefore, each of their
values is listed on a separate line, for example:
fetch:
usr DIST_PATH (undefined)
pkg MASTER_SITES = \
http://ftp.gnome.org/pub/GNOME/sources/glib/2.56/ \
ftp://ftp.gnome.org/pub/GNOME/sources/glib/2.56/ \
ftp://ftp.cse.buffalo.edu/pub/Gnome/sources/glib/2.56/ \
https://download.gnome.org/sources/glib/2.56/ \
# end of MASTER_SITES
pkg DIST_SUBDIR (undefined)
Do it for all packages that
* mention perl, or
* have a directory name starting with p5-*, or
* depend on a package starting with p5-
like last time, for 5.18, where this didn't lead to complaints.
Let me know if you have any this time.
Stop lying and drop maintainership of these packages. I have not
maintained them for a very long time already, so leave room for
fresh blood to take over them.
created with the proper permissions.
XXX sysutils/install-sh isn't quite right the same as /usr/bin/install
XXX because it ignores the umask for directory creation.
build and install phase. MAKE_ENV needs to remain the same for
packages that use scons. Otherwise, in the install step, scons will
take the new MAKE_ENV to mean that the build action changed, and it will
rebuild the package with the wrong configuration and install that.
