Age | Commit message (Collapse) | Author | Files | Lines |
|
dpkg-architecture ignores them and instead computes the expected value
(as if the variable had not been set in the first place).
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
fixup! Dh_Lib: Add on_each_pkg_in_parallel
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
This was accidentally removed when optimising a common case.
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Apparently, libtool+automake has a race condition in some cases, where
automake will try to relink a binary about be installed before the
library is installed. The bug is many years old and apparently not a
priority, so work around it in debhelper.
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
There is no point in loading a sequence that conditionally adds a tool
that no longer exists. Just give a deprecation warning for people
still explicitly loading it and drop the rest.
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Suggested-by: Michael Biebl <biebl@debian.org>
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
With "enough" packages, pkgfile is still a very expensive part of dh.
Mostly due to the can_skip check which often need to call pkgfile to
determine if a helper can be skipped. By doing a "bulk check" we can
reduce that overhead by about roughly 50% (20-25% of dh's total
runtime).
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
If we need to use Getopt::Long, we pay a ~0.020s overhead (per
helper). When doing dpkg-buildpackage -A/-B builds, dh will
unconditionally pass -i/-a in those cases making each helper cost a
bit more.
With this change, we short-cut -i and -a (but not their long variants)
so Dh_Lib can handle those directly. This reduces the overhead for
-A/-B so they are comparable to -b.
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Each time pkgfile is called, it will glob the debian/ dir (with a
filter). Obviously this adds up, but unfortunately, glob is a lot
cheaper in some cases than calling buildarch()/buildos(), which
requires Dpkg::Arch.
To solve this, we implement a cache of the globbing so we at most glob
once per file name. Depending on that result, we will now
unconditionally check for architecture qualified files or not.
This should make both cases faster for 2+ packages. It should also
apply to dh when checking if it can skip a command. When testing
about 500 empty "transitional" packages, we save about 7-8% run time
for dh_install.
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Minor tweak: Also remove the trailing "\s*" which is unnecessary as
there is a s/\s+$// just prior to all of these ifs.
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
That way, the generated files behave like other temporary debhelper
files (which is what they are designed to be).
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
This commit splits dh_install’s --list-missing and --fail-missing
functionality into the separate helper dh_missing.
To make this work, dh_install logs its source patterns and dh_missing
reads these patterns, treating them as installed. This allows us to
address #415396, i.e. recognize files installed by other helpers (e.g.
dh_installman) as installed.
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Meson is a new build system which is supposed to be extremly fast and,
even more importantly, as user friendly as possible.
It uses Ninja as backend to manage the actual build process.
Closes: #795253
Signed-off-by: Michael Biebl <biebl@debian.org>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Call dpkg-architecture at most once (per process) - even if
dpkg_architecture_value() is passed unknown variables.
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
|
|
When adding the makefile buildsystem cross variables, the intention was
that it would not affect non-makefile buildsystems (in particular no
downstream buildsystems). However, the decision which buildsystem to use
is done on a per-target basis. Thus a typical autoconf package will use
the autoconf buildsystem for configure and test, but fall back to the
makefile buildsystem for clean and build. Thus the cross variables were
added for autoconf build as well, which broke the cross build of icu.
The solution chosen here is to have autoconf take over build and clean
from makefile as well by inheriting its methods. Thus the semantics stay
unchanged with the exception of not adding the cross variables for
build.
All other children of the makefile buildsystem (including mkcmake)
already take over all targets, so this issue really only affects
autoconf.
Fixes: 7ea67c9aace4 ("makefile.pm: Set CC+CXX to the host compilers when
cross-building")
Signed-off-by: Helmut Grohne <helmut@subdivi.de>
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
They were not needed in the first place (as "install -d" DTRT).
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Co-Author: Helmut Grohne <helmut@subdivi.de>
Signed-off-by: Niels Thykier <niels@thykier.net>
|
|
Co-Author: Helmut Grohne <helmut@subdivi.de>
Signed-off-by: Niels Thykier <niels@thykier.net>
|