| Commit message (Collapse) | Author | Age |
|---|
| | |
|
| |
|
|
| |
still uses them.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
2 packages in the archive still use it, but both like this:
ifneq ($(shell dh_testversion '<<' 7.4.10 2>/dev/null && echo old),old)
dh_parallel = --parallel
endif
Happily, the way that was written, if the command doesn't exist,
it does the right thing and enables the behavior for the new debhelper.
Not that it would greatly matter if it did not, since --parallel is not
crucial.
|
| |
|
|
| |
it generates shlibs files for. This means that -X can be used to exclude libraries from processing by dpkg-gensymbols. It also means that libraries in unusual locations, where dpkg-gensymbols does not itself normally look will be passed to it, a behavior change which may break some packages. Closes: #557603
|
| |
|
|
| |
commands. (Unknown options in DH_OPTIONS still only result in warnings.)
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
#548382
|
| | |
|
| |
|
|
| |
after -- is much more sane. (However, -u will not go away any time soon.) Closes: #554509
|
| |
|
|
|
|
|
| |
The -a flag now does the same thing as the -s flag, so debhelper users do
not need to worry about using the -s flag when building a package that only
builds for some architectures, and dh will also work in that situation.
Closes: #540794
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
uses them.
Affected options are: -x, -r, -R, -l, -L, -m, --include-conffiles,
--no-restart-on-upgrade, --no-start, --restart-after-upgrade,
--init-script, --filename, --flavor, --autodest, --libpackage, --add-udeb,
--dpkg-shlibdeps-params, --dpkg-gencontrol-params, --update-rcd-params,
--major, --remove-d, --dirs-only, --keep-debug, --version-info,
--list-missing, --fail-missing, --language, --until, --after, --before,
--remaining, --with
* If any third-party debhelper commands use any of the above options,
they will be broken, and need to be changed to pass options to init().
* To avoid breaking rules files that pass options to commands that do not
use them, debhelper will now only warn if it encounters an unknown
option. This will be converted back to an error later.
|
| |
|
|
| |
passing a hash to init(), which is later passed on the Getopt::Long. Closes: #370823
|
| | |
|
| |
|
|
| |
in the prerm, and instead restarts it in the postinst, keeping its downtime minimal. Since some daemons could break if files are upgraded while they're running, it's not the default. It might become the default in a future (v8) compatability level. Closes: #471060
|
| |
|
|
| |
later).
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
preinst and postinst scripts. For example, if a package uses
dh_pysupport before dh_installinit, the prerm will first stop the init
script and then remove the python files.
* Introducing beginning of v6 mode.
* dh_installwm: In v6 mode, install a slave manpage link for
x-window-manager.1.gz. Done in v6 mode because some window managers
probably work around this longstanding debhelper bug by registering the
slave on their own. This bug was only fixable once programs moved out of
/usr/X11R6. Closes: #85963
* dh_builddeb: In v6 mode, fix bug in DH_ALWAYS_EXCLUDE handling, to work
the same as all the other code in debhelper. This could only be fixed in
v6 mode because packages may potentially legitimately rely on the old
buggy behavior. Closes: #242834
* dh_installman: In v6 mode, overwrite exsiting man pages. Closes: #288250
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
best of
a difficult situation.
* Revert all dh_python changes. Closes: #381389, #378604
* Conflict with python-support << 0.5.3 and python-central << 0.5.5.
* Make dh_python do nothing if debian/pycompat is found.
The new versions of dh_pysupport or dh_pycentral will take care of
everything dh_python used to do in this situation.
* dh_python is now deprecated. Closes: #358392, #253582, #189474
* Non-maintainer upload.
* Update of dh_python
- when buidling for a non-standard Python version, generate more
reasonable Depends like "python (>= X.Y) | pythonX.Y"
Closes: #375576
- fix handling of private extensions. Closes: #375948
- fix parsing of XS-Python-Version, it didn't work if only fixed versions
were listed in XS-Python-Version.
- fix use of unitialized value. Closes: #374776
- fix typos in POD documentation. Closes: #375936
* Non-maintainer upload.
* Update of dh_python
- vastly refactored, easier to understand, and the difference
between old policy and new policy is easier to grasp
- it supports an -X option which can be used to not scan some files
- uses debian/pyversions as reference source of information for
dependencies but also parse the XS-Python-Version header as fallback.
- ${python:Versions}'s default value is XS-Python-Version's value
instead of "all" when the package doesn't depend on a
specific python version. Closes: #373853
- always generate ${python:Provides} and leave the responsibility to the
maintainer to not use ${python:Provides} if he doesn't want the
provides.
- uses debian/pycompat or DH_PYCOMPAT as reference field to run in new
policy mode. The presence of XS-Python-Version will also trigger the
new policy mode (this is for short-term compatibility, it may be removed in
the not too-distant future).
DH_PYCOMPAT=1 is the default mode and is compatible to the old policy.
DH_PYCOMPAT=2 is the new mode and is compatible with the new policy.
* Use "grep ^Version:" instead of "grep Version:" on the output of
dpkg-parsechangelog since the above changelog entry matched "Version:" and
thus made the build fail.
* Non-maintainer upload.
* Integrate the new dh_python implementing the new Python policy. Closes: #370833
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
schemas from /etc/gconf/schemas. (Josselin Mouette)
* dh_gconf: kill gconfd-2 so that the newly installed schemas
are available straight away. (Josselin Mouette)
* dh_gconf: fix bashism in restart of gconfd-2
* dh_gconf: fix innaccuracy in man page; gconfd-2 is HUPPed, not
killed.
* dh_scrollkeeper: stop adding scrollkeeper to misc:Depends, since
the postinst will not run it if it's not installed, and a single run after
it's installed is sufficient to find all documents. Closes: #256745
* dh_fixperms: make .ali files mode 444 to prevent recompilation by GNAT.
For speed, only scan for .ali files in usr/lib/ada. Closes: #245211
* dh_python: check to make sure compileall.py is available before running it
in the postinst. Closes: #253112
* dh_installmodules: install debian/package.modprobe into etc/modprobe.d/
for module-init-tools. These files can sometimes need to differ from the
etc/modutils/ files. Closes: #204336, #234495
* dh_installmanpages is now deprecated.
* Add a test case for bug #244157, and fixed the inverted ok() parameters
in the others, and added a few new tests.
* dh_link: applied GOTO Masanori's patch to fix conversion of existing
relative symlinks between top level directories. Closes: #244157
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Closes: #112950
* dh_compress: deal with links pointing to links pointing to compressed
files, no matter what order find returns them. Closes: #204169
* dh_installmodules, dh_installpam, dh_installcron, dh_installinit,
dh_installogrotate: add --name= option, that can be used to specify
the name to use for the file(s) installed by these commands. For example,
dh_installcron --name=foo will install debian/package.foo.cron.daily to
etc/cron.daily/foo. Closes: #138202, #101003, #68545, #148844
(Thanks to Thomas Hood for connecting these bug reports.)
* dh_installinit: deprecated --init-script option in favor of the above.
* Add dh_installppp. Closes: #43403
|
| |
|
|
|
| |
* dh_link: Improve error message if link destination is a directory.
Closes: #206689
|
| | |
|
| |
|
|
| |
undocumented.7 links anymore.
|
| |
|
|
| |
Closes: #163128
|
| |
|
|
|
|
|
|
|
|
| |
* dh_strip: better documentation, removed extraneous "item" from SYNOPSIS.
Closes: #162493
* dh_strip: detect and don't strip debug/*.so files.
* Note that 4.1.11 changelog entry was incorrect, dh_perl worked fine
without that change, but the new behavior is less likely to break things
if dpkg-gencontrol changes.
* Various improvements to debhelper(1).
|
| |
|
|
|
| |
Do not use this verion of debhelper for woody backports!
* Removed dh_installxaw.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
from debstd, its interface and implementation suck, and I have maintained
it while never really deigning to use it. Now there is a remplacment:
dh_install, which ...
- copies files, doesn't move them. Closes: #75360, #82649
- doesn't have that whole annoying debian/package.files vs. debian/files
mess, as it uses debian/install.
- supports copying empty subdirs. Closes: #133037
- doesn't use tar, thus no error reproting problems. Closes: #112538
- files are listed relative to the pwd, debian/tmp need not be used at
all, so no globbing issues. Closes: #100404
- supports -X. Closes: #116902
- the whole concept of moving files out of a directory is gone, so this
bug doesn't really apply. Closes: #120026
- This is exactly what Bill Allombert asked for in #117383, even though I
designed it seemingly independantly. Thank you Bill! Closes: #117383
* Made debhelper's debian/rules a lot simpler by means of the above.
* Updated example rules file to use dh_install. Also some reordering and
other minor changes.
* dh_movefiles is lightly deprecated, and when you run into its bugs and
bad design, you are incouraged to just use dh_install instead.
* dh_fixperms: in v4 only, make all files in bin/ dirs +x. Closes: #119039
* dh_fixperms: in v4 only, make all files in etc/init.d executable (of
course there's -X ..)
* dh_link: in v4 only, finds existing, non-policy-conformant symlinks
and corrects them. This has the side effect of making dh_link idempotent.
* Added a -h/--help option. This seems very obvious, but it never occured to
me before..
* use v4 for building debhelper itself
* v4 mode is done, you may now use it without fear of it changing.
(This idea of this upload is to get v4 into woody so people won't run into
many issues backporting from sarge to woody later on. Packages targeted
for woody should continue to use whatever compatability level they are
using.)
|
| | |
|
