Your message dated Sat, 06 May 2017 13:48:45 +0000
with message-id <[email protected]>
and subject line Bug#857028: fixed in debhelper 10.3
has caused the Debian Bug report #857028,
regarding debhelper: dh_gencontrol should discard Multi-Arch: no
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)


-- 
857028: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=857028
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: debhelper
Version: 10.2.5
Severity: wishlist

I have been working on fixing multiarch metadata a bit. Using very
simple checks I was able to generate some multiarch hints[1] that are
correct with a very high precision. Unfortunately, there are lots of
them. In practise every 5th package has a hint. Clearly adoption is not
going to speed up any time soon, so it looks like we have a tooling
problem.

Given that many of these Multi-Arch headers can be autogenerated with a
very small risk of false positives, it seems obvious to just let some
central piece of build infrastructure (i.e. debhelper) add them. If we
do that, there is a small risk of false markings (but much lower than
the current human marking approach). Thus we will need a way to override
these automatic headers. Overrides are simple enough: Just add the
header and debhelper shouldn't change it. There is one problem though:
If the override needs to say that there shouldn't be a Multi-Arch
header, then we currently cannot express that as "Multi-Arch: no" is
rejected by dak[2].

As a first step, Niels was proposing that debhelper's dh_gencontrol
discards 'Multi-Arch: no' given a high enough compatibility level. (Is
that compat check even necessary, given that such a header is currently
forbidden?) Thus we could start adding "Multi-Arch: no" to a few
selected packages soon and have them continue working once automatic
insertion of such headers surfaces.

Here is the patch for that. Hope you like it.

Helmut


[1] https://wiki.debian.org/MultiArch/Hints
[2] I do not know why this is still the case. If anyone figures out,
    please tell.
diff --minimal -Nru debhelper-10.2.5/debian/changelog 
debhelper-10.2.5+nmu1/debian/changelog
--- debhelper-10.2.5/debian/changelog   2017-01-25 21:47:27.000000000 +0100
+++ debhelper-10.2.5+nmu1/debian/changelog      2017-03-07 07:56:52.000000000 
+0100
@@ -1,3 +1,10 @@
+debhelper (10.2.5+nmu1) UNRELEASED; urgency=medium
+
+  * Non-maintainer upload.
+  * Remove explicit "Multi-Arch: no" stanzas in compat 11. (Closes: #-1)
+
+ -- Helmut Grohne <[email protected]>  Tue, 07 Mar 2017 07:56:52 +0100
+
 debhelper (10.2.5) unstable; urgency=medium
 
   [ Translations ]
diff --minimal -Nru debhelper-10.2.5/dh_gencontrol 
debhelper-10.2.5+nmu1/dh_gencontrol
--- debhelper-10.2.5/dh_gencontrol      2016-12-07 19:46:58.000000000 +0100
+++ debhelper-10.2.5+nmu1/dh_gencontrol 2017-03-07 07:56:52.000000000 +0100
@@ -134,9 +134,15 @@
                push(@debug_info_params, "-DBuild-Ids=${build_ids}");
        }
 
+       my (@multiarch_params);
+       # Remove explicit "Multi-Arch: no" headers to avoid autorejects by dak.
+       push (@multiarch_params, '-UMulti-Arch')
+               if (!compat(10) and package_multiarch($package) eq 'no');
+
        # Generate and install control file.
        doit("dpkg-gencontrol", "-p$package", "-l$changelog", "-T$substvars",
-               "-P$tmp", @debug_info_params, @{$dh{U_PARAMS}});
+               "-P$tmp", @debug_info_params, @multiarch_params,
+               @{$dh{U_PARAMS}});
 
        # This chmod is only necessary if the user sets the umask to
        # something odd.

--- End Message ---
--- Begin Message ---
Source: debhelper
Source-Version: 10.3

We believe that the bug you reported is fixed in the latest version of
debhelper, which is due to be installed in the Debian FTP archive.

A summary of the changes between this version and the previous one is
attached.

Thank you for reporting the bug, which will now be closed.  If you
have further comments please address them to [email protected],
and the maintainer will reopen the bug report if appropriate.

Debian distribution maintenance software
pp.
Niels Thykier <[email protected]> (supplier of updated debhelper package)

(This message was generated automatically at their request; if you
believe that there is a problem with it please contact the archive
administrators by mailing [email protected])


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Format: 1.8
Date: Sat, 06 May 2017 13:29:31 +0000
Source: debhelper
Binary: debhelper dh-systemd
Architecture: source
Version: 10.3
Distribution: experimental
Urgency: medium
Maintainer: Debhelper Maintainers <[email protected]>
Changed-By: Niels Thykier <[email protected]>
Description:
 debhelper  - helper programs for debian/rules
 dh-systemd - debhelper add-on to handle systemd unit files - transitional pack
Closes: 415396 616070 770281 795253 800043 812248 837528 844077 844602 847138 
853881 856688 857028 860091 861627
Changes:
 debhelper (10.3) experimental; urgency=medium
 .
   [ Helmut Grohne ]
   * Supply PKG_CONFIG for cross compilation with the makefile buildsystem.
     (Closes: #853881)
   * Disable stripping during dh_auto_* in makefile buildsystem in compat 11.
     (Closes: #844077)
   * Remove explicit "Multi-Arch: no" stanzas as they are auto-rejected.
     (Closes: #857028)
 .
   [ Niels Thykier ]
   * Dh_Lib.pm: Make getpackages() produce the correct result independently
     of the order of the fields in a given paragraph of debian/control.
     (Closes: #847138)
   * dh_installman: Report installed manpages so the new dh_missing tool
     is informed about them.
   * dh_install: Deprecate --list-missing/--fail-missing in favor of the
     new dh_missing tool.  The options will be removed in compat 11.
   * dh: Run dh_missing by default (in no op mode).
   * dh_prep: Clean up generated files so they work like other temporary
     debhelper files.
   * dh_installinit: Clarify that it might make sense to skip dh_installinit
     for a package if it provides a systemd service but no sysvinit file.
     (Closes: #800043)
   * dh_installinit: Deprecate --no-restart-on-upgrade in favor of the new
     name --no-stop-on-upgrade, which does the same thing but is less
     likely to be confused with --no-restart-after-upgrade and is more
     descriptive of what it actually does.  Thanks to  Simon McVittie and
     Michael Biebl for the help.  (Closes: #837528)
   * debian/rules: Only apply the --no-parallel to dh_auto_test.  The rest
     of debhelper's build appears to work fine with --parallel.
   * dh_systemd_enable.1: Clarify that --no-enable does not control
     whether a service is started and that dh_systemd_start is the tool
     that does that.  (Closes: #770281)
   * autoconf.pm: Set --runstatedir to "/run" in compat 11.  Thanks to
     Michael Biebl for the suggestion.
   * dh_fixperms: Refactor to reduce the constant overhead a bit and
     extract some patterns into data list.  The latter hopefully reducing
     the need for tweaking dh_fixperms' logic.
   * dh_installdocs: Document that --link-doc has limitations on source and
     target package plus that compat 10 will reject unsupported
     combinations.
   * dh_gencontrol: Avoid adding Homepage fields to dbgsym packages.  Avoids
     non-overridable warnings from lintian when there is an issue in the
     homepage field of the original binary.  (Closes: #812248)
   * dh_strip: Fix a check that could cause dh_strip to generate dbgsym
     packages when it should not.
   * dh: Stop loading python-support in all compat levels as dh_pysupport
     has been removed in stretch.
   * python_support.pm: Remove the content with a deprecation warning as
     the module can no longer do anything.
   * dh_installman.1: Improve the manpage to clarify the scope of
     dh_installman and provide an examples.  (Closes: LP: #184156)
   * makefile.pm: Disable parallel installation if libtool is present to
     avoid a race-condition in automake when using libtool.
     (Closes: #861627)
 .
   [ Steven Chamberlain ]
   * When cross-compiling for GNU/kFreeBSD host systems, set
     CMAKE_SYSTEM_NAME to the correct value "kFreeBSD", not "FreeBSD".
     (Closes: #856688)
 .
   [ Felipe Sateler ]
   * dh_systemd_enable.1: Improve the wording of the documentation for
     the --no-enable option.  (Closes: #844602)
 .
   [ Michael Biebl ]
   * Add support for meson + ninja build systems.  (Closes: #795253)
 .
   [ Michael Stapelberg ]
   * dh_missing: New tool to handle --list-missing/--fail-missing.
     (Closes: #415396, #616070)
   * Dh_Lib.pm: Introduce a log_installed_files sub that can be used
     to inform dh_missing that a given file has been installed.
 .
   [ Johannes Schauer ]
   * dh_link.1: Improve the documentation and use ln(1) as a reference
     to make it simpler for people, who are used to the terms from ln.
     (Closes: #860091)
Checksums-Sha1:
 17cc7410f0f79ce9e289514d0f6c34409d8dc2cd 1725 debhelper_10.3.dsc
 2d80c50b69227ae6d252f4044eb20cc53bff0d79 398764 debhelper_10.3.tar.xz
 a04a9ea8b3896c6f504e7cc7c994fe11cef7aa23 4589 debhelper_10.3_source.buildinfo
Checksums-Sha256:
 38451d6b4db456a3d0ce1e0a27e81829f61ad1b47ec45643c5294aaa6999b150 1725 
debhelper_10.3.dsc
 d17c6d1d672113cc869b079015517034516238ec83c447e3be39dd9f10ebf32c 398764 
debhelper_10.3.tar.xz
 1c7885e72e7fea2a4ee61f937cbbafbceb3ca798d314f903ec469f6048106386 4589 
debhelper_10.3_source.buildinfo
Files:
 5105d8877292c9ab3b93642e3ef911dd 1725 devel optional debhelper_10.3.dsc
 83e5e673b4876de586d937c2ea2ec290 398764 devel optional debhelper_10.3.tar.xz
 3739033d54fe541404b3f7de04c2ea8a 4589 devel optional 
debhelper_10.3_source.buildinfo

-----BEGIN PGP SIGNATURE-----
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=OcJH
-----END PGP SIGNATURE-----

--- End Message ---

Reply via email to