Package: debhelper Version: 10.2.3 Severity: important Control: block 851041 by -1 Control: fixed -1 10.2.2 Control: affects -1 dh-elpa
Dear maintainer,
dh-elpa's dh sequencer file does this:
insert_after("dh_auto_test", "dh_elpa_test");
remove_command("dh_auto_test");
At compat 10 and with debhelper 10.2.2, this caused dh_elpa_test to be
executed once, during the `dh build` sequence.
However, with debhelper 10.2.3, there is an additional call to
dh_elpa_test during the `dh binary` sequence. Upstream test suites are
often unprepared to deal with fakeroot, and at least one package using
dh-elpa is FTBFSing as a result.
Is there a good reason why package test suites are now run under
fakeroot? More generally, is there some reason why they are now run
twice? When designing dh_elpa_test, we chose to require compat 10
precisely to avoid any test suites being run under fakeroot, because we
already knew that at least one of our packages couldn't handle it. So
this seems like a regression to the behaviour we observed at compat 9.
Thank you for your attention!
-- System Information:
Debian Release: stretch/sid
APT prefers testing
APT policy: (900, 'testing')
Architecture: i386 (i686)
Kernel: Linux 4.8.0-1-686-pae (SMP w/2 CPU cores)
Locale: LANG=en_GB.utf8, LC_CTYPE=en_GB.utf8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
Versions of packages debhelper depends on:
ii autotools-dev 20161112.1
ii binutils 2.27.51.20161212-1
ii dh-autoreconf 12
ii dh-strip-nondeterminism 0.028-1
ii dpkg 1.18.10
ii dpkg-dev 1.18.10
ii file 1:5.29-1
ii libdpkg-perl 1.18.10
ii man-db 2.7.5-2
ii perl 5.24.1~rc4-1
ii po-debconf 1.0.20
debhelper recommends no packages.
Versions of packages debhelper suggests:
ii dh-make 2.201608
-- no debconf information
--
Sean Whitton
signature.asc
Description: PGP signature
