mirabilos [Sun, 31 Jan 2021 17:17:59 +0000 (18:17 +0100)]
comment out procps, but note it; prepare for uploading
mirabilos [Sun, 31 Jan 2021 15:17:24 +0000 (16:17 +0100)]
.doc fails to be viewed in mc which thinks it’s a Microsoft product
mirabilos [Sun, 31 Jan 2021 15:33:13 +0000 (16:33 +0100)]
Merge branch 'mirbsd'
mirabilos [Sun, 31 Jan 2021 15:29:08 +0000 (16:29 +0100)]
Update to MirBSD CVS HEAD, final 0AB9.1 patchlevel
mirabilos [Sat, 30 Jan 2021 07:52:47 +0000 (08:52 +0100)]
re-add procps to run sanity
mirabilos [Sat, 30 Jan 2021 05:23:03 +0000 (06:23 +0100)]
Merge branch 'mirbsd'
mirabilos [Sat, 30 Jan 2021 05:21:25 +0000 (06:21 +0100)]
update to MirBSD CVS HEAD
mirabilos [Sat, 30 Jan 2021 04:50:12 +0000 (05:50 +0100)]
Merge branch 'mirbsd'
mirabilos [Sat, 30 Jan 2021 04:45:55 +0000 (05:45 +0100)]
update to latest CVS HEAD
mirabilos [Sat, 30 Jan 2021 03:04:13 +0000 (04:04 +0100)]
better description
mirabilos [Sat, 30 Jan 2021 03:01:27 +0000 (04:01 +0100)]
Merge branch 'mirbsd'; keep lib/error.c
mirabilos [Sat, 30 Jan 2021 02:54:21 +0000 (03:54 +0100)]
update from latest MirBSD CVS HEAD
mirabilos [Sat, 30 Jan 2021 01:44:23 +0000 (02:44 +0100)]
Merge branch 'for-mirbsd'
mirabilos [Sat, 30 Jan 2021 01:44:15 +0000 (02:44 +0100)]
we have TWO error()?
mirabilos [Sat, 30 Jan 2021 01:35:54 +0000 (02:35 +0100)]
rename fperrmsg to fpwarnmsg and remove its ability to exit
mirabilos [Sat, 30 Jan 2021 01:19:13 +0000 (02:19 +0100)]
Merge branch 'for-mirbsd'
mirabilos [Sat, 30 Jan 2021 01:18:58 +0000 (02:18 +0100)]
update copyright and other metadata
mirabilos [Sat, 30 Jan 2021 00:43:44 +0000 (01:43 +0100)]
add explicit prev/next/up info to some nodes:
the node “CVS command list” contains an @menu referencing “all” CVS
commands; a @menu is normally a navigation item though, and most of
them (e.g. “admin”) are normally an @node in the @menu CVS commands
(some are an @anchor only), and a node cannot live in two (or more)
menus, makeinfo’s linking will break, but hardcoding the correct(!)
refs manually makes it work
mirabilos [Sat, 30 Jan 2021 00:22:10 +0000 (01:22 +0100)]
make exit(3) available to the macro
mirabilos [Sat, 30 Jan 2021 00:22:05 +0000 (01:22 +0100)]
oops!
mirabilos [Sat, 30 Jan 2021 00:20:21 +0000 (01:20 +0100)]
update
(though not yet on the last point)
mirabilos [Sat, 30 Jan 2021 00:16:54 +0000 (01:16 +0100)]
Merge branch 'for-mirbsd'
mirabilos [Sat, 30 Jan 2021 00:16:48 +0000 (01:16 +0100)]
try to plug Underfull \hbox
mirabilos [Sat, 30 Jan 2021 00:12:49 +0000 (01:12 +0100)]
try to plug the Overfull \hbox
mirabilos [Fri, 29 Jan 2021 23:45:06 +0000 (00:45 +0100)]
there’s printfs out there that don’t handle nil for %s‽
mirabilos [Fri, 29 Jan 2021 23:41:55 +0000 (00:41 +0100)]
Revert "plug a fortify warning I don’t get, (size_t)-1 is checked above‽"
because the compiler couldn’t know it was exiting in the -1 case ☹
see commit
02b2c6f58dfc78a454d3e46bf3337eaf1f1d1d3f
This reverts commit
99e6916ed5dfc68688581cb727ff479930a54083.
mirabilos [Fri, 29 Jan 2021 23:38:56 +0000 (00:38 +0100)]
try to fix up the GNU error() API as much as possible:
they define an API error() whose first argument is the errorlevel
to exit with, BUT IF IT’S 0 IT DOES NOT EXIT! so it cannot be
marked __dead, throwing dead/alive branch analysis fully out of
the range of possibilities, leading to subsequent errors, such
as the one in commit
99e6916ed5dfc68688581cb727ff479930a54083 ☹
for CVS, rename the function to warning, drop its first argument,
introduce an error macro that does the if(status)exit(status)
dance, and don’t worry too much about constant arguments and rely
on modern compilers instead ☹
but nooooo <err.h> is not good enough for the GNU folks‽
mirabilos [Fri, 29 Jan 2021 23:14:10 +0000 (00:14 +0100)]
some glibc thing I don’t understand
mirabilos [Fri, 29 Jan 2021 23:09:16 +0000 (00:09 +0100)]
plug a fortify warning I don’t get, (size_t)-1 is checked above‽
mirabilos [Fri, 29 Jan 2021 22:56:06 +0000 (23:56 +0100)]
GNU has this weird idiom strncpy(dst, src, strlen(src)); fix it:
This is used with dynamically allocated buffers of at least the
length the third argument is and I’m wondering whether that’s
someone unknowing’s answer to being told strcpy is a no-no…
memcpy is, of course, the correct answer, in all these cases
(checked against the allocation size only, of course)
mirabilos [Fri, 29 Jan 2021 22:52:01 +0000 (23:52 +0100)]
quell ar(1) warnings
mirabilos [Fri, 29 Jan 2021 22:43:36 +0000 (23:43 +0100)]
this is the GNU version of “goto fail;” I guess; thankfully no effect
found by -Wmultistatement-macros in an else branch, but the
then branch was a goto already so no bug
mirabilos [Fri, 29 Jan 2021 22:38:13 +0000 (23:38 +0100)]
fix a Perl 5.32 error
mirabilos [Fri, 29 Jan 2021 12:36:01 +0000 (13:36 +0100)]
a round of lintian updates
mirabilos [Fri, 29 Jan 2021 12:24:09 +0000 (13:24 +0100)]
move a documentation file to please lintian’s doc-base checks
mirabilos [Fri, 29 Jan 2021 12:15:54 +0000 (13:15 +0100)]
bump version, otherwise lintian will complain
mirabilos [Fri, 29 Jan 2021 12:15:46 +0000 (13:15 +0100)]
fix whitespace
mirabilos [Fri, 29 Jan 2021 12:14:09 +0000 (13:14 +0100)]
also drop pre-wheezy Depends, Replaces/Provides/Conflicts, versioning in Suggests
Helmut Grohne [Fri, 29 Jan 2021 12:11:13 +0000 (13:11 +0100)]
reduce Build-Depends
mirabilos [Fri, 29 Jan 2021 12:08:08 +0000 (13:08 +0100)]
drop pre-epoch-2 maintainer script update code
mirabilos [Fri, 29 Jan 2021 12:06:34 +0000 (13:06 +0100)]
begin modernising packaging and review it
mirabilos [Tue, 5 Feb 2019 18:31:25 +0000 (19:31 +0100)]
update lintian overrides
mirabilos [Tue, 5 Feb 2019 18:28:46 +0000 (19:28 +0100)]
appease an idiotic, no-worth only-trouble-and-effort, lintian change
mirabilos [Tue, 5 Feb 2019 18:19:31 +0000 (19:19 +0100)]
prepare upload of new Debian patchlevel
mirabilos [Tue, 5 Feb 2019 18:18:09 +0000 (19:18 +0100)]
bump Policy
mirabilos [Tue, 5 Feb 2019 18:12:56 +0000 (19:12 +0100)]
hardcode /bin/mktemp for reproducible builds, usrmerge’s fault
mirabilos [Thu, 2 Aug 2018 22:14:17 +0000 (00:14 +0200)]
this is now required by Policy 4.2.0.0
mirabilos [Fri, 5 Jan 2018 19:07:57 +0000 (20:07 +0100)]
misc. updates
mirabilos [Fri, 5 Jan 2018 19:07:45 +0000 (20:07 +0100)]
fix reference
mirabilos [Sun, 19 Nov 2017 17:11:17 +0000 (18:11 +0100)]
version tag
mirabilos [Sun, 19 Nov 2017 17:04:58 +0000 (18:04 +0100)]
address lintian’s messages
mirabilos [Sun, 19 Nov 2017 12:47:21 +0000 (13:47 +0100)]
bump, S-V too
mirabilos [Sun, 19 Nov 2017 12:44:10 +0000 (13:44 +0100)]
Merge branch 'mirbsd'
mirabilos [Sun, 19 Nov 2017 12:40:46 +0000 (13:40 +0100)]
sync from MirBSD
mirabilos [Sat, 12 Aug 2017 20:18:56 +0000 (22:18 +0200)]
no autopkgtest, for now
mirabilos [Sat, 12 Aug 2017 19:29:46 +0000 (21:29 +0200)]
Merge branch 'mirbsd'
mirabilos [Sat, 12 Aug 2017 19:28:46 +0000 (21:28 +0200)]
update from MirBSD
mirabilos [Sat, 12 Aug 2017 01:48:00 +0000 (03:48 +0200)]
oops…
mirabilos [Sat, 12 Aug 2017 01:22:51 +0000 (03:22 +0200)]
dh_installdocs is n̲o̲t nodocs-clean…
mirabilos [Sat, 12 Aug 2017 01:10:29 +0000 (03:10 +0200)]
Merge branch 'mirbsd'
mirabilos [Sat, 12 Aug 2017 01:09:08 +0000 (03:09 +0200)]
update due to lintian
mirabilos [Sat, 12 Aug 2017 01:05:40 +0000 (03:05 +0200)]
Update lintian overrides, round 1
mirabilos [Sat, 12 Aug 2017 00:59:27 +0000 (02:59 +0200)]
lintian: autotools-dev is now included in debhelper
mirabilos [Sat, 12 Aug 2017 00:59:13 +0000 (02:59 +0200)]
update to latest Policy
mirabilos [Sat, 12 Aug 2017 00:38:34 +0000 (02:38 +0200)]
Merge branch 'mirbsd'
mirabilos [Sat, 12 Aug 2017 00:33:32 +0000 (02:33 +0200)]
update from MirBSD CVS
fixes CVE-2017-12836 (Debian #871810) but may break the testsuite yet
mirabilos [Fri, 28 Apr 2017 19:33:43 +0000 (21:33 +0200)]
false positive “ot”
mirabilos [Fri, 28 Apr 2017 18:35:18 +0000 (20:35 +0200)]
arrgh!
mirabilos [Fri, 28 Apr 2017 17:12:24 +0000 (19:12 +0200)]
bump
mirabilos [Fri, 28 Apr 2017 17:03:41 +0000 (19:03 +0200)]
Merge branch 'mirbsd'
mirabilos [Fri, 28 Apr 2017 16:58:26 +0000 (18:58 +0200)]
update from MirBSD CVS
mirabilos [Tue, 28 Mar 2017 18:36:48 +0000 (20:36 +0200)]
bump
mirabilos [Tue, 28 Mar 2017 18:19:12 +0000 (20:19 +0200)]
Merge branch 'mirbsd'
mirabilos [Tue, 28 Mar 2017 18:19:06 +0000 (20:19 +0200)]
missing pick
mirabilos [Tue, 28 Mar 2017 18:02:16 +0000 (20:02 +0200)]
document
mirabilos [Tue, 28 Mar 2017 17:49:25 +0000 (19:49 +0200)]
Merge branch 'mirbsd'
mirabilos [Tue, 28 Mar 2017 17:47:46 +0000 (19:47 +0200)]
reduce diff against previous version to only fixing #858769
mirabilos [Mon, 27 Mar 2017 16:57:57 +0000 (18:57 +0200)]
update from MirBSD CVS HEAD
mirabilos [Mon, 9 Jan 2017 23:35:42 +0000 (00:35 +0100)]
Revert "cvs-dbgsym can be xz compressed (default)", it does not work (WTF?!)
[…]
dh_md5sums
dh_builddeb -pcvs -- -Zgzip -z9
dpkg-deb: building package 'cvs-dbgsym' in '../cvs-dbgsym_1.12.13+real-21_amd64.deb'.
dpkg-deb: building package 'cvs' in '../cvs_1.12.13+real-21_amd64.deb'.
dh_builddeb --remaining-packages
dpkg-deb: building package 'cvs-dbgsym' in '../cvs-dbgsym_1.12.13+real-21_amd64.deb'.
dpkg-deb: building package 'cvs' in '../cvs_1.12.13+real-21_amd64.deb'.
dpkg-genbuildinfo
[…]
This reverts commit
9d2034eb02f2627bd7bccb8ee66c35a30ba2c011.
mirabilos [Mon, 9 Jan 2017 23:27:59 +0000 (00:27 +0100)]
cvs-dbgsym can be xz compressed (default)
mirabilos [Mon, 9 Jan 2017 23:19:56 +0000 (00:19 +0100)]
sanity is attempted to be run in parallel, which we can’t have
mirabilos [Mon, 9 Jan 2017 23:07:27 +0000 (00:07 +0100)]
release (provided it passes the sanity testsuthe)
mirabilos [Mon, 9 Jan 2017 23:06:43 +0000 (00:06 +0100)]
Merge branch 'mirbsd', mostly reverting commit
a4bdd3373a67c89fb350e619958b5a7e144c9c3b
mirabilos [Mon, 9 Jan 2017 23:01:52 +0000 (00:01 +0100)]
bring back most of what we reverted in master in
a4bdd3373a67c89fb350e619958b5a7e144c9c3b
mirabilos [Mon, 9 Jan 2017 22:51:21 +0000 (23:51 +0100)]
revert all local changes as we’ll reimport them via the mirbsd branch
mirabilos [Mon, 9 Jan 2017 22:34:32 +0000 (23:34 +0100)]
some tweaks to make it work… somewhat?
mirabilos [Mon, 9 Jan 2017 22:24:00 +0000 (23:24 +0100)]
maybe this is intelligent enough to #include_next automatically…
… but then, this is Teχ, so… probably not.
mirabilos [Mon, 9 Jan 2017 22:23:46 +0000 (23:23 +0100)]
let’s get wild…
mirabilos [Mon, 9 Jan 2017 22:21:22 +0000 (23:21 +0100)]
pdf14.sty is for LaTeX only ☹
mirabilos [Mon, 9 Jan 2017 18:34:32 +0000 (19:34 +0100)]
PDF/1.4 (try #4) and reorder dependencies:
• texlive-latex-recommended already included texlive-latex-base
• texlive-latex-extra includes texlive-latex-recommended
• autopoint was apparently unnecessary all the time…
mirabilos [Mon, 9 Jan 2017 18:28:21 +0000 (19:28 +0100)]
it’s either this, or B-D on texlive-latex-extra…
mirabilos [Mon, 9 Jan 2017 18:27:12 +0000 (19:27 +0100)]
not LaTeX, this is…
mirabilos [Mon, 9 Jan 2017 18:05:06 +0000 (19:05 +0100)]
ps2pdf currently generates PDF/1.4, so we’ll standardise on that
mirabilos [Mon, 9 Jan 2017 18:02:42 +0000 (19:02 +0100)]
further experiment with PA4 paper size
mirabilos [Mon, 9 Jan 2017 17:51:39 +0000 (18:51 +0100)]
experiment with PA4 paper size
mirabilos [Mon, 9 Jan 2017 17:46:55 +0000 (18:46 +0100)]
update
mirabilos [Mon, 9 Jan 2017 17:10:12 +0000 (18:10 +0100)]
Merge branch 'mirbsd'
mirabilos [Mon, 9 Jan 2017 16:27:12 +0000 (17:27 +0100)]
import latest CVS from MirBSD, most specifically the Savannah feature
mirabilos [Sun, 27 Nov 2016 14:47:44 +0000 (15:47 +0100)]
always use --build=/--host=; config.guess can be too smart otherwise
(see trying to build jupp against dietlibc, where the detected triplet
differs and does not work)
mirabilos [Wed, 9 Nov 2016 03:17:23 +0000 (04:17 +0100)]
Merge branch 'mirbsd'