- Arch-dev-public - lists.archlinux.org

[arch-dev-public] [devtools] [PATCH] install option (-I) does not require a directory containing a PKGBUILD
by Andrea Scarpino 11 Oct '10

11 Oct '10

When you install a package in a chroot using makechrootpkg -I, the script checks for a PKGBUILD in the current dir. This patch skip this check. >From 00009de3b189660e8d9729f9aead0c9a75bb22aa Mon Sep 17 00:00:00 2001 From: Andrea Scarpino <andrea(a)archlinux.org> Date: Mon, 11 Oct 2010 09:54:16 +0200 Subject: [PATCH] install option (-I) does not require a directory containing a PKGBUILD --- makechrootpkg | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/makechrootpkg b/makechrootpkg index 572a215..65db5c5 100755 --- a/makechrootpkg +++ b/makechrootpkg @@ -87,7 +87,7 @@ if [ "$EUID" != '0' ]; then exit 1 fi -if [ ! -f PKGBUILD ]; then +if [ ! -f PKGBUILD -a -z "$install_pkg" ]; then echo 'This must be run in a directory containing a PKGBUILD.' exit 1 fi -- 1.7.3.1 -- andreascarpino.it Arch Linux Developer

1 0

[arch-dev-public] new pkg for extra: LXDM
by Andreas Radke 10 Oct '10

10 Oct '10

LXDM had a nice stable release 0.3.0 a few weeks ago and is ready to use. It's the best light weight alternative we have next to gdm/kdm and slim/xdm. I know about Jürgens ask for help on lxde packages but I'd like to maintain lxdm in extra independent from lxde as an option for Xfce users. Any objections? -Andy

2 2

[arch-dev-public] tar based sync backend implemented!
by Allan McRae 10 Oct '10

10 Oct '10

Hi, I have implemented the reading directly from the db tarball for sync databases. See the patchset on by "backend" branch (http://projects.archlinux.org/users/allan/pacman.git/log/?h=backend) Some of the changes are quite big (e.g. 9 files changed, 2184 insertions, 1571 deletions), but that is mainly moving and duplicating code in the process of splitting the handling of local and sync dbs. There is still _a lot_ of clean-up to be done in the code, but it is fully functional. This breaks pactests... lots of pactests... but that is expected as the pactest suite does not understand a tar based sync db. I would like help fixing that as I get lost in the pactest code. Anyway, here is the timing for some operations. First time is with dropped caches, the second running it again with the db cached. pacman -Syy (download time ~15-20sec) old: 1m25.197s, 0m22.445s new: 0m19.391s, 0m22.084s (download speed variation...) pacman -Sl extra old: 0m6.973s, 0m0.193s new: 0m3.974s, 0m0.390s pacman -Ss pacman old: 0m11.521s, 0m0.395s new: 0m4.171s, 0m0.477s pacman -Si pacman old: 0m6.067s, 0m0.101s new: 0m1.929s, 0m0.437s echo "n" | pacman -Su (five packages to update) old: 0m17.823s, 0m0.306s new: 0m10.577s, 0m0.489s So the tar backend gets some quite significant speed gains in the uncached case, and has marginal losses when the db is cached. Note that the entire sync db is read and parsed no matter the (sync) operation. That explains the difference in the cases with the db cached. This could be adjusted to read the package list (pkgname, pkgver-pkgrel) first and read the desc/depends files for the entire db or a single package as needed but that is probably distant future stuff... Allan

1 1

[arch-dev-public] [signoff] sqlite3 - 3.7.3-1
by Andreas Radke 08 Oct '10

08 Oct '10

Upstream update. Please give signoffs. -Andy SQLite Release 3.7.3 On 2010 October 08 (3.7.3) Changes associated with this release include the following: * Added the sqlite3_create_function_v2() interface that includes a destructor callback. * Added support for custom r-tree queries using application-supplied callback routines to define the boundary of the query region. * The default page cache strives more diligently to avoid using memory beyond what is allocated to it by SQLITE_CONFIG_PAGECACHE. Or if using page cache is allocating from the heap, it strives to avoid going over the sqlite3_soft_heap_limit64(), even if SQLITE_ENABLE_MEMORY_MANAGEMENT is not set. * Added the sqlite3_soft_heap_limit64() interface as a replacement for sqlite3_soft_heap_limit(). * The ANALYZE command now gathers statistics on tables even if they have no indices. * Tweaks to the query planner to help it do a better job of finding the most efficient query plan for each query. * Enhanced the internal text-to-numeric conversion routines so that they work with UTF8 or UTF16, thereby avoiding some UTF16-to-UTF8 text conversions. * Fix a problem that was causing excess memory usage with large WAL transactions in win32 systems. * The interface between the VDBE and B-Tree layer is enhanced such that the VDBE provides hints to the B-Tree layer letting the B-Tree layer know when it is safe to use hashing instead of B-Trees for transient tables. * Miscellaneous documentation enhancements.

2 1

[arch-dev-public] [signoff] man-pages 3.28-1
by Andreas Radke 08 Oct '10

08 Oct '10

Upstream update. Please signoff. -Andy ==================== Changes in man-pages-3.28 ==================== Released: 2010-10-04, Munich Contributors ------------ The following people contributed notes, ideas, or patches that have been incorporated in changes in this release: Andries E. Brouwer <Andries.Brouwer(a)cwi.nl> Denis Barbier <bouzim(a)gmail.com> Jan Kara <jack(a)suse.cz> Landijk <landijk-user(a)yahoo.com> Lennart Poettering <mzjro(a)0pointer.net> Michael Haardt <michael(a)moria.de> Michael Kerrisk <mtk.manpages(a)gmail.com> Petr Baudis <pasky(a)suse.cz> Sam Varshavchik <mrsam(a)courier-mta.com> Apologies if I missed anyone! New and rewritten pages ----------------------- getaddrinfo_a.3 Petr Baudis New page documenting getaddrinfo_a() The page also documents gai_suspend(), gai_cancel(), and gai_error(). aio.7 Michael Kerrisk New page providing an overview of POSIX asynchronous I/O Newly documented interfaces in existing pages --------------------------------------------- exec.3 Michael Kerrisk Document execpe() This function was added to glibc in version 2.11. Also various other small rewrites in the page. New and changed links --------------------- gai_cancel.3 gai_error.3 gai_suspend.3 Petr Baudis New links to new getaddrinfo_a.3 page error_one_per_line.3 Michael Kerrisk Fix misnamed link file (was error_on_per_line.3) execvpe.3 Michael Kerrisk New link to exec.3 sigstack.3 Michael Kerrisk New link to sigaltstack.2 No new programs should use sigstack(3). Point the user to the better sigalstack(2), whose man page briefly mentions sigstack(3). vlimit.3 Michael Kerrisk New link to getrlimit.2 No new programs should use vlimit(3). Point the user to the better setrlimit(2), whose man page briefly mentions vlimit(3). vtimes.3 Michael Kerrisk New link to getrusage.2 No new programs should use vtimes(3). Point the user to the better getrusage(2), whose man page briefly mentions vtimes(3). Global changes -------------- Various pages Michael Kerrisk Switch to American usage: "-wards" ==> "-ward" American English uses "afterward" in preference to "afterwards", and so on chdir.2 chmod.2 chown.2 gethostname.2 getsid.2 pread.2 setpgid.2 sigaltstack.2 stat.2 truncate.2 wait.2 dirfd.3 getsubopt.3 mkdtemp.3 mkstemp.3 siginterrupt.3 strdup.3 Michael Kerrisk Simplify feature test macro requirements Changes to individual pages --------------------------- getrlimit.2 Michael Kerrisk Add mention of the ancient vlimit() function getrusage.2 Michael Kerrisk Add mention of the ancient vtimes() function io_cancel.2 io_destroy.2 io_getevents.2 io_setup.2 io_submit.2 Michael Kerrisk SEE ALSO: add aio(7) sched_setscheduler.2 Michael Kerrisk ERRORS: note that NULL 'param' yields EINVAL stat.2 Michael Kerrisk Note feature test macro requirements for blkcnt_t and blksize_t timer_create.2 Michael Kerrisk Standardize on name 'sevp' for sigevent argument truncate.2 Michael Kerrisk Correct and simplify ftruncate() feature test macro requirements The glibc 2.12 feature test macro requirements for ftruncate() are buggy; see http://sourceware.org/bugzilla/show_bug.cgi?id=12037. Corrected the requirements in the SYNOPSIS, and added a BUGS section describing the problem in glibc 2.12. aio_cancel.3 Michael Kerrisk Add pointer to aio(7) for example program Refer the reader to aio(7) for a description of the aiocb structure CONFORMING TO: Add POSIX.1-2008; add VERSIONS section aio_error.3 Michael Kerrisk Wording improvements in RETURN VALUE Add pointer to aio(7) for example program Refer the reader to aio(7) for a description of the aiocb structure CONFORMING TO: Add POSIX.1-2008; add VERSIONS section aio_fsync.3 Michael Kerrisk Refer the reader to aio(7) for a description of the aiocb structure CONFORMING TO: Add POSIX.1-2008; add VERSIONS section aio_read.3 Michael Kerrisk Various minor rewordings and additions Add pointer to sigevent(7) for details of notification of I/O completion Add pointer to aio(7) for example program Refer the reader to aio(7) for a description of the aiocb structure CONFORMING TO: Add POSIX.1-2008; add VERSIONS section aio_return.3 Michael Kerrisk Improve description in RETURN VALUE Add pointer to aio(7) for example program Refer the reader to aio(7) for a description of the aiocb structure CONFORMING TO: Add POSIX.1-2008; add VERSIONS section aio_suspend.3 Michael Kerrisk Various additions and rewordings. Give some arguments more meaningful names. More explicitly describe the 'nitems' argument. Explicitly note that return is immediate if an I/O operation has already completed. Note that aio_error(3) should be used to scan the aiocb list after a successful return. Add references to other relevant pages. Various other pieces rewritten. Refer the reader to aio(7) for a description of the aiocb structure CONFORMING TO: Add POSIX.1-2008; add VERSIONS section aio_write.3 Michael Kerrisk Add pointer to sigevent(7) for details of notification of I/O completion Various minor rewordings and additions Refer the reader to aio(7) for a description of the aiocb structure CONFORMING TO: Add POSIX.1-2008; add VERSIONS section clearenv.3 Michael Kerrisk Fix error in feature test macro requirements dysize.3 Michael Kerrisk Remove crufty statement about old SCO bug exec.3 Michael Kerrisk Add feature test macro requirements for execvpe() Rewrite description of PATH and mention _CS_PATH Note execvp() and execlp() behavior for filename containing a slash getaddrinfo.3 Michael Kerrisk Add SEE ALSO reference to new getaddrinfo_a.3 page gethostbyname.3 Michael Kerrisk Fix formatting of feature test macros getw.3 Michael Kerrisk Fix feature test macros malloc.3 Landijk Remove editorializing comments on memory overcommitting See https://bugzilla.kernel.org/show_bug.cgi?id=19332 Michael Kerrisk Various minor reorganizations and wording fix-ups mq_notify.3 Michael Kerrisk Standardize on name 'sevp' for sigevent argument nl_langinfo.3 Michael Haardt Make it clear that nl_langinfo() interacts with setlocale() Add an example program posix_openpt.3 Michael Kerrisk Fix feature test macro requirements rand.3 Michael Kerrisk Remove duplicate #include in example program strtok.3 Petr Baudis Add reference to strtok() example in getaddrinfo(3) inotify.7 Michael Kerrisk Added section noting limitations and caveats of inotify sigevent.7 Michael Kerrisk Add SEE ALSO reference to new getaddrinfo_a.3 page Add SEE ALSO referring to new aio(7) page suffixes.7 Michael Kerrisk Change explanation of ".rpm" to "RPM software package"

2 1

[arch-dev-public] [signoff] glibc-2.12.1-2
by Allan McRae 08 Oct '10

08 Oct '10

A few maintenance changes: - pull latest upstream snapshot from 2.12 branch (only a couple of minor bugfixes) - Fedora patch to not optimize so much during linking for those "i686" machines that do not handle it. Real fix will be in binutils-2.21 - Gentoo patch to fix static linking issues. Not quite accepted upstream yet but at least one of the main contributors has pulled it to his branch. - make-3.82 fix - ldd on x86_64 should work nicely with multilib packages (and not affect anyone not using them) Signoff both, Allan

5 4

[arch-dev-public] Mirror country names (was: Mirror status page)
by Dan McGee 08 Oct '10

08 Oct '10

2 5

[arch-dev-public] [signoff] binutils-2.20.1-4
by Allan McRae 08 Oct '10

08 Oct '10

Added /usr/include/demangle.h header as it is useful sometimes... Signoff both, Allan

3 2

[arch-dev-public] News Draft: mirror status
by Dan McGee 07 Oct '10

07 Oct '10

Mirror status and customized mirror lists The website now provides a [mirror status page](/mirrors/status/) that regularly polls all known mirrors and gathers statistics like delay, check duration, and availability. Please consult the page for more details on what is collected and how it works. In addition, you can now [update your mirrorlist](/mirrorlist/) and choose to have it automatically sorted using the up-to-date mirror status information.

2 1

[arch-dev-public] Namcap 2.7 in [testing]
by Dan McGee 07 Oct '10

07 Oct '10

Since the python rebuild is in progress, this has to go there anyway, but let me know if you see any problems. There are a few good changes in this release: http://projects.archlinux.org/namcap.git/log/ http://projects.archlinux.org/namcap.git/tree/NEWS -Dan

1 0