bash 4.1.2 and "broken" regular expression matching conditional operator (=~)
by Roman Rakus
Hi all,
please take a look at bash faq section E - E14. Quoting:
E14) Why does quoting the pattern argument to the regular expression matching
conditional operator (=~) cause regexp matching to stop working?
In versions of bash prior to bash-3.2, the effect of quoting the regular
expression argument to the [[ command's =~ operator was not specified.
The practical effect was that double-quoting the pattern argument required
backslashes to quote special pattern characters, which interfered with the
backslash processing performed by double-quoted word expansion and was
inconsistent with how the == shell pattern matching operator treated
quoted characters.
In bash-3.2, the shell was changed to internally quote characters in single-
and double-quoted string arguments to the =~ operator, which suppresses the
special meaning of the characters special to regular expression processing
(`.', `[', `\', `(', `), `*', `+', `?', `{', `|', `^', and `$') and forces
them to be matched literally. This is consistent with how the `==' pattern
matching operator treats quoted portions of its pattern argument.
Since the treatment of quoted string arguments was changed, several issues
have arisen, chief among them the problem of white space in pattern arguments
and the differing treatment of quoted strings between bash-3.1 and bash-3.2.
Both problems may be solved by using a shell variable to hold the pattern.
Since word splitting is not performed when expanding shell variables in all
operands of the [[ command, this allows users to quote patterns as they wish
when assigning the variable, then expand the values to a single string that
may contain whitespace. The first problem may be solved by using backslashes
or any other quoting mechanism to escape the white space in the patterns.
Please, take a look in your bash scripts and fix them.
RR
14 years, 3 months
Orphaning xerces-c
by Peter Lemenkov
Hello All!
I'm no longer using Xerces-C, so I believe, I can't maintain it
anymore. Feel free to pick it up. Any volunteers?
--
With best regards, Peter Lemenkov.
14 years, 3 months
KDE-SIG meeting report (04/2010)
by Jaroslav Reznik
This is a report of the weekly KDE-SIG-Meeting with a summary of the
topics that were discussed. If you want to add a comment please reply
to this email or add it to the related meeting page.
----------------------------------------------------------------------------------
= Weekly KDE Summary =
Week: 04/2010
Time: 2010-01-26 14:00 UTC
Meeting page:
https://fedoraproject.org/wiki/SIGs/KDE/Meetings/2010-01-26
Meeting minutes:
http://meetbot.fedoraproject.org/fedora-meeting/2010-01-26/kde-
sig.2010-01-26-14.02.html
Meeting log:
http://meetbot.fedoraproject.org/fedora-meeting/2010-01-26/kde-
sig.2010-01-26-14.02.log.html
----------------------------------------------------------------------------------
= Participants =
* BenBoeckel
* JaroslavReznik
* KevinKofler
* LukasTinkl
* RexDieter
* StevenParrish
* ThanNgo
* Nucleo
----------------------------------------------------------------------------------
= Agenda =
* (Kevin_Kofler) drop ABRT from the KDE spin
* (rdieter) kde-l10n template, first draft
* (rdieter) kde-4.3.5, kde-4.3.95 status updates
= Summary =
roll call kde-4.3.5, kde-4.3.95 status updates
* 4.3.95 f11 builds are done @ kde-redhat/unstable
kde-l10n-sl Slovenian vs Sierra Leone
* AGREED: with kde-4.4, work to rename translations to kde-l10n-<locale>
* AGREED: for f13+, also rename translations to kde-i18n-<locale>
* kde-l10n-sl is really Slovenian
webkitpart, update or not for 4.3.5
* ACTION: Kevin Kofler to patch/revert kdebase to support older/existing
webkitpart
kde-l10n template
* Rex Dieter prepared kde-l10n template [1]
* Issues:
o Include/exclude/blacklist logic
o Bodhi overhead with more than 50 packages
* ACTION: Rex Dieter to followup with lmacken on pkg split impact on bodhi
drop ABRT from the KDE spin?
* Kevin Kofler brought some concerns:
o most of our stuff is KDE apps which use KCrash/DrKonqi anyway
o it reports crashes to downstream instead of upstream where they
belong
o it's a GTK+-based service running in the background
* KDE SIG to sort down requirements for Abrt developers
----------------------------------------------------------------------------------
= Next Meeting =
https://fedoraproject.org/wiki/SIGs/KDE/Meetings/2010-02-02
----------------------------------------------------------------------------------
= Links =
[1] http://rdieter.fedorapeople.org/rpms/kde-l10n.spec
--
Jaroslav Řezník <jreznik(a)redhat.com>
Software Engineer - Base Operating Systems Brno
Office: +420 532 294 275
Mobile: +420 731 455 332
Red Hat, Inc. http://cz.redhat.com/
14 years, 3 months
Disabling drm/ttm/radeon module loading for debugging
by Pasi Kärkkäinen
Hello,
How can I make sure drm/radeon/ttm modules are not loaded,
until I do it manually myself?
I extracted the initrd image, edited the 'init' script and
commented out loading of the modules, repacked the initrd,
and booted but something (plymouth? udev?) still loads the modules..
I'd like to load them later manually, for some debugging purposes..
-- Pasi
14 years, 3 months
gvfs causes hangs
by Steve Grubb
Hello,
I have been running into something on F-12 that is really annoying and was
wondering if anyone else is seeing this. When I use kmail and want to attach a
file that is not in my Documents folder and go up one level to my homedir, it
hangs. I can't do anything with kmail except kill the email I was composing.
Digging into this further, if you run lsof, it hangs when it gets to ~/.gvfs:
alarm(15) = 0
write(5, "\240(A\0\0\0\0\0", 8) = 8
write(5, "\23\0\0\0", 4) = 4
write(5, "/home/sgrubb/.gvfs\0", 19) = 19
write(5, "\0\20\0\0", 4) = 4
read(6, 0x7fff9cc98428, 4) = ? ERESTARTSYS (To be restarted)
--- SIGALRM (Alarm clock) @ 0 (0) ---
alarm(0) = 0
rt_sigaction(SIGALRM, {SIG_DFL, [ALRM], SA_RESTORER|SA_RESTART,
0x7ff0359ac740}, {0x412140, [ALRM], SA_RESTORER|SA_RESTART, 0x7ff0359ac740}, 8)
= 0
close(5) = 0
close(6) = 0
rt_sigaction(SIGALRM, {0x412140, [ALRM], SA_RESTORER|SA_RESTART,
0x7ff0359ac740}, {SIG_DFL, [ALRM], SA_RESTORER|SA_RESTART, 0x7ff0359ac740}, 8) =
0
alarm(5) = 0
wait4(-1,
And also find ~ -name anything also hangs:
open("..", O_RDONLY|O_NOCTTY|O_NONBLOCK|O_DIRECTORY|O_NOFOLLOW) = 5
fstat(5, {st_mode=S_IFDIR|0700, st_size=4096, ...}) = 0
fchdir(5) = 0
close(5) = 0
newfstatat(AT_FDCWD, ".gvfs",
Is this a defective file system or are a whole bunch of apps needing to be
fixed? Also, do other people notice the same thing?
-Steve
14 years, 3 months
F12 to Rawhide upgrade
by Mike Chambers
Over the weekend I tried this via yum and the upgrade/install part went
fine. But once I rebooted, the gui wouldn't work. Meaning, at the
gdm/login screen, it was garbled. I downgraded xorg-x11-server-comment
(was 1 or 2 packages total) to the F12 (may have been upgrade/test one)
version, and that got me the login screen (although it didn't show my
username, just "other). But once at the login screen, my mouse/keyboard
(wireless desktop) wouldn't work at all. I then aborted and reinstalled
F12. Nouveau chipset/card for video.
Not big priority to run it, but just wanted to report what I initially
ran into.
Mike Chambers
Madisonville, KY
"Best lil town on Earth!"
14 years, 3 months
Unable to start X
by Paul F. Johnson
Hi,
I'm using the nvidia 7600 video card with the neuveau driver (not using the
commerical driver, just the fedora shipped one). I did a rawhide update on
Saturday and it looks like something there has borked my system. I had a power
outage last night and had to reboot. It doesn't matter which version of the
kernel I try to load, but when I try to boot X I'm told that the kernel has
claimed a driver and the server won't start until the kernel has unclaimed it.
I've tried passing nomodeset and vesa in the boot line, but no good. If I boot
to runlevel 3, I can get a command line login, so the system is up and
working, but issuing startx gives me the error about the claimed module.
Any ideas on how to get my desktop to boot or do I need to downgrade the
xorg-x11-drv-neuvaux (or however it's spelt!)?
TTFN
Paul
--
It's only me, only me and no-one else.
14 years, 3 months
Fedora Release Engineering meeting summary for 2010-01-25
by Jesse Keating
Minutes:
http://meetbot.fedoraproject.org/fedora-meeting/2010-01-25/fedora-releng....
Minutes (text):
http://meetbot.fedoraproject.org/fedora-meeting/2010-01-25/fedora-releng....
Log:
http://meetbot.fedoraproject.org/fedora-meeting/2010-01-25/fedora-releng....
Meeting summary
---------------
* roll call (Oxf13, 18:05:43)
* members present: poelcat wwoods warren notting rdieter and Oxf13
(Oxf13, 18:10:34)
* Meeting practices (Oxf13, 18:10:53)
* lmacken also in attendance (Oxf13, 18:11:12)
* dgilmore also in attendance (Oxf13, 18:12:31)
* discussed closing each topic out with info lines before moving on.
(Oxf13, 18:12:49)
* use tickets with 'meeting' keyword to get topics on meeting agenda.
(Oxf13, 18:13:33)
* Test images for QA (Oxf13, 18:15:30)
* got past the rawhide does not build problem (Oxf13, 18:17:01)
* images delivered to alt.fp.o and noted in the ticket (Oxf13,
18:17:09)
* AGREED: we will use dated directories for test image delivery and
maintain symlinks. (Oxf13, 18:22:51)
* ACTION: Oxf13 will work with poelcat to update the SOP reflecting
this (Oxf13, 18:23:07)
* ACTION: Oxf13 will work with QA team to determine symlink names and
set expectations (Oxf13, 18:23:31)
* Standard Operating Procedures (Oxf13, 18:25:40)
* LINK:
https://fedoraproject.org/wiki/Release_Engineering_Release_Tickets
(poelcat, 18:27:31)
* https://fedoraproject.org/wiki/Composing_test_images (poelcat,
18:28:48)
* https://fedoraproject.org/wiki/Create_release_signing_key (poelcat,
18:29:00)
* jwb is working on documenting the steps for doing a bodhi updates
push (Oxf13, 18:30:53)
* ACTION: Oxf13 will work on documenting the steps for purging orphans
(Oxf13, 18:32:10)
* ACTION: All releng members who perform releng tasks are encouraged
to document the steps if no SOP currently exists for your task.
(Oxf13, 18:33:39)
* if you don't have time to format for wiki send to poelcat the
wiki-monkey (poelcat, 18:34:01)
* ACTION: Oxf13 will review the two pending SOP documents today
(Oxf13, 18:34:03)
* open floor (Oxf13, 18:36:36)
* New /mnt/koji (Oxf13, 18:37:37)
* new /mnt/koji hardware should be delivered on Friday (Oxf13,
18:37:47)
* Discussed new hardware incoming, and the need to start load testing
it hopefully next week or hte week after (Oxf13, 18:44:55)
* dist-git update (Oxf13, 18:45:01)
* repos were refreshed, and some trimming was done of kernel.git
(Oxf13, 18:48:55)
* SmootherFrOgZ has contributions to give to fedpkg, just have to work
out patch submission (Oxf13, 18:49:12)
* dist-git project will soon need to test building with Koji in stg
(Oxf13, 18:52:31)
* koji.stg.fp.o will be ready soon for public use (Oxf13, 18:52:42)
* We may be able to combine new /mnt/koji hardware testing needs and
dist-git building needs into the same testing (Oxf13, 18:53:07)
* open floor (Oxf13, 18:54:22)
* No Frozen Rawhide (Oxf13, 18:56:37)
* waiting on koji.stg for further testing (Oxf13, 18:58:29)
Meeting ended at 19:00:42 UTC.
Action Items
------------
* Oxf13 will work with poelcat to update the SOP reflecting this
* Oxf13 will work with QA team to determine symlink names and set
expectations
* Oxf13 will work on documenting the steps for purging orphans
* All releng members who perform releng tasks are encouraged to document
the steps if no SOP currently exists for your task.
* Oxf13 will review the two pending SOP documents today
--
Jesse Keating RHCE (http://jkeating.livejournal.com)
Fedora Project (http://fedoraproject.org/wiki/JesseKeating)
GPG Public Key (geek.j2solutions.net/jkeating.j2solutions.pub)
identi.ca (http://identi.ca/jkeating)
14 years, 3 months
What is the pulseaudio way of recording?
by Paulo Cavalcanti
Hi,
I have been recording using arecord and aplay this way:
arecord -D default -d 0 -f cd | aplay -f cd -D default &
The problem is that, while this works fine on rhel5, the same is no true
for Fedora 12 using pulseaudio, since I clearly hear some random noise
(specially at low volumes).
Maybe this is related to the fact that the arecord process is consuming 15%
of my CPU (a quadcore, 2.4MHz). Compared to Audacious 2.2, for instance,
playing an mp3 file, which consumes only 4% of the CPU, this is really too
much.
I concluded that the problem has to do with recording with pulseaudio
active.
I know there are parecord and paplay, but they are just symbolic links to
pacat,
and I do not know how to combine them in a similar way.
Therefore, my question is what is the best way of recording with pulseadio,
without wasting too much of my CPU and without noise?
Thanks.
--
Paulo Roma Cavalcanti
LCG - UFRJ
14 years, 3 months
yelp failing to build with new ld linker
by Roland Grunberg
Hi,
Upon doing a mock build of yelp-2.28.1-2.fc13 with the new version of binutils
(containing the proposed DSO link changes to ld) from
http://roland.fedorapeople.org/ld-test/, the build fails.
/usr/bin/ld.bfd: /usr/lib/xulrunner-sdk-1.9.2/lib/libxpcomglue.a(nsGlueLinkingDlopen.o): undefined reference to symbol 'dlopen@(a)GLIBC_2.1'
/usr/bin/ld.bfd: note: 'dlopen@(a)GLIBC_2.1' is defined in DSO /usr/lib/gcc/i686-redhat-linux/4.4.3/../../../libdl.so so try adding it to the linker command line
In the linker line we can see "-ldl -lxpcomeglue" present. -lxpcomglue uses
symbols from -ldl, while yelp does not. Since -lxpcomglue is a static library,
the -ldl was added to satisfy this.
The problem is with the order in which libraries are declared. Since -ldl is
declared before -lxpcomglue in the link line, and since nothing in yelp uses
symbols from -ldl, it is dropped at that point leaving -lxpcomglue with undefined
symbols from -ldl.
The following potential solutions were proposed by Roland McGrath to this issue :
Rearranging the link line would fix this ("-lxpcomglue -ldl"), or
xulrunner-devel could add a libxpcomglue.so linker script containing :
AS_NEEDED ( libxpcomglue.a -ldl )
This way any package using -lxpcomglue would not need to know whether
it is a shared/static library, and what other libraries it makes use of.
Thank-You,
--
Roland Grunberg
14 years, 3 months
