Bug#832875: Avoid "utility" in package description
Package: python-json-schema-validator Version: 2.3.1-3 Greetings! The package description is "JSON schema validation utility", which is a bit misleading. The problem is that it doesn't provide a utility with which you can validate JSON agains a schema. Instead, it provides a "Python module" or a "Python library" for integrating validation. Also, minor typo, there's a "the" missing here: "This package provides a JSON Schema validator conforming to a subset of second draft of the specification." Further, some questions here: "JSON files can be verified against a schema expressed in the python source code, producing either a SchemaError exception or ValidationError exception." Can you only hard-code the schema in Python or also read the schema (which itself is JSON) from a file? Also, can you only validate files or can you validate in-memory data, too? Thanks to you and the Debian team for your effort! Uli
Bug#831554: cyrus-nntpd-2.4: nntp not working, regression of #734648?
Package: cyrus-nntpd-2.4 Version: 2.4.17+nocaldav-0~deb8u1 Severity: grave Justification: renders package unusable Dear Maintainer, I have upgraded my server to from Debian 7 to Debian 8. After this upgrade it is not possible anymore to feed news into my imap spool. In the logs I get the following output from fetchnews: cyrus/fetchnews[6474]: fetchnews: nntp.aioe.org offered 4; localhost rejected 0, accepted 0, failed 4 This problem was registered as bugreport on cyrus and fixed in 2.4.18: https://bugzilla.cyrusimap.org/show_bug.cgi?id=3833 . A detailed description of the problem can be found there. Also I had a similar problem during upgrading from debian 6 to debian 7: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=734648 I compiled the cyrus nntpd from cyrus 2.4.18. With this nntpd I could successfully deliver news into my imap spool. -- System Information: Debian Release: 8.5 APT prefers stable APT policy: (500, 'stable') Architecture: i386 (i686) Kernel: Linux 3.2.41-042stab111.12 (SMP w/2 CPU cores) Locale: LANG=de_DE.utf8, LC_CTYPE=de_DE.utf8 (charmap=UTF-8) (ignored: LC_ALL set to de_DE.utf8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) Versions of packages cyrus-nntpd-2.4 depends on: ii cyrus-nntpd 2.4.17+nocaldav-0~deb8u1 ii dpkg 1.17.27 cyrus-nntpd-2.4 recommends no packages. cyrus-nntpd-2.4 suggests no packages. -- no debconf information
Bug#576829: dependency on KDE?
Per Olofssonwrote: > xdg-utils does not depend on KDE, actually. It is supposed to work > with any desktop environment. > > Which desktop environment (if any) are you using? Hmmm, question is rather which DE I was using (on my old computer at that time ...), but I think I used plain fluxbox at that time, which is also what I'm running now, although on a different machine which was setup from zero. I do have kde-runtime installed though (I don't know why), which provides kreadconfig. At the moment, I can't reproduce the issue in the original bug report any more, although there still seems to be some kind of dependency on KDE: uli@zox ~> xdg-settings get default-web-browser /usr/bin/xdg-settings: 1: /usr/bin/xdg-settings: kde-config: not found chrome.desktop The first line of output was actually to stderr, the second, expected line was to stdout. However, after that, I tried what the bug reporter initially complained about, and lo and behold, the error went away! By pure luck, I happened to run the suggested "sh -x /usr/bin/xdg-settings get default-web-browser" before that though: uli@zox ~> sh -x /usr/bin/xdg-settings get default-web-browser + check_common_commands get default-web-browser + [ 2 -gt 0 ] + parm=get + shift + [ 1 -gt 0 ] + parm=default-web-browser + shift + [ 0 -gt 0 ] + [ -z ] + unset XDG_UTILS_DEBUG_LEVEL + [ 0 -lt 1 ] + xdg_redirect_output= > /dev/null 2> /dev/null + [ xget = x--list ] + [ xget != x ] + [ xdefault-web-browser != x ] + [ xget = xget -o x != x ] + op=get + parm=default-web-browser + shift 2 + [ xget != xget -a xget != xcheck -a xget != xset ] + detectDE + unset GREP_OPTIONS + [ -n ] + [ x = x ] + [ xtrue = xtrue ] + DE=kde + [ xkde = x ] + [ xkde = x ] + [ xkde = xgnome ] + dispatch_specific + [ xget = xget ] + get_browser_kde + read_kde_browser + read_kde_config kdeglobals General BrowserApplication + configfile=kdeglobals + configsection=General + configkey=BrowserApplication + kreadconfig --file kdeglobals --group General --key BrowserApplication + application= + [ x != x ] + kde-config --path config + cut/usr/bin/xdg-settings: 1: /usr/bin/xdg-settings: kde-config: not found -d : -f 1 + configfile_dir= + configfile_path=/kdeglobals + [ ! -f /kdeglobals ] + return + browser= + [ x = x ] + get_browser_mime + test -z + MIME=text/html + xdg_mime_fixup + [ kde = kde -a -z ] + ktradertest text/html Application + XDG_MIME_FIXED=yes + xdg-mime query default text/html chrome.desktop + [ 0 -eq 0 ] + exit_success + [ 0 -gt 0 ] + exit 0 I then changed the default-web-browser to test the original bug report, but couldn't reproduce the issue. However, after that, even the error message from xdg-settings is gone, I haven't rebooted or restarted the session though! Anyhow, here's how the trace looks afterwards: uli@zox ~> sh -x /usr/bin/xdg-settings get default-web-browser + check_common_commands get default-web-browser + [ 2 -gt 0 ] + parm=get + shift + [ 1 -gt 0 ] + parm=default-web-browser + shift + [ 0 -gt 0 ] + [ -z ] + unset XDG_UTILS_DEBUG_LEVEL + [ 0 -lt 1 ] + xdg_redirect_output= > /dev/null 2> /dev/null + [ xget = x--list ] + [ xget != x ] + [ xdefault-web-browser != x ] + [ xget = xget -o x != x ] + op=get + parm=default-web-browser + shift 2 + [ xget != xget -a xget != xcheck -a xget != xset ] + detectDE + unset GREP_OPTIONS + [ -n ] + [ x = x ] + [ xtrue = xtrue ] + DE=kde + [ xkde = x ] + [ xkde = x ] + [ xkde = xgnome ] + dispatch_specific + [ xget = xget ] + get_browser_kde + read_kde_browser + read_kde_config kdeglobals General BrowserApplication + configfile=kdeglobals + configsection=General + configkey=BrowserApplication + kreadconfig --file kdeglobals --group General --key BrowserApplication + application=chrome.desktop + [ xchrome.desktop != x ] + echo chrome.desktop + browser=chrome.desktop + [ xchrome.desktop = x ] + resolve_kde_browser_desktop + [ -z chrome.desktop ] + echo chrome.desktop chrome.desktop + [ 0 -eq 0 ] + exit_success + [ 0 -gt 0 ] + exit 0 The difference starts after the call to kreadconfig, which seems to yield the selected browser after setting it once. In both cases, there is a line "DE=kde", which is bogus. Looking at "xprop -root", I see KDE_FULL_SESSION is set. This comes from my fluxbox startup script where it was used to work around some bug in kded (used to be part of KDE3, I used to run some KDE applications back then). I just deactivated the workaround in fluxbox startup, and now at least the DE is detected correctly: uli@zox ~> sh -x /usr/bin/xdg-settings get default-web-browser + check_common_commands get default-web-browser + [ 2 -gt 0 ] + parm=get + shift + [ 1 -gt 0 ] + parm=default-web-browser + shift + [ 0 -gt 0 ] + [ -z ] + unset XDG_UTILS_DEBUG_LEVEL + [ 0 -lt 1 ] + xdg_redirect_output= > /dev/null 2> /dev/null + [ xget = x--list ] + [ xget != x ] + [ xdefault-web-browser != x ] + [ xget = xget -o x != x ] + op=get + parm=default-web-browser + shift 2 + [ xget != xget -a xget != xcheck
Bug#755724: i3lock: segfault on start
Package: i3lock Version: 2.6-1 Severity: important Dear Maintainer, I just tried to try out i3lock. When I run it, I expect it to lock my X11 session until I unlock it with a password. Instead, it just dies with a segfault. I'm not using i3 (which seems to be the primary target) but fluxbox. I also had xscreensaver running before. Stopping it doesn't make any difference, but I don't know what persistant settings it leaves in X. GDB backtrace: #0 0x76a6d198 in ?? () from /usr/lib/x86_64-linux-gnu/libxkbcommon-x11.so.0 #1 0x76a6d3ae in ?? () from /usr/lib/x86_64-linux-gnu/libxkbcommon-x11.so.0 #2 0x76a6ce0c in ?? () from /usr/lib/x86_64-linux-gnu/libxkbcommon-x11.so.0 #3 0x76a6ac1c in xkb_x11_keymap_new_from_device () from /usr/lib/x86_64-linux-gnu/libxkbcommon-x11.so.0 #4 0x00404079 in ?? () #5 0x00403288 in ?? () #6 0x75d88b45 in __libc_start_main (main=0x402fa0, argc=1, argv=0x7fffe3d8, init=optimized out, fini=optimized out, rtld_fini=optimized out, stack_end=0x7fffe3c8) at libc-start.c:287 #7 0x00403819 in ?? () Tail of strace output: socket(PF_FILE, SOCK_STREAM|SOCK_CLOEXEC, 0) = 3 connect(3, {sa_family=AF_FILE, path=@/tmp/.X11-unix/X0}, 20) = 0 getpeername(3, {sa_family=AF_FILE, path=@/tmp/.X11-unix/X0}, [20]) = 0 [...] access(/usr/share/X11/xkb, R_OK|X_OK) = 0 stat(/home/uli/.xkb, 0x7fffa12da830) = -1 ENOENT (No such file or directory) poll([{fd=3, events=POLLIN|POLLOUT}], 1, 4294967295) = 1 ([{fd=3, revents=POLLOUT}]) writev(3, [{\207\1\4\0\3\0\7\0\0\0\7\0\337\0\337\0\207\30\4\0\0\1\0\0\0\0\0\0\0\0\0\0, 32}], 1) = 32 poll([{fd=3, events=POLLIN}], 1, 4294967295) = 1 ([{fd=3, revents=POLLIN}]) recvmsg(3, {msg_name(0)=NULL, msg_iov(1)=[{\1\3\5\0\6\0\0\0\0\0\36\0\0\0\0\0\0\0\0\0\0\1\0\0\0\377\0\0\0\0\0\0..., 4096}], msg_controllen=0, msg_flags=0}, 0) = 56 --- SIGSEGV (Segmentation fault) @ 0 (0) --- +++ killed by SIGSEGV +++ Segmentation fault Looking at strace, it seems as if XCB was reading into a buffer of insufficient size, so it might be a bug there. Let me know if there's anything else I can do. Cheers! Uli -- System Information: Debian Release: jessie/sid APT prefers unstable APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 3.13-1-amd64 (SMP w/4 CPU cores) Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages i3lock depends on: ii libc6 2.19-5 ii libcairo2 1.12.16-2 ii libev4 1:4.15-3 ii libpam0g1.1.8-3 ii libxcb-dpms01.10-3 ii libxcb-image0 0.3.9-1 ii libxcb-shm0 1.10-3 ii libxcb-util00.3.8-2 ii libxcb-xinerama01.10-3 ii libxcb-xkb1 1.10-3 ii libxcb1 1.10-3 ii libxkbcommon-x11-0 0.4.1-1 ii libxkbcommon0 0.4.0-1 i3lock recommends no packages. i3lock suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#755724: i3lock: segfault on start
Michael Stapelberg stapelb...@debian.org wrote: Ulrich Eckhardt ulrich.eckha...@base-42.de writes: ii libxkbcommon-x11-0 0.4.1-1 ii libxkbcommon0 0.4.0-1 Note that you have version 0.4.1-1 of libxkbcommon-x11-0 but 0.4.0-1 of libxkbcommon0. That _should_ not be a problem, but you can you please try using 0.4.1-1 of both and (if that doesn’t help) 0.4.0-1 of both? Good catch, Michael! I upgraded libxkbcommon0 to 0.4.1-1 and i3lock now works. Thanks! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#747860: Can't switch CRTC between different outputs
Package: arandr Version: 0.1.7.1-1 Severity: wishlist Hi! I have a setup with a laptop, a docking station and two external monitors connected to the docking station. Since the laptop can only drive two displays at a time (it only has two CRTCs), I use xrandr to turn off the internal LVDS display and activate the two external displays. This works, but is cumbersome and requires three steps. Using arandr, I was hoping that it could turn off the builtin display and activate the external displays in one step and vice versa. However, this doesn't work, since activating the second external display fails because the required CRTC is already in use. It would be nice if arandr could activate the displays in the required number of steps automatically. Just for the record, a similar issue exists for lxrandr, the other GUI for xrandr in Debian. Thanks! Uli
Bug#747861: Can't switch CRTC between different outputs
Package: lxrandr Version: 0.1.2-3 Severity: wishlist Hi! I have a setup with a laptop, a docking station and two external monitors connected to the docking station. Since the laptop can only drive two displays at a time (it only has two CRTCs), I use xrandr to turn off the internal LVDS display and activate the two external displays. This works, but is cumbersome and requires three steps. Using lxrandr, I was hoping that it could turn off the builtin display and activate the external displays in one step and vice versa. However, this doesn't work, since activating the second external display fails because the required CRTC is already in use. It would be nice if lxrandr could activate the displays in the required number of steps automatically. Just for the record, a similar issue exists for arandr, the other GUI for xrandr in Debian. Thanks! Uli
Bug#746462: Password entry via keyboard shortcut messed up
package: keepass2 version: 2.25+dfsg-1 The keyboard shortcut to auto-type a password (control+V) is causing me problems, sometimes it works, sometimes not. I have the feeling that keepass starts typing the password while I still have the control key held, which changes the meaning of the generated keypresses. keepass2 should wait until the user released the control key. I think that an earlier version (the one in testing from ~6 months ago) didn't show these problems. That version had the problem that the control key remained stuck, i.e. X thought that the key was still held and only pressing the according control key worked to get X in sync with reality again. As a workaround, using the auto-type function from the context menu seems to work without issues. Cheers! Uli
Bug#734648: cyrus-nntpd-2.4: nntp not working after upgrade to wheezy
Package: cyrus-nntpd-2.4 Version: 2.4.16-4+deb7u1 Severity: grave Tags: upstream Justification: renders package unusable Dear Maintainer, I have upgraded my server to from Debian 6 (with cyrus 2.2) to Debian 7 (with cyrus 2.4.16). After this upgrade it is not possible anymore to feed news into my imap-spool. This problem was registered as bugreport on cyrus: https://bugzilla.cyrusimap.org/show_bug.cgi?id=3833 . A detailed description of the problem can be found there. This bug was resolved quickly by the cyrus team but planned for 2.4.18. I was able to incorporate the two commits of the solution into the source package cyrus-imapd-2.4_2.4.16-4+deb7u1 and build a new working nntpd for cyrus 2.4.16. The following patch can be applied to fix this problem: --- --- nntpd.c.deb 2014-01-07 20:33:00.827064145 +0100 +++ nntpd.c 2014-01-08 19:10:03.242389940 +0100 @@ -1065,7 +1065,7 @@ be = backend_current; if (arg1.len (!is_newsgroup(arg1.s) || -(r = open_group(arg1.s, 1, be, NULL goto nogroup; +(r = open_group(arg1.s, 0, be, NULL goto nogroup; else if (be) { prot_printf(be-out, %s, cmd.s); if (arg1.len) { @@ -1780,6 +1780,8 @@ if (!has_prefix) { snprintf(mailboxname, sizeof(mailboxname), %s%s, newsprefix, name); name = mailboxname; + + if (!is_newsgroup(name)) return IMAP_MAILBOX_NONEXISTENT; } if (!r) r = mlookup(name, newserver, acl, NULL); @@ -2544,7 +2546,7 @@ } } else { - prot_printf(nntp_out, %s %u %u %c\r\n, name, + prot_printf(nntp_out, %s %u %u %c\r\n, name+strlen(newsprefix), group_state-exists ? index_getuid(group_state, group_state-exists) : group_state-mailbox-i.last_uid, group_state-exists ? index_getuid(group_state, 1) : @@ -3000,7 +3002,7 @@ if (!rcpt) return -1; /* construct the mailbox name */ - sprintf(rcpt, %.*s, (int) n, p); + sprintf(rcpt, %s%.*s, newsprefix, (int) n, p); /* skip mailboxes that we don't serve as newsgroups */ if (!is_newsgroup(rcpt)) continue; -- System Information: Debian Release: 7.3 APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'stable') Architecture: i386 (i686) Kernel: Linux 2.6.32-042stab078.27 (SMP w/1 CPU core) Locale: LANG=de_DE.utf8, LC_CTYPE=de_DE.utf8 (charmap=UTF-8) (ignored: LC_ALL set to de_DE.utf8) Shell: /bin/sh linked to /bin/bash Versions of packages cyrus-nntpd-2.4 depends on: ii cyrus-common-2.4 2.4.16-4+deb7u1 ii libc6 2.13-38 ii libcomerr21.42.5-1.1 ii libdb5.1 5.1.29-5 ii libsasl2-22.1.25.dfsg1-6+deb7u1 ii libssl1.0.0 1.0.1e-2+deb7u3 ii libwrap0 7.6.q-24 ii zlib1g1:1.2.7.dfsg-13 cyrus-nntpd-2.4 recommends no packages. cyrus-nntpd-2.4 suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#418630: Similar issues for a Samsung printer
Hi! I'm using 1.0.23-3 here and I'm facing similar issues. The device is detected as | uli@zox:~$ scanimage -L | Failed cupsGetDevices | device `smfp:usb;04E8;344F;Z72TB8KD4F00NFW' is a SAMSUNG SCX-3400 Series on USB Scanner (I'm not sure what the CUPS-output is doing there...). Any --mode argument I tried is rejected, both with and without the connecting = and with arguments Gray, Grayscale and Color. I couldn't get the --preview option to work either. BTW: Have you already forwarded this bug to upstream? I sense that this mixture of different argument styles isn't something that can be resolved here in Debian but rather something that requires a decision from upstream first. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#711447: clicking on tabs doesn't focus windows
Hi! I have just discovered what seems to be another aspect of this bug. If you switch to click tab focus (Configuration/Focus Model/Click Tab Focus), the window doesn't get focused when clicking on a tab. Note that this requires external tabs, with internal tabs you can click to change focus. Also, this bug is not present in stable (1.3.2-4) but only in testing (1.3.5-1). Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#576829: dependency on KDE?
Hi! I just stumbled across the same problem using 1.1.0~rc1+git20111210-7. I'm not sure if that is the cause, but kfmclient is the KDE File Manager client, as far as I know, so it seems that xdg-utils actually depend on KDE. Could that be the reason? Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#711447: clicking on tabs doesn't raise windows
Package: fluxbox Version: 1.3.5-1 Severity: wishlist Hi! I believe that since a recent update of fluxbox in testing, you can't click on window tabs any more in order to raise the window. Similarly, though not as grave, you can't drag a tab onto a group in order to add the window to the group or to reorder the tabs. I'm now wondering if this was by accident or intentional... Greetings from Hamburg! Uli -- System Information: Debian Release: jessie/sid APT prefers testing APT policy: (500, 'testing'), (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 3.2.0-4-amd64 (SMP w/4 CPU cores) Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages fluxbox depends on: ii libc6 2.17-3 ii libfontconfig1 2.9.0-7.1 ii libfreetype62.4.9-1.1 ii libfribidi0 0.19.5-2 ii libgcc1 1:4.8.0-7 ii libice6 2:1.0.8-2 ii libimlib2 1.4.5-1 ii libsm6 2:1.2.1-2 ii libstdc++6 4.8.0-7 ii libx11-62:1.5.0-1+deb7u1 ii libxext62:1.3.1-2+deb7u1 ii libxft2 2.3.1-1 ii libxinerama12:1.1.2-1+deb7u1 ii libxpm4 1:3.5.10-1 ii libxrandr2 2:1.3.2-2+deb7u1 ii libxrender1 1:0.9.7-1+deb7u1 ii menu2.1.46 ii zlib1g 1:1.2.8.dfsg-1 Versions of packages fluxbox recommends: ii feh 2.3-2 ii xfonts-terminus 4.35-1 Versions of packages fluxbox suggests: pn fbautostart none pn fbdesk none pn fbpager none -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#708853: support browsing remote filesystems via SSH
Package: thunar Version: 1.2.3-4+b1 Severity: wishlist I'm trying to access a filesystem on a different system using SSH. I know that thunar supports this, but I can't get it to work or even extract any meaningful error message. Just to make that clear: I'm not running XFCE, just thunar inside a fluxbox session! I've been doing some research, and it seems that thunar requires some external packages in order to use SSH, but I don't know which and they are not hard dependencies. In any case, all recommends and suggests are installed, apart from xfce4-panel. I have also installed gvfs-bin and gvfs-fuse, but neither helped. I guess that it's just a matter of documenting these dependencies (README.Debian perhaps?), if these are the problem at all. That said, the error messages are not really helpful. If I put ssh://192.168.1.2/ into thunar's location field, it just displays a red icon next to it, without any further error message. If I start it from a terminal, I get a popup and the same text echoed on the console: ~$ thunar ssh://192.168.1.2/ Thunar: Failed to open ssh://192.168.1.2/: The specified location is not supported This is very similar to gvfs: ~$ gvfs-mount -m ssh://192.168.1.2/ Error mounting location: The specified location is not supported I'm confident I could find out what's wrong myself, but I think a user shouldn't be required to read the sources and search the web in order to get this to work, hence the wishlist bug. Greetings from Hamburg! Uli -- System Information: Debian Release: jessie/sid APT prefers testing APT policy: (500, 'testing') Architecture: amd64 (x86_64) Kernel: Linux 3.2.0-4-amd64 (SMP w/4 CPU cores) Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages thunar depends on: ii desktop-file-utils 0.20-0.1 ii exo-utils 0.6.2-5 ii libatk1.0-0 2.4.0-2 ii libc6 2.13-38 ii libcairo2 1.12.2-3 ii libdbus-1-3 1.6.8-1 ii libdbus-glib-1-20.100.2-1 ii libexo-1-0 0.6.2-5 ii libgdk-pixbuf2.0-0 2.26.1-1 ii libglib2.0-02.33.12+really2.32.4-5 ii libgtk2.0-0 2.24.10-2 ii libgudev-1.0-0 175-7.2 ii libice6 2:1.0.8-2 ii libnotify4 0.7.5-2 ii libpango1.0-0 1.30.0-1 ii libsm6 2:1.2.1-2 ii libthunarx-2-0 1.2.3-4+b1 ii libxfce4ui-1-0 4.8.1-1 ii libxfce4util4 4.8.2-1 ii shared-mime-info1.0-1+b1 ii thunar-data 1.2.3-4 Versions of packages thunar recommends: ii dbus-x111.6.8-1 ii gvfs1.12.3-4 ii libfontconfig1 2.9.0-7.1 ii libfreetype62.4.9-1.1 ii thunar-volman 0.6.1-1 ii tumbler 0.1.25-1+b1 ii xdg-user-dirs 0.14-1 ii xfce4-panel 4.8.6-4 Versions of packages thunar suggests: ii thunar-archive-plugin 0.3.0-4 ii thunar-media-tags-plugin 0.2.0-1 -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#652228: Works with Chrome, fails with Iceweasel
I just tried dragging an image from thunar (1.2.3-4+b1, running without XFCE) onto the demo field in both Iceweasel (10.0.12esr-1+nmu1) and Chromium (26.0.1410.43-1). It works only in Chromium. Just my 2cc. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#708853: [Pkg-xfce-devel] Bug#708853: support browsing remote filesystems via SSH
Yves-Alexis Perez wrote: Actually you miss one suggests (from gvfs): gvfs-backends. Can you install it and report back? Ah yes, that one was indeed missing, and it works now! Can you please add a note concerning this to /usr/share/doc/README.Debian? That would have helped immensely! Other than that, it's still a useless error message, though I'm not sure if that is from thunar or gvfs. Thank you very much! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#696221: Patch to provide more detailed info on errors
Package: cutecom Version: 0.22.0-2 When opening one of the serial ports fails, you only get the error that it failed but not why, which is frustrating and not helpful. The typical faults are that a port isn't there or that you don't have access permissions. The following piece of code changes the error handling after open() so that the actual error becomes available to the user: m_fd=open(dev.toLatin1(), flags | O_NDELAY); if (m_fd0) { int e = errno; char* es = strerror(e); std::cerropening failed with errno= e ( es ) std::endl; m_fd=-1; QMessageBox::information(this, tr(Error), tr(Could not open %1:\n error #%2 (%3)).arg(dev).arg(e).arg(es)); return; } Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#691693: Tabs in Python sourcefiles
Package: geany Version: 1.22+dfsg-2 With the default configuration, Geany uses tabs to indent Python sources. According to Python's coding guidelines (PEP 8), sources should not use any tabs but exactly four spaces. As a workaround, you can manually switch to spaces under Document Indent Type Spaces, which already uses the correct width. Two notes: 1. I'm not sure if the Python template file is generated on the fly or loaded, but the template also uses tabs. 2. A similar issue might be Makefiles, which actually rely on being indented with tabs. I haven't tested those though. Thank you! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#687696: Improper handling of non-ASCII paths
Package: svn-load Version: 1.3-1 I'm trying to load a directory containing non-ASCII filenames, which fails. I'm pretty sure that the encoding is proper UTF-8 and that the locale is set up correctly. I have manually verified the encoding and tools like e.g. ls and thunar also handle it. If you want to verify yourself, the files are part of the ca-certificates package and stored in /etc/ssl. What I did was this: uli@zox:~$ svnadmin create tmp-repo uli@zox:~$ svn-load file:///home/uli/tmp-repo ssl /etc/ssl Adding /tmp/svn-load2yNV6x/working/openssl.cnf Adding /tmp/svn-load2yNV6x/working/certs Adding /tmp/svn-load2yNV6x/working/certs/Entrust.net_Secure_Server_CA.pem Adding /tmp/svn- load2yNV6x/working/certs/WellsSecure_Public_Root_Certificate_Authority.pem Adding /tmp/svn-load2yNV6x/working/certs/5ad8a5d6.0 Adding /tmp/svn-load2yNV6x/working/certs/SwissSign_Platinum_CA_-_G2.pem Traceback (most recent call last): File /usr/bin/svn-load, line 649, in module overlay_files(client, workingdir, d) File /usr/bin/svn-load, line 438, in overlay_files client.add(counterpath, ignore=False) pysvn._pysvn_2_7.ClientError: Error during add of '/tmp/svn- load2yNV6x/working/certs' Can't convert string from native encoding to 'UTF-8': AC_Ra?\195?\173z_Certic?\195?\161mara_S.A..pem uli@zox:~$ svn-load file:///home/uli/tmp-repo ssl /etc/ssl/certs Adding /tmp/svn-loadyTTNvb/working/Entrust.net_Secure_Server_CA.pem Adding /tmp/svn- loadyTTNvb/working/WellsSecure_Public_Root_Certificate_Authority.pem Adding /tmp/svn-loadyTTNvb/working/5ad8a5d6.0 Adding /tmp/svn-loadyTTNvb/working/SwissSign_Platinum_CA_-_G2.pem Traceback (most recent call last): File /usr/bin/svn-load, line 649, in module overlay_files(client, workingdir, d) File /usr/bin/svn-load, line 420, in overlay_files client.add(counterpath, ignore=False) UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 33: ordinal not in range(128) As you see, the actual error differs a bit depending on whether I load /etc/ssl or /etc/ssl/certs. Source browsing a bit seems to indicate that in the first case the file is added due to a recursive add while in the second case it is added directly. In both cases though, it seems that the passed parameter is treated as a string, while it is actually an array of bytes already encoded as UTF-8. As a workaround, using svn_load_dirs from the subversion-tools packages seems to do the job. Greetings! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#679035: Useless warning about disconnected USB device
Package: smartmontools Version: 5.41+svn3365-1 I have a portable hard drive, a WD my passport. Recently, after disconnecting, smart-notifier started giving me regular warnings about not being able to access the disconnected device. Since the device is disconnected, this is little surprising and little helpful. I haven't tried whether this is reliably reproducible. One thing making this case possibly special is that the harddisk actually has bad blocks that would be a reason for warning. A single warning about a disconnected device is probably useless, but doesn't hurt. Repeatedly warning about a device that is disconnected is not only useless but annoying. In partictular since everytime I get a popup, I first have to unfold the part that contains the actual error. If I get too many, I'll start not looking at the errors. The net result is that important errors are shadowed by the useless warnings. Please disable these notifications! Thank you! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#671021: broken link in docs
Package: python-pyinotify Version: 0.9.3-1 If you import the pyinotify module and invoke help(pyinotify), you get the following output: | MODULE DOCS | http://docs.python.org/library/pyinotify This link doesn't exist, pyinotify is not (yet?) part of upstream's standard library. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#665700: thunar scrolls erratically when hover-selecting files
Package: thunar Version: 1.2.3-3 There are two observations which are related, I guess. I'm using the detailed list view (Control-2). The columns displayed are Name, Date Modified, Size. I have checked Automatically expand columns as needed in the configuration. Case I: Here, the size of the Name column is less than the width of the window, but the other two columns exceed the width (just wondering, doesn't that already contradict the settings?), so that a scroll bar is visible at the bottom. When I scroll right, and hover-select a file (different or same as before), the scrollbar is reset to the leftmost position. Case II: Here, the size of the Name column already exceeds the width of the window. If I change the selected file by hovering the mouse over it, the position is scrolled horizontally, alternating between the leftmost and a middle position. The middle position is so that the rightmost part of the Name column is aligned to the right border of the window. How Thunar should behave: It shouldn't scroll horizontally at all. If I look at a column and select files there, it shouldn't scroll away from the column. Further notes: - I'm not sure if the auto-expand code behaves correctly when there simply isn't enough space. - I'm not sure if the auto-expand code works at all. I have seen it change layout sometimes, but I couldn't find a rule to it. - I'm using Fluxbox as window manager, in case that matters (parenting, resize notifications...). Thanks for maintaining this software, btw! Uli -- System Information: Debian Release: wheezy/sid APT prefers testing APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: powerpc (ppc) Kernel: Linux 3.2.0-1-powerpc Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages thunar depends on: ii desktop-file-utils 0.18-1 ii exo-utils 0.6.2-4 ii libatk1.0-0 2.2.0-2 ii libc6 2.13-27 ii libcairo2 1.10.2-7 ii libdbus-1-3 1.4.18-1 ii libdbus-glib-1-20.98-1 ii libexo-1-0 0.6.2-4 ii libgdk-pixbuf2.0-0 2.24.1-1 ii libglib2.0-02.30.2-6 ii libgtk2.0-0 2.24.10-1 ii libgudev-1.0-0 175-3.1 ii libice6 2:1.0.7-2 ii libnotify4 0.7.4-1 ii libpango1.0-0 1.29.4-3 ii libsm6 2:1.2.0-2 ii libthunarx-2-0 1.2.3-3 ii libxfce4ui-1-0 4.8.1-1 ii libxfce4util4 4.8.2-1 ii shared-mime-info0.90-1 ii thunar-data 1.2.3-3 Versions of packages thunar recommends: pn dbus-x111.4.18-1 pn gvfs1.10.1-2+b1 pn libfontconfig1 2.8.0-3.1 pn libfreetype62.4.8-1 pn thunar-volman 0.6.0-4+b1 pn tumbler 0.1.23-1 pn xdg-user-dirs 0.14-1 pn xfce4-panel none Versions of packages thunar suggests: ii thunar-archive-plugin 0.3.0-4 ii thunar-media-tags-plugin 0.2.0-1 -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#665700: [Pkg-xfce-devel] Bug#665700: thunar scrolls erratically when hover-selecting files
On Sunday 25 March 2012 14:05:28 Yves-Alexis Perez wrote: forwarded 665700 https://bugzilla.xfce.org/show_bug.cgi?id=8597 thanks On dim., 2012-03-25 at 11:40 +0200, Ulrich Eckhardt wrote: Package: thunar Version: 1.2.3-3 There are two observations which are related, I guess. I'm using the detailed list view (Control-2). The columns displayed are Name, Date Modified, Size. I have checked Automatically expand columns as needed in the configuration. Case I: Here, the size of the Name column is less than the width of the window, but the other two columns exceed the width (just wondering, doesn't that already contradict the settings?), Exactly the opposite, if you read the help text. That means the columns are expanded so the text is visible. So yes, that might means overflowing the window. Yes, that's the fault in my expectations. I was expecting it to automatically resize the columns but within the limits set by the window. As a programmer, I guess that this would be very hard to program in a way that it works correctly under all corner cases. Further notes: - I'm not sure if the auto-expand code behaves correctly when there simply isn't enough space. Why? Wrong expectations, see above. - I'm not sure if the auto-expand code works at all. I have seen it change layout sometimes, but I couldn't find a rule to it. That's pretty not clear at all. Care to explain? Wrong expectations, I was assuming that Thunar auto-adjusted when resizing the window, but as you said it only adjusts to the required size of the columns. Thanks for the quick feedback! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#663329: has_capture_channel() returns wrong value
Package: libasound2 Version: 1.0.25-2 snd_mixer_selem_has_capture_channel() should return 0 or 1, depending on whether the element has a capture channel. However, it also regularly returns -22 (-EINVAL) here for mixers that are output-only. This not only differs from the documentation, but also from the complementary snd_mixer_selem_has_playback_channel(), which never does this, even for input- only channels. Looking at the sources (slightly reformatted to fit into a mail), the problem seems obvious: int snd_mixer_selem_has_capture_channel( snd_mixer_elem_t *elem, snd_mixer_selem_channel_id_t channel) { CHECK_BASIC(elem); CHECK_DIR(elem, SM_CAP_CVOLUME|SM_CAP_CSWITCH); return sm_selem_ops(elem)-is( elem, SM_CAPT, SM_OPS_IS_CHANNEL, channel); } int snd_mixer_selem_has_playback_channel( snd_mixer_elem_t *elem, snd_mixer_selem_channel_id_t channel) { CHECK_BASIC(elem); return sm_selem_ops(elem)-is( elem, SM_PLAY, SM_OPS_IS_CHANNEL, (int)channel); } The difference is that has_capture_channel() has an additional CHECK_DIR() which contains a return -EINVAL when the element doesn't have either the CVOLUME or the CSWITCH capability. My gut feeling is that the CHECK_DIR() needs to be removed, but I haven't tried that and leave the to people that fully understand the API. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#660461: persistent app settings applied to wrong window
Package: fluxbox Version: 1.3.2-2 Following steps to reproduce the problem: 1. install fspanel (0.7-10) and alsamixergui (0.9.0rc2-1-9.1) 2. start alsamixergui 3. right-click on its titlebar - remember - position 4. start fspanel Initially, fspanel is now positioned on top of the ALSA mixer. The problem is also visible in the apps file after making the position persistent: [app] (name=) (class=) [Position](UPPERLEFT) {1438 17} [end] Of course the actual bug is in both fspanel and alsamixergui, i.e. that they both don't set the mandatory WM_CLASS[1]. However, here is a point where the behaviour of fluxbox could be improved. In the presence of a broken client like those two, it should simply refuse to store any resource information, because the stored information would match every such client. Cheers! Uli [1] http://tronche.com/gui/x/icccm/sec-4.html says This property must be present when the window leaves the Withdrawn state -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#660333: failure to switch desktops and hidden by other windows
Here's a patch against the Debian sources. The modifications go beyond just this bug, I discovered a few others on the way. Changes: 1. Set WM_CLASS property. See #660461 for a reason why. 2. Fix some indention mistakes and typos. 3. Handle sticky windows (desktop=-1) as visible everywhere 4. EWMH compliance: a) test _NET_SUPPORTED to detect compliant WM b) use _NET_NUMBER_OF_DESKTOPS and _NET_CURRENT_DESKTOP for switching c) use _NET_WM_DESKTOP to locate client windows d) prefer _NET_WM_NAME, fallback to WM_NAME for client name e) use _NET_WM_STATE to position fspanel above all other windows f) use _NET_WM_WINDOW_TYPE to mark fspanel as dektop part g) use _NET_CLIENT_LIST instead of _WIN_CLIENT_LIST Note that I neither use fspanel regularly nor have I tested those changes extensively. BTW: I found out that when clicking on a task in fspanel, the window is not raised to the top. Only if the window was previously minimized/iconified it is restored and raised. This might be a fluxbox quirk or configuration, but if you can reproduce this with any other window managers I would file a bug about it, so it isn't forgotten. As a workaround, clicking three times helps, the first focuses, the second minimizes and the third restores and raises the window. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#660333: Patch...
--- fspanel-0.7/fspanel.c 2012-02-19 13:32:54.0 +0100 +++ fspanel.c 2012-02-19 13:59:12.0 +0100 @@ -38,6 +38,14 @@ Display *dd; Window root_win; +/* WM supports EWMH +This flag is set if the window manager supports the EWMH protocol for e.g. +switching workspaces. The fallback if this is not supported is to use the +Gnome variant. This is determined by looking for the presence of the +_NET_SUPPORTED property of the root window. Note that this is only used +for communication with the WM, whether each client supports this protocol +is up to the individual client. */ +int wm_use_ewmh; Pixmap generic_icon; Pixmap generic_mask; GC fore_gc; @@ -72,7 +80,17 @@ char *atom_names[] = { _WIN_CLIENT_LIST, _WIN_WORKSPACE_COUNT, _WIN_STATE, - WM_STATE + WM_STATE, + _NET_NUMBER_OF_DESKTOPS, + _NET_CURRENT_DESKTOP, + _NET_WM_STATE, + _NET_WM_STATE_ABOVE, + _NET_SUPPORTED, + _NET_WM_WINDOW_TYPE, + _NET_WM_WINDOW_TYPE_DESKTOP, + _NET_WM_DESKTOP, + _NET_WM_NAME, + UTF8_STRING, }; #define ATOM_COUNT (sizeof (atom_names) / sizeof (atom_names[0])) @@ -89,6 +107,16 @@ Atom atoms[ATOM_COUNT]; #define atom__WIN_WORKSPACE_COUNT atoms[7] #define atom__WIN_STATE atoms[8] #define atom_WM_STATE atoms[9] +#define atom__NET_NUMBER_OF_DESKTOPS atoms[10] +#define atom__NET_CURRENT_DESKTOP atoms[11] +#define atom__NET_WM_STATE atoms[12] +#define atom__NET_WM_STATE_ABOVE atoms[13] +#define atom__NET_SUPPORTED atoms[14] +#define atom__NET_WM_WINDOW_TYPE atoms[15] +#define atom__NET_WM_WINDOW_TYPE_DESKTOP atoms[16] +#define atom__NET_WM_DESKTOP atoms[17] +#define atom__NET_WM_NAME atoms[18] +#define atom_UTF8_STRING atoms[19] void * @@ -132,8 +160,8 @@ fill_rect (taskbar *tb, int x, int y, in void scale_icon (task *tk) { -int xx, yy, x, y; -unsigned int w, h, d, bw; + int xx, yy, x, y; + unsigned int w, h, d, bw; Pixmap pix, mk = None; XGCValues gcv; GC mgc = None; @@ -223,13 +251,40 @@ get_task_kdeicon (task *tk) } } +/* returns whether the window is visible on the desktop */ int -find_desktop (Window win) +is_visible_on_desktop (Window win, int desk) +{ + int client_desk = -1; + unsigned long *data; + + if (wm_use_ewmh) + data = get_prop_data (win, atom__NET_WM_DESKTOP, XA_CARDINAL, 0); + else + data = get_prop_data (win, atom__WIN_WORKSPACE, XA_CARDINAL, 0); + + if (data) + { + client_desk = *data; + XFree (data); + } + + /* If the client_desk is -1, it is visible on all desktops */ + return (client_desk == -1) || (client_desk == desk); +} + +/* index of the currently displayed desktop */ +int +get_current_desktop () { int desk = 0; unsigned long *data; - data = get_prop_data (win, atom__WIN_WORKSPACE, XA_CARDINAL, 0); + if (wm_use_ewmh) + data = get_prop_data (root_win, atom__NET_CURRENT_DESKTOP, XA_CARDINAL, 0); + else + data = get_prop_data (root_win, atom__WIN_WORKSPACE, XA_CARDINAL, 0); + if (data) { desk = *data; @@ -271,19 +326,36 @@ is_iconified (Window win) return ret; } +/* window name +The returned pointer must be freed using XFree() after use. + +TODO: The encoding for WM_NAME can be STRING or COMPOUND_TEXT. In any case +this encoding should be normalized before returning from this function. */ +char* +get_window_name (Window win) +{ + char* res = NULL; + /* try EWMH's _NET_WM_NAME first */ + res = get_prop_data (win, atom__NET_WM_NAME, atom_UTF8_STRING, 0); + if (!res) + /* fallback to WM_NAME */ + res = get_prop_data (win, XA_WM_NAME, XA_STRING, 0); + return res; +} + void add_task (taskbar * tb, Window win, int focus) { task *tk, *list; - /* is this window on a different desktop? */ - if (tb-my_desktop != find_desktop (win) || is_hidden (win)) + /* is this window on a different desktop or hidden? */ + if (!is_visible_on_desktop (win, tb-my_desktop) || is_hidden (win)) return; tk = calloc (1, sizeof (task)); tk-win = win; tk-focused = focus; - tk-name = get_prop_data (win, XA_WM_NAME, XA_STRING, 0); + tk-name = get_window_name (win); tk-iconified = is_iconified (win); get_task_kdeicon (tk); @@ -331,6 +403,7 @@ gui_create_taskbar (void) { taskbar *tb; Window win; + XClassHint wm_class; MWMHints mwm; XSizeHints size_hints; XWMHints wmhints; @@ -353,11 +426,30 @@ gui_create_taskbar (void) /*value mask*/ CWBackPixel | CWEventMask, /* attribs */ att); - /* don't let any windows cover fspanel */ - set_prop (win, atom__WIN_LAYER, 10); /* WIN_LAYER_ABOVE_DOCK */ + /* set name and class */ + wm_class.res_name = fspanel; + wm_class.res_class = fspanel; + XSetClassHint (dd, win, wm_class); - set_prop (win, atom__WIN_STATE, WIN_STATE_STICKY | - WIN_STATE_FIXED_POSITION); + /* don't let any windows cover fspanel */ + if (wm_use_ewmh) + { + XChangeProperty (dd, win, + atom__NET_WM_STATE, XA_ATOM, 32, + PropModeReplace, + (unsigned char *) atom__NET_WM_STATE_ABOVE, 1); + + XChangeProperty (dd, win, +
Bug#660333: failure to switch desktops and hidden by other windows
Package: fspanel Version: 0.7-10 When using a contemporary fluxbox (1.3.2) as window manager, fspanel fails in at least two ways: 1. It gets hidden by other windows, although it should be on top of everything but fullscreen apps. 2. The two up/down buttons to switch desktops don't work. Note that this still worked with fluxbox 1.1. However, I'm not filing this against fluxbox, because they simply Removed (obsolete) Gnome window manager hints (quoting their changelog). fspanel should instead use EWMH[1] to set itself on top and to switch desktops. Also, one by one, other WMs will also drop support for this, so this isn't only a problem with fluxbox. I'm currently facing similar bugs in a different program (cobweb panel, not yet packaged in Debian). I've already fixed issue 1 above, maybe I'll backport the patch from XCB to Xlib, unless someone beats me to it. Cheers! Uli [1] http://standards.freedesktop.org/wm-spec/wm-spec-latest.html -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#636632: Still present...
...in the new 1.3.2-2, verified on one PowerPC system and another x86 system. I'm wondering, is there a problem with either the description of the problem or the patch? Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#405702: sdl-net1.2 bug #405702 fixed long ago
On Thursday 29 December 2011 01:20:43 you wrote: Sorry for the big delay, there are new efforts to try to maintain SDL packages in better shape. This bug seems to have been fixed long ago (Thu Apr 20 16:48:25 2006 +, about 9 months before this Debian bug report was submitted), in this commit upstream: http://hg.libsdl.org/SDL_net/rev/8551dd6a2235 There is one thing that strikes me though, and that is that I only consider the fix above as a workaround, not as a fix. There are problems with it: 1. the ignored signal is ignored process-wide This means that other code which relies on the signal being raised will fail when libSDL switches the process to ignoring it. 2. the signal is not ignored if a handler is in place If other areas actually registered a handler, the signal will not be ignored, the code in SDL checks this. This could mean that SDL then again starts to fail (I don't think so though). 3. the wrong handler is restored if the signal was ignored If the calling code ignored the signal process-wide, SDL will not ignore it again, but when uninitialized it will again switch it to the default, while the remaining code might still rely on it being ignored. What I don't get is why send() isn't simply called with MSG_NOSIGNAL as I suggested. I switched to dropping SDL_net and creating my own network wrapper doing just that and I never had problems again. I can't provide proper tests though, since I have abandoned that code altogether. Anyway, this bug report was fixed long ago althought it was never properly explained, documented or closed... so doing that right now :) Thank you for this work, Manuel! This bug was nagging me everytime I looked at the list of bugs I filed, because I never understood how it could linger for so long without more people complaining. The fact that a functioning workaround is in place explains it. Note that I don't object to closing this bug, as obviously the fix is good enough, but I wanted to raise the concerns so they are at least documented. Cheers! :) Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#651207: whitespace handling in commandline parsing
Package: aptsh Version: 0.0.7+nmu2 Entering remove foo tries to remove foo. Entering remove foo complains about not being able to find e, remove foo complains about not being able to finde ve. It seems like after detecting the remove token, it simply skips ahead six chars from the beginning before looking for the next token. Cheers and thanks for this tool! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#631127: very strange KMail error
Hi! I guess the very strange KMail error mentioned in [1] is actually described in [2], right? In any case, I can also confirm that in testing, current kmail (4:4.4.11.1+l10n-1) with libsasl ( 2.1.24~rc1.dfsg1+cvs2011) is still broken. I've tried version 2.1.25 from [3], too, but it doesn't fix those two bugs either. What I'm still puzzled by is the issue with the domain in /etc/resolv. What exactly do I have to do to get this to work again? Also, is that a workaround for the libsasl2 bug or is it a bug in my /etc/resolf that just happened to work with older libsasl versions? Cheers! Uli [1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=636534 [2] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=631127 [3] ftp.de.debian.org:/debian/pool/main/c/cyrus-sasl2/libsasl2-2_2.1.25.dfsg1-2_powerpc.deb -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#639837: UI hangs for 30s while loading pages
On Tuesday 30 August 2011 22:32:31 Mike Hommey wrote: How big is the ~/.mozilla/firefox/*/sessionstore.js file ? uli@fruitbox:~$ ll ~/.mozilla/firefox/*/sessionstore.js -rw--- 1 uli uli 305k Aug 30 21:43 /home/uli/.mozilla/firefox/xzqx6a0f.default/sessionstore.js 305k on one line .. is that much/little/normal? Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#639837: UI hangs for 30s while loading pages
On Wednesday 31 August 2011 08:18:02 Mike Hommey wrote: Another thing I can think of that could explain your symptoms is if you have ipv6 enabled, but ipv6 doesn't actually work. I'd suggest you try to disable ipv6 (about:config, network.dns.disableIPv6) and see how it goes from there. Tried that, no change. Maybe a tiny bit less often, but still regularly present. In any case, it would not explain why the UI freezes (I'd hope that Iceweasel doesn't do blocking networking requests in the UI loop) or the presence of mentioned error messages about FD abuse in the logfile. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#639837: UI hangs for 30s while loading pages
Package: iceweasel Version: 5.0-6 Severity: important Iceweasel regularly hangs while loading pages. Many pages show the symptoms, though never regularly. I haven't determined if some do and some don't. It could also be that some advertisments that are being randomly served influence the issue. Following symptoms: 1. Complete UI freeze No switching to other tabs, aborting loading of the website, not even redrawing the window. 2. There is always some pending IO The lower-left status popup (where e.g. the URL you're hovering over is displayed) shows the state of the loading operation. While the UI hangs, this is typically in the state Waiting for www.example.com 3. In the output (~/.xsession-errors), I get these four lines: ** (firefox-bin:29217): WARNING **: Serious fd usage error 14 ** (firefox-bin:29217): WARNING **: Serious fd usage error 10 Yes, two empty ones and two with these warnings! These always (or usually) come in such pairs. I have seen errors 10, 13, 14 (according to errno those are ECHILD, EACCES, EFAULT) there. 4. Neither high CPU (htop) nor IO (iostat) load occurs. 5. After 30s, loading and UI operation resumes, without any faults. The interesting thing here is that these 30s are reproducible, not randomly. I watched the clock after a few such hangs to find that this is systematic indeed. Good luck! Uli -- Package-specific info: -- Extensions information Name: Default themefalse Location: /usr/lib/iceweasel/extensions/{972ce4c6-7e08-4474-a285-3208198ce6fd} Package: iceweasel Status: enabled Name: DownloadHelper extensionfalse Location: ${PROFILE_EXTENSIONS}/{b9db16a4-6edc-47ec-a1f4-b86292ed211d} Status: enabled -- Plugins information Name: Shockwave Flash Location: /usr/lib/mozilla/plugins/libflash-mozplugin.so Package: libflash-mozplugin Status: enabled -- Addons package information ii iceweasel 5.0-6 Web browser based on Firefox ii libflash-mozpl 0.4.13-9 GPL Flash (SWF) Library - Mozilla-compatible -- System Information: Debian Release: wheezy/sid APT prefers testing APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: powerpc (ppc) Kernel: Linux 2.6.39-rc7-powerpc Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages iceweasel depends on: ii debianutils 4.0.2 Miscellaneous utilities specific t ii fontconfig2.8.0-3generic font configuration library ii libc6 2.13-16Embedded GNU C Library: Shared lib ii libgcc1 1:4.6.1-4 GCC support library ii libgdk-pixbuf2.0-02.23.5-3 GDK Pixbuf library ii libglib2.0-0 2.28.6-1 The GLib library of C routines ii libgtk2.0-0 2.24.4-3 The GTK+ graphical user interface ii libnspr4-0d 4.8.9-1NetScape Portable Runtime Library ii libstdc++64.6.1-4GNU Standard C++ Library v3 ii procps1:3.2.8-11 /proc file system utilities ii xulrunner-5.0 5.0-6 XUL + XPCOM application runner iceweasel recommends no packages. Versions of packages iceweasel suggests: ii libgssapi-krb5-21.9.1+dfsg-1 MIT Kerberos runtime libraries - k pn mozplugger none (no description available) ii ttf-lyx 2.0.0-1 TrueType versions of some TeX font pn ttf-mathematica4.1 none (no description available) pn xfonts-mathml none (no description available) pn xprint none (no description available) Versions of packages xulrunner-5.0 depends on: ii libasound2 1.0.24.1-2 shared library for ALSA applicatio ii libatk1.0-0 2.0.1-2 ATK accessibility toolkit ii libbz2-1.0 1.0.5-6 high-quality block-sorting file co ii libc6 2.13-16 Embedded GNU C Library: Shared lib ii libcairo2 1.10.2-6.1 The Cairo 2D vector graphics libra ii libdbus-1-3 1.4.14-1 simple interprocess messaging syst ii libevent-1.4-2 1.4.13-stable-1 An asynchronous event notification ii libfontconfig1 2.8.0-3 generic font configuration library ii libfreetype62.4.6-1 FreeType 2 font engine, shared lib ii libgcc1 1:4.6.1-4GCC support library ii libgdk-pixbuf2.0-0 2.23.5-3 GDK Pixbuf library ii libglib2.0-02.28.6-1 The GLib library of C routines ii libgtk2.0-0 2.24.4-3 The GTK+ graphical user interface ii libhunspell-1.2-0 1.2.14-4 spell checker and morphological an ii libjpeg62 6b1-2Independent JPEG Group's JPEG runt ii libmozjs5d 5.0-6Mozilla SpiderMonkey JavaScript li ii libnspr4-0d
Bug#636632: fbrun history is broken
Package: fluxbox Version: 1.3.1~dfsg1-2 The history management of fbrun is broken. The most obvious symptom is that after starting something with it, the next time fbrun is started the cursor up button doesn't reproduce that command. While one could argue that this is still subjective, there is another thing that isn't. The part that is more of a proof is the smashed content of the history file. For that try the following: 0. remove ~/.fluxbox/fbrun_history (or back it up) 1. run fbrun and enter xterm as command 2. run fbrun and enter sed as command 3. run fbrun and enter grep as command 4. run fbrun and enter xterm as command 5. run fbrun and enter xterm as command Don't use the history for entering xterm the second and third time, type them manually! The contents of fbrun_history on my machine are now: | sexterm | | xterm I've been looking around the sources a bit, and I'll try to provide a patch. I have already found one error and one thing that at least makes no sense to me. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#636632: Patch...
There are two changes in this patch: 1. Loading the history, the code used while (!infile.eof()) { getline(infile, ...); ... } That's simply wrong, because it will read one superfluous empty line at the end. You need to check if reading succeeded instead of checking if the stream already encountered EOF. 2. Writing back the history, the code tried to skip elements until it found the current one, then write the rest and then finally the current command. This avoids writing the beginning of the file, but firstly it doesn't work, and secondly it is IMHO a completely unnecessary optimization because a) it isn't called often enough and b) the file is 1kB here after years of use. I'm using the most simple implementation and just write the whole history every time, skipping all duplicates and finally append the current command. There was one thing I'm not sure how the code was supposed to work and which wasn't documented either. The question is how editing of previously entered commands is reflected in the history. There are two approaches: 1. If I enter xter (if I pressed enter too early), going back in history and then appending the m, it replaces the xter with xterm in history. 2. If I enter xter, going back in history and appending the m yields two entries xter and xterm. I chose approach 1 in the patch. However, that means that if a user wants to have two entries like xterm sh and xterm bash, they can't achieve that by just editing the history, they must enter them completely separately instead. If you don't like that, just skip the code leading to the second continue in the look that writes the history. ;) Cheers! Uli Index: FbRun.cc === --- FbRun.cc (revision 34) +++ FbRun.cc (revision 39) @@ -137,43 +137,22 @@ hide(); // hide gui // save command history to file -if (text().size() != 0) { // no need to save empty command +if (!command.empty()) { // no need to save empty command -// don't allow duplicates into the history file, first -// look for a duplicate -if (m_current_history_item m_history.size() - text() == m_history[m_current_history_item]) { -// m_current_history_item is the duplicate -} else { -m_current_history_item = 0; -for (; m_current_history_item m_history.size(); - ++m_current_history_item) { -if (m_history[m_current_history_item] == text()) -break; -} -} - -fstream inoutfile(m_history_file.c_str(), ios::in|ios::out); -if (inoutfile) { -// now m_current_history_item points at the duplicate, or -// at m_history.size() if no duplicate -if (m_current_history_item != m_history.size()) { -unsigned int i = 0; -// read past history items before current -for (; inoutfile.good() i m_current_history_item; i++) -inoutfile.ignore(1, '\n'); - -// write the history items that come after current -for (i++; i m_history.size(); i++) -inoutfilem_history[i]endl; - -} else { -// set put-pointer at end of file -inoutfile.seekp(0, ios::end); +ofstream outfile(m_history_file.c_str()); +if (outfile) { +for (unsigned i = 0; i != m_history.size(); ++i) { +// don't allow duplicates into the history file +if (m_history[i] == command) +continue; +// skip currently selected item, this will be appended but is +// not caught as duplicate when it was modified by the user +if (i == m_current_history_item) +continue; +outfilem_history[i]endl; } // append current command to the file -inoutfilecommandendl; - +outfilecommandendl; } else cerrFbRun Warning: Can't write command history to file: m_history_fileendl; } @@ -197,8 +176,7 @@ m_history.clear(); // each line is a command string line; -while (!infile.eof()) { -getline(infile, line); +while (getline(infile, line)) { if (line.size()) // don't add empty lines m_history.push_back(line); }
Bug#598456: Solved...
Hi! It seems there is a setting that tells fluxbox to ignore or not ignore the tabs when maximizing, it is under the Tab Options and called Maximize Over. With that deactivated, windows are now maximized as expected. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#633157: Using wrong configuration file leads to data loss
Package: fluxbox Version: 1.3.1~dfsg1-2 I can tell fluxbox to use a different configuration file: $ fluxbox -help [...] -rc stringuse alternate resource file. [...] This works to the extent that if I specify a different path, it will actually use it. However, and that is inacceptable, inside the file which is autogenerated if it doesn't exist, there are still references to my other configuration: session.styleOverlay: ~/.fluxbox/overlay session.keyFile:/home/uli/.fluxbox/keys session.appsFile: ~/.fluxbox/apps session.slitlistFile: /home/uli/.fluxbox/slitlist session.styleFile: /home/uli/.fluxbox/styles/Sn33z-mod If I save anything in e.g. the apps file, which happens during normal WM interaction, the configuration there is lost, although I specified a different path exactly in order to avoid overwriting it. Some more notes: 1. It looks weird that some paths above are absolute, while some are relative to the homedir. 2. Fluxbox doesn't have a configuration file, but a configuration dir. The commandline should honour that, i.e. not expect just a single init file but a directory. 3. If I specify a path to a file that doesn't exist, I'd expect to either get an error or that the file is created. This works to some extent, but if I specify a file where the parent dir doesn't exist, I get errors on stdout (most users won't see that) and fluxbox uses the default configuration from my home dir, which again can lead to data loss. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#632688: ConfigureWindow not handled correctly
Package: fluxbox Version: 1.3.1~dfsg1-2 The attached file is an excerpt from a program. It creates a single window, undecorated, on top of other windows and in the upper left corner of the screen. At least it should do that, and it used to. At the moment, it only achieves parts of this, the resulting position is wrong. The program uses XCB, but I have tried an earlier version of the program that uses Xlib, and it suddenly has the same problems, so that is not the cause for problems. I have tried openbox, sawfish and icewm, and all three the WMs work as expected. I have also tried fluxbox 1.1.1+dfsg2-1 and it shows the same wrong behaviour as the current version. If that had always been the case, I'm pretty sure that I would have noticed before, because I've been running fluxbox for ages. That again makes me wonder if this is really just a problem with fluxbox, at least I suspect other influences, like possibly in the X server. Cheers! Uli // g++ -g -O0 -W test.cpp $(pkg-config xcb xcb-icccm xcb-shape xcb-image --libs --cflags) #include xcb/xcb.h #include xcb/xcb_icccm.h #include stdint.h #include stdexcept #include iostream #include string.h #include stdlib.h #include iomanip xcb_connection_t* connection = 0; xcb_atom_t resolve_atom_name(char const* name) { size_t const len = strlen(name); if(static_castuint16_t(len) != len) throw std::runtime_error(xatom: name too long); xcb_intern_atom_cookie_t cookie = xcb_intern_atom(connection, 0, // only_if_exists static_castuint16_t(len), name); xcb_generic_error_t* error = 0; xcb_intern_atom_reply_t* reply = xcb_intern_atom_reply(connection, cookie, error); if(error) throw std::runtime_error(xcb_intern_atom_reply failed with error); if(!reply) throw std::runtime_error(xcb_intern_atom_reply failed without reply); xcb_atom_t const res = reply-atom; free(reply); return res; } int main() { try { /* open connection with the server */ connection = xcb_connect(NULL,NULL); if (xcb_connection_has_error(connection)) throw std::runtime_error(xcb_connect() failed); /* retrieve the setup */ xcb_setup_t const* setup = xcb_get_setup(connection); if(!setup) throw std::runtime_error(xcb_get_setup() failed); /* get the first screen */ xcb_screen_t* screen = xcb_setup_roots_iterator(setup).data; if(!screen) throw std::runtime_error(missing root screen); /* create window */ uint32_t const win_values[] = { screen-white_pixel, ( XCB_EVENT_MASK_EXPOSURE |XCB_EVENT_MASK_BUTTON_PRESS |XCB_EVENT_MASK_BUTTON_RELEASE) }; xcb_window_t window_id = xcb_generate_id(connection); xcb_create_window(connection, XCB_COPY_FROM_PARENT, // depth window_id, screen-root, // parent 0, 0, // position 100, 100, // size 0, // border width XCB_WINDOW_CLASS_INPUT_OUTPUT, // class XCB_COPY_FROM_PARENT, // visual XCB_CW_BACK_PIXEL | XCB_CW_EVENT_MASK, win_values); // resolve properties, (IDs, types and values) xcb_atom_t const wm_name_atom = resolve_atom_name(_NET_WM_NAME); xcb_atom_t const wm_window_type_atom = resolve_atom_name(_NET_WM_WINDOW_TYPE); xcb_atom_t const win_layer_atom = resolve_atom_name(_WIN_LAYER); xcb_atom_t const utf8_string_atom = resolve_atom_name(UTF8_STRING); xcb_atom_t const cardinal_atom = resolve_atom_name(CARDINAL); xcb_atom_t const atom_atom = resolve_atom_name(ATOM); xcb_atom_t const wm_window_type_desktop_atom = resolve_atom_name(_NET_WM_WINDOW_TYPE_DESKTOP); // set window name xcb_change_property(connection, XCB_PROP_MODE_REPLACE, window_id, wm_name_atom, utf8_string_atom, 8, strlen(test), test); // set layer to 10 (above dock) uint32_t layer = 10; xcb_change_property(connection, XCB_PROP_MODE_REPLACE, window_id, win_layer_atom, cardinal_atom, 32, 1, layer); // set window type to desktop element to avoid decoration xcb_change_property(connection, XCB_PROP_MODE_REPLACE, window_id,
Bug#630148: Failure to clean up resources in temporary dir
Package: konqueror Version: 4:4.4.5-3 Note up front: This is probably not only a problem with konqueror but also with other KDE parts, but it hit me most with konqueror. The problem is that konqueror caches files in /var/tmp/ from webbrowsing, which are never cleaned up. I recently removed a folder /var/tmp/kdecache- uli/http which took up 1.5GiB(!) of space there. Note that rebooting or anything like that doesn't help, these files are never cleaned up AFAICT. Two wishes here: 1. If applications cache things, they also need to clean up their cache. 2. If it is not the applications, the surrounding system itself should clean up regularly. Thank you! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#630150: lnusertemp fails to honour TEMPDIR
Package: kdelibs4c2a Version: 4:3.5.10.dfsg.1-5 TEMPDIR is the canonical Unix way to specify the location for temporary files. lnusertemp doesn't honour this environment variable and simply creates these files in /tmp or /var/tmp. As a workaround, you can set KDETMP and KDEVARTMP, as per manpage. Thanks! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#629581: Zombie windows after closing terminal
Package: fluxbox Version: 1.3.1~dfsg1-2 Problem: The simple procedure to reproduce the problem is to run xterm -e false in a shell (xterm -e false works, too). This will create an xterm that is immediately closed again. Sometimes, this window will get stuck. Checking the output of ps ux|grep xterm, I see that the xterm process was already terminated, so the window is left without the according X client. Workaround: In order to remove that window, just repeat the command that actually caused it. In rare cases, you will end up with one more zombie, but now and then the zombies are removed. Notes: - What makes this special is that the client only lives for a very short time. Due to the asynchronous nature of X, it is possible that the client terminates even before the window manager has fully mapped the window that was created. At least that's what I would look at, but I don't really know. - This might be related to[1], but there it mentions tab groups. Cheers! Uli [1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=543387 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#598456: Still present...
...using version 1.3.1~dfsg1-2. Also, just wondering, was this forwarded upstream yet? Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#465301: Partially fixed...
Hi! I'm running icewm 1.3.7-1 here and blast 1.1-22 here and windows don't close their holes again when they are moved. When they are resized, they will close the holes though. That said, I'm not sure I would call this behaviour a bug though. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#628203: System load tooltip in panel displays wrong data
Package: icewm Version: 1.3.7-1 In the panel/taskbar/system tray is a small applet displaying the CPU load. If you hover the mouse over it, you get a tooltip window displaying various information textually: 1. CPU Load: 0.58 0.97 1.04, 270 I can recognize the first three and know where to look up their meaning (Though I really doubt Joe average User can!), but the last value is a mystery to me. 2. Ram: 1007.03/80.86M Firstly, it should probably be RAM, not Ram. Secondly, I guess it should be 80.86/1007.03M instead, i.e. the free RAM. Or maybe it should be 926.17/1007.03M, which is the used RAM just as the CPU and swap usage display. As optional feature, I would display the usage in either percent, like the CPU load and/or in MiB instead of MB, in order to avoid fractional values. 3. Swap: 1953.15/1868.25M Again, I would swap these two values like with the RAM above or even make this a percentage. 4. ACPI Temp: I'm on a PowerPC system, but ACPI is a x86-ism, AFAIK. Hence also no temperature here, but I'd rather say the whole entry should be removed when no temperature is available. 5. CPU Freq: 0.000GHz This value is simply wrong, this is also the main issue that prompted me to write this bug report, before I actually really looked at the values there and started thinking. ;) Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#628417: Missing menu entries to start application
Package: kmail Version: 4:4.4.7-3 Hi! Since quite some time (roughly since the transition of testing to KDE 4, I believe) there haven't been any menu entries for most KDE applications, like dolphin, konqueror, knode and kmail. Interestingly k3b and ktorrent have menu entries, but I believe those are not parts of the KDE core. It would be great to have those menu entries back! ;) Thanks! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#442908: Seems fixed....
I tried versions 1:3.6.1-4 (unstable) and 1:3.6.99svn11761-1 (experimental). Neither of these three bugs were reproducible with these two versions. Good job, thank you! :) Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#257641: Update?
On Tuesday 10 May 2011 20:45:32 Pierre Habouzit wrote: I'm unable to reproduce. valgrind is actually a shell script that does what it has to wrt LD_LIBRARY_PATH, are you sure you can reproduce? what do you set in your LD_LIBRARY_PATH, please give steps to reproduce. I'm not using LD_LIBRARY_PATH and it still segfaults, but I'm on a PowerPC system, see #603961. Maybe Martin is on a PowerPC system, too, and LD_LIBRARY_PATH is actually irrelevant? Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#626009: Should suggest evas-engine
Package: libevas1 Version: 1.0.0-1 Hi! Firstly, thanks for packaging this, I'm thrilled to try out what the Enlightenment team has created. For that reason, I installed python-evas and started toying with it. However, none of the examples I tried worked, and that for the simple reason that I didn't have any rendering engines installed. I'd Suggests: libevas1-engines-core|libevas1-engines-x or something like that. Alternatively, I'd maybe add a Provides: libevas1-engine to these two packages and a Depends: libevas1-engine to libevas1, but I guess you will find the best way to assure that at least some engine can be loaded. BTW: There is a bug in libevas1 that it reports all possible engines in evas_render_method_list(), even those that are not installed and that can't be loaded. Only when trying to load one it returns zero, but the Python wrapper doesn't interpret that as error. These two separate issues contributed to my problems getting things to run. Forcing me to install at least one engine would have helped me a lot! Thank you! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#567793: your mail
On Saturday 26 February 2011 12:29:37 you wrote: is it going better in squeeze or sid? I just switched to 24BPP and EXA, and the artifacts are still there. Switching to XAA fixes the issues again. The version of xserver-xorg-video-radeon is 1:6.13.1-2+squeeze1. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#610269: Could not write to file newsserver when sending posting
Package: knode Version: 4:4.4.7-2 Since a few days, knode fails to send postings to at least some newsgroup. The group in question that doesn't work is comp.lang.c++. A group that does work is de.test. When sending, I get Could not write to file news.hansenet.de, which is my newsserver!? Further notes: - Around the time that happened first, I ran out of space on the root partition (Just wondering, why does KDE store stuff there even though I have set $TMPDIRCo to something else?). I did not get any disk full errors though and this is also no longer the case. - I have quit and restarted knode. - I have unsubscribed from that group and resubscribed again. - I subscribed to de.test just for testing. - I have not yet tried any other newsgroups. - I have started knode from the commandline, no errors visible there expect: Connecting to deprecated signal QDBusConnectionInterface::serviceOwnerChanged(QString,QString,QString) knode(16839)/kdecore (KConfigSkeleton) KCoreConfigSkeleton::writeConfig: Any ideas how to fix this or what else to try? Thanks! Uli -- System Information: Debian Release: 6.0 APT prefers testing APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: powerpc (ppc) Kernel: Linux 2.6.37-rc5-powerpc Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages knode depends on: ii kdebase-runtime 4:4.4.5-1 runtime components from the offici ii kdepim-runtime4:4.4.7-1 Runtime components for akonadi-kde ii kdepimlibs-kio-plugins4:4.4.5-2 kio slaves used by KDE PIM applica ii libc6 2.11.2-7 Embedded GNU C Library: Shared lib ii libgcc1 1:4.4.5-8 GCC support library ii libkabc4 4:4.4.5-2 library for handling address book ii libkde3support4 4:4.4.5-2 the KDE 3 Support Library for the ii libkdecore5 4:4.4.5-2 the KDE Platform Core Library ii libkdepim44:4.4.7-2 KDE PIM library ii libkdeui5 4:4.4.5-2 the KDE Platform User Interface Li ii libkhtml5 4:4.4.5-2 the KHTML Web Content Rendering En ii libkio5 4:4.4.5-2 the Network-enabled File Managemen ii libkmime4 4:4.4.5-2 library for handling MIME data ii libkontactinterface4 4:4.4.5-2 Kontact interface library ii libkparts44:4.4.5-2 the Framework for the KDE Platform ii libkpgp4 4:4.4.7-2 gpg based crypto library ii libkpimidentities44:4.4.5-2 library for managing user identiti ii libkpimtextedit4 4:4.4.5-2 library that provides a textedit w ii libkpimutils4 4:4.4.5-2 library for dealing with email add ii libkutils44:4.4.5-2 various utility classes for the KD ii libmailtransport4 4:4.4.5-2 mail transport service library ii libqt4-dbus 4:4.6.3-4 Qt 4 D-Bus module ii libqt4-qt3support 4:4.6.3-4 Qt 3 compatibility library for Qt ii libqt4-xml4:4.6.3-4 Qt 4 XML module ii libqtcore44:4.6.3-4 Qt 4 core module ii libqtgui4 4:4.6.3-4 Qt 4 GUI module ii libstdc++64.4.5-8The GNU Standard C++ Library v3 knode recommends no packages. knode suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#608034: Useless error message on startup failure
Package: pavucontrol Version: 0.9.9-1 When trying to start pavucontrol, it gave me the error Connection failed: connection refused. This is rather unhelpful, since the problem is that the pulseaudio package is not installed (shouldn't it depend on it?) and that it therefore can't start the pulseaudio daemon, I guess. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#541538: Sound volume jumps on play
On Saturday 25 December 2010 19:26:44 you wrote: It looks like the symptom that you describe is likely a sound driver issue. Depending on what you call a driver, I agree. The point is that plain ALSA output without pulseaudio works, so that part of the driver stack is not the problem. However, adding pulseaudio as intermediate driver layer causes problems. Are you still able to reproduce it in current testing? I'm not experiencing the jumps in the output level, though I haven't tested it extensively. The output sliders twitch when starting, but neither is the effect permanent nor audible. However, I also mentioned that the problem was not 100% reproducible even at that time (see point 5). Considering point one, there's still a similar effect in place. If I run alsamixergui in parallel, changing the PCM volume there affects both the sliders for Amarok on the Playback tab and those on the Output Devices tab of pavucontrol. Using pavucontrol, I can change both pairs of sliders separately though. The behaviour of alsamixergui is to follow changes on both to some extent. Considering point 6, this effect is still present. However, the absolute limite of 43% percent actually depends on the the settings of both the sliders on the Playback and Output Devices tabs of pavucontrol. In any case, the static begins when turning the PCM output to close to zero on alsamixergui or when it reaches this level through the combined pavucontrol slider levels. That and the occasional cracks and plops made stop using pulseaudio, so I can't really say whether the main part of the bug report is still applicable with current versions. https://wiki.ubuntu.com/Audio/AlsaInfo See attachment, hope that helps! Cheers and thanks! Uli upload=truescript=truecardinfo= !! !!ALSA Information Script v 0.4.59 !! !!Script ran on: Sun Dec 26 11:03:13 UTC 2010 !!Linux Distribution !!-- Debian GNU/Linux squeeze/sid \n \l !!DMI Information !!--- Manufacturer: Product Name: !!Kernel Information !!-- Kernel release:2.6.37-rc5-powerpc Operating System: GNU/Linux Architecture: ppc Processor: unknown SMP Enabled: No !!ALSA Version !! Driver version: 1.0.23 Library version:1.0.23 Utilities version: 1.0.23 !!Loaded ALSA modules !!--- snd_aoa_fabric_layout !!Sound Servers on this system !! Pulseaudio: Installed - Yes (/usr/bin/pulseaudio) Running - Yes ESound Daemon: Installed - Yes (/usr/bin/esd) Running - No aRts: Installed - Yes (/usr/bin/artsd) Running - No !!Soundcards recognised by ALSA !!- 0 [SoundByLayout ]: AppleOnbdAudio - SoundByLayout SoundByLayout !!PCI Soundcards installed in the system !!-- !!Advanced information - PCI Vendor/Device/Susbsystem ID's !! !!Modprobe options (Sound related) !! snd-atiixp-modem: index=-2 snd-intel8x0m: index=-2 snd-via82xx-modem: index=-2 snd-pcsp: index=-2 snd-usb-audio: index=-2 snd-powermac: index=0 !!Loaded sound module options !!-- !!Module: snd_aoa_fabric_layout * : !!ALSA Device nodes !!- crw-rw+ 1 root audio 116, 4 Dec 24 18:58 /dev/snd/controlC0 crw-rw+ 1 root audio 116, 6 Dec 24 18:58 /dev/snd/pcmC0D0c crw-rw+ 1 root audio 116, 5 Dec 26 11:21 /dev/snd/pcmC0D0p crw-rw+ 1 root audio 116, 3 Dec 24 18:58 /dev/snd/seq crw-rw+ 1 root audio 116, 2 Dec 24 18:58 /dev/snd/timer /dev/snd/by-path: total 0 drwxr-xr-x 2 root root 60 Dec 24 18:58 . drwxr-xr-x 3 root root 160 Dec 24 18:58 .. lrwxrwxrwx 1 root root 12 Dec 24 18:58 pci-0001:10:17.0 - ../controlC0 !!Aplay/Arecord output !! APLAY List of PLAYBACK Hardware Devices card 0: SoundByLayout [SoundByLayout], device 0: Master [] Subdevices: 0/1 Subdevice #0: subdevice #0 ARECORD List of CAPTURE Hardware Devices card 0: SoundByLayout [SoundByLayout], device 0: Master [] Subdevices: 1/1 Subdevice #0: subdevice #0 !!Amixer output !!- !!---Mixer controls for card 0 [SoundByLayout] Card hw:0 'SoundByLayout'/'SoundByLayout' Mixer name: 'SoundByLayout' Components: '' Controls : 5 Simple ctrls : 5 Simple mixer control 'Headphone',0 Capabilities: pswitch pswitch-joined penum Playback channels: Mono Mono: Playback [on] Simple mixer control 'Headphone Detect Autoswitch',0 Capabilities: pswitch pswitch-joined penum Playback channels: Mono Mono: Playback [on] Simple mixer control 'Headphone Detected',0 Capabilities: pswitch pswitch-joined penum Playback channels: Mono Mono: Playback [on] Simple mixer control 'Speakers',0
Bug#544112: Fixed?
Hi! I recently noticed that the problem seems to be gone. I then downloaded and installed fluxbox_1.1.1+dfsg2-1_powerpc.deb, restarted fluxbox, and the problem reappeared, so that version is affected. Installing fluxbox_1.1.1+git20100908.df2f51b-2_powerpc.deb again, the effect is gone again, so something in between must have fixed the issue. Note that I picked this package directly from the FTP server, APT doesn't seem to know about it. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#605988: Can't undo all changes without first selecting a specific one
Package: kompare Version: 4:4.4.5-1 I'm looking at a kompare window opened by kdesvn when selecting diff local changes there. 'ps aux' says the commandline is kompare -on -. The button to revert all changes to the file (Apply All) is deactivated until I select a specific change in the file, it should always be activated if there is any change to revert. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#605998: Text View behaves weirdly
Package: kompare Version: 4:4.4.5-1 Under the Settings menu, there is an option Show Text View. There are several things surprising with this text view: 1. The option is not saved anywhere, so if I select the text view, close kompare and reopen it, the text view is gone and I have to activate it again. I would have expected that kompare remembers things selected under Settings. 2. The position in the text view is the last difference, regardless of which position in the source/target views I selected previously. The position also doesn't move, neither the position in the source/target view when moving through the text view nor the one in the text view when moving through the source/target views. 3. I can edit text in the text view. I have looked around a bit, but I couldn't find any lasting effect of these edits. They neither show up in source/target view nor after saving the files. 4. I can not really edit text in the text view. For example, trying to type a space triggers a hotkey that is bound to Apply Difference instead. 5. I can't detach the text view from the main window. If I do, it looses all its content and decoration. Further, I get a stream of errors on stdout: X Error: BadMatch (invalid parameter attributes) 8 Major opcode: 62 (X_CopyArea) Resource id: 0x58053fc These come every second or so while the mouse is inside the detached window which I can still move and resize or reattach, which restores the content and stops the errors. I tried with dolphin, which uses similar detachable windows and there I don't have any such problems. I tried with both fluxbox and icewm in order to rule out the WM as the problem for this. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#606014: various errors printed to console
Package: kompare Version: 4:4.4.5-1 When I just start kompare without any arguments, it outputs the following two errors on the terminal window: kompare(2276)/kdecore (KLibrary) findLibraryInternal: plugins should not have a 'lib' prefix: libkomparepart.so kompare(2276)/kdecore (KLibrary) findLibraryInternal: plugins should not have a 'lib' prefix: libkomparenavtreepart.so When running it with two files to compare, it gives me this: kompare(2466)/kdecore (KLibrary) findLibraryInternal: plugins should not have a 'lib' prefix: libkomparepart.so kompare(2466)/kdecore (KLibrary) findLibraryInternal: plugins should not have a 'lib' prefix: libkomparenavtreepart.so kompare(2466)/kdecore (services) KMimeTypeFactory::parseMagic: Now parsing /usr/share/mime/magic load glyph failed err=6 face=0x108a32e0, glyph=512 QFontEngine: Glyph neither outline nor bitmap format=0 The last two lines are then repeated over and over while the program is running. Just in case it matters, Unicode codepoint 512 (u200) is Latin capital letter A with double grave. Since I'm comparing text files that consist of a subset of ASCII, their content shouldn't be the reason for these errors. Greetings! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#606026: handle_event.c:565: handle_breakpoint: Assertion `sbp' failed.
Package: ltrace Version: 0.5.3-2.1 Severity: important Consider this three-liner: #include iostream int main() {} Compiled with g++ x.cpp and run with ltrace ./a.out, I get the following output: _start(1, 0xbfb13834, 0xbfb1383c, 0xbfb138d4, 0x4800f930 unfinished ... __libc_start_main(1, 0xbfb13834, 0xbfb1383c, 0xbfb138d4, 0x4800f930 unfinished ... _ZNSt8ios_base4InitC1Ev(0x10010b50, 65535, 0xbfb1383c, 0xbfb138d4, 0ltrace: handle_event.c:565: handle_breakpoint: Assertion `sbp' failed. Aborted If I remove the first line, no such crash happens. g++ is at version 4:4.4.5-1, libstdc++6 is version 4.4.5-8, just in case that matters. I also tried downgrading to ltrace to 0.5-3.1 from stable, with that version I don't get any crash either. Cheers! Uli -- System Information: Debian Release: squeeze/sid APT prefers testing APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: powerpc (ppc) Kernel: Linux 2.6.33-2-powerpc Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages ltrace depends on: ii libc6 2.11.2-7 Embedded GNU C Library: Shared lib ii libelfg0 0.8.13-1 an ELF object file access library ltrace recommends no packages. ltrace suggests no packages. -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#603961: valgrind: segmentation fault on PowerPC
Package: valgrind Version: 1:3.6.0~svn11254+nmu1 Severity: important Hi! Valgrind is broken for PowerPC, just starting it without any further command to run already causes a segmentation fault. Running it on any executable has the same problem. There is a patch available, the simple recipe mentioned in https://bugs.launchpad.net/ubuntu/+source/valgrind/+bug/484048 or http://bugs.kde.org/show_bug.cgi?id=214223 works. Or maybe one should say it works around the issue, which could be nonportable valgrind code or a GCC bug. Note: This is not related to http://bugs.debian.org/cgi- bin/bugreport.cgi?bug=529016, which also reports breakage on PowerPC, but the problem is a different one. With the change mentioned above, valgrind is again usable on PowerPC. Thanks! Uli -- System Information: Debian Release: squeeze/sid APT prefers testing APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: powerpc (ppc) Kernel: Linux 2.6.33-2-powerpc Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages valgrind depends on: ii libc6 2.11.2-7 Embedded GNU C Library: Shared lib ii libc6-dbg 2.11.2-7 Embedded GNU C Library: detached d Versions of packages valgrind recommends: ii gdb 7.0.1-2+b1 The GNU Debugger Versions of packages valgrind suggests: pn alleyoop none (no description available) pn kcachegrind none (no description available) pn valkyrie none (no description available) -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#257641: Update?
Greetings! I'm facing the same problem, but: 1. valgrind's version is 1:3.6.0~svn11254+nmu1 2. libc6-dbg's version is 2.11.2-7 3. valgrind actually depends on libc6-dbg, so uninstalling is not a viable workaround. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#529016: Still broken...
FWIW, libc6-dbg's version is 2.11.2-7 here, valgrind's is 1:3.6.0~svn11254+nmu1. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#533468: Workaround
Since there obviously is no activity on Splix towards fixing this bug, I'd like to point out that the Foomatic driver now supports the ML-1610, so installing that instead of Splix gets you a working printer. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#458018: This doesn't only happen to short files
Hi! I'm just evaluating VLC as alternative to xine or mplayer, and I stumbled across the same bug. However, in my case the file wasn't called a.avi but داخلحماممغربي.flv, and picks up a bunch of files called *.txt or *.utf8. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#483836: Can't reproduce this
Command line is vlc http://dump.com/wp-content/uploads/golf.flv There's a bunch of spurious warnings and errors and the video output is broken (but it's like that for all videos, which is a different bug), but VLC works without crashing. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#602708: Video output is X-compressed to 25% and colors are off
Package: vlc Version: 1.1.3-1 Greetings! Any video I try to play is reduced in size and only on the left quarter of the video area, the rest of the video area is black. Further, the colors in the video are off, you can still recognize the content though. Lastly, there are horizontal stripes, somehow like interlacing gone wrong. When I just run vlc, I get this: | u...@fruitbox:~$ vlc | VLC media player 1.1.3 The Luggage (revision exported) | Blocked: call to unsetenv(DBUS_ACTIVATION_ADDRESS) | Blocked: call to unsetenv(DBUS_ACTIVATION_BUS_TYPE) | Warning: call to signal(13, 0x1) | [0x10103918] main libvlc: Running vlc with the default interface. Use 'cvlc' to use vlc without interface. | Blocked: call to setlocale(6, ) | Blocked: call to sigaction(17, 0x484dda48, 0x484ddad4) | Warning: call to rand() | Warning: call to rand() | Warning: call to rand() | Warning: call to rand() | Warning: call to rand() When I drag'n'drop a video onto the window, I get this: | swScaler: pal8 is not supported as output pixel format | [0x103aaf58] swscale scale error: could not init SwScaler and/or allocate memory | [0x103bb5d8] xcb_xv generic error: no available XVideo adaptor | object not accessible | [swscaler @ 0x1042cf30] No accelerated colorspace conversion found from yuv420p to rgb4. | object not accessible The first three lines of this are in red, I'm not sure what this means. In any any case, VLC itself doesn't display any error, this is only on the commandline. About the system, it's an old MacMini G4, a big-endian CPU, using a Radeon graphic chip. I've also attached the output of xvidinfo in case this gives some insight. Cheers! Uli -- System Information: Debian Release: squeeze/sid APT prefers testing APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: powerpc (ppc) Kernel: Linux 2.6.33-2-powerpc Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages vlc depends on: ii libaa1 1.4p5-38 ascii art library ii libc6 2.11.2-7 Embedded GNU C Library: Shared lib ii libfreetype62.4.2-1 FreeType 2 font engine, shared lib ii libfribidi0 0.19.2-1 Free Implementation of the Unicode ii libgcc1 1:4.4.5-4GCC support library ii libgl1-mesa-glx [libgl1 7.7.1-4 A free implementation of the OpenG ii libqtcore4 4:4.6.3-4Qt 4 core module ii libqtgui4 4:4.6.3-4Qt 4 GUI module ii libsdl-image1.2 1.2.10-2+b1 image loading library for Simple D ii libsdl1.2debian 1.2.14-6.1 Simple DirectMedia Layer ii libstdc++6 4.4.5-4 The GNU Standard C++ Library v3 ii libtar 1.2.11-6 C library for manipulating tar arc ii libvlccore4 1.1.3-1 base library for VLC and its modul ii libx11-62:1.3.3-3X11 client-side library ii libx11-xcb1 2:1.3.3-3Xlib/XCB interface library ii libxcb-keysyms1 0.3.6-1 utility libraries for X C Binding ii libxcb-randr0 1.6-1X C Binding, randr extension ii libxcb-shm0 1.6-1X C Binding, shm extension ii libxcb-xv0 1.6-1X C Binding, xv extension ii libxcb1 1.6-1X C Binding ii libxext62:1.1.2-1X11 miscellaneous extension librar ii ttf-freefont20090104-7 Freefont Serif, Sans and Mono True ii vlc-nox 1.1.3-1 multimedia player and streamer (wi ii zlib1g 1:1.2.3.4.dfsg-3 compression library - runtime Versions of packages vlc recommends: ii vlc-plugin-notify 1.1.3-1LibNotify plugin for VLC ii vlc-plugin-pulse 1.1.3-1PulseAudio plugin for VLC Versions of packages vlc suggests: pn mozilla-plugin-vlcnone (no description available) pn videolan-doc none (no description available) Versions of packages vlc-nox depends on: ii liba52-0.7.4 0.7.4-14 library for decoding ATSC A/52 str ii libasound2 1.0.23-2.1shared library for ALSA applicatio ii libass40.9.9-1 library for SSA/ASS subtitles rend ii libavahi-client3 0.6.27-2 Avahi client library ii libavahi-common3 0.6.27-2 Avahi common library ii libavc1394-0 0.5.3-1+b2control IEEE 1394 audio/video devi ii libavcodec52 5:0.6~svn20100726-0.1 library to encode decode multimedi ii libavformat52 5:0.6~svn20100726-0.1 ffmpeg file format library ii libavutil494:0.5.2-6 ffmpeg utility library ii libc6 2.11.2-7 Embedded GNU C Library: Shared lib ii libcaca0 0.99.beta17-1 colour ASCII art library ii
Bug#521275: Close bug?
Hi! I'm just looking through bugs I filed, and this one is IMHO fixed or invalid. As far as I'm concerned it was a user error, not a bug in the package. Also sorry about the Reply-To in the initial mail, I had the wrong personality selected in my mail client. Thanks! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#598456: Maximizing windows doesn't account for tabs, pushes them offscreen
Package: fluxbox Version: 1.1.1+git20100908.df2f51b-2 Hi! The description of this is really that short, if you have tabs activated and maximize a window, the window frame is resized to the screen's size (title bar touching the top), but the tabs are sitting on top of this title bar, so they are pushed off the top of the screen. I would expect the tabs to still be visible inside the screen. Thanks! Uli -- System Information: Debian Release: squeeze/sid APT prefers testing APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: powerpc (ppc) Kernel: Linux 2.6.32-5-powerpc Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages fluxbox depends on: ii libc6 2.11.2-6 Embedded GNU C Library: Shared lib ii libfontconfig1 2.8.0-2.1generic font configuration library ii libfreetype62.4.2-1 FreeType 2 font engine, shared lib ii libfribidi0 0.19.2-1 Free Implementation of the Unicode ii libgcc1 1:4.4.4-8GCC support library ii libice6 2:1.0.6-1X11 Inter-Client Exchange library ii libimlib2 1.4.2-8+b2 powerful image loading and renderi ii libsm6 2:1.1.1-1X11 Session Management library ii libstdc++6 4.4.4-8 The GNU Standard C++ Library v3 ii libx11-62:1.3.3-3X11 client-side library ii libxext62:1.1.2-1X11 miscellaneous extension librar ii libxft2 2.1.14-2 FreeType-based font drawing librar ii libxinerama12:1.1-3 X11 Xinerama extension library ii libxpm4 1:3.5.8-1X11 pixmap library ii libxrandr2 2:1.3.0-3X11 RandR extension library ii libxrender1 1:0.9.6-1X Rendering Extension client libra ii menu2.1.43 generates programs menu for all me ii zlib1g 1:1.2.3.4.dfsg-3 compression library - runtime Versions of packages fluxbox recommends: ii xfonts-terminus 4.30-2 Fixed-width fonts for fast reading Versions of packages fluxbox suggests: pn fbdesknone (no description available) pn fbpager none (no description available) -- no debconf information -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#593364: X11 SHAPE extension not updating windows correctly
Package: xcompmgr Version: 1.1.5-1 Hi! I'm writing a program that uses the SHAPE extensions to give a window a non-rectangular shape. This works up to the point where I try to modify the shape at runtime. The attached example program creates a simple window of 50x50 pixels. It then creates a moving square of 10x10 pixels inside that window which is not part of the window, using the SHAPE extension, so that you can watch the background or underlying window through this hole. While xcompmgr is running, the window isn't updated correctly. I tried this using fluxbox (which itself has some problems with X11 SHAPE), but it also happens with openbox, icewm or windowmaker, it even happens when not using any windowmanager at all. I tried using compiz as compositing and window manager, but failed getting it to run even halfway. The system is an old Mac Mini, using an ATI Radeon videochip and running a current Debian/testing system. Give me some feedback if you can't reproduce this, I'll try to circle in the other factors then. Cheers! Uli -- System Information: Debian Release: squeeze/sid APT prefers testing APT policy: (500, 'testing'), (500, 'stable'), (1, 'experimental') Architecture: powerpc (ppc) Kernel: Linux 2.6.32-5-powerpc Locale: LANG=en_GB.UTF-8, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Versions of packages xcompmgr depends on: ii libc6 2.11.2-2 Embedded GNU C Library: Shared lib ii libx11-6 2:1.3.3-3 X11 client-side library ii libxcomposite11:0.4.2-1 X11 Composite extension library ii libxdamage1 1:1.1.3-1 X11 damaged region extension libra ii libxfixes31:4.0.5-1 X11 miscellaneous 'fixes' extensio ii libxrender1 1:0.9.6-1 X Rendering Extension client libra // Example program for bug in xcompmgr // // This should display a window with a moving hole in it, but with xcompmgr // running, it isn't updated correctly. // // g++ -Wall `pkg-config xcb xcb-shape --cflags --libs` bug-shape-changer.cpp -o bug-shape-changer // #include stdlib.h #include string.h #include xcb/xcb.h #include xcb/shape.h #include iostream #include stdexcept #include assert.h /* global objects, the connection to the X server and the window created there. */ xcb_connection_t* connection = 0; xcb_window_t win; unsigned const window_size = 50; unsigned const hole_size = 10; int hole_position = 0; void init_connection() { /* open connection to the server */ connection = xcb_connect(NULL,NULL); if (xcb_connection_has_error(connection)) throw std::runtime_error(xcb_connect() failed); } void check_shape_extension() { // query for availability xcb_query_extension_reply_t const* reply = xcb_get_extension_data( connection, xcb_shape_id); if(!reply) throw std::runtime_error(xcb_get_extension_data failed); if(!reply-present) throw std::runtime_error(X11 shape extension not present); // query version xcb_shape_query_version_cookie_t cookie = xcb_shape_query_version_unchecked(connection); xcb_shape_query_version_reply_t* version = xcb_shape_query_version_reply(connection, cookie, 0); if(!version) throw std::runtime_error(xcb_shape_query_version failed); } void create_window(int16_t x, int16_t y, uint16_t width, uint16_t height) { /* retrieve the first screen from the setup */ xcb_setup_t const* setup = xcb_get_setup(connection); if(!setup) throw std::runtime_error(xcb_get_setup() failed); xcb_screen_t* screen = xcb_setup_roots_iterator(setup).data; if(!screen) throw std::runtime_error(missing root screen); // create window xcb_window_t id = xcb_generate_id(connection); uint32_t const win_values[] = { screen-black_pixel, XCB_EVENT_MASK_EXPOSURE }; xcb_create_window(connection, XCB_COPY_FROM_PARENT, // depth id, screen-root, // parent x, y, width, height, // position, size 0, // border width XCB_WINDOW_CLASS_INPUT_OUTPUT, // class XCB_COPY_FROM_PARENT, // visual XCB_CW_BACK_PIXEL | XCB_CW_EVENT_MASK, win_values); win = id; } void set_window_mask() { // set rectangualar base outline first xcb_rectangle_t const rect = {0, 0, window_size, window_size}; xcb_shape_rectangles(connection, XCB_SHAPE_SO_SET, XCB_SHAPE_SK_BOUNDING, 0, win, 0, 0, // offset 1, rect); // cut out moving rectangle xcb_rectangle_t const hole = {hole_position, hole_position, hole_size, hole_size}; xcb_shape_rectangles(connection, XCB_SHAPE_SO_SUBTRACT, XCB_SHAPE_SK_BOUNDING, 0, win, 0, 0, // offset 1, hole); xcb_flush(connection); } int main(int argc, char** argv) { try { init_connection(); create_window(0, 0,
Bug#544112: Example program
Hi! I noticed in a different context that fluxbow was sending superfluous expose events when losing and gaining focus. These events are normally sent when parts of a window were exposed and thus need redrawing. I guess these could be responsible for the anomalous flickering. The attached example program just creates a simple window using the X11 SHAPE extension. You can see that when losing focus, the window receives two expose events and when the window gains focus it receives four expose events. All these events are seemingly unnecessary, since nothing in the window was actually exposed. Notes: - I also tried the same program under icewm, where these events are not sent. - I have noticed that both gmplayer and xine suffer from the flickering, but I don't actually know if it is caused by the expose events. - I'm not sure if this is related, but #593061 is also about focus handling. I noticed here that when entering the window created by the example program, the window doesn't always receive focus. In particular when I cross the border quickly it seems to sometimes miss this. I hope this helps a bit. Uli // Example program for Debian bug #544112 // // g++ -Wall `pkg-config xcb xcb-shape --cflags --libs` // #include stdlib.h #include string.h #include xcb/xcb.h #include xcb/shape.h #include iostream #include stdexcept #include assert.h /* global objects, the connection to the X server and the window created there. */ xcb_connection_t* connection = 0; xcb_window_t win; void init_connection() { /* open connection to the server */ connection = xcb_connect(NULL,NULL); if (xcb_connection_has_error(connection)) throw std::runtime_error(xcb_connect() failed); } void check_shape_extension() { // query for availability xcb_query_extension_reply_t const* reply = xcb_get_extension_data( connection, xcb_shape_id); if(!reply) throw std::runtime_error(xcb_get_extension_data failed); if(!reply-present) throw std::runtime_error(X11 shape extension not present); // query version xcb_shape_query_version_cookie_t cookie = xcb_shape_query_version_unchecked(connection); xcb_shape_query_version_reply_t* version = xcb_shape_query_version_reply(connection, cookie, 0); if(!version) throw std::runtime_error(xcb_shape_query_version failed); } void create_window(int16_t x, int16_t y, uint16_t width, uint16_t height) { /* retrieve the first screen from the setup */ xcb_setup_t const* setup = xcb_get_setup(connection); if(!setup) throw std::runtime_error(xcb_get_setup() failed); xcb_screen_t* screen = xcb_setup_roots_iterator(setup).data; if(!screen) throw std::runtime_error(missing root screen); // create window xcb_window_t id = xcb_generate_id(connection); uint32_t const win_values[] = { screen-black_pixel, XCB_EVENT_MASK_EXPOSURE }; xcb_create_window(connection, XCB_COPY_FROM_PARENT, // depth id, screen-root, // parent x, y, width, height, // position, size 0, // border width XCB_WINDOW_CLASS_INPUT_OUTPUT, // class XCB_COPY_FROM_PARENT, // visual XCB_CW_BACK_PIXEL | XCB_CW_EVENT_MASK, win_values); win = id; } int main(int argc, char** argv) { try { init_connection(); create_window(0, 0, 50, 50); // set rectangualar base outline xcb_rectangle_t const rect = {0, 0, 50, 50}; xcb_shape_rectangles(connection, XCB_SHAPE_SO_SET, XCB_SHAPE_SK_BOUNDING, 0, win, 0, 0, // offset 1, rect); // actually display window xcb_map_window(connection, win); int const e = xcb_flush(connection); if(e=0) // TODO: provide more detailed diagnostics if possible throw std::runtime_error(xcb_flush failed); // poll connection's file descriptor for events int const ss = xcb_get_file_descriptor(connection); while(true) { // wait on the connection socket fd_set fd; FD_ZERO(fd); FD_SET( ss, fd); timeval timeout; timeout.tv_usec = 0; timeout.tv_sec = 1; switch(select( ss+1, fd, 0, 0, timeout)) { case -1: throw std::runtime_error(select failed); case 0: std::cout select timed out std::endl; break; case 1: break; default: std::cout select return unexpected value std::endl; break; } // handle all available events xcb_generic_event_t* e = 0; while((e = xcb_poll_for_event(connection))) { switch (e-response_type ~0x80) { case XCB_EXPOSE:
Bug#593061: Pointer moving between windows not detected correctly
Package: fluxbox Version: 1.1.1+dfsg2-1 When moving between two windows and at the same time between groups, the focus is not updated correctly. Following steps illustrate the setup: * Delete your .fluxbox, I'm assuming a default configuration. * Start fluxbox. * In the Tab Options (Right-click on the background) turn off Tabs in Titlebar to get external tabs. * In the Focus Model, select Mouse Focus and Mouse Tab Focus. You now have a setup where you change the focus with just the mouse movement, no clicks involved. It allows you to switch between unrelated windows and, by moving over the tabs, to switch within one group of windows. Now, in order to reproduce the buggy behavior, use following steps: * Create three windows A, B, C (I used three uxterms, but it doesn't matter). * Combine two of the three to a group, by middle-dragging the tab of one onto the other, so you get A, [B, C] * Top-align all three windows so that they overlap partially: [A*] [B|C] ,-+, [ ]] In the picture, A has the focus and is on top. The important part is that you can move seamlessly between these windows, hence the alignment. You can also position B and C below, but the above setup is easiest to explain and reproduce. * Now, move the pointer from the A window onto either of the B or C tabs. The last step should have the effect of switching focus away from A to the according window in the B/C group. However, it only switches the focus to the B/C group, it does not switch to the tab's window within that group. Note that this is a rather old bug, it didn't come up with the version that recently entered testing. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#551102: Can't reproduce
Hi! I just tried to reproduce this, but failed. The installed programs are a current Debian testing: fluxbox 1.1.1+dfsg2-1 tilda 0.09.6-2 conky-all 1.8.0-1 Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#592520: missing keybord mapping example file
Package: stormbaancoureur Version: 2.1.6-1 Hi! The README mentions a .stormbaancoureur.keys.example file, but that is apparently not part of the package. It would probably also be nice for Debian beginners if it was not hidden, even if it must be stored as hidden file eventually. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#592075: patch: add mixer bindings for mousewheel
Package: alsamixergui Version: 0.9.0rc2-1-9 Severity: whishlist The attached patch allows using the mousewheel to increase/decrease the volume. Cheers! Uli --- alsamixergui-0.9.0rc2-1.orig/src/Fl_AM.cxx 2002-05-09 17:41:35.0 +0200 +++ src/Fl_AM.cxx 2010-08-06 20:15:06.0 +0200 @@ -385,6 +384,28 @@ public: } int handle(int e) { +switch(e) { +case FL_MOUSEWHEEL: +{ +if(!Fl::event_inside(this)) +break; +Fl_AMGroup* g = (Fl_AMGroup*) this-parent(); +mixer_focus_elem = g-elem_id; +double new_value = increment(value(), Fl::event_dy()); +if (this==g-volumeL || g-lock-value()) { +g-volumeL-value(new_value); +mixer_volume_hard[MIXER_CHN_LEFT] = new_value; +} +if (this==g-volumeR || g-lock-value()) { +g-volumeR-value(new_value); +mixer_volume_hard[MIXER_CHN_RIGHT] = new_value; +} +gui_update(); +return 1; +} +default: +break; +} return Fl_Slider::handle(e); } };
Bug#587999: turning off radeonfb crashes system on console switch
Hi! I got around to testing with video=radeonfb:off on the kernel commandline. What you then get is a completely hanging system when switching terminale (chvt), when trying to stop the login manager KDM or when logging out of an X session. Note that this means that rebooting the system becomes impossible, similarly partitions aren't unmounted properly either. I also tried activating KMS explicitly then, but that crashes even earlier during boot, even before init is started. The kernel seems to detect that my video card has 256MiB of VRAM, while it only has 32. Some other memory layouts it detects are similarly off. If I find a free hour I'll try to disable KDM and then try without radeonfb, if that is broken, too, or if it is only in combination with X. I'll also try to copy the kernel output when using KMS. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#587999: further info
On Sunday 11 July 2010 10:05:08 Brice Goglin wrote: What about the original problem (pixel columns switched when using XV, with 16 or 24 bits color depth), is it gone now? No, not really, but it doesn't affect me now. If I understand correctly, what happened was that KMS broke DRI for X, so it switched back to non-accelerated rendering. Non-accelerated rendering gets the pixel columns wrong. In addition, it seems that DRI with 24BPP requires more memory now that I have (can someone confirm/explain that?), which made diagnosis more complicated. After turning off KMS and upgrading, I was to switch back to 16BPP, which works accelerated now, so the bug doesn't affect me. However, this is rather a workaround than a fix. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#567793: EXA causes arifacts
On Sunday 11 July 2010 10:12:23 Brice.Goglin wrote: On Sun, Jan 31, 2010 at 02:14:33PM +0100, Ulrich Eckhardt wrote: I have artifacts with colors being off and sometimes areas being filled with black/white noise. [...] Assuming this is the problem you referred to at the end of http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=587999#66 could you give an update to the status of this exact bug report too? No, the artifacts that prompted me to file this bug were different. However, I'll give EXA another try with the new driver and report the current state. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#587999: further info
On Friday 09 July 2010 17:49:57 Michel Dänzer wrote: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=565313 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567616 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567793 The depth 16 issue could be fixed in xserver-xorg-video-radeon 1:6.13.1-1, not sure about the other EXA issue. Thanks, that seems to do the job. I downloaded 6.13.1-1 from unstable and now I can again use 16 BPP without the greenish tint. Further, if I deactivate KMS (modprobe radeon modeset=0), X also initialises DRI (even though it complains about this and that) and bug 587999 doesn't appear, so it's indeed only the non-DRI code it seems. Just for the record, I had the effect that areas were not recognized as newly exposed with EXA and the new radeon driver, though only with a KDE app. Switching back to XAA seems to work around that. I'll try to reproduce this and file a separate report if I get to it. Thank you for your help, I owe you a beer! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#587999: further info
On Friday 09 July 2010 14:56:39 Michel Dänzer wrote: This is the reason for the DRI being disabled now: (II) RADEON(0): Memory manager initialized to (0,0) (1728,4854) (II) RADEON(0): Reserved area from (0,1680) to (1728,1682) (II) RADEON(0): Largest offscreen area available: 1728 x 3172 (II) RADEON(0): Will use front buffer at offset 0x0 (II) RADEON(0): Will use back buffer at offset 0x9cf000 (II) RADEON(0): Will use depth buffer at offset 0x14e2000 (II) RADEON(0): Will use 0 kb for textures at offset 0x1ff5000 (EE) RADEON(0): Static buffer allocation failed. Disabling DRI. (EE) RADEON(0): At least 34020 kB of video memory needed at this resolution and depth. Maybe it'll work without Option AccelMethod XAA I'll try that. (or with explicit EXA instead), otherwise you'll need to reduce the maximum desktop size via the Virtual directive or run in depth 16. EXA and 16 BPP are out, except if using neither of EXA and XAA changes anything there: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=565313 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567616 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567793 Hopefully... Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#587999: further info
On Monday 05 July 2010 09:03:53 you wrote: On Sam, 2010-07-03 at 23:58 +0200, Ulrich Eckhardt wrote: (EE) RADEON(0): [dri] RADEONDRIGetVersion failed to open the DRM [dri] Disabling DRI. It should work better with the DRI enabled. Good catch, but shouldn't it work even without DRI? Do you have the firmware-linux package installed? Yes. If so, please post the output of running sudo modprobe radeon as well as the messages appearing in the dmesg output after it. The module is already loaded. I can unload it though, even with a running X11 session, lsmod shows that its refcount is zero. When loading it, it outputs [drm] radeon kernel modesetting enabled. in dmesg. However, when starting the X server, it outputs radeonfb :00:10.0: Invalid ROM contents there. Note that this error is nothing new. X11 has been complaining about a missing ROM since I got this machine (MacMini), but has always worked nonetheless. If I understand correctly, the graphics adapter doesn't have a ROM and doesn't need one, because it is initialized by the firmware. Thanks for your effort! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#587999: Xv switches pixel columns
Package: xserver-xorg Version: 1:7.5+6 Using mplayer and xine (didn't try others) with Xv as video output driver, I have the effect that the order of pixel columns is switched. If I use 24 bits color depth, the order is 1,0,3,2 etc. Using 16 bits color depth, the order is 3,2,1,0,7,6,5,4. Notes: * I'm using the radeon driver backend. * I'm on a big-endian machine (PPC). As a workaround, I can use xshm as video output driver. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#587999: further info
-- Package-specific info: /var/lib/x11/X.roster does not exist. /var/lib/x11/X.md5sum does not exist. X server symlink status: lrwxrwxrwx 1 root root 13 Dec 10 2007 /etc/X11/X - /usr/bin/Xorg -rwxr-xr-x 1 root root 1940408 Jun 3 18:28 /usr/bin/Xorg /var/lib/x11/xorg.conf.roster does not exist. VGA-compatible devices on PCI bus: :00:10.0 VGA compatible controller: ATI Technologies Inc RV280 [Radeon 9200] (rev 01) /var/lib/x11/xorg.conf.md5sum does not exist. Xorg X server configuration file status: -rw-r--r-- 1 root root 6148 Jul 3 16:47 /etc/X11/xorg.conf Contents of /etc/X11/xorg.conf: # This configfile demonstrates the broken behaviour of xorg # The problem is that this file as configuration is capable of starting X in # the configured layout, but once shut down, this does _not_ work a second # time. # # Note: # Option UseFBDev true # will enable startup. # # Configuration: # xserver-xorg-video-ati 6.6.3-2 # Monitor: # SyncMaster 225BW on VGA # Machine: # Mac Mini PPC Section Files FontPath/usr/share/fonts/X11/misc FontPath/usr/X11R6/lib/X11/fonts/misc FontPath/usr/share/fonts/X11/cyrillic FontPath/usr/X11R6/lib/X11/fonts/cyrillic FontPath/usr/share/fonts/X11/100dpi/:unscaled FontPath/usr/X11R6/lib/X11/fonts/100dpi/:unscaled FontPath/usr/share/fonts/X11/75dpi/:unscaled FontPath/usr/X11R6/lib/X11/fonts/75dpi/:unscaled FontPath/usr/share/fonts/X11/Type1 FontPath/usr/X11R6/lib/X11/fonts/Type1 FontPath/usr/share/fonts/X11/100dpi FontPath/usr/X11R6/lib/X11/fonts/100dpi FontPath/usr/share/fonts/X11/75dpi FontPath/usr/X11R6/lib/X11/fonts/75dpi # path to defoma fonts FontPath/var/lib/defoma/x-ttcidfont-conf.d/dirs/TrueType EndSection Section Module Loadbitmap Loaddbe Loadddc Loaddri Loadextmod #Load freetype Loadglx Loadint10 Loadvbe EndSection #Section Extensions #Option Composite enable #EndSection Section InputDevice Identifier Generic Keyboard Driver kbd Option CoreKeyboard Option XkbRules xorg Option XkbModel pc104 Option XkbLayout dvorak Option XkbVariantdvorak Option XkbOptionslv3:ralt_switch EndSection Section InputDevice Identifier Configured Mouse Driver mouse Option CorePointer Option Device/dev/input/mice EndSection Section Device Identifier Builtin Radeon Driver radeon # select model with integrated DVI adapter Option MacModel mini-internal # Option MonitorLayout TMDS,NULL # Option PanelSize 1680x1050 # ignore monitor mode lines received via DDC but use DDC to detect the monitor itself Option IgnoreEDID true # dynamically scale the chip clock to reduce energy consumption #Option DynamicClocks true # TODO: try this on # Option DefaultTMDSPLL true # Option AGPMode 4 #Option XAANoOffscreenPixmaps true # Option UseFBDev false # Option UseFBDev true # Option AGPFastWrite off # Option DDCMode true # Option AccelMethod EXA # Option AccelMethod XAA # Option ReverseDDCtrue ### # BusID PCI:0:16:0 # All options supported by the driver (2008-08-24) # #Option SWcursor boolean #Option SWcursor true #Option NoAccel boolean #Option Dac6Bit boolean #Option VideoKey integer #Option ScalerWidth integer #Option AGPMode integer #Option AGPFastWrite boolean #Option BusType string #Option DDCMode boolean #Option DisplayPriority string #Option ColorTiling boolean #Option IgnoreEDID boolean #Option PanelSize string #Option EnablePageFlip boolean #Option ForceMinDotClock frequency #Option RenderAccel boolean #Option AccelMethod string # old X accel Option AccelMethod XAA # new X accel #Option AccelMethod EXA #Option AccelDFS boolean #Option FBTexPercent integer #Option DepthBits integer #Option DMAForXv boolean #Option SubPixelOrder string #Option DynamicClocks boolean #Option VGAAccess boolean #Option ReverseDDC boolean #Option LVDSProbePLL boolean #Option TVDACLoadDetect boolean #Option DefaultTMDSPLL boolean #Option DefaultTVDACAdj boolean #Option DRI boolean #Option DefaultConnectorTable boolean #Option MacModel string #Option TVStandard string #Option ForceTVOut boolean #Option IgnoreLidStatus boolean #Option Int10 boolean
Bug#586919: Missing manpage for apctest
Package: apcupsd Version: 3.14.5-2 There is a manpage source file in package apcupsd-doc, installed as apctest.man, which should probably be compiled and installed as manpage instead. There is also a file called apcupsd.man in the same folder, which seems to be the source file for the supplied manpage and which (I'd say) doesn't have to be installed there. Cheers! Uli -- Sator Laser GmbH, Fangdieckstraße 75a, 22547 Hamburg, Deutschland Geschäftsführer: Thorsten Föcking, Amtsgericht Hamburg HR B62 932 ** Sator Laser GmbH, Fangdieckstraße 75a, 22547 Hamburg, Deutschland Geschäftsführer: Thorsten Föcking, Amtsgericht Hamburg HR B62 932 ** Visit our website at http://www.satorlaser.de/ ** Diese E-Mail einschließlich sämtlicher Anhänge ist nur für den Adressaten bestimmt und kann vertrauliche Informationen enthalten. Bitte benachrichtigen Sie den Absender umgehend, falls Sie nicht der beabsichtigte Empfänger sein sollten. Die E-Mail ist in diesem Fall zu löschen und darf weder gelesen, weitergeleitet, veröffentlicht oder anderweitig benutzt werden. E-Mails können durch Dritte gelesen werden und Viren sowie nichtautorisierte Änderungen enthalten. Sator Laser GmbH ist für diese Folgen nicht verantwortlich. ** -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#584850: Provide better diagnostic for directories [patch]
Package: mmv Version: 1.01b-15 I just tried to rename a bunch of directories: $ mmv -v DSide-* DSide #1 DSide-* - DSide #1 : no match. Nothing done. This is completely unhelpful, as the error no match seems to indicate the match string was wrong. Only after many fruitless attempts and thoroughly searching the manpage I found the -r option that (for reasons I don't understand) is necessary to convince mmv to move directories. At the very least I would have expected the -v to tell me that the matching directory entries are not considered in the rename. Looking at the sources, I'd do something like the following to function keepmatch(): #ifdef IS_MSDOS - if ((ffrom-fi_attrib FA_DIREC) ? !dirs : !fils) + if ((ffrom-fi_attrib FA_DIREC) ? !dirs : !fils) { #else getstat(pathbuf, ffrom); - if ((ffrom-fi_stflags FI_ISDIR) ? !dirs : !fils) + if ((ffrom-fi_stflags FI_ISDIR) ? !dirs : !fils) { #endif + if (verbose) + printf(ignoring directory %s\n, ffrom-fi_name); return(0); + } Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#581531: Can't disable text in image
Package: webcam Version: 3.95.dfsg.1-8.1 'webcam' adds a text containing a timestamp by default. This text can not be turned off, if you specify an empty text in the configuration file, you only get a message telling you that there's a syntax error on that line and the value is reverted to the default. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#551124: Bug still present, overall condition even worse
Firstly, I still can't play consecutive Flash videos. Note that this only applies to Flash videos, not other types of videos, in case that wasn't clear. Further, when while xine is idle (either after finishing one or on startup) and then I drag'n'drop another file onto xine, it takes around ten seconds until the video output starts, while the audio output comes much earlier. During that time, the UI is unresponsive, stopping, closing, volume all don't work. After that, audio and video are not out of sync, the video only starts being displayed later. This applies to all videos btw, not just flash, and it doesn't apply to videos started directly from the commandline. Also, when starting via the play button, the video starts correctly. For the record, here are the now installed packages and versions: xine-ui0.99.6-1 libc6 2.10.2-6 libcurl3-gnutls7.20.0-3 liblircclient0 0.8.3-5 libpng12-0 1.2.43-1 libreadline6 6.1-1 libx11-6 2:1.3.3-2 libxext6 2:1.1.1-3 libxft22.1.14-2 libxine1 1.1.18.1-1+b1 libxine1-ffmpeg1.1.18.1-1+b1 libxine1-x 1.1.18.1-1+b1 libxinerama1 2:1.1-3 libxtst6 2:1.1.0-2 libxv1 2:1.0.5-1 libxxf86vm11:1.1.0-2 Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#575583: Unusable on non-QWERT keyboards
Package: tint Version: 0.03b TINT is not usable e.g. on a Dvorak keyboard, it assumes that the keys JKL are arranged in a way similar to how the cursor keys are arranged. By default, it should use the cursor keys instead of JKL, their arrangement is more portable between different keyboards. Thanks! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#570830: Patch to avoid invalid fstat calls when opening empty files
Package: libxine1-bin Version: 1.1.17-1+b1 Hi! When opening empty files with xine, it reports: This is xine (X11 gui) - a free video player v0.99.6cvs. (c) 2000-2007 The xine Team. system call fstat: Bad file descriptor system call fstat: Bad file descriptor call failed object not accessible object not accessible call failed The attached patch fixes the two error messages from fstat() calls and a typo. The problem was that when detecting an empty file, it is closed, but the file descriptor in struct file_input_plugin_t is still left at the previous value, causing the errors when it is used later. Enjoy! Uli --- xine-lib-1.1.17/src/input/input_file.c 2009-11-30 21:55:45.0 +0100 +++ ../xine-lib-1.1.17/src/input/input_file.c 2010-02-21 19:46:26.0 +0100 @@ -292,7 +292,7 @@ static uint32_t file_plugin_get_blocksiz } /* - * Return 1 is filepathname is a directory, otherwise 0 + * Return 1 if filepathname is a directory, otherwise 0 */ static int is_a_dir(char *filepathname) { struct stat pstat; @@ -409,6 +409,7 @@ static int file_plugin_open (input_plugi if (file_plugin_get_length (this_gen) == 0) { _x_message(this-stream, XINE_MSG_FILE_EMPTY, this-mrl, NULL); close (this-fh); + this-fh = -1; xine_log (this-stream-xine, XINE_LOG_MSG, _(input_file: File empty: %s\n), this-mrl); return -1;
Bug#569591: patch
Greetings! After some hacking, I think I found the culprit, it is the algorithm that linebreaks text in order to fit it into the given width of a dialog box. When a single word (e.g. a path) is so long it doesn't fit into a line on its own, it tries to step back to the last whitespace first, but that isn't reset after each line. Cheers! Uli --- xine-ui-0.99.5/src/xitk/xine-toolkit/image.c 2007-03-30 01:07:56.0 +0200 +++ ../xine-ui-0.99.5/src/xitk/xine-toolkit/image.c 2010-02-21 18:52:34.0 +0100 @@ -515,6 +515,7 @@ xitk_image_t *xitk_image_create_image_wi xitk_font_string_extent(fs, bp, lbearing, rbearing, NULL, NULL, NULL); linew = rbearing - lbearing; + lastws = 0; } p++;
Bug#561779: Bug fixed
On Thursday 18 February 2010 14:54:27 Ana Guerrero wrote: After a quick chat with Antonis by IRC he told me he has tried to reproduce the bug several times with a new version (4:4.3.4-1) and the bug was not reproducible for it. Ulrich, you reported it against 4:4.3.2-1, could you check with this newer version? I have tested it here with 4:4.3.4-1, and it still exists and with the same steps for reproduction. What I have noticed is that the visible marker (highlighter) for the selected message sometimes disappears when changing sorting order, but I'm not sure when. Just wondering, did you reproduce the problem with an older version? In any case, I have just taken a few screenshots to illustrate the problem, just in case my description was unclear. I'll clean these up and post them later. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#569591: Filename error handling
Package: xine-ui Version: 0.99.5+cvs2007 Hi! I was looking at bug #510606 and started trying to reproduce it. However, I can't see exactly the behaviour the original reporter described, but I see that xine gets a bit picky with long filenames, empty files and invalid paths, so I decided to write that up separately. # test 1 $ touch 1234567890 $ xine 1234567890 This is xine (X11 gui) - a free video player v0.99.6cvs. (c) 2000-2007 The xine Team. system call fstat: Bad file descriptor system call fstat: Bad file descriptor object not accessible call failed The weird things there are the calls to fstat with bad file descriptors. While this seems to be caught, it is an indicator of something going very wrong. Note that the filename is really only ten characters long, it rather seems to be related to the fact that the file is empty. When I ran xine on a logfile (just for testing), it didn't produce those two lines. Anyway, what xine does here is to produce two popups. The first one is *sight*, unknown error. (File is empty: 1234567890) Well, obviously the error is known, so that doesn't make sense. (Also it probably should be *sigh* not *sight*.) The second popup is - xine engine error - Input plugin failed to open mrl '1234567890' This error seems a) wrong, since it did manage to open the file and b) superfluous since the first popup already explains all. Another weird things is that these two popups don't come in a fixed order! Most of the time the first one is in front of the second one, but not always! # test 2 $ touch 1234567890 $ xine `pwd`1234567890 This is xine (X11 gui) - a free video player v0.99.6cvs. (c) 2000-2007 The xine Team. object not accessible call failed Note here that the resulting argument does not actually resolve to a file, since 'pwd' doesn't have a trailing slash in its output. Again, I get two error popups, while one is enough. # test 3 # these are actually 200 characters below! $ touch 1234567890...1234567890 $ xine `pwd`1234567890...1234567890 This is xine (X11 gui) - a free video player v0.99.6cvs. (c) 2000-2007 The xine Team. xiTK received SIGSEGV signal, RIP. Aborted xdg-screensaver: Window 71303875 does not exist Here, xine actually aborted with a segmentation violation. For a split second, it displayed the message - xine engine error - Input plugin failed to open mrl '/home/uli/tmp12345678901234567 Note here that the text really ends with a cut off path. Neither the full path is given nor even the closing single quote! Running the same in GDB produces the following output: Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 0x4f0f1480 (LWP 4765)] _wordcopy_bwd_dest_aligned (dstp=1318006776, srcp=1318006768, len=1071651524) at wordcopy.c:391 object not accessible 391 wordcopy.c: No such file or directory. in wordcopy.c (gdb) bt #0 _wordcopy_bwd_dest_aligned (dstp=1318006776, srcp=1318006768, len=1071651524) at wordcopy.c:391 #1 0x0faa89dc in *__GI_memmove (dest=0x4f0ec4f2, src=value optimized out, len=4294967282) at memmove.c:99 #2 0xe010069b in ?? () Cannot access memory at address 0xf44f0ee8 Things to note here are that xine is trying to copy one billion (hex 0x3fe01ac4) words, which is more that a process in a 32-bit environment has as virtual address space, I think. Further up the stack, it is doing a __GI_memmove() which has as an actual size the value 0xfff2. How that value converts to a length of 0x3fe01ac4 in the call to _wordcopy_bwd_dest_aligned() is beyond me, and possibly worth a bug for some other code. Further testing with GDB turned up a few more things: 1. (gdb) run /home/uli/tmp1234567890 This creates an error message (no surprise here) but otherwise works as expected. 2. (gdb) run /home/uli/tmp12345678901234567890 This creates the same error messages, but on shutdown, xine receives a SIGSEGV in _int_free(). In the call stack is XCloseDisplay() though, so that might be a bug in Xlib. 3. (gdb) run /home/uli/tmp123456789012345678901234567890 Same as case 2. 4. (gdb) run /home/uli/tmp1234567890123456789012345678901234567890 At this size, it suddenly runs normally! I tried multiple times, I didn't see the same effect as with cases 2 and 3. 5. (gdb) run /home/uli/tmp12345678901234567890123456789012345678901234567890 Here we get the effects described above, with the SIGSEGV in the call to _wordcopy_bwd_dest_aligned(). The effects are all easy to reproduce. I'm going to compile a version locally with debug symbols and see if I can create more info or even a patch tomorrow. Cheers and good evening! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact
Bug#567616: Some more info...
Hi! Since an update yesterday (I'm running testing here), I'm seeing the same effect here. I can also work around this by switching to 24 bits color depth. Anyway, some info on this system here: - Mac Mini with a G4 PowerPC CPU - Radeon graphics chip, 'lspci' says VGA compatible controller: ATI Technologies Inc RV280 [Radeon 9200] (rev 01) - Some installed packages: xserver-xorg1:7.5+2 xserver-xorg-core 2:1.7.4-2 xserver-xorg-video-ati 1:6.12.4-2 xserver-xorg-video-mach64 6.8.2-2 xserver-xorg-video-r128 6.8.1-2 xserver-xorg-video-radeon 1:6.12.4-2 HTH Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#532616: Translation...
The title would translate to nv nvidia driver: refresh rate from 60 to 50. mein Problem: die Wiederholrate wird beim nv-Treiber auf 60Hz gesetzt und bei dem nvidia-Treiber auf 50Hz gesetzt. das führt dazu, das die Zeichen plötzlich verdoppelt werden. My problem: The refresh rate is set to 60Hz with the 'nv' driver and to 50Hz with the 'nvidia' driver. This leads to the characters suddenly being doubled. For the record, I can't tell what the reporter actually meant with characters being doubled, the description isn't clear in German either. HTH Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#567616: Duplicate bug report?
Hi! I think that 16 bit colour on M64 gives green tinted screen http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=565313 new update has problem using DefaultDepth 16 in xorgs config http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=567616 actually describe the same problem. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#566622: Wrong URL in manpage
Package: scrounge-ntfs Version: 0.9-6 Simple patch: - http://memberwebs.com/swalter/software/scrounge/ + http://memberwebs.com/stef/software/scrounge/ For the record, the URL on the package's page at Debian's is correct. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#563162: ownership problems give bogus error message
Package: cryptsetup Version: 2:1.1.0~rc2-1 I tried activating an encrypted partition as normal user, and it gave me this: u...@fruitbox:~$ cryptdisks_start hda5 Starting crypto disk...hda5 (skipped, device /dev/hda5 does not exist)...failed. u...@fruitbox:~$ ls /dev/hda hda hda1 hda2 hda3 hda4 hda5 As you see, the partition exists: u...@fruitbox:~$ cat /etc/crypttab # target name source device key file options hda5/dev/hda5 noneluks,noauto And it works as root: r...@fruitbox:/home/uli# cryptdisks_start hda5 Starting crypto disk...hda5 (starting)... Unlocking the disk /dev/hda5 (hda5) Enter passphrase: Key slot 0 unlocked. hda5 (started)...done. I guess the code is simply assuming ENOENT when open fails, but it should also handle EACCESS and similar errors. Cheers! Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#560126: Similar problems here...
Mine aren't as severe, but at least they are enough to crash my X server, see http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=561331. Further, also with a radeon card, I'm suffering a randomly corrupted cursor when moving between two windows ... I'll file a separate report unless already done by someone else. Uli -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Bug#506666: Can't reproduce here...
I tried the simple SDL_Init/SDL_Quit example from [1] and it doesn't crash. I also tried running tuxpaint, not crashing my machine either. The system is a current testing system, also on a Mac Mini G4. Attached is the xorg.conf file, notable differences are: - The whole Module section. - In the Device section the MacModel option, I'm not sure if this is important though. - In the Device section the IgnoreEDID option. My monitor actually reports more (60Hz) than it is really capable of (58Hz), so I have to program the timing manually. - I'm running at a display depth of 16 bit instead of 8. Actually, what depth are you running? I assumed it would use the first one, i.e. 8BPP in your case, is that true? However, I think this is mostly a mode for legacy code, as all others now use some kind of RGB encoding. I would definitely suggest trying 15/16/24 bits, too, that might already fix your problem. Uli [1] http://lazyfoo.net/SDL_tutorials/lesson01/linux/cli/index.php Section Files FontPath/usr/share/fonts/X11/misc FontPath/usr/X11R6/lib/X11/fonts/misc FontPath/usr/share/fonts/X11/cyrillic FontPath/usr/X11R6/lib/X11/fonts/cyrillic FontPath/usr/share/fonts/X11/100dpi/:unscaled FontPath/usr/X11R6/lib/X11/fonts/100dpi/:unscaled FontPath/usr/share/fonts/X11/75dpi/:unscaled FontPath/usr/X11R6/lib/X11/fonts/75dpi/:unscaled FontPath/usr/share/fonts/X11/Type1 FontPath/usr/X11R6/lib/X11/fonts/Type1 FontPath/usr/share/fonts/X11/100dpi FontPath/usr/X11R6/lib/X11/fonts/100dpi FontPath/usr/share/fonts/X11/75dpi FontPath/usr/X11R6/lib/X11/fonts/75dpi # path to defoma fonts FontPath/var/lib/defoma/x-ttcidfont-conf.d/dirs/TrueType EndSection Section Module Loadbitmap Loaddbe Loadddc Loaddri Loadextmod Loadglx Loadint10 Loadvbe EndSection Section InputDevice Identifier Generic Keyboard Driver kbd Option CoreKeyboard Option XkbRules xorg Option XkbModel pc104 Option XkbLayout dvorak Option XkbVariantdvorak Option XkbOptionslv3:ralt_switch EndSection Section InputDevice Identifier Configured Mouse Driver mouse Option CorePointer Option Device/dev/input/mice EndSection Section Device Identifier Builtin Radeon Driver radeon # select model with integrated DVI adapter Option MacModel mini-internal # ignore monitor mode lines received via DDC but use DDC to detect the monitor itself Option IgnoreEDID true Option AccelMethod XAA EndSection Section Monitor Identifier SyncMaster 225BW on VGA Option DPMS EndSection Section Monitor Identifier SyncMaster 225BW on DVI ModelName SyncMaster # 1680x1050 57.96 Hz (CVT) hsync: 63.06 kHz; pclk: 141.25 MHz Modeline 1680x1...@58 141.25 1680 1784 1960 2240 1050 1053 1059 1088 -hsync +vsync Modeline 1680x1...@57 140.25 1680 1784 1960 2240 1050 1053 1059 1088 -hsync +vsync Option PreferredMode 1680x1...@57 EndSection Section Screen Identifier Default Screen Device Builtin Radeon Monitor SyncMaster 225BW on DVI DefaultDepth16 SubSection Display Depth 1 Modes 1680x1050 1280x1024 1024x768 800x600 640x480 EndSubSection SubSection Display Depth 4 Modes 1680x1050 1280x1024 1024x768 800x600 640x480 EndSubSection SubSection Display Depth 8 Modes 1680x1050 1280x1024 1024x768 800x600 640x480 EndSubSection SubSection Display Depth 15 Modes 1680x1050 1280x1024 1024x768 800x600 640x480 EndSubSection SubSection Display Depth 16 Modes 1680x1050 1280x1024 1024x768 800x600 640x480 EndSubSection SubSection Display Depth 24 Modes 1680x1050 1280x1024 1024x768 800x600 640x480 EndSubSection EndSection Section ServerLayout Identifier Default Layout Screen Default Screen InputDevice Generic Keyboard InputDevice Configured Mouse EndSection Section DRI Mode0666 EndSection
Bug#562971: Alternative patch with ls-like '--si' option
Well, I think all involved parties made clear what their position on that matter is. I hope no hard feelings remain. That said, behave like GNU ls as an explicit goal is a good idea regardless. With that in mind, I just hacked together exactly that, see attached patch. Uli diff -upwr tree-1.5.3.orig/man/tree.1 tree-1.5.3/man/tree.1 --- tree-1.5.3.orig/man/tree.1 2009-12-31 15:54:48.0 +0100 +++ tree-1.5.3/man/tree.1 2009-12-31 16:27:24.0 +0100 @@ -115,7 +115,13 @@ Print the size of each file in bytes alo .B -h Print the size of each file but in a more human readable way, e.g. appending a size letter for kilobytes (K), megabytes (M), gigabytes (G), terrabytes (T), -petabytes (P) and exabytes (E). +petabytes (P) and exabytes (E). These are similar to how 'ls' works, i.e. using +a base of 1024. +.PP +.TP +.B --si +Like '-h', print the size of each file in a more human readable way, but using +SI prefixes and a base of 1000. .PP .TP .B -u @@ -131,16 +137,16 @@ Print the date of the last modification .PP .TP .B --inodes -Prints the inode number of the file or directory +Prints the inode number of the file or directory. .PP .TP .B --device -Prints the device number to which the file or directory belongs +Prints the device number to which the file or directory belongs. .PP .TP .B -F Append a `/' for directories, a `=' for socket files, a `*' for executable files -and a `|' for FIFO's, as per ls -F +and a `|' for FIFO's, as per ls -F. .PP .TP .B -q Only in tree-1.5.3.orig: strverscmp.c.orig Only in tree-1.5.3: tree diff -upwr tree-1.5.3.orig/tree.c tree-1.5.3/tree.c --- tree-1.5.3.orig/tree.c 2009-12-31 15:54:48.0 +0100 +++ tree-1.5.3/tree.c 2009-12-31 16:08:24.0 +0100 @@ -182,7 +182,7 @@ int strverscmp (const char *s1, const ch /* Globals */ int dflag, lflag, pflag, sflag, Fflag, aflag, fflag, uflag, gflag; -int qflag, Nflag, Dflag, inodeflag, devflag, hflag; +int qflag, Nflag, Dflag, inodeflag, devflag, hflag, siflag; int noindent, force_color, nocolor, xdev, noreport, nolinks, flimit; char *pattern = NULL, *ipattern = NULL, *host = NULL, *title = Directory Tree, *sp = ; const char *charset=NULL; @@ -206,7 +206,7 @@ int main(int argc, char **argv) q = p = dtotal = ftotal = 0; aflag = dflag = fflag = lflag = pflag = sflag = Fflag = uflag = gflag = FALSE; - Dflag = qflag = Nflag = Hflag = Rflag = hflag = FALSE; + Dflag = qflag = Nflag = Hflag = Rflag = hflag = siflag = FALSE; noindent = force_color = nocolor = xdev = noreport = nolinks = FALSE; inodeflag = devflag = FALSE; flimit = 0; @@ -390,6 +390,12 @@ int main(int argc, char **argv) cmpfunc = dirsfirstsort; break; } + if (!strcmp(--si,argv[i])) { + j = strlen(argv[i])-1; + siflag = TRUE; + sflag = TRUE; /* Assume they also want -s */ + break; + } if (!strncmp(--filelimit,argv[i],11)) { j = 11; if (*(argv[i]+11) == '=') { @@ -1293,14 +1299,18 @@ void printit(char *s) void psize(char *buf, off_t size) { char *unit=BKMGTPEZY; + char *si_unit= kMGTPEZY; int idx; - if (!hflag) sprintf(buf, sizeof(off_t) == sizeof(long long)? %11lld : %9ld, size); - else { + if (hflag) { for (idx=size1024?0:1; size = (1024*1024); idx++,size=10); if (!idx) sprintf(buf, %4d, (int)size); else sprintf(buf, ((size10) = 10)? %3.0f%c : %3.1f%c, (float)size/(float)1024,unit[idx]); - } + } else if (siflag) { +for (idx=size1000?0:1; size = (100); idx++,size/=1000); +if (!idx) sprintf(buf, %4d, (int)size); +else sprintf(buf, (size = 1)? %3.0f%c : %3.1f%c, (float)size/1000.0f,si_unit[idx]); + } else sprintf(buf, sizeof(off_t) == sizeof(long long)? %11lld : %9ld, size); } void html_encode(FILE *fd, char *s) Only in tree-1.5.3: tree.c~ Only in tree-1.5.3: tree.o
Bug#562971: Wrong unit when using -h [patch]
Package: tree Version: 1.5.2.2-1 Actually there are several aspects to this bug, which is also related to #556186[1]: 1. The prefix for thousand is k, not K, all others are correctly capitalized. 2. The prefix k means one thousand. Abusing it for 1024 is just that - abuse[2]. Note that the manpage 'documented' this bug using the offensive wording reported in #556186. This IMHO doubly asinine phrase was removed in one of the newer releases, but without replacing it with some honest explanation of the prefix abuse. The attached patch fixes both issues, the latter by using the correct meaning of the SI prefixes and not by documenting the bug in the manpage. It was made against the pristine upstream 1.5.3 sources. Cheers! Uli [1] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=556186 [2] I'm aware that some people disagree, this is kind of a religious matter. --- tree.c.orig 2009-12-29 17:13:47.0 +0100 +++ tree.c 2009-12-29 18:09:36.0 +0100 @@ -1292,14 +1292,14 @@ void printit(char *s) void psize(char *buf, off_t size) { - char *unit=BKMGTPEZY; + char *unit= kMGTPEZY; int idx; if (!hflag) sprintf(buf, sizeof(off_t) == sizeof(long long)? %11lld : %9ld, size); else { -for (idx=size1024?0:1; size = (1024*1024); idx++,size=10); +for (idx=size1000?0:1; size = 100; idx++,size/=1000); if (!idx) sprintf(buf, %4d, (int)size); -else sprintf(buf, ((size10) = 10)? %3.0f%c : %3.1f%c, (float)size/(float)1024,unit[idx]); +else sprintf(buf, (size=1) ? %3.0f%c : %3.1f%c, size/1000.0f,unit[idx]); } }