Bug#826013: gcalcli: fails with "ERROR: Missing module - cannot import name run"

2016-06-01 Thread Andrei Paskevich
Package: gcalcli Version: 3.3.2-1 Severity: grave Dear Maintainer, since the last upgrade of python-googleapi in unstable, gcalcli refuses to start, making the package unusable: % /usr/bin/gcalcli agenda ERROR: Missing module - cannot import name run After downgrading to python-googleapi 1.

Bug#809546: conky-std: conky is compiled without BUILD_MATH (-cli,-std,-all)

2016-01-02 Thread Andrei Paskevich
Hi Vincent, You are right, this is indeed an upstream bug, which seems to have been fixed in December, see https://github.com/brndnmtthws/conky/commit/2fab2d38207d73533d29062883b284d80e5325a2 Best, Andrei

Bug#809546: conky-std: conky is compiled without BUILD_MATH (-cli,-std,-all)

2015-12-31 Thread Andrei Paskevich
Package: conky-std Version: 1.10.1-2 Severity: normal Dear Maintainer, first of all, happy New Year! I found that the gauge indicators are broken (gauge line missing) and the -l option (logarithmic scale) is ignored in the graphs. According to the man page, this is caused by conky being compiled

Bug#661465: Please reassign (was: Please remove ATK-Wrapper dependency)

2012-03-13 Thread Andrei Paskevich
> This topic could be closed by removing the gconf dependency of > libatk-wrapper-java-jni. Should this be done by reassigning this > bug or by opening a new one? I agree with your analysis, but don't forget that a direct dependency on libgnome and libgnomevfs was introduced in openjdk-7-jre since

Bug#661465: openjdk-7-jre: dependency on gconf2

2012-03-06 Thread Andrei Paskevich
> well, then you can use openjdk-7-jre-headless Unfortunately, we cannot use openjdk-7-jre-headless when we need any of two hundred packages that depend on -jre, such as eclipse. I can only repeat myself: I believe that it should be possible to use the default Java (or any other programming langua

Bug#661465: openjdk-7-jre: dependency on gconf2

2012-02-27 Thread Andrei Paskevich
Package: openjdk-7-jre Version: 7~u3-2.1-1 Severity: normal Dear Maintainer, the new version of openjdk-7 in unstable introduces dependency on libatk-wrapper-java-jni which depends on gconf2. I believe that it should be possible to use the default Java implementation in Debian without installing

Bug#606849: rxvt-unicode: -tr option is broken since last update

2010-12-12 Thread Andrei Paskevich
Package: rxvt-unicode Version: 9.09-3 Severity: normal When I launch 'urxvt -tr' I obtain a usual urxvt window instead of getting a fake-transparent window with the root pixmap in the background. Apparently, either -tr option is ignored in the last version of urxvt or it just fails to retrieve

Bug#603100: bsdmainutils: ncal ignores -M when -b is used

2010-11-10 Thread Andrei Paskevich
Package: bsdmainutils Version: 8.0.15 Severity: normal The last version of ncal supports the cal-styled output with -b (thanks a lot for this!). However, ncal -b ignores -M, making the bugs #590592 and #598385 actually unfixed. -- System Information: Debian Release: squeeze/sid APT prefers un

Bug#582247: gcalcli fails with AttributeError on all operations

2010-05-19 Thread Andrei Paskevich
Package: gcalcli Version: 1.4-3 Severity: grave Tags: upstream Justification: renders package unusable The issue and a simple solution are fully described on http://code.google.com/p/gcalcli/issues/detail?id=51 I'm not the original reporter (randerson55), but my symptoms are exactly the same and

Bug#526574: fbreader: toggle toolbar with a shortcut, save in the UI state

2009-05-01 Thread Andrei Paskevich
Package: fbreader Version: 0.10.7dfsg-1 Severity: wishlist It would be very handy to show/hide the toolbar with a shortcut and to have the current state saved in ui.xml. Currently, the toolbar is always visible at the start (in the windowed mode). -- System Information: Debian Release: squeez

Bug#526569: fbreader: grabs key release when started from command line

2009-05-01 Thread Andrei Paskevich
Package: fbreader Version: 0.10.7dfsg-1 Severity: important When I launch FBReader from command line, it starts quickly enough to get focus before I release the key. FBReader interprets key releases rather than key presses and thus dutifully toggles the fullscreen mode on every start. Current

Bug#468254: _get_cword drops spaces at the end of the completion word

2008-03-23 Thread Andrei Paskevich
Please, put $cur in double quotes in the last 'echo' in _get_cword: --- /usr/share/bash/bash_completion 2008-03-23 01:09:58.0 +0200 +++ /etc/bash_completion2008-03-23 20:56:42.0 +0200 @@ -211,7 +211,7 @

Bug#472132: bash-completion: _get_cword works incorrectly in UTF-8 locales

2008-03-22 Thread Andrei Paskevich
Package: bash-completion Version: 20060301-3 Severity: normal Tags: patch In the definition of _get_cword in bash_completion, both $COMP_POINT and $word_start stand for offset in bytes, not in characters. However, in the ${variable:offset} expression, 'offset' must be in characters. In a UTF-8

Bug#468254: bash-completion: strange behavior on ls and spaces

2008-03-11 Thread Andrei Paskevich
The problem can be partially cured by the following one-line change in _get_cword: = --- bash_completion.old 2008-03-12 00:52:00.0 +0200 +++ /etc/bash_completion2008-03-12 00:53:23.0 +0200 @@ -188,7 +1

Bug#470539: bash-completion: completion function _umount sets IFS incorrectly

2008-03-11 Thread Andrei Paskevich
Package: bash-completion Version: 20060301-3 Severity: normal Tags: patch Currently, completion for umount does not work. The attached patch fixes the problem. -- System Information: Debian Release: lenny/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: i386 (i686) Kerne

Bug#467231: bash_completion is big and loads slowly; load-by-need proposed

2008-02-23 Thread Andrei Paskevich
Package: bash Version: 3.1dfsg-9 Severity: wishlist The default /etc/bash_completion in Debian/sid takes 216kb, and a couple of scripts in /etc/bash_completion.d add another 100kb. To load this, my system needs half a second, a perceivable and quite annoying delay on every shell start. I wrote