Bug#908747: dpkg-source: Default -I and -i option should not exclude .ignore

2021-11-03 Thread Osamu Aoki
I agree with Ian. If package maintainer wants to have a few extra ignore rules locally, why not use > .git/info/exclude file. I think its high time to fix this. Osamu

Bug#615807: Side effects of case insensitive "dpkg-query -s"

2019-03-16 Thread Osamu Aoki
control: severity -1 wishlist thanks This behavior was not much of real problem but how this happens was a bit of mastery for me. Here is what I found out. https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=615807 If any kind of search term normalization is in place, search result shouldn't

Bug#615807: dpkg-www case sesitivity issue

2019-03-15 Thread Osamu Aoki
Hi, The situation of case sensitivity can be summarized for web browser page opened by the "dpkg-www" as: Search term with tailing "*": case sensitive Search term without tailing "*": case insensitive As I read Policy 4.3.0

Bug#817040: dpkg-deb -c: sort list by name

2016-03-08 Thread Osamu Aoki
Hi, On Tue, Mar 08, 2016 at 01:10:49AM +0100, Guillem Jover wrote: > I don't think the attached patch does what you intended though? --sort > only works when packing (i.e. when reading from the filesystem). Very true. Excuse me. Maybe basic command such as dpkg-deb shoudl not filter output

Bug#817040: dpkg-deb -c: sort list by name

2016-03-07 Thread Osamu Aoki
Package: dpkg Version: 1.18.4 Severity: wishlist Tags: patch Making dpkg-deb to be deterministic by fixing #719845 was good for reproducible build. But "dpkg-deb -c" output is not human friendly. Basically, we want it to be sorted by name instead of internal data order. Please consider adding

Bug#796671: dpkg-source: warn removal only when directory is non-empty

2015-08-23 Thread Osamu Aoki
Package: dpkg-dev Version: 1.17.25 Severity: normal This is about dpkg-source -x ... for multiple upstream tarbells. In short, if there are directories matching component names of secondary tarballs, dpkg-source should not warn them if such directories contain nothing. (I checked the pertinent

Bug#721256: dpkg: multiarch friendly dpkg -L

2013-08-29 Thread Osamu Aoki
Package: dpkg Version: 1.17.1 Severity: wishlist In short, dpkg -L is not friendly to multiarch. Let's make dpkg -L as friendly as dpkg -l without causing problems to others. Currently, if I run dpkg -L on a multiarch package, it returns very unfriendly response. For example: | $ dpkg -L

Bug#699807: dpkg: [U-A] display for the auto mode of the same priority

2013-02-05 Thread Osamu Aoki
Package: dpkg Version: 1.16.9 Severity: normal [This may be a minor bug.] Command: update-alternatives ---config etc. The displayed selection of auto mode with the same priority currently seems to be based on the first alphabetic order one. But the reality for the same priority case seems to

Bug#590472: dpkg-dev: deb-control needs a bit more udates for Breaks

2010-07-26 Thread Osamu Aoki
Package: dpkg-dev Version: 1.15.7.2 Severity: minor deb-control(5) has been updated for new 3.9 policy but it is not perfect yet. I attach a sure fix as patch and discuss other points below: Question: Do we need tomention Breaks in below in place of Conflicts or list both? Replaces:

Bug#141325: Fix for lenny possible?? : Bug#141325: [u-a] simple fix for single manual choice under dangling link and --config

2008-10-04 Thread Osamu Aoki
Hi dpkg folks, I think this old update-alternatives bug needs to be fixed for lenny. I merged many bug reports. Many of which were reassigned from other packages. So impact to annoyance of users exists. If you want to be very conservative patch for lenny, you should use just following parts of

Bug#141325: [u-a] simple fix for single manual choice under dangling link and --config

2008-10-01 Thread Osamu Aoki
severity 141325 important tags 141325 patch forcemerge 141325 199426 203582 282290 285686 311288 311288 thanks Hi, When alternative system has broken link for some reason and only one package uses that alternative system in manual mode, update-alternative does not update link properly and exits