Bug#946055: /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution

2019-12-18 Thread Antoine Beaupré
On 2019-12-18 22:59:54, Thorsten Glaser wrote: >> I encourage you to engage with joeyh anyways, if you can. ;) > > We engaged in real life during our trip to DebConf in Bosnia, > which was enjoyable (if a lengthy road trip), I can relate ;) > I just had trouble with the website (it’s not my medium

Bug#946055: /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution

2019-12-18 Thread Thorsten Glaser
On Wed, 18 Dec 2019, Antoine Beaupré wrote: > It's not a static page: there's a field at the top of the page where you > can input a page name and then you get served a login form and so on. Ah, okay. Perhaps it didn’t show up in my browser or so. > But nevermind that, I filed it: > >

Bug#946055: /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution

2019-12-18 Thread Antoine Beaupré
On 2019-12-18 20:42:11, Thorsten Glaser wrote: >> https://etckeeper.branchable.com/todo/ > > This seems to be a static page? I don’t find out how to do that. It's not a static page: there's a field at the top of the page where you can input a page name and then you get served a login form and so

Bug#946055: /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution

2019-12-18 Thread Antoine Beaupré
Control: forward -1 https://etckeeper.branchable.com/todo/vcs-commit_hook_is_not_POSIX-friendly/ On 2019-12-18 20:32:41, Thorsten Glaser wrote: > On Wed, 18 Dec 2019, Thorsten Glaser wrote: > >> The patch fixes the issue by reimplementing the -mmessage feature >> introduced in 1.18.11 in

Bug#946055: /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution

2019-12-18 Thread Thorsten Glaser
Hi Antoine, > This package, like all of mine, is LowNMU so upload whenever you feel > ready. OK, thanks! > > ++ sed '1s/^-m \{0,1\}//' >"$logfile" <<-EOF > > ++ $* > > ++ EOF > > ... isn't there a simpler way than piping this through sed? I’ve wondered about

Bug#946055: /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution

2019-12-18 Thread Thorsten Glaser
On Wed, 18 Dec 2019, Thorsten Glaser wrote: > The patch fixes the issue by reimplementing the -mmessage feature > introduced in 1.18.11 in portable shell *and* fixing issues with > messages starting with -n or -c or containing backslashes as well. I can confirm that a package built with this

Bug#946055: /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution

2019-12-18 Thread Antoine Beaupré
On 2019-12-18 20:16:50, Thorsten Glaser wrote: > Package: etckeeper > Version: 1.18.12-1 > Followup-For: Bug #946055 > > Find a patch attached. I intend to NMU if this isn’t fixed RSN. This package, like all of mine, is LowNMU so upload whenever you feel ready. > The patch fixes the issue by

Bug#946055: /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution

2019-12-18 Thread Thorsten Glaser
Package: etckeeper Version: 1.18.12-1 Followup-For: Bug #946055 Find a patch attached. I intend to NMU if this isn’t fixed RSN. The patch fixes the issue by reimplementing the -mmessage feature introduced in 1.18.11 in portable shell *and* fixing issues with messages starting with -n or -c or

Bug#946055: /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution

2019-12-03 Thread Thorsten Glaser
Package: etckeeper Version: 1.18.12-1 Severity: serious Justification: Policy 10.4 /etc/etckeeper/commit.d/50vcs-commit[22]: ${@#-m}: bad substitution This is because doing trim operations on an array are not implemented in mksh, and unspecified in POSIX: […] If parameter is '#', '*', or '@',