control: tag -1 -moreinfo +confirmed control: noowner -1 I consider d648533e5f1d9942a07cb9e907eec2d58b1965fb ready for upload. On Mon, Nov 14, 2016 at 04:47:59PM +0800, Boyuan Yang wrote: > A missing dependency libqt4-sql-sqlite caused this problem. It seems that > the automatic dh_shlibdeps failed to list it inside Depends list (it only > listed libqt4-sql). A explicit dependency was added the latest commit. Okay. Perhaps dlopen() is used to access libqt4-sql-sqlite. > > > * Copyright information update for beta10 version. > > > > 3. Are you installing the plugins at present, 'hunspell' and 'webcam'? > > Contents of package suggests not. Upstream seems to think they belong > > in separate binary packages. What do you think about that? > > Added debian/README.Debian file to explain the situation. The plugin structure > won't support Qt5, plus upstream placed the plugin inside /usr/share, which is > not acceptable. The building system (CMake) for plugins is not in standard > status, too. My choice is not to install the plugins and wait until upstream > provides with a better solution. > > Patching may be possible but that would be painful. Yes, this seems like an upstream problem. I hope they can resolve it. > > > Blhc on debomatic reported some missing CXXFLAGS (-fPIE). However > > > since new gcc has already set -fPIE by default since 6.2.0-7, this > > > should be harmless. > > > > Yes. Lintian would be complaining if it wasn't being passed. > > Again this needs patching, so I would choose to look into it later until some > real problem comes up. What needs patching? If it was needed, Lintian would have emitted a warning. > > 4. I'm curious why you nuked gbp.conf. Entirely up to you, of > > course. > > I found git-buildpackage not reading the "debian/gbp.conf" at all unless I > list it explicitly inside environment variable, which is different from words > written in man 5 gbp.conf. Command line arguments should be a better choice > for me. That's really weird. I've never seen that. You might want to report a bug. > > 5. Your Forwarded: headers got eaten at some point. Would be nice > > to restore them. > > Well it is a problem hiding in gbp pq, which will eat non-mandatory headers > when exporting patches because git commit does not preserve those information. > Maybe I fill file a bug report (feature request?) later. Already exists: #785274 A workaround is to put the Forwarded: headed at the end of the patch description, above the '---' line. -- Sean Whitton
Attachment:
signature.asc
Description: PGP signature