Re: How to deal with meaningless SOVERSION bumps from upstream?

2019-03-19 Thread Bernd Zeimetz
On 3/19/19 2:25 AM, Mo Zhou wrote: > Hello guys, > > I'm talking about the src:double-conversion package (popcon >= 70k), > and a choice for the post-Buster stage. > > The upstream doesn't follow semantic versioning convention at all. > Recently they have changed the SOVERSION to 3: > https://git

Re: How to deal with meaningless SOVERSION bumps from upstream?

2019-03-19 Thread Ondřej Surý
> > On 19 Mar 2019, at 11:15, Florian Weimer wrote: > > * Mo Zhou: > >> Should I bump the SOVERSION and trigger a transition after the Buster >> release? (I think it's pointless, as it triggers pointless rebuild for >> many packages including Qt) Or ignore the upstream SOVERSION bump and >> tra

Re: Reg: Debian Source code

2019-03-19 Thread rajudev
Srinivas Rao writes: > Hi Debian devel team, > > I am planning to work on small IoT kind of project using Debian OS. I would > like to customize Debian OS. could you please tell me, where can I get or > download Debian source code and how can I create build system. > > please let me know if an

Re: How to deal with meaningless SOVERSION bumps from upstream?

2019-03-19 Thread Florian Weimer
* Sam Hartman: > There's fairly significant advantages to keeping ABI consistent with > upstream when possible. > So, unless the Debian ABI has diverged from upstream already, I'd > recommend that you track the soname change and accept the rebuilds > post-buster. It's possible to support both son

Re: How to deal with meaningless SOVERSION bumps from upstream?

2019-03-19 Thread Florian Weimer
* Mo Zhou: > Should I bump the SOVERSION and trigger a transition after the Buster > release? (I think it's pointless, as it triggers pointless rebuild for > many packages including Qt) Or ignore the upstream SOVERSION bump and > track the changes by myself? I've been able to convince libidn2 to

Re: How to deal with meaningless SOVERSION bumps from upstream?

2019-03-19 Thread Sam Hartman
There's fairly significant advantages to keeping ABI consistent with upstream when possible. So, unless the Debian ABI has diverged from upstream already, I'd recommend that you track the soname change and accept the rebuilds post-buster. If you've already had to break ABI with upstream, then be as