Re: [Development] Binary compatibility for qtestlib

2015-05-20 Thread Thiago Macieira
On Wednesday 20 May 2015 11:28:26 Knoll Lars wrote: > >So do we maintain binary compatibility for testlib‎? > > I don’t see a strong use case for staying binary compatible with testlib. There is one use-case, which is to run regression tests without recompiling the unit test. All the tests that

Re: [Development] Binary compatibility for qtestlib

2015-05-20 Thread Hausmann Simon
on Cc: development@qt-project.org Subject: Re: [Development] Binary compatibility for qtestlib On Wed, May 20, 2015 at 9:16 PM, Hausmann Simon mailto:simon.hausm...@theqtcompany.com>> wrote: ‎Hi, Lately development of testlib picked up again and I've been wondering: the api consists of

Re: [Development] Binary compatibility for qtestlib

2015-05-20 Thread Knoll Lars
On 20/05/15 13:16, "Hausmann Simon" wrote: >‎Hi, > > >Lately development of testlib picked up again and I've been wondering: >the api consists of a fair amount of macros that call "internal" >functions. It would be convenient to change the signature > of those while maintaining source compatibil

Re: [Development] Binary compatibility for qtestlib

2015-05-20 Thread Jason McDonald
On Wed, May 20, 2015 at 9:16 PM, Hausmann Simon < simon.hausm...@theqtcompany.com> wrote: > ‎Hi, > > Lately development of testlib picked up again and I've been wondering: > the api consists of a fair amount of macros that call "internal" functions. > It would be convenient to change the signatu

[Development] Binary compatibility for qtestlib

2015-05-20 Thread Hausmann Simon
‎Hi, Lately development of testlib picked up again and I've been wondering: the api consists of a fair amount of macros that call "internal" functions. It would be convenient to change the signature of those while maintaining source compatibility, however it would naturally break the ABI. On t