Bug#800460: Ping
Hello Dave, Στις Wednesday 20 of January 2016 18:18:28 γράψατε: > Hi Dimitrios, > > > It looks like there is still a problem with the latest git head. It > seems that there may have been a mishap with the patching, and the patch > has been reversed at some point? I can see in b7027842 that the C++11 > has been set back to ON in the CMakeLists file. I did that intentionally, thinking that problems have been solved and we should apply as many features as we can. It seems that this is not the case. So I will do a new upload with the C++11 features disablesd in the CMakeLists file. > > I personally keep a debian/source/local-options file like so, to enforce > patches-unapplied in my other repository: > > https://anonscm.debian.org/cgit/debian-science/packages/3depict.git/tree/deb > ian/source/local-options?id=7fc15a0ad61ab38f0c1e4c1b2085dadf212d693c > > Last time I checked this was the recommended thing to do, as quilt + gbp > dont play nicely together. > > I'm not totally clear what is going on here, but removing the .pc > directory and going back to a patches-unapplied git seems to fix the > problem for me. I started keeping the .pc directory some time ago, because I noticed that the deletion of it somehow reverted the effect of the patches. I do not see any side eefects up to now. > > I dont want to push any changes until we agree on what both the cause of > the problem and what the solution is. > > > Thanks! Best regards Dimitris
Bug#800460: Ping
Hi Dmitirios, Thanks for the quick response. My current opinion is that we should lock-step with the C++11 transition in Debian, which occurs with gcc6. https://wiki.debian.org/GCC5#Prepare_for_GCC_6 Otherwise, we are simply risking bugs like 798858, 800460 and 80953, as C++11 is not the Debian default at this time. On 21/01/16 12:00, Dimitrios Eftaxiopoulos wrote: > Hello Dave, > > > Στις Wednesday 20 of January 2016 18:18:28 γράψατε: >> Hi Dimitrios, >> >> >> It looks like there is still a problem with the latest git head. It >> seems that there may have been a mishap with the patching, and the patch >> has been reversed at some point? I can see in b7027842 that the C++11 >> has been set back to ON in the CMakeLists file. > > I did that intentionally, thinking that problems have been solved and we > should apply as many features as we can. It seems that this is not the case. > So I will do a new upload with the C++11 features disablesd in the CMakeLists > file. > >> >> I personally keep a debian/source/local-options file like so, to enforce >> patches-unapplied in my other repository: >> >> https://anonscm.debian.org/cgit/debian-science/packages/3depict.git/tree/deb >> ian/source/local-options?id=7fc15a0ad61ab38f0c1e4c1b2085dadf212d693c >> >> Last time I checked this was the recommended thing to do, as quilt + gbp >> dont play nicely together. >> >> I'm not totally clear what is going on here, but removing the .pc >> directory and going back to a patches-unapplied git seems to fix the >> problem for me. > > I started keeping the .pc directory some time ago, because I noticed that the > deletion of it somehow reverted the effect of the patches. I do not see any > side eefects up to now. > >> >> I dont want to push any changes until we agree on what both the cause of >> the problem and what the solution is. >> >> >> Thanks! > > Best regards > Dimitris >
Bug#800460: Ping
Hi Dimitrios, It looks like there is still a problem with the latest git head. It seems that there may have been a mishap with the patching, and the patch has been reversed at some point? I can see in b7027842 that the C++11 has been set back to ON in the CMakeLists file. I personally keep a debian/source/local-options file like so, to enforce patches-unapplied in my other repository: https://anonscm.debian.org/cgit/debian-science/packages/3depict.git/tree/debian/source/local-options?id=7fc15a0ad61ab38f0c1e4c1b2085dadf212d693c Last time I checked this was the recommended thing to do, as quilt + gbp dont play nicely together. I'm not totally clear what is going on here, but removing the .pc directory and going back to a patches-unapplied git seems to fix the problem for me. I dont want to push any changes until we agree on what both the cause of the problem and what the solution is. Thanks!
