Hi Ted, I can only repeat if you suspect that something is wrong: Please file a bug report. Otherwise the information will bit-rot in my inbox, sorry.
Kind regards Andreas. Am Wed, Jan 26, 2022 at 04:19:00PM -0500 schrieb Ted Ralphs: > Hi Andreas, > > Thanks for the kind reply and sorry for the direct e-mail. As a fellow > maintainer, I also get direct e-mails that I wish would be posted to bug > trackers or mailing lists. Honestly, I was in a hurry and couldn't easily > figure out the right place for this kind of report. Anyway, thanks for your > patient explanation, it makes sense. The executable does indeed seem to be > stripped in my local installation. However, it does seem as though NDEBUG > is not defined as it is supposed to be (and would be with the default set > of flags that the configure script produces). I guess this is because we > are adding this in a non-standard way? It's probably not a big deal, but it > means there are asserts and other things that are not intended to be run in > production code that are being run. Is this something that would be > appropriate to raise on the maintainer's list? I may just be missing some > basic knowledge of the packaging system. > > Cheers, > > Ted > > On Wed, Jan 26, 2022 at 1:18 AM Andreas Tille <andr...@an3as.eu> wrote: > > > Hi Ted, > > > > Am Tue, Jan 25, 2022 at 10:37:49PM -0500 schrieb Ted Ralphs: > > > Hi Andreas, > > > > > > I am the author of SYMPHONY, which you so kindly packaged on Deiban. > > > > It was not really me but the Debian Science team. Please note: > > Communication > > via our mailing list or via the bug tracking system (by calling > > reportbug coinor-symphony > > ) is prefered over contacting single maintainers. The rationale is that > > I personally have my hands in more than 1000 Debian packages and might not > > be as responsible as I want to be. > > > > > Thanks > > > so much for that, it's a huge help! I recently became aware that SYMPHONY > > > is built with "-g" on Debian for some odd reason. > > > > That's not odd but by policy. There is no harm done by using the -g > > option since Debian automatically provides a dbgsym package including > > debug symbols. The resulting binary is stripped afterwards in the > > packaging process (man dh_strip). > > > > > This is actually true for > > > some other COIN-OR projects and I'm trying to track down the reason. > > There > > > is a discussion about this same issue in Cbc here: > > > > > > https://github.com/coin-or/Cbc/issues/474 > > > > > > I think it has somehow to do with the autoreconf, which seems to result > > in > > > different (and wrong) default flags being selected, but it's a little > > hard > > > for me to tell what's happening. Do you have any idea? > > > > Please checkout the resulting binary if you are not convinced that debug > > symbols are removed. > > > > > By the way, I plan > > > to release a 5.7.0 version soon, which will incorporate a completely > > > revamped build system, which may fix this issue. Thanks again! > > > > I hope it will not "fix" the chance to build with -g in the first place. > > ;-) Feel free to ping on the maintainers list or create a bug report > > with the title "New version available" once this is out. > > > > Thank you in any case for your feedback about the Debian package > > > > Andreas. > > > > > > > > -- > > http://fam-tille.de > > > > > -- > Dr. Ted Ralphs > Professor, Industrial and Systems Engineering > Lehigh University > (610) 628-1280 > ted 'at' lehigh 'dot' edu > coral.ie.lehigh.edu/~ted -- http://fam-tille.de -- debian-science-maintainers mailing list debian-science-maintainers@alioth-lists.debian.net https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-science-maintainers