[removing upstream from CC since my message is about Debian] Dear Dmitry,
On Tue, Dec 06, 2016 at 08:28:22AM +0300, Dmitry Bogatov wrote: > Thank you for report. Thank you for your reply! > [Added upstream maintainer into CC: frank-fisc...@shadow-soft.de] > > I just installed emacs25 and now my /usr/bin/emacs points to > /usr/bin/emacs25-nox, and I can confirm, that `make test' do fail. Dear > upstream maintainer, can you please take a look? I tested `hg tip` and reported a bug upstream: https://bitbucket.org/lyro/evil/issues/731/test-suite-fails-with-emacs-25 > [Back to Debian affairs.] > > Right now it seems impossible to uninstall emacs24 without breaking a > lot of things: > > ; sudo apt-get purge emacs24-nox emacs24-lucid emacs24 > [... reflowed a bit ...] > The following packages will be REMOVED: > dash-el* dh-elpa* elpa-aggressive-indent* > elpa-elisp-slime-nav* elpa-epl* elpa-evil* elpa-evil-leader* > elpa-evil-paredit* elpa-flycheck* elpa-git-commit* > elpa-goto-chg* elpa-magit* elpa-magit-popup* elpa-paredit* > elpa-pkg-info* elpa-undo-tree* elpa-with-editor* emacs* > emacs24-nox* haskell-mode* > [...] > Do you want to continue? [Y/n] ^C > > But I just switched to emacs25 and everything is fine, including > evil. So whatever issues test suite discovers, they do not disrupt my > emacs/evil workflow. So, if we would fail to fix test suite till emacs25 > becomes the only emacs, I would comment-out tests. Unprincipled -- yes, > but RM would be much greater problem. At least for me. We won't remove emacs24 until is no longer has any reverse dependencies. The ftp-masters wouldn't allow it, for one thing. Once I update dh-elpa, all the elpa-* packages you list would be fixed. We have filed bugs against the others, and might NMU to bump s/emacs24/emacs25/. I have not yet updated dh-elpa because of evil-el's test suite, i.e., this bug is blocking updating dh-elpa. If the upstream fix proves to be very difficult and upstream is unable to provide a fix soon, we have two options: 1. Update dh-elpa anyway, causing evil-el to FTBFS, and bump this bug to RC-severity to keep evil-el out of stretch. Not ideal, but evil-el has never been in testing anyway thanks to #829299. 2. Update dh-elpa and comment out the failing tests in evil-el. I don't think we should proceed with option (2) unless we are confident that elpa-evil works fine for everyday usage under emacs25, and #829299 gets fixed. (If #829299 is not fixed, option (2) achieves nothing.) I'd like to ask you to do two things: - Since you use evil, could you run elpa-evil under emacs25 for the next two or three weeks for your regular usage, and report back? - Could you fix #829299 based on the private e-mails we exchanged a few weeks ago, please? Thanks again. -- Sean Whitton
signature.asc
Description: PGP signature
