{"thread":{"id":"34937","subject":"1.8.4 rebase regression?","startedAt":"2013-09-15T23:57:39Z","lastAt":"2013-09-17T10:59:26Z","messageCount":10,"participants":["Patrick Welche","Matthieu Moy"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"227703","messageId":"20130915235739.GA712@quark","threadId":"34937","inReplyTo":null,"subject":"1.8.4 rebase regression?","fromName":"Patrick Welche","fromEmail":"prlw1@cam.ac.uk","sentAt":"2013-09-15T23:57:39Z","receivedAt":"2013-09-15T23:57:39Z","isPatch":false,"sender":{"key":"prlw1@cam.ac.uk","avatar":null},"body":"I just upgraded (via pkgsrc) from git 1.8.3.4 to 1.8.4. With 1.8.4, I had\nlocal changes in glib, did a git pull --rebase. Some of my changes\nconflicted, but\n\n$ git rebase --abort\nNo rebase in progress?\n\nso somehow the usual process of amending the edit, or skipping the patch\nno longer works.\n\nI found a similar report at:\nhttp://mail-index.netbsd.org/pkgsrc-users/2013/09/14/msg018646.html\n\n\nAnother less important regression is that in an xterm, with 1.8.3.4 I see\n\n$ /tmp/bin/git diff \ndiff --cc glib/gmain.c\nindex 738e69c,5aaebd0..0000000\n--- a/glib/gmain.c\n+++ b/glib/gmain.c\n@@@ -4953,32 -4921,32 +4953,48 @@@ g_unix_signal_watch_dispatch (GSourc\n  }\n...\n\nbut with 1.8.4, I see\n\n$ git diff\nESC[1mdiff --cc glib/gmain.cESC[m\nESC[1mindex 738e69c,5aaebd0..0000000ESC[m\nESC[1m--- a/glib/gmain.cESC[m\nESC[1m+++ b/glib/gmain.cESC[m\nESC[36m@@@ -4953,32 -4921,32 +4953,48 @@@ESC[m ESC[mg_unix_signal_watch_dispatch (GSourcESC[m\n\n\n(same xterm, no change of TERM in both invocations above)\ngit status in 1.8.4 does show red, so colour does work...\n\nThoughts on how to help debug?\n\nCheers,\n\nPatrick\n"},{"id":"227706","messageId":"vpqsix5b3uk.fsf@anie.imag.fr","threadId":"34937","inReplyTo":"20130915235739.GA712@quark","subject":"Re: 1.8.4 rebase regression?","fromName":"Matthieu Moy","fromEmail":"matthieu.moy@grenoble-inp.fr","sentAt":"2013-09-16T07:52:35Z","receivedAt":"2013-09-16T07:52:35Z","isPatch":false,"sender":{"key":"matthieu.moy@grenoble-inp.fr","avatar":"https://gravatar.com/avatar/72c8a2705971a25dfaff23cece15130d405685845d911aedd5667ace277f3fc5?d=mp&s=160"},"body":"Patrick Welche <prlw1@cam.ac.uk> writes:\n\n> I just upgraded (via pkgsrc) from git 1.8.3.4 to 1.8.4. With 1.8.4, I had\n> local changes in glib, did a git pull --rebase. Some of my changes\n> conflicted, but\n>\n> $ git rebase --abort\n> No rebase in progress?\n>\n> so somehow the usual process of amending the edit, or skipping the patch\n> no longer works.\n>\n> I found a similar report at:\n> http://mail-index.netbsd.org/pkgsrc-users/2013/09/14/msg018646.html\n\nDoes the following patch fix this issue?\n\n  https://github.com/git/git/commit/99855dd\n\n-- \nMatthieu Moy\nhttp://www-verimag.imag.fr/~moy/\n"},{"id":"227707","messageId":"20130916102832.GB489@quark","threadId":"34937","inReplyTo":"vpqsix5b3uk.fsf@anie.imag.fr","subject":"Re: 1.8.4 rebase regression?","fromName":"Patrick Welche","fromEmail":"prlw1@cam.ac.uk","sentAt":"2013-09-16T10:28:32Z","receivedAt":"2013-09-16T10:28:32Z","isPatch":false,"sender":{"key":"prlw1@cam.ac.uk","avatar":null},"body":"On Mon, Sep 16, 2013 at 09:52:35AM +0200, Matthieu Moy wrote:\n> Patrick Welche <prlw1@cam.ac.uk> writes:\n> \n> > I just upgraded (via pkgsrc) from git 1.8.3.4 to 1.8.4. With 1.8.4, I had\n> > local changes in glib, did a git pull --rebase. Some of my changes\n> > conflicted, but\n> >\n> > $ git rebase --abort\n> > No rebase in progress?\n> >\n> > so somehow the usual process of amending the edit, or skipping the patch\n> > no longer works.\n> >\n> > I found a similar report at:\n> > http://mail-index.netbsd.org/pkgsrc-users/2013/09/14/msg018646.html\n> \n> Does the following patch fix this issue?\n> \n>   https://github.com/git/git/commit/99855dd\n\nYes thanks!\n\nCONFLICT (content): Merge conflict in glib/gmain.c\nFailed to merge in the changes.\n\n:-)\n\nAny thoughts of the cosmetic ESC[31m change?\n\nCheers,\n\nPatrick\n"},{"id":"227708","messageId":"vpqioy1815z.fsf@anie.imag.fr","threadId":"34937","inReplyTo":"20130915235739.GA712@quark","subject":"Re: 1.8.4 rebase regression?","fromName":"Matthieu Moy","fromEmail":"matthieu.moy@grenoble-inp.fr","sentAt":"2013-09-16T11:18:48Z","receivedAt":"2013-09-16T11:18:48Z","isPatch":false,"sender":{"key":"matthieu.moy@grenoble-inp.fr","avatar":"https://gravatar.com/avatar/72c8a2705971a25dfaff23cece15130d405685845d911aedd5667ace277f3fc5?d=mp&s=160"},"body":"Patrick Welche <prlw1@cam.ac.uk> writes:\n\n> $ git diff\n> ESC[1mdiff --cc glib/gmain.cESC[m\n> ESC[1mindex 738e69c,5aaebd0..0000000ESC[m\n> ESC[1m--- a/glib/gmain.cESC[m\n> ESC[1m+++ b/glib/gmain.cESC[m\n> ESC[36m@@@ -4953,32 -4921,32 +4953,48 @@@ESC[m ESC[mg_unix_signal_watch_dispatch (GSourcESC[m\n>\n>\n> (same xterm, no change of TERM in both invocations above)\n> git status in 1.8.4 does show red, so colour does work...\n>\n> Thoughts on how to help debug?\n\nCan you try:\n\ngit -c color.ui=never diff\ngit -c color.ui=auto diff\ngit -c color.ui=always diff\n\n?\n\nIf you have a bit of time, you can use \"git bisect\" on a clone of\ngit.git to find out the guilty commit.\n\n-- \nMatthieu Moy\nhttp://www-verimag.imag.fr/~moy/\n"},{"id":"227730","messageId":"vpqd2o77wbk.fsf@anie.imag.fr","threadId":"34937","inReplyTo":"vpqioy1815z.fsf@anie.imag.fr","subject":"Re: 1.8.4 rebase regression?","fromName":"Matthieu Moy","fromEmail":"matthieu.moy@grenoble-inp.fr","sentAt":"2013-09-17T07:15:43Z","receivedAt":"2013-09-17T07:15:43Z","isPatch":false,"sender":{"key":"matthieu.moy@grenoble-inp.fr","avatar":"https://gravatar.com/avatar/72c8a2705971a25dfaff23cece15130d405685845d911aedd5667ace277f3fc5?d=mp&s=160"},"body":"Matthieu Moy <Matthieu.Moy@grenoble-inp.fr> writes:\n\n> Patrick Welche <prlw1@cam.ac.uk> writes:\n>\n>> $ git diff\n>> ESC[1mdiff --cc glib/gmain.cESC[m\n>> ESC[1mindex 738e69c,5aaebd0..0000000ESC[m\n>> ESC[1m--- a/glib/gmain.cESC[m\n>> ESC[1m+++ b/glib/gmain.cESC[m\n>> ESC[36m@@@ -4953,32 -4921,32 +4953,48 @@@ESC[m ESC[mg_unix_signal_watch_dispatch (GSourcESC[m\n>>\n>>\n>> (same xterm, no change of TERM in both invocations above)\n>> git status in 1.8.4 does show red, so colour does work...\n>>\n>> Thoughts on how to help debug?\n>\n> Can you try:\n>\n> git -c color.ui=never diff\n> git -c color.ui=auto diff\n> git -c color.ui=always diff\n>\n> ?\n\n... and Junio suggested offline to look for a broken pager, so, you can\ntry this too:\n\ngit --no-pager diff\n\n-- \nMatthieu Moy\nhttp://www-verimag.imag.fr/~moy/\n"},{"id":"227739","messageId":"20130917091333.GB289@quark","threadId":"34937","inReplyTo":"vpqioy1815z.fsf@anie.imag.fr","subject":"Re: 1.8.4 rebase regression?","fromName":"Patrick Welche","fromEmail":"prlw1@cam.ac.uk","sentAt":"2013-09-17T09:13:33Z","receivedAt":"2013-09-17T09:13:33Z","isPatch":false,"sender":{"key":"prlw1@cam.ac.uk","avatar":null},"body":"On Mon, Sep 16, 2013 at 01:18:48PM +0200, Matthieu Moy wrote:\n> Patrick Welche <prlw1@cam.ac.uk> writes:\n> \n> > $ git diff\n> > ESC[1mdiff --cc glib/gmain.cESC[m\n> > ESC[1mindex 738e69c,5aaebd0..0000000ESC[m\n> > ESC[1m--- a/glib/gmain.cESC[m\n> > ESC[1m+++ b/glib/gmain.cESC[m\n> > ESC[36m@@@ -4953,32 -4921,32 +4953,48 @@@ESC[m ESC[mg_unix_signal_watch_dispatch (GSourcESC[m\n> >\n> >\n> > (same xterm, no change of TERM in both invocations above)\n> > git status in 1.8.4 does show red, so colour does work...\n> >\n> > Thoughts on how to help debug?\n> \n> Can you try:\n> \n> git -c color.ui=never diff\n> git -c color.ui=auto diff\n> git -c color.ui=always diff\n> \n> ?\n\nGot it: the change between 1.8.3.4 and 1.8.4 is that colour is on by\ndefault. If I take 1.8.3.4 and git -c color.ui=always log, I see\nthe same ESC codes => not a regression! I'll just have to sort my\nbox out if I want colour. (The only oddity is that git status is\ncorrectly colourful.)\n\n(\n> If you have a bit of time, you can use \"git bisect\" on a clone of\n> git.git to find out the guilty commit.\n\ncommit e5be297279e8ee8c503eb59da21ab17edc40e748\nMerge: a3bc3d0 6897a64\n\nbut that is presumably just when the default changed\n)\n\nSorry for the confusion - it just looked like a regression to me,\nbut isn't!\n\nCheers,\n\nPatrick\n"},{"id":"227740","messageId":"20130917091508.GC289@quark","threadId":"34937","inReplyTo":"vpqd2o77wbk.fsf@anie.imag.fr","subject":"Re: 1.8.4 rebase regression?","fromName":"Patrick Welche","fromEmail":"prlw1@cam.ac.uk","sentAt":"2013-09-17T09:15:08Z","receivedAt":"2013-09-17T09:15:08Z","isPatch":false,"sender":{"key":"prlw1@cam.ac.uk","avatar":null},"body":"On Tue, Sep 17, 2013 at 09:15:43AM +0200, Matthieu Moy wrote:\n> Matthieu Moy <Matthieu.Moy@grenoble-inp.fr> writes:\n> \n> > Patrick Welche <prlw1@cam.ac.uk> writes:\n> >\n> >> $ git diff\n> >> ESC[1mdiff --cc glib/gmain.cESC[m\n> >> ESC[1mindex 738e69c,5aaebd0..0000000ESC[m\n> >> ESC[1m--- a/glib/gmain.cESC[m\n> >> ESC[1m+++ b/glib/gmain.cESC[m\n> >> ESC[36m@@@ -4953,32 -4921,32 +4953,48 @@@ESC[m ESC[mg_unix_signal_watch_dispatch (GSourcESC[m\n> >>\n> >>\n> >> (same xterm, no change of TERM in both invocations above)\n> >> git status in 1.8.4 does show red, so colour does work...\n> >>\n> >> Thoughts on how to help debug?\n> >\n> > Can you try:\n> >\n> > git -c color.ui=never diff\n> > git -c color.ui=auto diff\n> > git -c color.ui=always diff\n> >\n> > ?\n> \n> ... and Junio suggested offline to look for a broken pager, so, you can\n> try this too:\n> \n> git --no-pager diff\n\nEven better - colour works here too => I must look at my pager.\n\nThank you!\n\nPatrick\n"},{"id":"227741","messageId":"vpq8uyv6btk.fsf@anie.imag.fr","threadId":"34937","inReplyTo":"20130917091333.GB289@quark","subject":"Re: 1.8.4 rebase regression?","fromName":"Matthieu Moy","fromEmail":"matthieu.moy@grenoble-inp.fr","sentAt":"2013-09-17T09:23:51Z","receivedAt":"2013-09-17T09:23:51Z","isPatch":false,"sender":{"key":"matthieu.moy@grenoble-inp.fr","avatar":"https://gravatar.com/avatar/72c8a2705971a25dfaff23cece15130d405685845d911aedd5667ace277f3fc5?d=mp&s=160"},"body":"Patrick Welche <prlw1@cam.ac.uk> writes:\n\n> Got it: the change between 1.8.3.4 and 1.8.4 is that colour is on by\n> default. If I take 1.8.3.4 and git -c color.ui=always log, I see\n> the same ESC codes => not a regression! I'll just have to sort my\n> box out if I want colour. (The only oddity is that git status is\n> correctly colourful.)\n\nMaybe you disabled the pager for \"git status\"?\n\n> (\n>> If you have a bit of time, you can use \"git bisect\" on a clone of\n>> git.git to find out the guilty commit.\n>\n> commit e5be297279e8ee8c503eb59da21ab17edc40e748\n> Merge: a3bc3d0 6897a64\n>\n> but that is presumably just when the default changed\n> )\n\nIt doesn't seem so.\n\nOn my box, color work just fine when piped to less, but I guess not all\npagers accept this, and perhaps a wrong value of $LESS can cause less to\nreject it.\n\n-- \nMatthieu Moy\nhttp://www-verimag.imag.fr/~moy/\n"},{"id":"227742","messageId":"20130917100726.GF289@quark","threadId":"34937","inReplyTo":"vpq8uyv6btk.fsf@anie.imag.fr","subject":"Re: 1.8.4 rebase regression?","fromName":"Patrick Welche","fromEmail":"prlw1@cam.ac.uk","sentAt":"2013-09-17T10:07:26Z","receivedAt":"2013-09-17T10:07:26Z","isPatch":false,"sender":{"key":"prlw1@cam.ac.uk","avatar":null},"body":"On Tue, Sep 17, 2013 at 11:23:51AM +0200, Matthieu Moy wrote:\n> Patrick Welche <prlw1@cam.ac.uk> writes:\n> \n> > Got it: the change between 1.8.3.4 and 1.8.4 is that colour is on by\n> > default. If I take 1.8.3.4 and git -c color.ui=always log, I see\n> > the same ESC codes => not a regression! I'll just have to sort my\n> > box out if I want colour. (The only oddity is that git status is\n> > correctly colourful.)\n> \n> Maybe you disabled the pager for \"git status\"?\n\nInteresting: I didn't disable it on purpose, but indeed, it is not\nused for status.\n\nCan't believe this:\n\n$ echo $PAGER\nmore\n\nunset PAGER, and git diff is fine...\n\nThanks for all the help!\n\nCheers,\n\nPatrick\n"},{"id":"227743","messageId":"vpq1u4n67e9.fsf@anie.imag.fr","threadId":"34937","inReplyTo":"20130917100726.GF289@quark","subject":"Re: 1.8.4 rebase regression?","fromName":"Matthieu Moy","fromEmail":"matthieu.moy@grenoble-inp.fr","sentAt":"2013-09-17T10:59:26Z","receivedAt":"2013-09-17T10:59:26Z","isPatch":false,"sender":{"key":"matthieu.moy@grenoble-inp.fr","avatar":"https://gravatar.com/avatar/72c8a2705971a25dfaff23cece15130d405685845d911aedd5667ace277f3fc5?d=mp&s=160"},"body":"Patrick Welche <prlw1@cam.ac.uk> writes:\n\n> On Tue, Sep 17, 2013 at 11:23:51AM +0200, Matthieu Moy wrote:\n>> Patrick Welche <prlw1@cam.ac.uk> writes:\n>> \n>> > Got it: the change between 1.8.3.4 and 1.8.4 is that colour is on by\n>> > default. If I take 1.8.3.4 and git -c color.ui=always log, I see\n>> > the same ESC codes => not a regression! I'll just have to sort my\n>> > box out if I want colour. (The only oddity is that git status is\n>> > correctly colourful.)\n>> \n>> Maybe you disabled the pager for \"git status\"?\n>\n> Interesting: I didn't disable it on purpose, but indeed, it is not\n> used for status.\n\nAh, my bad. Whether the pager should be enabled by default for status\nled to a lot of debates here, and I thought the conclusion was yes. But\nI'm the one having it enabled by default:\n\n[pager]\n        status = true\n\n> Can't believe this:\n>\n> $ echo $PAGER\n> more\n>\n> unset PAGER, and git diff is fine...\n\nless will be a much better pager than more, indeed. The default behavior\nof less is sometimes anoying (open full-page, and 'q' restores the\nterminal, which is very inconvenient for short output), but if you\ndidn't set $LESS, then Git will set it for you to something appropriate\nfor git pager.\n\n-- \nMatthieu Moy\nhttp://www-verimag.imag.fr/~moy/\n"}]}