Re: [BUG] Index.lock error message regression in git 2.11.0
From: Robbie Iannucci <hidden>
Date: 2016-12-03 01:56:23
Apparently I'm not supposed to send attachments >_<. Here's the script in non-attachement form: -------- #!/bin/bash make -j 8 2>&1 > /dev/null if [[ $? != 0 ]]; then # skip this version exit 125 fi git=`realpath ./git` rm -rf .test_repo || true mkdir .test_repo cd .test_repo $git init echo HELLO > a $git add a $git commit -m 'initial_commit' $git branch --track new_branch $git checkout master echo HELLO >> a $git add a $git commit -m 'second_commit' $git checkout new_branch touch .git/index.lock if $git merge --ff-only master 2>&1 | grep -F "index.lock" ; then echo OK exit 0 fi echo Message is missing exit 1 On Fri, Dec 2, 2016 at 5:44 PM, Robbie Iannucci [off-list ref] wrote:
Hello, I just upgraded to 2.11.0 from 2.10.2, and I noticed that some commands no longer print an error message when the `index.lock` file exists (such as `git merge --ff-only`). It appears this bug was introduced in 55f5704da69d3e6836620f01bee0093ad5e331e8 (sequencer: lib'ify checkout_fast_forward()). I determined this by running the attached bisect script (on OS X, but I don't think that matters; I've also seen the error message missing on Linux): $ cd /path/to/git/src $ git bisect start v2.11.0-rc0 v2.10.2 $ git bisect run /path/to/bisect.test.sh (my original version of the script also includes some other makefile parameters to get a modern version of gettext and openssl too, but they're not relevant to this ML). I'm not certain that I have enough context to propose a meaningful patch though :/. Cheers, Robbie