Gnus development mailing list
 help / color / mirror / Atom feed
From: Steinar Bang <sb@dod.no>
To: ding@gnus.org
Subject: TIP: using git-bisect (Was: nnimap no workee once again)
Date: Tue, 03 Jan 2012 08:39:45 +0100	[thread overview]
Message-ID: <8739bxqloe.fsf_-_@dod.no> (raw)
In-Reply-To: <87d3b1sr1i.fsf@newsguy.com>

>>>>> Harry Putnam <reader@newsguy.com>:
> Steinar Bang <sb@dod.no> writes:
>> Have you tried using git bisect to narrow it down to a specific commit?

> I've never bisected a git... I Disected a few frogs once : )

Googling for git-bisect should give you the docs
 http://book.git-scm.com/5_finding_issues_-_git_bisect.html

> Honestly I have no idea what you are talking about... but it did get
> me thinking to update gnus and see if the problem persists.

Ok.  Sounds like the not working didn't come because of a "git pull"...?
If so, it's a good idea to say so.  It narrows down the possibilities.

Using git bisect applies when a "git pull" leaves you with a broken
gnus.  What I do in that case, is:
 1. Find the SHA1 from before the last pull (the contents of
    .git/ORIG_HEAD), this will be "good"
 2. Start the command
     git bisect start bad good
    where bad is a sha1 hash, or tag, that you know is bad, eg. HEAD and
    good is a sha1 hash, or tag, that you know is good.  If I've done
    nothing except pulling from upstream in my repo, and the previous
    version was ok, I can actually do this
     git bisect start HEAD ORIG_HEAD
    then you loop, doing this:
     a. do "make clean" followed by "make" (or "./configure && make" if
        that's your preference)
     b. start a new emacs and start gnus and see if the problem persist.
        If the problem persists, do
         git bisect bad
        If the problem isn't there, do
         git bisect good
        If "git bisect" tells me that it has found the first offending
        commit, then go to 3, else go to a.
 3. Copy the text from "git bisect" where it found the offending commit,
    and send it to the gnus group
 4. Use "git bisect reset" to set the working directory back to normal,
    then decide if I want to try to fix the issue yourself, or pull
    back to the last good version, until it can be fixed, and then "make
    clean" and "make" gnus


 




  reply	other threads:[~2012-01-03  7:39 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-01-02 15:28 nnimap no workee once again Harry Putnam
2012-01-02 16:03 ` Steinar Bang
2012-01-02 22:00   ` Harry Putnam
2012-01-03  7:39     ` Steinar Bang [this message]
2012-01-03 20:10       ` TIP: using git-bisect Harry Putnam
2012-01-02 16:27 ` nnimap no workee once again Tassilo Horn
2012-01-07  1:43 ` Ted Zlatanov
2012-01-09 19:13   ` Harry Putnam
2012-01-10 14:50     ` Ted Zlatanov
2012-01-11  3:16       ` Harry Putnam
2012-01-11  3:32         ` Harry Putnam

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8739bxqloe.fsf_-_@dod.no \
    --to=sb@dod.no \
    --cc=ding@gnus.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).