git@vger.kernel.org list mirror (unofficial, one of many)
 help / color / mirror / code / Atom feed
From: Joachim Kuebart <joachim.kuebart@gmail.com>
To: Junio C Hamano <gitster@pobox.com>
Cc: Joachim Kuebart via GitGitGadget <gitgitgadget@gmail.com>,
	git@vger.kernel.org, Luke Diamand <luke@diamand.org>
Subject: Re: [PATCH] git-p4: speed up search for branch parent
Date: Thu, 29 Apr 2021 09:48:34 +0200	[thread overview]
Message-ID: <CAJGkkrQJFaLPfCBTVn6k1v9cCwF4wEUxr+ZyzebUBQJB8qLaWg@mail.gmail.com> (raw)
In-Reply-To: <xmqq5z05akyf.fsf@gitster.g>

On Thu, 29 Apr 2021 at 04:22, Junio C Hamano <gitster@pobox.com> wrote:
>
> "Joachim Kuebart via GitGitGadget" <gitgitgadget@gmail.com> writes:
>
> > From: Joachim Kuebart <joachim.kuebart@gmail.com>
>
> Thanks.  As git-p4 is not in my area of expertise, I'll make a style
> critique first, while pinging Luke as an area expert (you can learn
> who they are with "git shortlog --no-merges --since=18.months.ago
> git-p4.py").

Hi Junio, thanks for your timely and thorough review and for putting
up with my greenhorn mistakes ;-)

> > Previously, the code iterated through the parent branch commits and
> > compared each one to the target tree using diff-tree.
>
> It is customary in this project to describe the problem in the
> present tense.  In other words, whoever is writing the log message
> still lives in the world without this patch applied to the system.

I will rephrase the commit message and give better details as you
mentioned. Thanks a lot for your suggestions!

> When making a claim on performance, it is helpful to our readers to
> give some numbers, even in a limited test, e.g.
>
>     In a sample history where ~100 commits needed to be traversed to
>     find the fork point on my Windows box, the current code took
>     10.4 seconds to complete, while the new code yields the same
>     result in 1.8 seconds, which is a significant speed-up.
>
> or something along these lines.

I will add some measurements.

> > Signed-off-by: Joachim Kuebart <joachim.kuebart@gmail.com>
> >  git-p4.py | 22 +++++++++++-----------
> >  1 file changed, 11 insertions(+), 11 deletions(-)
> >
> > diff --git a/git-p4.py b/git-p4.py
> > index 09c9e93ac401..dbe94e6fb83b 100755
> > --- a/git-p4.py
> > +++ b/git-p4.py
> > @@ -3600,19 +3600,19 @@ def importNewBranch(self, branch, maxChange):
> >          return True
> >
> >      def searchParent(self, parent, branch, target):
> > +        for tree in read_pipe_lines(["git", "rev-parse",
> > +                                     "{}^{{tree}}".format(target)]):
> > +            targetTree = tree.strip()
>
> It looks very strange to run a commit that you expect a single line
> of output, and read the result in a loop.  Doesn't git-p4.py supply
> a more suitable helper to read a single line output from a command?

You're absolutely right that this isn't very readable. I had a quick
look around for a function that reads a single-line response, but I'll
look again and come up with a clearer solution.

> > +        for blob in read_pipe_lines(["git", "rev-list", "--format=%H %T",
> >                                       "--no-merges", parent]):
>
> This is not a new problem you introduced, but when we hear word
> "blob" in the context of this project, it reminds us of the "blob"
> object, while the 'blob' variable used in this loop has nothing to
> do with it.  Perhaps rename it to say 'line' or something?

Will do, thanks!

> > +            if blob[:7] == "commit ":
> > +                continue
>
> Perhaps blob.startswith("commit ") to avoid hardcoded string length?

Yes, that's the name of the function that I can never think of when I need it.

Thanks again for your comments,

Joachim

  reply	other threads:[~2021-04-29  7:49 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-28 20:06 Joachim Kuebart via GitGitGadget
2021-04-29  2:22 ` Junio C Hamano
2021-04-29  7:48   ` Joachim Kuebart [this message]
2021-04-29  8:22     ` Luke Diamand
2021-04-29  8:31       ` Junio C Hamano
2021-04-29 19:31         ` Joachim Kuebart
2021-04-29 11:30       ` Joachim Kuebart
2021-05-05 11:56 ` [PATCH v2 0/2] " Joachim Kuebart via GitGitGadget
2021-05-05 11:56   ` [PATCH v2 1/2] git-p4: ensure complex branches are cloned correctly Joachim Kuebart via GitGitGadget
2021-05-05 11:56   ` [PATCH v2 2/2] git-p4: speed up search for branch parent Joachim Kuebart via GitGitGadget

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

  List information: http://vger.kernel.org/majordomo-info.html

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

  git send-email \
    --in-reply-to=CAJGkkrQJFaLPfCBTVn6k1v9cCwF4wEUxr+ZyzebUBQJB8qLaWg@mail.gmail.com \
    --to=joachim.kuebart@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=gitgitgadget@gmail.com \
    --cc=gitster@pobox.com \
    --cc=luke@diamand.org \
    --subject='Re: [PATCH] git-p4: speed up search for branch parent' \
    /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

Code repositories for project(s) associated with this inbox:

	https://80x24.org/mirrors/git.git

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 read-only IMAP folder(s) and NNTP newsgroup(s).