On Tue, Oct 31, 2023 at 5:49 AM Taylor Blau <me@xxxxxxxxxxxx> wrote:
>
> On Tue, Oct 17, 2023 at 07:37:46PM +0800, Hongyi Zhao wrote:
> > I want to calculate a certain developer's contribution based on
> > different standards of code line count and the importance of the code.
>
> I agree with brian that "number of lines added/removed" is not a perfect
> measure of productivity ;-).
>
> But I think that there is a slightly cleaner way to compute the result
> you're after, like so:
>
> git rev-list --author="$who" origin/main |
> git diff-tree --stdin -r --numstat --no-commit-id |
> awk '{ s += $1 + $2 } END { print s }'
See below:
werner@X10DAi:~/Public/repo/github.com/OrderN/CONQUEST-release.git$
git log --author="JianboLin" --stat --summary origin/f-mlff | awk 'NF
==4 && $2 =="|" && $3 ~/[0-9]+/ && $4 ~/[+-]+|[+]+|[-]+/ {s+=$3} END
{print s}'
8519
werner@X10DAi:~/Public/repo/github.com/OrderN/CONQUEST-release.git$
git rev-list --author="JianboLin" origin/f-mlff | git diff-tree
--stdin -r --numstat --no-commit-id | awk '{ s += $1 + $2 } END {
print s }'
8519
So, your method and my original one give exactly the same result.
Therefore, I can't see what their fundamental difference is.
> Thanks,
> Taylor
Regards,
Zhao
