Re: Antw: [EXT] Re: git add --interactive patch improvement for split hunks

Jeff King [off-list ref] writes:

    It looks like Junio carried it in "What's Cooking" for almost a
    year, marked as "waiting for re-roll" to handle the squash, but then
    eventually discarded it as stale. :(

Heh, thanks for digging.

Is the moral of the story that we should merge down unfinished
topics more aggressively (hoping that the untied loose ends would be
tied after they hit released version), we should prod owners of
stalled topics with sharper stick more often, or something else?

On Tue, Jun 29, 2021 at 11:09:19PM -0700, Junio C Hamano wrote:

Jeff King [off-list ref] writes:

quoted

    It looks like Junio carried it in "What's Cooking" for almost a
    year, marked as "waiting for re-roll" to handle the squash, but then
    eventually discarded it as stale. :(

Heh, thanks for digging.

Is the moral of the story that we should merge down unfinished
topics more aggressively (hoping that the untied loose ends would be
tied after they hit released version), we should prod owners of
stalled topics with sharper stick more often, or something else?

I'm not sure. I think the topic would have graduated if either you had
just applied the squash and merged it down, or if the original author
had checked back in over the intervening year to say "hey, what happened
to my patch" (either by reading "what's cooking" or manually).

I suspect drive-by contributors might not realize they need to do the
latter in some cases, but I wouldn't have counted 2014-era Ævar in that
boat. So I dunno.

-Peff

On Wed, Jun 30 2021, Jeff King wrote:

On Tue, Jun 29, 2021 at 11:09:19PM -0700, Junio C Hamano wrote:

quoted

Jeff King [off-list ref] writes:

quoted

    It looks like Junio carried it in "What's Cooking" for almost a
    year, marked as "waiting for re-roll" to handle the squash, but then
    eventually discarded it as stale. :(

Heh, thanks for digging.

Is the moral of the story that we should merge down unfinished
topics more aggressively (hoping that the untied loose ends would be
tied after they hit released version), we should prod owners of
stalled topics with sharper stick more often, or something else?

I'm not sure. I think the topic would have graduated if either you had
just applied the squash and merged it down, or if the original author
had checked back in over the intervening year to say "hey, what happened
to my patch" (either by reading "what's cooking" or manually).

I suspect drive-by contributors might not realize they need to do the
latter in some cases, but I wouldn't have counted 2014-era Ævar in that
boat. So I dunno.

Or maybe the moral of the story that it's a net addition of complexity
to git-add--interactive.perl. If I didn't care enough to remember or
notice the issue again maybe it wasn't all that important to begin with.

Likewise when it got ejected nobody else seemed to notice/care enough to
say "hey I liked that feature" & to pick it up.

I'd entirely forgotten I wrote that. Now that I'm reminded of it I don't
care enough myself to rebase it, test it again, and especially not to
figure out if/how it's going to interact with the new C implementation /
add and adjust a test for the two.

But maybe someone else will, it would be neat if someone has more of an
itch from the lack of that feature & wants to pick it up.

On Wed, Jun 30, 2021 at 10:27:16AM +0200, Ævar Arnfjörð Bjarmason wrote:

quoted

I'm not sure. I think the topic would have graduated if either you had
just applied the squash and merged it down, or if the original author
had checked back in over the intervening year to say "hey, what happened
to my patch" (either by reading "what's cooking" or manually).

I suspect drive-by contributors might not realize they need to do the
latter in some cases, but I wouldn't have counted 2014-era Ævar in that
boat. So I dunno.

Or maybe the moral of the story that it's a net addition of complexity
to git-add--interactive.perl. If I didn't care enough to remember or
notice the issue again maybe it wasn't all that important to begin with.

Likewise when it got ejected nobody else seemed to notice/care enough to
say "hey I liked that feature" & to pick it up.

Yeah, that's probably a fair interpretation, too. :)

I'd entirely forgotten I wrote that. Now that I'm reminded of it I don't
care enough myself to rebase it, test it again, and especially not to
figure out if/how it's going to interact with the new C implementation /
add and adjust a test for the two.

But maybe someone else will, it would be neat if someone has more of an
itch from the lack of that feature & wants to pick it up.

I can probably save you a little time/mental energy here: the C version
already does what your patch was trying to do. Once we switch to it as
the default, your patch would be obsolete anyway. :)

-Peff