Skip to content

Accommodate for pu having been renamed to seen #668

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Conversation

dscho
Copy link
Member

@dscho dscho commented Jun 23, 2020

This patch series adjusts Git's own source code to reflect that change.

Please note that even with these patches, there are still a couple places where pu is used:

  • In the translations. These are legitimate words in languages that are not English (as in "gpg n'a pas pu signer les données" where "pu" is French for the English "could").
  • In upload-pack.c, where a variable named pu is short form for "pack-objects updates".

Changes since v2:

  • One accidental quoting change in v1 was reverted.

  • Rebased onto maint (no merge conflicts, so it does not actually change anything).

Changes since v1:

Cc: Denton Liu liu.denton@gmail.com

@dscho
Copy link
Member Author

dscho commented Jun 23, 2020

/submit

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 23, 2020

Error: git fetch https://github.com/gitgitgadget/git -- +refs/notes/gitgitgadget:refs/notes/gitgitgadget +refs/heads/maint:refs/remotes/upstream/maint +refs/heads/master:refs/remotes/upstream/master +refs/heads/next:refs/remotes/upstream/next +refs/heads/pu:refs/remotes/upstream/pu +refs/pull/668/head:refs/pull/668/head +refs/pull/668/merge:refs/pull/668/merge failed: 128,
fatal: couldn't find remote ref refs/heads/pu

@dscho
Copy link
Member Author

dscho commented Jun 23, 2020

/submit

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 23, 2020

@@ -1179,8 +1179,8 @@ look at the section below this one for some context.)
[[after-approval]]
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

On the Git mailing list, Đoàn Trần Công Danh wrote (reply to this):

Hi Dscho,

On 2020-06-23 15:04:13+0000, Johannes Schindelin via GitGitGadget <gitgitgadget@gmail.com> wrote:
> diff --git a/Documentation/git-ls-remote.txt b/Documentation/git-ls-remote.txt
> index 0a5c8b7d493..492e573856f 100644
> --- a/Documentation/git-ls-remote.txt
> +++ b/Documentation/git-ls-remote.txt
> @@ -101,9 +101,9 @@ f25a265a342aed6041ab0cc484224d9ca54b6f41	refs/tags/v0.99.1
>  7ceca275d047c90c0c7d5afb13ab97efdf51bd6e	refs/tags/v0.99.3
>  c5db5456ae3b0873fc659c19fafdde22313cc441	refs/tags/v0.99.2
>  0918385dbd9656cab0d1d81ba7453d49bbc16250	refs/tags/junio-gpg-pub
> -$ git ls-remote http://www.kernel.org/pub/scm/git/git.git master pu rc
> +$ git ls-remote http://www.kernel.org/pub/scm/git/git.git master seen rc

rc is not with us anymore.

Should we replace it with next, too?

>  5fe978a5381f1fbad26a80e682ddd2a401966740	refs/heads/master
> -c781a84b5204fb294c9ccc79f8b3baceeb32c061	refs/heads/pu
> +c781a84b5204fb294c9ccc79f8b3baceeb32c061	refs/heads/seen
>  $ git remote add korg http://www.kernel.org/pub/scm/git/git.git
>  $ git ls-remote --tags korg v\*
>  d6602ec5194c87b0fc87103ca4d67251c76f233a	refs/tags/v0.99

-- 
Danh

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

On the Git mailing list, Junio C Hamano wrote (reply to this):

Đoàn Trần Công Danh  <congdanhqx@gmail.com> writes:

> Hi Dscho,
>
> On 2020-06-23 15:04:13+0000, Johannes Schindelin via GitGitGadget <gitgitgadget@gmail.com> wrote:
>> diff --git a/Documentation/git-ls-remote.txt b/Documentation/git-ls-remote.txt
>> index 0a5c8b7d493..492e573856f 100644
>> --- a/Documentation/git-ls-remote.txt
>> +++ b/Documentation/git-ls-remote.txt
>> @@ -101,9 +101,9 @@ f25a265a342aed6041ab0cc484224d9ca54b6f41	refs/tags/v0.99.1
>>  7ceca275d047c90c0c7d5afb13ab97efdf51bd6e	refs/tags/v0.99.3
>>  c5db5456ae3b0873fc659c19fafdde22313cc441	refs/tags/v0.99.2
>>  0918385dbd9656cab0d1d81ba7453d49bbc16250	refs/tags/junio-gpg-pub
>> -$ git ls-remote http://www.kernel.org/pub/scm/git/git.git master pu rc
>> +$ git ls-remote http://www.kernel.org/pub/scm/git/git.git master seen rc
>
> rc is not with us anymore.
>
> Should we replace it with next, too?

I do not think so.  I think we never had 'rc'.

I think what the above example is demonstrating is this.

    SYNOPSIS calls the last command line arguments <refs>; they are
    actually mere patterns (which is how these command line
    arguments are described in the documentation).  It is *not* an
    error if no refs match a particular pattern.

And because we have no refs that match the pattern "rc", we only see
"master" and "pu" (now "seen") from the command.

I see a couple of possible improvements here:

 - The "<refs>...::" documentation should explain what kind of
   pattern match is performed here.  I recall these originally were
   just tail matches, but the rule might have been made more
   flexible over time.

 - The example should first explain the setting.  The first sample
   depends on the current (./.) repository having these tags or it
   would not work (showing the sample upfront and explaining the
   outcome shown in the sample would work well in this case,
   e.g. "we can see that in the current repository, there are tags
   X, Y and Z").  The second one at least needs to say two things:
   the sample repository does not have a branch called 'rc' and that
   is why it is not shown, and it is not an error for patterns to
   produce no match.

Thanks.

>
>>  5fe978a5381f1fbad26a80e682ddd2a401966740	refs/heads/master
>> -c781a84b5204fb294c9ccc79f8b3baceeb32c061	refs/heads/pu
>> +c781a84b5204fb294c9ccc79f8b3baceeb32c061	refs/heads/seen
>>  $ git remote add korg http://www.kernel.org/pub/scm/git/git.git
>>  $ git ls-remote --tags korg v\*
>>  d6602ec5194c87b0fc87103ca4d67251c76f233a	refs/tags/v0.99

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

On the Git mailing list, Johannes Schindelin wrote (reply to this):

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.

--8323328-1867953520-1592947936=:54
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Junio & Danh,

On Tue, 23 Jun 2020, Junio C Hamano wrote:

> =C4=90o=C3=A0n Tr=E1=BA=A7n C=C3=B4ng Danh  <congdanhqx@gmail.com> write=
s:
>
> > On 2020-06-23 15:04:13+0000, Johannes Schindelin via GitGitGadget <git=
gitgadget@gmail.com> wrote:
> >> diff --git a/Documentation/git-ls-remote.txt b/Documentation/git-ls-r=
emote.txt
> >> index 0a5c8b7d493..492e573856f 100644
> >> --- a/Documentation/git-ls-remote.txt
> >> +++ b/Documentation/git-ls-remote.txt
> >> @@ -101,9 +101,9 @@ f25a265a342aed6041ab0cc484224d9ca54b6f41	refs/tag=
s/v0.99.1
> >>  7ceca275d047c90c0c7d5afb13ab97efdf51bd6e	refs/tags/v0.99.3
> >>  c5db5456ae3b0873fc659c19fafdde22313cc441	refs/tags/v0.99.2
> >>  0918385dbd9656cab0d1d81ba7453d49bbc16250	refs/tags/junio-gpg-pub
> >> -$ git ls-remote http://www.kernel.org/pub/scm/git/git.git master pu =
rc
> >> +$ git ls-remote http://www.kernel.org/pub/scm/git/git.git master see=
n rc
> >
> > rc is not with us anymore.
> >
> > Should we replace it with next, too?
>
> I do not think so.  I think we never had 'rc'.

Indeed, and the context given in the patch demonstrates that no `rc` is
shown, so I assumed the same things as Junio explained here:

> I think what the above example is demonstrating is this.
>
>     SYNOPSIS calls the last command line arguments <refs>; they are
>     actually mere patterns (which is how these command line
>     arguments are described in the documentation).  It is *not* an
>     error if no refs match a particular pattern.
>
> And because we have no refs that match the pattern "rc", we only see
> "master" and "pu" (now "seen") from the command.

Precisely.

> I see a couple of possible improvements here:
>
>  - The "<refs>...::" documentation should explain what kind of
>    pattern match is performed here.  I recall these originally were
>    just tail matches, but the rule might have been made more
>    flexible over time.
>
>  - The example should first explain the setting.  The first sample
>    depends on the current (./.) repository having these tags or it
>    would not work (showing the sample upfront and explaining the
>    outcome shown in the sample would work well in this case,
>    e.g. "we can see that in the current repository, there are tags
>    X, Y and Z").  The second one at least needs to say two things:
>    the sample repository does not have a branch called 'rc' and that
>    is why it is not shown, and it is not an error for patterns to
>    produce no match.

Those sound like wonderful #leftoverbits to me.

Thank you,
Dscho

>
> Thanks.
>
> >
> >>  5fe978a5381f1fbad26a80e682ddd2a401966740	refs/heads/master
> >> -c781a84b5204fb294c9ccc79f8b3baceeb32c061	refs/heads/pu
> >> +c781a84b5204fb294c9ccc79f8b3baceeb32c061	refs/heads/seen
> >>  $ git remote add korg http://www.kernel.org/pub/scm/git/git.git
> >>  $ git ls-remote --tags korg v\*
> >>  d6602ec5194c87b0fc87103ca4d67251c76f233a	refs/tags/v0.99
>

--8323328-1867953520-1592947936=:54--

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

On the Git mailing list, Đoàn Trần Công Danh wrote (reply to this):

On 2020-06-23 12:31:45-0700, Junio C Hamano <gitster@pobox.com> wrote:
> Đoàn Trần Công Danh  <congdanhqx@gmail.com> writes:
> > On 2020-06-23 15:04:13+0000, Johannes Schindelin via GitGitGadget <gitgitgadget@gmail.com> wrote:
> >> diff --git a/Documentation/git-ls-remote.txt b/Documentation/git-ls-remote.txt
> >> index 0a5c8b7d493..492e573856f 100644
> >> --- a/Documentation/git-ls-remote.txt
> >> +++ b/Documentation/git-ls-remote.txt
> >> @@ -101,9 +101,9 @@ f25a265a342aed6041ab0cc484224d9ca54b6f41	refs/tags/v0.99.1
> >>  7ceca275d047c90c0c7d5afb13ab97efdf51bd6e	refs/tags/v0.99.3
> >>  c5db5456ae3b0873fc659c19fafdde22313cc441	refs/tags/v0.99.2
> >>  0918385dbd9656cab0d1d81ba7453d49bbc16250	refs/tags/junio-gpg-pub
> >> -$ git ls-remote http://www.kernel.org/pub/scm/git/git.git master pu rc
> >> +$ git ls-remote http://www.kernel.org/pub/scm/git/git.git master seen rc
> >
> > rc is not with us anymore.
> >
> > Should we replace it with next, too?
> 
> I do not think so.  I think we never had 'rc'.

You're right.
Actually, I didn't read the context after the diff but went to
215a7ad1ef (Big tool rename., 2005-09-07) instead.

In that tree, there's a line for refs/heads/rc:

	b1d096f2926c4e37c9c0b6a7bf2119bedaa277cb        refs/heads/rc

Not sure why it was there in the past but it's fixed in
6077d36299 (ls-remote doc: fix example invocation on git.git,
2013-06-22)


> I think what the above example is demonstrating is this.
> 
>     SYNOPSIS calls the last command line arguments <refs>; they are
>     actually mere patterns (which is how these command line
>     arguments are described in the documentation).  It is *not* an
>     error if no refs match a particular pattern.
> 
> And because we have no refs that match the pattern "rc", we only see
> "master" and "pu" (now "seen") from the command.
> 
> I see a couple of possible improvements here:
> 
>  - The "<refs>...::" documentation should explain what kind of
>    pattern match is performed here.  I recall these originally were
>    just tail matches, but the rule might have been made more
>    flexible over time.
> 
>  - The example should first explain the setting.  The first sample
>    depends on the current (./.) repository having these tags or it
>    would not work (showing the sample upfront and explaining the
>    outcome shown in the sample would work well in this case,
>    e.g. "we can see that in the current repository, there are tags
>    X, Y and Z").  The second one at least needs to say two things:
>    the sample repository does not have a branch called 'rc' and that
>    is why it is not shown, and it is not an error for patterns to
>    produce no match.

I guess this is a leftover from 6077d36299 (ls-remote doc: fix example
invocation on git.git, 2013-06-22), if this was documented together
with said commit, it'll be less confusing.

> 
> Thanks.
> 
> >
> >>  5fe978a5381f1fbad26a80e682ddd2a401966740	refs/heads/master
> >> -c781a84b5204fb294c9ccc79f8b3baceeb32c061	refs/heads/pu
> >> +c781a84b5204fb294c9ccc79f8b3baceeb32c061	refs/heads/seen
> >>  $ git remote add korg http://www.kernel.org/pub/scm/git/git.git
> >>  $ git ls-remote --tags korg v\*
> >>  d6602ec5194c87b0fc87103ca4d67251c76f233a	refs/tags/v0.99

-- 
Danh

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

This branch is now known as js/pu-to-seen.

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

This patch series was integrated into seen via git@d4243b4.

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

This patch series was integrated into seen via git@c0f4b66.

@dscho dscho force-pushed the accommodate-for-pu-having-been-renamed-to-seen branch from 17adbd5 to c8e356c Compare June 24, 2020 14:18
@dscho dscho changed the base branch from seen to master June 24, 2020 14:21
@dscho
Copy link
Member Author

dscho commented Jun 24, 2020

/submit

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

@@ -1179,8 +1179,8 @@ look at the section below this one for some context.)
[[after-approval]]
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

On the Git mailing list, Denton Liu wrote (reply to this):

Hi Dscho,

On Wed, Jun 24, 2020 at 02:48:38PM +0000, Johannes Schindelin via GitGitGadget wrote:
> diff --git a/Documentation/gitworkflows.txt b/Documentation/gitworkflows.txt
> index abc0dc6bc7..0965b60884 100644
> --- a/Documentation/gitworkflows.txt
> +++ b/Documentation/gitworkflows.txt
> @@ -85,15 +85,15 @@ As a given feature goes from experimental to stable, it also
>  
>  There is a fourth official branch that is used slightly differently:
>  
> -* 'pu' (proposed updates) is an integration branch for things that are
> -  not quite ready for inclusion yet (see "Integration Branches"
> -  below).
> +* `seen` (patches seen by the maintainer) is an integration branch for
> +  things that are not quite ready for inclusion yet (see "Integration
> +  Branches" below).

Tiny nit: we should use sq instead of backticks to match the style of
the rest of the document.

-Denton

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

On the Git mailing list, Denton Liu wrote (reply to this):

Hi Dscho,

On Wed, Jun 24, 2020 at 02:48:37PM +0000, Johannes Schindelin via GitGitGadget wrote:
> Changes since v1:
> 
>  * Rebased onto master (no conflicts, so it is safe, and it is more robust
>    than basing the patches on seen which already contains v1 of these
>    patches).

Out of curiosity, why would we ever want to base any patches on `seen`?
I understand there are some very rare edge-cases where it might be
appropriate to base patches on `next` (where the series based on
many, many other topics) but I'm not sure if I know of any situations
where it'd make sense to base topics on `seen`.

Thanks,

Denton

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

On the Git mailing list, Junio C Hamano wrote (reply to this):

"Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com>
writes:

> Changes since v1:
>
>  * Rebased onto master (no conflicts, so it is safe, and it is more robust
>    than basing the patches on seen which already contains v1 of these
>    patches).

Thanks, I actually wanted to include it in 'maint', so I'll queue on
the same base (no conflicts, so it is safe, and it will be in a
maintenance release if we are going to issue one).

>  * Adjusted the quoting to match 
>    https://lore.kernel.org/git/e250f1bb100aca94c914f1b2d38a3849c2566aea.1592909867.git.liu.denton@gmail.com/

I know I mentioned it and I think the patch to SubmittingPatches
does improve by doing `seen` because it matches the way how the
nearby `git pull --rebase` is quoted.

But I am not sure about the patch to gitworkflows.txt, where the
text around the new `seen` mention 'master' and 'next'.  I think
your v1 was more (locally) consistent.

I am on the fence to the change to giteveryday.txt, where `pu` got
changed to `seen`; your v1 had "(patches seen by the maintainer)" as
an explanation after the `seen`.  I guess it is inconsistent to
explain only why `seen` is `seen` without doing the same for `next`,
so I would say v2 is an improvement over v1.

In short,

>  1:  dc6f971290 ! 1:  35e3dafd6a docs: adjust for the recent rename of `pu` to `seen`
>      @@ Documentation/SubmittingPatches: their trees themselves.
>          patches, and will let you know. This works only if you rebase on top
>          of the branch in which your patch has been merged (i.e. it will not
>       -  tell you if your patch is merged in pu if you rebase on top of
>      -+  tell you if your patch is merged in 'seen' if you rebase on top of
>      ++  tell you if your patch is merged in `seen` if you rebase on top of
>          master).

Good.

>        * Read the Git mailing list, the maintainer regularly posts messages
>      @@ Documentation/giteveryday.txt: $ git push --follow-tags ko <13>
>        <2> see which branches haven't been merged into `master` yet.
>        Likewise for any other integration branches e.g. `maint`, `next`
>       -and `pu` (potential updates).
>      -+and `seen` (patches seen by the maintainer).
>      ++and `seen`.

Probably good.

>        <3> read mails, save ones that are applicable, and save others
>        that are not quite ready (other mail readers are available).
>        <4> apply them, interactively, with your sign-offs.
>      @@ Documentation/gitworkflows.txt: As a given feature goes from experimental to sta
>       -* 'pu' (proposed updates) is an integration branch for things that are
>       -  not quite ready for inclusion yet (see "Integration Branches"
>       -  below).
>      -+* 'seen' (patches seen by the maintainer) is an integration branch for
>      ++* `seen` (patches seen by the maintainer) is an integration branch for

Not---'seen' was more consistent relative to the surrounding text.

Thanks.

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

On the Git mailing list, Junio C Hamano wrote (reply to this):

Denton Liu <liu.denton@gmail.com> writes:

> Hi Dscho,
>
> On Wed, Jun 24, 2020 at 02:48:37PM +0000, Johannes Schindelin via GitGitGadget wrote:
>> Changes since v1:
>> 
>>  * Rebased onto master (no conflicts, so it is safe, and it is more robust
>>    than basing the patches on seen which already contains v1 of these
>>    patches).
>
> Out of curiosity, why would we ever want to base any patches on `seen`?

Never.  Even bulding on top of 'next' is discouraged.  

Either "prepare a merge on top of 'master' with all the topics in
flight that you depend on, and base your series on top of it,
risking that any one of these topics can take your series hostage"
or "wait until these topics graduate and then base your topic on
'master'".  I'd vastly prefer the latter, as it would become
cumbersome if one of the topics you base your series on gets
rerolled.



@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

On the Git mailing list, Junio C Hamano wrote (reply to this):

Junio C Hamano <gitster@pobox.com> writes:

> "Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com>
> writes:
>
>> Changes since v1:
>>
>>  * Rebased onto master (no conflicts, so it is safe, and it is more robust
>>    than basing the patches on seen which already contains v1 of these
>>    patches).
>
> Thanks, I actually wanted to include it in 'maint', so I'll queue on
> the same base (no conflicts, so it is safe, and it will be in a
> maintenance release if we are going to issue one).

By the way, I find myself typing 'pu' all the time, even though I've
been using 'seen' for almost 48 hours by now.  My private tooling
all have been updated to work with 'seen', but it seems that it
takes time to retrain muscle memory.  I'll see if I can fully adjust
before the next week starts.

I do not know how many of you regularly have interacted with 'pu'
and now need to go through the same adjustment as I do.  Sorry for
using you as a guinea pig for an experiment for you know what to
gauge the cost.

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

On the Git mailing list, Denton Liu wrote (reply to this):

On Wed, Jun 24, 2020 at 10:05:43AM -0700, Junio C Hamano wrote:
> I do not know how many of you regularly have interacted with 'pu'
> and now need to go through the same adjustment as I do.  Sorry for
> using you as a guinea pig for an experiment for you know what to
> gauge the cost.

Heh, I was wondering if you had any ulterior motives ;)

Since we're on the topic of the cost of renaming branches, I was reading
a reply from you back in 2011 about how HEAD symrefs are the only valid
ones[0]. I'm not sure if the situation has changed since then but
perhaps we could officially expand the scope of symrefs to allow users
to essentially alias branches? It might reduce the cost of performing
branch renames by having a backwards compatible option.

[0]: https://lore.kernel.org/git/7vsjvpq0jk.fsf@alter.siamese.dyndns.org/

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

On the Git mailing list, Junio C Hamano wrote (reply to this):

Denton Liu <liu.denton@gmail.com> writes:

> On Wed, Jun 24, 2020 at 10:05:43AM -0700, Junio C Hamano wrote:
>> I do not know how many of you regularly have interacted with 'pu'
>> and now need to go through the same adjustment as I do.  Sorry for
>> using you as a guinea pig for an experiment for you know what to
>> gauge the cost.
>
> Heh, I was wondering if you had any ulterior motives ;)
>
> Since we're on the topic of the cost of renaming branches, I was reading
> a reply from you back in 2011 about how HEAD symrefs are the only valid
> ones[0]. I'm not sure if the situation has changed since then but
> perhaps we could officially expand the scope of symrefs to allow users
> to essentially alias branches? It might reduce the cost of performing
> branch renames by having a backwards compatible option.

It would be one way to transition, adding a symref in refs/heads/pu
pointing at refs/heads/seen, but that unfortunately defeats the
whole point of the rename, to make room for pu/<topic> hierarchy for
contributors with names, in which P and U appear as the first and
the last capital letters, respectively.

So, no, that won't be a solution, unfortunately.

I have an unused branch 'pu/nomore' in the primary repository I work
in, so that my accidental "git checkout -B pu jch" will fail, which
also takes advantage of this D/F conflict preventing a ref from
being created.



@gitgitgadget
Copy link

gitgitgadget bot commented Jun 24, 2020

This patch series was integrated into seen via git@61ad2ba.

dscho added 3 commits June 25, 2020 13:54
As of "What's cooking in git.git (Jun 2020, #4; Mon, 22)", there is no
longer any `pu` branch, but a `seen` branch.

While we technically do not even need to update the manual pages, it
makes sense to update them because they clearly talk about branches in
git.git.

Please note that in two instances, this patch not only updates the
branch name, but also the description "(proposed updates)".

Where appropriate, quotes have been added for readability.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
This patch tries to rewrite history a bit: the mail contents that have
been added to Git's source code are actually fixed, we cannot change
them in hindsight.

But as the `pu` branch _was_ renamed, and as the documents were added to
Git's source code not so much as historical record, but to describe the
status quo, let's pretend that we have a time machine and adjust the
provided information accordingly.

Where appropriate, quotes were added for readability.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
As our test suite partially reflects how we work in the Git project, it
is natural that the branch name `pu` was used in a couple places.

Since that branch was renamed to `seen`, let's use the new name
consistently.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
@dscho dscho force-pushed the accommodate-for-pu-having-been-renamed-to-seen branch from c8e356c to e38ade2 Compare June 25, 2020 11:54
@dscho dscho changed the base branch from master to maint June 25, 2020 11:55
@dscho
Copy link
Member Author

dscho commented Jun 25, 2020

/submit

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 25, 2020

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 25, 2020

On the Git mailing list, Junio C Hamano wrote (reply to this):

"Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com>
writes:

> This patch series adjusts Git's own source code to reflect that change.
>
> Please note that even with these patches, there are still a couple places
> where pu is used:
>
>  * In the translations. These are legitimate words in languages that are not
>    English (as in "gpg n'a pas pu signer les données" where "pu" is French
>    for the English "could").
>  * In upload-pack.c, where a variable named pu is short form for
>    "pack-objects updates".
>
> Changes since v2:
>
>  * One accidental quoting change in v1 was reverted.

Thanks for being thorough.  

You could have just told me that the fixup queued on 'seen' looks
good to you and squash it in the first step instead to save one
roundtrip, but replacing with a new set of three patches is not so
bad, either ;-)

Will replace.

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 25, 2020

This patch series was integrated into seen via git@acc59a7.

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 25, 2020

This patch series was integrated into next via git@06c9e5a.

@gitgitgadget gitgitgadget bot added the next label Jun 25, 2020
@gitgitgadget
Copy link

gitgitgadget bot commented Jun 26, 2020

On the Git mailing list, Kaartic Sivaraam wrote (reply to this):

On 25-06-2020 17:48, Johannes Schindelin via GitGitGadget wrote:
> This patch series adjusts Git's own source code to reflect that change.
> 
> Please note that even with these patches, there are still a couple places
> where pu is used:
> 

This reminds me. The ProGit book references the `pu` branch in several
places in the text and images IIRC. Is it fine to change them now?
Would it be premature?

>  * In the translations. These are legitimate words in languages that are not
>    English (as in "gpg n'a pas pu signer les données" where "pu" is French
>    for the English "could").
>  * In upload-pack.c, where a variable named pu is short form for
>    "pack-objects updates".

-- 
Sivaraam

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 29, 2020

This patch series was integrated into seen via git@9c496a2.

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 30, 2020

On the Git mailing list, Johannes Schindelin wrote (reply to this):

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.

--8323328-1420957276-1593438059=:56
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi,

On Wed, 24 Jun 2020, Junio C Hamano wrote:

> Denton Liu <liu.denton@gmail.com> writes:
>
> > On Wed, Jun 24, 2020 at 02:48:37PM +0000, Johannes Schindelin via GitG=
itGadget wrote:
> >> Changes since v1:
> >>
> >>  * Rebased onto master (no conflicts, so it is safe, and it is more r=
obust
> >>    than basing the patches on seen which already contains v1 of these
> >>    patches).
> >
> > Out of curiosity, why would we ever want to base any patches on `seen`=
?
>
> Never.  Even bulding on top of 'next' is discouraged.
>
> Either "prepare a merge on top of 'master' with all the topics in
> flight that you depend on, and base your series on top of it,
> risking that any one of these topics can take your series hostage"
> or "wait until these topics graduate and then base your topic on
> 'master'".  I'd vastly prefer the latter, as it would become
> cumbersome if one of the topics you base your series on gets
> rerolled.

I recall having had to base a patch on `seen` (n=C3=A9e `pu`) because it
would otherwise have needed "two base branches". In another instance, I
made a patch to fix incorrectly-resolved merge conflicts.

There _are_ occasions when you want to base your patch on `seen`,
admittedly not very common occasions.

Ciao,
Dscho

--8323328-1420957276-1593438059=:56--

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 30, 2020

On the Git mailing list, Johannes Schindelin wrote (reply to this):

Hi Junio,

On Wed, 24 Jun 2020, Junio C Hamano wrote:

> Junio C Hamano <gitster@pobox.com> writes:
>
> > "Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com>
> > writes:
> >
> >> Changes since v1:
> >>
> >>  * Rebased onto master (no conflicts, so it is safe, and it is more robust
> >>    than basing the patches on seen which already contains v1 of these
> >>    patches).
> >
> > Thanks, I actually wanted to include it in 'maint', so I'll queue on
> > the same base (no conflicts, so it is safe, and it will be in a
> > maintenance release if we are going to issue one).
>
> By the way, I find myself typing 'pu' all the time, even though I've
> been using 'seen' for almost 48 hours by now.  My private tooling
> all have been updated to work with 'seen', but it seems that it
> takes time to retrain muscle memory.  I'll see if I can fully adjust
> before the next week starts.

I do understand the issue. If you're as addicted to tab-completion as I
am, something like `pulse` might have made that transition.

Ciao,
Dscho

> I do not know how many of you regularly have interacted with 'pu'
> and now need to go through the same adjustment as I do.  Sorry for
> using you as a guinea pig for an experiment for you know what to
> gauge the cost.

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 30, 2020

On the Git mailing list, Johannes Schindelin wrote (reply to this):

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.

--8323328-1172079787-1593439636=:56
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

Hi Junio,

On Thu, 25 Jun 2020, Junio C Hamano wrote:

> "Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com>
> writes:
>
> > This patch series adjusts Git's own source code to reflect that change=
.
> >
> > Please note that even with these patches, there are still a couple pla=
ces
> > where pu is used:
> >
> >  * In the translations. These are legitimate words in languages that a=
re not
> >    English (as in "gpg n'a pas pu signer les donn=C3=A9es" where "pu" =
is French
> >    for the English "could").
> >  * In upload-pack.c, where a variable named pu is short form for
> >    "pack-objects updates".
> >
> > Changes since v2:
> >
> >  * One accidental quoting change in v1 was reverted.
>
> Thanks for being thorough.
>
> You could have just told me that the fixup queued on 'seen' looks
> good to you and squash it in the first step instead to save one
> roundtrip, but replacing with a new set of three patches is not so
> bad, either ;-)

To be honest, the GitGitGadget-based workflow makes it quicker for me to
just submit a new iteration. In fact, I did not even see your fixup until
I read your mail.

Ciao,
Dscho

>
> Will replace.
>
>

--8323328-1172079787-1593439636=:56--

@gitgitgadget
Copy link

gitgitgadget bot commented Jun 30, 2020

On the Git mailing list, Junio C Hamano wrote (reply to this):

Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:

>> You could have just told me that the fixup queued on 'seen' looks
>> good to you and squash it in the first step instead to save one
>> roundtrip, but replacing with a new set of three patches is not so
>> bad, either ;-)
>
> To be honest, the GitGitGadget-based workflow makes it quicker for me to
> just submit a new iteration.

I do not mind seeing a new iteration that gives easier time for
others to comment on the version that is closer to the final than
the previous round.  The offer was only for contributors who find
it easier to just say "yeah, I am happy with that change" than
submitting a new round.

> In fact, I did not even see your fixup until I read your mail.

This I actually would mind a bit more.  The reason why I publish
'seen' is to make it easier for authors of individual topics how
their work would play with other topics in flight, and it diminishes
the value of it if contributors do not pay attention to what is
queued there.  I expect contributors to fetch and look at what is
queued in origin/seen.

There may be evil merges that reveal subtle interactions between
topics, some of which may involve the topic an author may care
about.  There may be fixups for problems that were not found during
review but only found during the integration process.  I try to
communicate these back on the list when possible, but the thing is,
a day does not have sufficient number of minutes for me to always do
so.

Thanks.

@gitgitgadget
Copy link

gitgitgadget bot commented Jul 1, 2020

On the Git mailing list, Johannes Schindelin wrote (reply to this):

Hi Junio,

On Tue, 30 Jun 2020, Junio C Hamano wrote:

> Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:
>
> >> You could have just told me that the fixup queued on 'seen' looks
> >> good to you and squash it in the first step instead to save one
> >> roundtrip, but replacing with a new set of three patches is not so
> >> bad, either ;-)
> >
> > To be honest, the GitGitGadget-based workflow makes it quicker for me to
> > just submit a new iteration.
>
> I do not mind seeing a new iteration that gives easier time for
> others to comment on the version that is closer to the final than
> the previous round.  The offer was only for contributors who find
> it easier to just say "yeah, I am happy with that change" than
> submitting a new round.
>
> > In fact, I did not even see your fixup until I read your mail.
>
> This I actually would mind a bit more.  The reason why I publish
> 'seen' is to make it easier for authors of individual topics how
> their work would play with other topics in flight, and it diminishes
> the value of it if contributors do not pay attention to what is
> queued there.  I expect contributors to fetch and look at what is
> queued in origin/seen.
>
> There may be evil merges that reveal subtle interactions between
> topics, some of which may involve the topic an author may care
> about.  There may be fixups for problems that were not found during
> review but only found during the integration process.  I try to
> communicate these back on the list when possible, but the thing is,
> a day does not have sufficient number of minutes for me to always do
> so.

I understand. And I am trying my best to accommodate.

Ciao,
Dscho

>
> Thanks.
>

@gitgitgadget
Copy link

gitgitgadget bot commented Jul 2, 2020

This patch series was integrated into seen via git@e311cab.

@gitgitgadget
Copy link

gitgitgadget bot commented Jul 7, 2020

This patch series was integrated into seen via git@8a78e4d.

@gitgitgadget
Copy link

gitgitgadget bot commented Jul 7, 2020

This patch series was integrated into next via git@8a78e4d.

@gitgitgadget
Copy link

gitgitgadget bot commented Jul 7, 2020

This patch series was integrated into master via git@8a78e4d.

@gitgitgadget gitgitgadget bot added the master label Jul 7, 2020
@gitgitgadget gitgitgadget bot closed this Jul 7, 2020
@gitgitgadget
Copy link

gitgitgadget bot commented Jul 7, 2020

Closed via 8a78e4d.

@dscho dscho deleted the accommodate-for-pu-having-been-renamed-to-seen branch July 7, 2020 07:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant