threads / patch / 56568

patchdoc/technical: remove outdated MIDX default note

Subject: [PATCH] doc/technical: remove outdated MIDX default note

## tl;dr

9 messages between Sep 22, 2021 and Sep 25, 2021. Diffs are folded; open one to read it.

replies: 8people: 4as markdown or json

Eric Wong· Sep 22, 2021, 22:13 UTC · lore

Multi-pack-index files are used by default since commit 18e449f86b74bab35b150549c8342d252fe7ae00, so the outdated note was misleading.

Signed-off-by: Eric Wong <e@80x24.org>
---
 Documentation/technical/multi-pack-index.txt | 2 --
 1 file changed, 2 deletions(-)

Show changes to Documentation/technical/multi-pack-index.txt +0 −2

diff --git a/Documentation/technical/multi-pack-index.txt b/Documentation/technical/multi-pack-index.txt
index 1a73c3ee20..2038fb7995 100644
--- a/Documentation/technical/multi-pack-index.txt
+++ b/Documentation/technical/multi-pack-index.txt
@@ -36,8 +36,6 @@ Design Details
   directory of an alternate. It refers only to packfiles in that
   same directory.
 
-- The core.multiPackIndex config setting must be on to consume MIDX files.
-
 - The file format includes parameters for the object ID hash
   function, so a future change of hash algorithm does not require
   a change in format.

Derrick Stolee· Sep 23, 2021, 13:47 UTC · re: Eric Wong · lore

Re: [PATCH] doc/technical: remove outdated MIDX default note

On 9/22/2021 6:13 PM, Eric Wong wrote:

> Multi-pack-index files are used by default since
> commit 18e449f86b74bab35b150549c8342d252fe7ae00, so the
> outdated note was misleading.

You are correct that the note is misleading, but it is still correct.

A user can _disable_ reading MIDX files by setting core.multiPackIndex to false. If this was in the user-facing docs, not the technical docs, then I might push back on trying to make that distinction.

> -- The core.multiPackIndex config setting must be on to consume MIDX files.
> -

As it is, this deletion is fine.

A nit on your commit message: format commit references like this:

  18e449f86 (midx: enable core.multiPackIndex by default, 2020-09-25)

Thanks, -Stolee

Taylor Blau· Sep 23, 2021, 17:45 UTC · re: Derrick Stolee · lore

Re: [PATCH] doc/technical: remove outdated MIDX default note

On Thu, Sep 23, 2021 at 09:47:03AM -0400, Derrick Stolee wrote:

Show 10 quoted lines

> On 9/22/2021 6:13 PM, Eric Wong wrote:
> > Multi-pack-index files are used by default since
> > commit 18e449f86b74bab35b150549c8342d252fe7ae00, so the
> > outdated note was misleading.
>
> You are correct that the note is misleading, but it is still correct.
>
> A user can _disable_ reading MIDX files by setting core.multiPackIndex
> to false. If this was in the user-facing docs, not the technical docs,
> then I might push back on trying to make that distinction.

It may present a fuller picture to instead say:

  The core.multiPackIndex config setting must be on (which is the default)
  to consume MIDX files. Setting it to `false` prevents Git from reading a
  MIDX file, even if one exists.

Thanks, Taylor

Eric Wong· Sep 23, 2021, 20:12 UTC · re: Taylor Blau · lore

[PATCH v2] doc/technical: update note about core.multiPackIndex

Taylor Blau <me@ttaylorr.com> wrote:

Show 17 quoted lines

> On Thu, Sep 23, 2021 at 09:47:03AM -0400, Derrick Stolee wrote:
> > On 9/22/2021 6:13 PM, Eric Wong wrote:
> > > Multi-pack-index files are used by default since
> > > commit 18e449f86b74bab35b150549c8342d252fe7ae00, so the
> > > outdated note was misleading.
> >
> > You are correct that the note is misleading, but it is still correct.
> >
> > A user can _disable_ reading MIDX files by setting core.multiPackIndex
> > to false. If this was in the user-facing docs, not the technical docs,
> > then I might push back on trying to make that distinction.
> 
> It may present a fuller picture to instead say:
> 
>   The core.multiPackIndex config setting must be on (which is the default)
>   to consume MIDX files. Setting it to `false` prevents Git from reading a
>   MIDX file, even if one exists.

Ah, thanks, I'll just use your text.

---------8<----------
Subject: [PATCH] doc/technical: update note about core.multiPackIndex

MIDX files are used by default since commit d0fa539bc569 (doc/technical: remove outdated MIDX default note, 2021-09-22)

Helped-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Eric Wong <e@80x24.org>
---
 Documentation/technical/multi-pack-index.txt | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

Show changes to Documentation/technical/multi-pack-index.txt +3 −1

diff --git a/Documentation/technical/multi-pack-index.txt b/Documentation/technical/multi-pack-index.txt
index 1a73c3ee203e..86f40f24909a 100644
--- a/Documentation/technical/multi-pack-index.txt
+++ b/Documentation/technical/multi-pack-index.txt
@@ -36,7 +36,9 @@ Design Details
   directory of an alternate. It refers only to packfiles in that
   same directory.
 
-- The core.multiPackIndex config setting must be on to consume MIDX files.
+- The core.multiPackIndex config setting must be on (which is the
+  default) to consume MIDX files.  Setting it to `false` prevents
+  Git from reading a MIDX file, even if one exists.
 
 - The file format includes parameters for the object ID hash
   function, so a future change of hash algorithm does not require

Taylor Blau· Sep 23, 2021, 20:21 UTC · re: Eric Wong · lore

Re: [PATCH v2] doc/technical: update note about core.multiPackIndex

On Thu, Sep 23, 2021 at 08:12:52PM +0000, Eric Wong wrote:

Show 8 quoted lines

> Subject: [PATCH] doc/technical: update note about core.multiPackIndex
>
> MIDX files are used by default since commit d0fa539bc569
> (doc/technical: remove outdated MIDX default note, 2021-09-22)
>
> Helped-by: Taylor Blau <me@ttaylorr.com>
> Signed-off-by: Eric Wong <e@80x24.org>
> ---

Thanks; this version looks good to me.

Thanks, Taylor

Eric Wong· Sep 24, 2021, 11:11 UTC · re: Eric Wong · lore

[PATCH v3] doc/technical: update note about core.multiPackIndex

MIDX files are used by default since commit 18e449f86b74 (midx: enable core.multiPackIndex by default, 2020-09-25)

Helped-by: Taylor Blau <me@ttaylorr.com>
Signed-off-by: Eric Wong <e@80x24.org>
---
 v3: I cited the wrong commit in the commit message of v2 :x

 Documentation/technical/multi-pack-index.txt | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

Show changes to Documentation/technical/multi-pack-index.txt +3 −1

diff --git a/Documentation/technical/multi-pack-index.txt b/Documentation/technical/multi-pack-index.txt
index 1a73c3ee203e..86f40f24909a 100644
--- a/Documentation/technical/multi-pack-index.txt
+++ b/Documentation/technical/multi-pack-index.txt
@@ -36,7 +36,9 @@ Design Details
   directory of an alternate. It refers only to packfiles in that
   same directory.
 
-- The core.multiPackIndex config setting must be on to consume MIDX files.
+- The core.multiPackIndex config setting must be on (which is the
+  default) to consume MIDX files.  Setting it to `false` prevents
+  Git from reading a MIDX file, even if one exists.
 
 - The file format includes parameters for the object ID hash
   function, so a future change of hash algorithm does not require

Derrick Stolee· Sep 25, 2021, 17:17 UTC · re: Eric Wong · lore

Re: [PATCH v3] doc/technical: update note about core.multiPackIndex

On 9/24/2021 7:11 AM, Eric Wong wrote:

Show 7 quoted lines

> MIDX files are used by default since commit 18e449f86b74
> (midx: enable core.multiPackIndex by default, 2020-09-25)
> 
> Helped-by: Taylor Blau <me@ttaylorr.com>
> Signed-off-by: Eric Wong <e@80x24.org>
> ---
>  v3: I cited the wrong commit in the commit message of v2 :x

Good catch.

Show 18 quoted lines

>  Documentation/technical/multi-pack-index.txt | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/Documentation/technical/multi-pack-index.txt b/Documentation/technical/multi-pack-index.txt
> index 1a73c3ee203e..86f40f24909a 100644
> --- a/Documentation/technical/multi-pack-index.txt
> +++ b/Documentation/technical/multi-pack-index.txt
> @@ -36,7 +36,9 @@ Design Details
>    directory of an alternate. It refers only to packfiles in that
>    same directory.
>  
> -- The core.multiPackIndex config setting must be on to consume MIDX files.
> +- The core.multiPackIndex config setting must be on (which is the
> +  default) to consume MIDX files.  Setting it to `false` prevents
> +  Git from reading a MIDX file, even if one exists.
>  
>  - The file format includes parameters for the object ID hash
>    function, so a future change of hash algorithm does not require

This looks good to me, too!

Thanks, -Stolee

Junio C Hamano· Sep 23, 2021, 20:27 UTC · re: Taylor Blau · lore

Re: [PATCH] doc/technical: remove outdated MIDX default note

Taylor Blau <me@ttaylorr.com> writes:

Show 17 quoted lines

> On Thu, Sep 23, 2021 at 09:47:03AM -0400, Derrick Stolee wrote:
>> On 9/22/2021 6:13 PM, Eric Wong wrote:
>> > Multi-pack-index files are used by default since
>> > commit 18e449f86b74bab35b150549c8342d252fe7ae00, so the
>> > outdated note was misleading.
>>
>> You are correct that the note is misleading, but it is still correct.
>>
>> A user can _disable_ reading MIDX files by setting core.multiPackIndex
>> to false. If this was in the user-facing docs, not the technical docs,
>> then I might push back on trying to make that distinction.
>
> It may present a fuller picture to instead say:
>
>   The core.multiPackIndex config setting must be on (which is the default)
>   to consume MIDX files. Setting it to `false` prevents Git from reading a
>   MIDX file, even if one exists.

Do you want to replace the one-line sentence with that paragraph, instead of removing it, in a replacement patch?

Taylor Blau· Sep 23, 2021, 20:32 UTC · re: Junio C Hamano · lore

Re: [PATCH] doc/technical: remove outdated MIDX default note

On Thu, Sep 23, 2021 at 01:27:31PM -0700, Junio C Hamano wrote:

Show 22 quoted lines

> Taylor Blau <me@ttaylorr.com> writes:
>
> > On Thu, Sep 23, 2021 at 09:47:03AM -0400, Derrick Stolee wrote:
> >> On 9/22/2021 6:13 PM, Eric Wong wrote:
> >> > Multi-pack-index files are used by default since
> >> > commit 18e449f86b74bab35b150549c8342d252fe7ae00, so the
> >> > outdated note was misleading.
> >>
> >> You are correct that the note is misleading, but it is still correct.
> >>
> >> A user can _disable_ reading MIDX files by setting core.multiPackIndex
> >> to false. If this was in the user-facing docs, not the technical docs,
> >> then I might push back on trying to make that distinction.
> >
> > It may present a fuller picture to instead say:
> >
> >   The core.multiPackIndex config setting must be on (which is the default)
> >   to consume MIDX files. Setting it to `false` prevents Git from reading a
> >   MIDX file, even if one exists.
>
> Do you want to replace the one-line sentence with that paragraph,
> instead of removing it, in a replacement patch?

That's what I was hoping for, and Eric did that here:

    https://lore.kernel.org/git/20210923201252.GA25980@dcvr/

Thanks, Taylor