Date: Fri, 11 Feb 2022 15:50:39 +0100 From: Kristof Provost <kp@FreeBSD.org> To: Drew Gallatin <gallatin@netflix.com> Cc: Andrew Gallatin <gallatin@cs.duke.edu>, "Alexander V. Chernikov" <melifaro@freebsd.org>, src-committers@freebsd.org, dev-commits-src-all@freebsd.org, dev-commits-src-main@freebsd.org, Conrad Meyer <cem@freebsd.org>, Andriy Gapon <avg@freebsd.org> Subject: Re: git: b1f7154cb125 - main - gitignore: ignore vim swap files & .rej/.orig Message-ID: <6CFFB341-C17E-427D-99B8-E640505B83CB@FreeBSD.org> In-Reply-To: <CADLQ3sKdAvci47RxK=Vqg4JA0Wff2z9dcuDrByWNdendQu=M%2Bw@mail.gmail.com> References: <202201170935.20H9ZM3T069547@gitrepo.freebsd.org> <72061959-88f8-d2ee-b4e9-f09ca084411e@cs.duke.edu> <4FC1EDC4-9F20-4271-B468-715A1727AC8A@FreeBSD.org> <CADLQ3sKdAvci47RxK=Vqg4JA0Wff2z9dcuDrByWNdendQu=M%2Bw@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--=_MailMate_DE4680ED-4562-4D6E-882A-8C0E25C4755B_=
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
On 11 Feb 2022, at 15:38, Drew Gallatin wrote:
> The problem seems to be that all of these solutions are processed
> *FIRST*,
> before a .gitignore in the repo, so they cannot be used to negate a
> gitignore rule in the repo. Eg:
>
The man page claims this:
When deciding whether
to ignore a path, Git normally checks gitignore patterns from
multiple
sources, with the following order of precedence, from highest to
lowest
(within one level of precedence, the last matching pattern decides
the
outcome):
And then lists command line, .gitignore file in the same directory as
the path, or in any parent directory, GIT_DIR/info/exclude and
core.excludesFile.
So it’s does check them in the order you’d expect, but apparently
does not let GIT_DIR/info/exclude overrule .gitignore. That’s a bit
unfortunate. I’d really have expected a ‘!*.orig’ in
GIT_DIR/info/exclude to overrule the .gitignore, but I’ve also tested
it and it doesn’t.
> This makes a stronger case that those who want to ignore .orig and
> .rej
> files can add them to *their* local .gitignore, and we should back
> them out
> of the .gitignore for the project.
>
I don’t really like it (for the usual reasons: .orig & friends should
pretty much never get checked in, so .gitignore is appropriate. Your
workflow is a bit unusual, so ideally you’d do the special thing.),
but yeah, it looks like we should indeed remove them from .gitignore in
the repo.
Kristof
--=_MailMate_DE4680ED-4562-4D6E-882A-8C0E25C4755B_=
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE html>
<html>
<head>
<meta http-equiv=3D"Content-Type" content=3D"text/xhtml; charset=3Dutf-8"=
>
</head>
<body><div style=3D"font-family: sans-serif;"><div class=3D"markdown" sty=
le=3D"white-space: normal;">
<p dir=3D"auto">On 11 Feb 2022, at 15:38, Drew Gallatin wrote:</p>
</div><div class=3D"plaintext" style=3D"white-space: normal;"><blockquote=
style=3D"margin: 0 0 5px; padding-left: 5px; border-left: 2px solid #136=
BCE; color: #136BCE;"><p dir=3D"auto">The problem seems to be that all of=
these solutions are processed *FIRST*,
<br>
before a .gitignore in the repo, so they cannot be used to negate a
<br>
gitignore rule in the repo. Eg:</p>
<br></blockquote></div>
<div class=3D"markdown" style=3D"white-space: normal;">
<p dir=3D"auto">The man page claims this:</p>
<pre style=3D"margin-left: 15px; margin-right: 15px; padding: 5px; border=
: thin solid gray; overflow-x: auto; max-width: 90vw; background-color: #=
E4E4E4;"><code>When deciding whether
to ignore a path, Git normally checks gitignore patterns from multiple
sources, with the following order of precedence, from highest to lowest
(within one level of precedence, the last matching pattern decides the
outcome):
</code></pre>
<p dir=3D"auto">And then lists command line, .gitignore file in the same =
directory as the path, or in any parent directory, GIT_DIR/info/exclude a=
nd core.excludesFile.</p>
<p dir=3D"auto">So it=E2=80=99s does check them in the order you=E2=80=99=
d expect, but apparently does not let GIT_DIR/info/exclude overrule .giti=
gnore. That=E2=80=99s a bit unfortunate. I=E2=80=99d really have expected=
a =E2=80=98!*.orig=E2=80=99 in GIT_DIR/info/exclude to overrule the .git=
ignore, but I=E2=80=99ve also tested it and it doesn=E2=80=99t.</p>
</div><div class=3D"plaintext" style=3D"white-space: normal;"><blockquote=
style=3D"margin: 0 0 5px; padding-left: 5px; border-left: 2px solid #136=
BCE; color: #136BCE;"><p dir=3D"auto">This makes a stronger case that tho=
se who want to ignore .orig and .rej
<br>
files can add them to *their* local .gitignore, and we should back them o=
ut
<br>
of the .gitignore for the project.</p>
<br></blockquote></div>
<div class=3D"markdown" style=3D"white-space: normal;">
<p dir=3D"auto">I don=E2=80=99t really like it (for the usual reasons: .o=
rig & friends should pretty much never get checked in, so .gitignore =
is appropriate. Your workflow is a bit unusual, so ideally you=E2=80=99d =
do the special thing.), but yeah, it looks like we should indeed remove t=
hem from .gitignore in the repo.</p>
<p dir=3D"auto">Kristof</p>
</div></div></body>
</html>
--=_MailMate_DE4680ED-4562-4D6E-882A-8C0E25C4755B_=--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?6CFFB341-C17E-427D-99B8-E640505B83CB>