diff mbox series

[2/2] checkpatch: Don't check for unified diff format in git sendemail headers

Message ID 20240222051539.3001988-4-saravanak@google.com
State New
Headers
Series [1/2] checkpatch: Don't check for 75 chars per line for create/delete mode lines |

Commit Message

Saravana Kannan Feb. 22, 2024, 5:15 a.m. UTC 
  When checkpatch is used as a git sendemail-validate hook, it's also passed
in the email header for sanity check. These headers are, as expected, not
in unified diff format. So, don't complain about unified diff format for
these header files.

Signed-off-by: Saravana Kannan <saravanak@google.com>
---
 scripts/checkpatch.pl | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Comments

Joe Perches Feb. 22, 2024, 8:54 a.m. UTC | #1 
On Wed, 2024-02-21 at 21:15 -0800, Saravana Kannan wrote:
> When checkpatch is used as a git sendemail-validate hook, it's also passed
> in the email header for sanity check.

Why?

If so, why not use a front-end script to stop/remove
the file from being scanned by checkpatch?

> These headers are, as expected, not
> in unified diff format. So, don't complain about unified diff format for
> these header files.
Saravana Kannan Feb. 23, 2024, 12:45 a.m. UTC | #2 
On Thu, Feb 22, 2024 at 12:54 AM Joe Perches <joe@perches.com> wrote:
>
> On Wed, 2024-02-21 at 21:15 -0800, Saravana Kannan wrote:
> > When checkpatch is used as a git sendemail-validate hook, it's also passed
> > in the email header for sanity check.
>
> Why?
>
> If so, why not use a front-end script to stop/remove
> the file from being scanned by checkpatch?

Sure, I could do that. But this also makes it easier for people to
start using checkpatch. Or I can put up a git hook wrapper script in
here for people to symlink into their .git/hooks that does this.

I'd prefer the lazy route of not creating a 1 line wrapper script :)

-Saravana

>
> > These headers are, as expected, not
> > in unified diff format. So, don't complain about unified diff format for
> > these header files.
>
Joe Perches Feb. 23, 2024, 1:01 a.m. UTC | #3 
On Thu, 2024-02-22 at 16:45 -0800, Saravana Kannan wrote:
> On Thu, Feb 22, 2024 at 12:54 AM Joe Perches <joe@perches.com> wrote:
> > 
> > On Wed, 2024-02-21 at 21:15 -0800, Saravana Kannan wrote:
> > > When checkpatch is used as a git sendemail-validate hook, it's also passed
> > > in the email header for sanity check.
> > 
> > Why?
> > 
> > If so, why not use a front-end script to stop/remove
> > the file from being scanned by checkpatch?
> 
> Sure, I could do that. But this also makes it easier for people to
> start using checkpatch. Or I can put up a git hook wrapper script in
> here for people to symlink into their .git/hooks that does this.
> 
> I'd prefer the lazy route of not creating a 1 line wrapper script :)

I'd not.  checkpatch is for _patches_.
Don't feed stuff to it that isn't patches and expect good results.
Saravana Kannan Feb. 23, 2024, 1:10 a.m. UTC | #4 
On Thu, Feb 22, 2024 at 5:01 PM Joe Perches <joe@perches.com> wrote:
>
> On Thu, 2024-02-22 at 16:45 -0800, Saravana Kannan wrote:
> > On Thu, Feb 22, 2024 at 12:54 AM Joe Perches <joe@perches.com> wrote:
> > >
> > > On Wed, 2024-02-21 at 21:15 -0800, Saravana Kannan wrote:
> > > > When checkpatch is used as a git sendemail-validate hook, it's also passed
> > > > in the email header for sanity check.
> > >
> > > Why?
> > >
> > > If so, why not use a front-end script to stop/remove
> > > the file from being scanned by checkpatch?
> >
> > Sure, I could do that. But this also makes it easier for people to
> > start using checkpatch. Or I can put up a git hook wrapper script in
> > here for people to symlink into their .git/hooks that does this.
> >
> > I'd prefer the lazy route of not creating a 1 line wrapper script :)
>
> I'd not.  checkpatch is for _patches_.
> Don't feed stuff to it that isn't patches and expect good results.

Would you be open to being a maintainer if I add a git hook
sendemail-validate wrapper? It feels silly to add myself as a
maintainer for a 1-line script. I'd rather give it to you :)

-Saravana
Joe Perches Feb. 23, 2024, 1:22 a.m. UTC | #5 
On Thu, 2024-02-22 at 17:10 -0800, Saravana Kannan wrote:
> On Thu, Feb 22, 2024 at 5:01 PM Joe Perches <joe@perches.com> wrote:
> > 
> > On Thu, 2024-02-22 at 16:45 -0800, Saravana Kannan wrote:
> > > On Thu, Feb 22, 2024 at 12:54 AM Joe Perches <joe@perches.com> wrote:
> > > > 
> > > > On Wed, 2024-02-21 at 21:15 -0800, Saravana Kannan wrote:
> > > > > When checkpatch is used as a git sendemail-validate hook, it's also passed
> > > > > in the email header for sanity check.
> > > > 
> > > > Why?
> > > > 
> > > > If so, why not use a front-end script to stop/remove
> > > > the file from being scanned by checkpatch?
> > > 
> > > Sure, I could do that. But this also makes it easier for people to
> > > start using checkpatch. Or I can put up a git hook wrapper script in
> > > here for people to symlink into their .git/hooks that does this.
> > > 
> > > I'd prefer the lazy route of not creating a 1 line wrapper script :)
> > 
> > I'd not.  checkpatch is for _patches_.
> > Don't feed stuff to it that isn't patches and expect good results.
> 
> Would you be open to being a maintainer if I add a git hook
> sendemail-validate wrapper? It feels silly to add myself as a
> maintainer for a 1-line script. I'd rather give it to you :)

<shrug>

I think you need a local script and not one in the tree.
Or maybe some different git invocation command.  Dunno.
Lukas Bulwahn Feb. 23, 2024, 7:05 p.m. UTC | #6 
On Fri, Feb 23, 2024 at 2:10 AM Saravana Kannan <saravanak@google.com> wrote:
>
> On Thu, Feb 22, 2024 at 5:01 PM Joe Perches <joe@perches.com> wrote:
> >
> > On Thu, 2024-02-22 at 16:45 -0800, Saravana Kannan wrote:
> > > On Thu, Feb 22, 2024 at 12:54 AM Joe Perches <joe@perches.com> wrote:
> > > >
> > > > On Wed, 2024-02-21 at 21:15 -0800, Saravana Kannan wrote:
> > > > > When checkpatch is used as a git sendemail-validate hook, it's also passed
> > > > > in the email header for sanity check.
> > > >
> > > > Why?
> > > >
> > > > If so, why not use a front-end script to stop/remove
> > > > the file from being scanned by checkpatch?
> > >
> > > Sure, I could do that. But this also makes it easier for people to
> > > start using checkpatch. Or I can put up a git hook wrapper script in
> > > here for people to symlink into their .git/hooks that does this.
> > >
> > > I'd prefer the lazy route of not creating a 1 line wrapper script :)
> >
> > I'd not.  checkpatch is for _patches_.
> > Don't feed stuff to it that isn't patches and expect good results.
>
> Would you be open to being a maintainer if I add a git hook
> sendemail-validate wrapper? It feels silly to add myself as a
> maintainer for a 1-line script. I'd rather give it to you :)
>

I agree with Joe's shrug. We got enough scripts, where very few
(actually: probably nobody) know what they are good for.

However, Saravana, if it helps you, feel free to add a section in the
checkpatch documentation where you describe which workflow you have
and in which files you need to set up what.
So, in case you forget, you will find it in the documentation and
possibly it is also helpful to others---if they read the
documentation, or some AI bot reads the documentation in the future
and then suggests it to someone asking that AI bot---well, that is our
brave new world nowadays...

I will maintain that section in the checkpatch documentation for you,
if you submit the documentation change as a proper patch.

Lukas
diff mbox series

Patch

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index f306634a938c..4312166ca828 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -7689,7 +7689,8 @@  sub process {
 		exit(0);
 	}
 
-	if (!$is_patch && $filename !~ /cover-letter\.patch$/) {
+	if (!$is_patch && $filename !~ /cover-letter\.patch$/ &&
+	    $filename !~ /\.git\/\.gitsendemail\.header\.\w+$/) {
 		ERROR("NOT_UNIFIED_DIFF",
 		      "Does not appear to be a unified-diff format patch\n");
 	}