Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1163601 > unrolled thread

diffs in changelogs

Back to article view | Back to linux.kernel

Contents

  diffs in changelogs Andrew Morton <akpm@linux-foundation.org> - 2015-06-11 22:50 +0200
    Re: diffs in changelogs Joe Perches <joe@perches.com> - 2015-06-12 05:20 +0200
      Re: diffs in changelogs Andrew Morton <akpm@linux-foundation.org> - 2015-06-12 05:20 +0200
        Re: diffs in changelogs Joe Perches <joe@perches.com> - 2015-06-12 06:00 +0200

#1163601 — diffs in changelogs

People often put diff snippets in changelogs.  This causes problems
when one tries to apply a file containing both the changelog and the
diff because patch(1) tries to apply the diff which it found in the
changelog.

eg, something like

	git show d24a6e1087030b6da | patch -p1

will go haywire.

So can we please have a checkpatch test warning people away from doing
this?


patch(1) seems to be really promiscuous in its detection of a patch.  I
haven't had much success searching for "^--- " and similar.  What works
best for me is searching for "^[whitespace]@@ -".

Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1163738

On Thu, 2015-06-11 at 13:40 -0700, Andrew Morton wrote:
> People often put diff snippets in changelogs.  This causes problems
> when one tries to apply a file containing both the changelog and the
> diff because patch(1) tries to apply the diff which it found in the
> changelog.
That 
> eg, something like
> 
> 	git show d24a6e1087030b6da | patch -p1
> 
> will go haywire.
> 
> So can we please have a checkpatch test warning people away from doing
> this?
> 
> 
> patch(1) seems to be really promiscuous in its detection of a patch.  I
> haven't had much success searching for "^--- " and similar.  What works
> best for me is searching for "^[whitespace]@@ -".

I don't think that's a good test.
Coccinelle uses @@

And how did that commit actually get applied?

I tried applying it to a new branch checked out at
ce2b3f595e1c56639085645e0130426e443008c0, it fails.

Anyway, maybe:
---
 scripts/checkpatch.pl | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 69c4716..2d87e37 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -2399,6 +2399,18 @@ sub process {
 			$in_commit_log = 1;
 		}
 
+# Check if the commit log has a diff which confuse patch
+
+		print("icl: <$in_commit_log> line: <$line>\n");
+		if ($in_commit_log &&
+		    (($line =~ m@^\s+diff\b.*a/[\w/]+@ &&
+		      $line =~ m@^\s+diff\b.*a/([\w/]+)\s+b/$1\b@) ||
+		     $line =~ m@^\s*(?:\-\-\-\s+a/|\+\+\+\s+b/)@ ||
+		     $line =~ m/^\s*\@\@ \-\d+,\d+ \+\d+,\d+ \@\@/)) {
+			ERROR("DIFF_IN_COMMIT_MSG",
+			      "It seems a diff exists in the commit message. This can confuse patch\n" . $herecurr);
+		}
+
 # Check if there is UTF-8 in a commit log when a mail header has explicitly
 # declined it, i.e defined some charset where it is missing.
 		if ($in_header_lines &&


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1163739

On Thu, 11 Jun 2015 20:12:59 -0700 Joe Perches <joe@perches.com> wrote:

> On Thu, 2015-06-11 at 13:40 -0700, Andrew Morton wrote:
> > People often put diff snippets in changelogs.  This causes problems
> > when one tries to apply a file containing both the changelog and the
> > diff because patch(1) tries to apply the diff which it found in the
> > changelog.
> That 
> > eg, something like
> > 
> > 	git show d24a6e1087030b6da | patch -p1
> > 
> > will go haywire.
> > 
> > So can we please have a checkpatch test warning people away from doing
> > this?
> > 
> > 
> > patch(1) seems to be really promiscuous in its detection of a patch.  I
> > haven't had much success searching for "^--- " and similar.  What works
> > best for me is searching for "^[whitespace]@@ -".
> 
> I don't think that's a good test.
> Coccinelle uses @@
> 
> And how did that commit actually get applied?

Good question.  Maybe `git apply' is smarter about this than patch(1)?

> I tried applying it to a new branch checked out at
> ce2b3f595e1c56639085645e0130426e443008c0, it fails.

There are tons of them:

z:/usr/src/git26> git log | grep "^[    ]*@@ -"  | wc -l
120

> Anyway, maybe:

Looks good, thanks.  -ENOCHANGELOG.  Please send it for real when convenient.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [next] | [standalone]

#1163750

On Thu, 2015-06-11 at 20:21 -0700, Andrew Morton wrote:
> On Thu, 11 Jun 2015 20:12:59 -0700 Joe Perches <joe@perches.com> wrote:
> 
> > On Thu, 2015-06-11 at 13:40 -0700, Andrew Morton wrote:
> > > People often put diff snippets in changelogs.  This causes problems
> > > when one tries to apply a file containing both the changelog and the
> > > diff because patch(1) tries to apply the diff which it found in the
> > > changelog.
> > That 
> > > eg, something like
> > > 
> > > 	git show d24a6e1087030b6da | patch -p1
> > > 
> > > will go haywire.
> > > 
> > > So can we please have a checkpatch test warning people away from doing
> > > this?
> > > 
> > > 
> > > patch(1) seems to be really promiscuous in its detection of a patch.  I
> > > haven't had much success searching for "^--- " and similar.  What works
> > > best for me is searching for "^[whitespace]@@ -".
> > 
> > I don't think that's a good test.
> > Coccinelle uses @@
> > 
> > And how did that commit actually get applied?
> 
> Good question.  Maybe `git apply' is smarter about this than patch(1)?

Dunno.  git am failed.
My guess is it was applied normally and then
git commit --amend added the other diff content

> > I tried applying it to a new branch checked out at
> > ce2b3f595e1c56639085645e0130426e443008c0, it fails.
> 
> There are tons of them:
> 
> z:/usr/src/git26> git log | grep "^[    ]*@@ -"  | wc -l
> 120
> 
> > Anyway, maybe:
> 
> Looks good, thanks.  -ENOCHANGELOG.  Please send it for real when convenient.

RFCPATCH




--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web