diff mbox

Fix regressions caused by renaming README

Message ID 1468950439-16259-1-git-send-email-sw@weilnetz.de (mailing list archive)
State New, archived
Headers show

Commit Message

Stefan Weil July 19, 2016, 5:47 p.m. UTC
Commit e5dfc5e8e715c572aea44ac4d96c43941d4741c7 renamed README but did not
update scripts/checkpatch.pl. Fix this.

An update was also needed for qemu.nsi (Windows).

Reported-by: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: Stefan Weil <sw@weilnetz.de>
---
 qemu.nsi              | 4 ++--
 scripts/checkpatch.pl | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

Comments

Pranith Kumar July 19, 2016, 6:15 p.m. UTC | #1
On Tue, Jul 19, 2016 at 1:47 PM, Stefan Weil <sw@weilnetz.de> wrote:
> Commit e5dfc5e8e715c572aea44ac4d96c43941d4741c7 renamed README but did not
> update scripts/checkpatch.pl. Fix this.
>
> An update was also needed for qemu.nsi (Windows).
>
> Reported-by: Daniel P. Berrange <berrange@redhat.com>
> Signed-off-by: Stefan Weil <sw@weilnetz.de>

Thank you for fixing this.

--
Pranith
Peter Maydell July 19, 2016, 8:53 p.m. UTC | #2
On 19 July 2016 at 18:47, Stefan Weil <sw@weilnetz.de> wrote:
> Commit e5dfc5e8e715c572aea44ac4d96c43941d4741c7 renamed README but did not
> update scripts/checkpatch.pl. Fix this.
>
> An update was also needed for qemu.nsi (Windows).
>
> Reported-by: Daniel P. Berrange <berrange@redhat.com>
> Signed-off-by: Stefan Weil <sw@weilnetz.de>
> ---
>  qemu.nsi              | 4 ++--
>  scripts/checkpatch.pl | 2 +-
>  2 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/qemu.nsi b/qemu.nsi
> index a20f6ef..c037017 100644
> --- a/qemu.nsi
> +++ b/qemu.nsi
> @@ -116,7 +116,7 @@ Section "${PRODUCT} (required)"
>      File "${SRCDIR}\Changelog"
>      File "${SRCDIR}\COPYING"
>      File "${SRCDIR}\COPYING.LIB"
> -    File "${SRCDIR}\README"
> +    File "${SRCDIR}\README.md"
>      File "${SRCDIR}\VERSION"
>
>      File "${BINDIR}\*.bmp"
> @@ -206,7 +206,7 @@ Section "Uninstall"
>      Delete "$INSTDIR\Changelog"
>      Delete "$INSTDIR\COPYING"
>      Delete "$INSTDIR\COPYING.LIB"
> -    Delete "$INSTDIR\README"
> +    Delete "$INSTDIR\README.md"
>      Delete "$INSTDIR\VERSION"
>      Delete "$INSTDIR\*.bmp"
>      Delete "$INSTDIR\*.bin"
> diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
> index afa7f79..8247305 100755
> --- a/scripts/checkpatch.pl
> +++ b/scripts/checkpatch.pl
> @@ -286,7 +286,7 @@ sub top_of_kernel_tree {
>
>         my @tree_check = (
>                 "COPYING", "MAINTAINERS", "Makefile",
> -               "README", "docs", "VERSION",
> +               "README.md", "docs", "VERSION",
>                 "vl.c"
>         );

Does raise the question of whether we should be renaming
the file in the first place. README is the traditional
name and fits with all our other basically-plain-text
document names like COPYING, MAINTAINERS, HACKING.

thanks
-- PMM
Pranith Kumar July 19, 2016, 9:17 p.m. UTC | #3
On Tue, Jul 19, 2016 at 4:53 PM, Peter Maydell <peter.maydell@linaro.org> wrote:
> Does raise the question of whether we should be renaming
> the file in the first place. README is the traditional
> name and fits with all our other basically-plain-text
> document names like COPYING, MAINTAINERS, HACKING.
>

I was hoping that this change would be helpful since mark
down(at-least the way we have it) still looks like plain text if you
open in a text editor. The advantage is that github repositories look
way better since mark down is formatted well there. I am open to
getting it reverted if it really is bothering.
Peter Maydell July 19, 2016, 9:33 p.m. UTC | #4
On 19 July 2016 at 22:17, Pranith Kumar <bobby.prani@gmail.com> wrote:
> On Tue, Jul 19, 2016 at 4:53 PM, Peter Maydell <peter.maydell@linaro.org> wrote:
>> Does raise the question of whether we should be renaming
>> the file in the first place. README is the traditional
>> name and fits with all our other basically-plain-text
>> document names like COPYING, MAINTAINERS, HACKING.
>>
>
> I was hoping that this change would be helpful since mark
> down(at-least the way we have it) still looks like plain text if you
> open in a text editor. The advantage is that github repositories look
> way better since mark down is formatted well there. I am open to
> getting it reverted if it really is bothering.

On the other hand we don't use github as our primary git
repo, and we don't try to render the markdown anywhere
in our build process. The documentation really is a
plain text file in my view, and I'm not hugely enthusiastic
about switching the filename to placate a website we
don't even use except as a backup/mirror.

thanks
-- PMM
Markus Armbruster July 20, 2016, 6:37 a.m. UTC | #5
Peter Maydell <peter.maydell@linaro.org> writes:

> On 19 July 2016 at 22:17, Pranith Kumar <bobby.prani@gmail.com> wrote:
>> On Tue, Jul 19, 2016 at 4:53 PM, Peter Maydell <peter.maydell@linaro.org> wrote:
>>> Does raise the question of whether we should be renaming
>>> the file in the first place. README is the traditional
>>> name and fits with all our other basically-plain-text
>>> document names like COPYING, MAINTAINERS, HACKING.
>>>
>>
>> I was hoping that this change would be helpful since mark
>> down(at-least the way we have it) still looks like plain text if you
>> open in a text editor. The advantage is that github repositories look
>> way better since mark down is formatted well there. I am open to
>> getting it reverted if it really is bothering.
>
> On the other hand we don't use github as our primary git
> repo, and we don't try to render the markdown anywhere
> in our build process. The documentation really is a
> plain text file in my view, and I'm not hugely enthusiastic
> about switching the filename to placate a website we
> don't even use except as a backup/mirror.

I'd less unenthusiastic if MarkDown worked the same everywhere.  Its
common core does, but how would we ensure we stick to the common core?
How would we even know what the core common to the various MarkDown
dialects is?  Or are we ready to commit to GitHub's dialect?
Paolo Bonzini July 20, 2016, 7 a.m. UTC | #6
On 20/07/2016 08:37, Markus Armbruster wrote:
> I'd less unenthusiastic if MarkDown worked the same everywhere.  Its
> common core does, but how would we ensure we stick to the common core?
> How would we even know what the core common to the various MarkDown
> dialects is?

README.md uses exactly three features:

* monospace text

* headings

* lists

> Or are we ready to commit to GitHub's dialect?

We probably want to rely on GitHub's stricter interpretation of
underscores, where they only start/end italic formatting if they are
respectively the first or the last character in the word.  This is a
common extension (for example Stack Exchange does the same).

The question is: is anyone actually viewing formatted markdown anywhere
else than on github?

Paolo
Fam Zheng July 20, 2016, 7:07 a.m. UTC | #7
On Wed, 07/20 09:00, Paolo Bonzini wrote:
> We probably want to rely on GitHub's stricter interpretation of
> underscores, where they only start/end italic formatting if they are
> respectively the first or the last character in the word.  This is a
> common extension (for example Stack Exchange does the same).
> 
> The question is: is anyone actually viewing formatted markdown anywhere
> else than on github?

It may have slightly better syntax highlighting in editors.

Out of curiosity, I'm wondering if it's acceptable to commit a "ln -s README.md
README" change in the repo, at all?

Fam
Paolo Bonzini July 20, 2016, 7:09 a.m. UTC | #8
On 20/07/2016 09:07, Fam Zheng wrote:
>> > The question is: is anyone actually viewing formatted markdown anywhere
>> > else than on github?
> It may have slightly better syntax highlighting in editors.
> 
> Out of curiosity, I'm wondering if it's acceptable to commit a "ln -s README.md
> README" change in the repo, at all?

Yes, that's a good idea.

Paolo
Fam Zheng July 20, 2016, 7:18 a.m. UTC | #9
On Wed, 07/20 09:09, Paolo Bonzini wrote:
> 
> 
> On 20/07/2016 09:07, Fam Zheng wrote:
> >> > The question is: is anyone actually viewing formatted markdown anywhere
> >> > else than on github?
> > It may have slightly better syntax highlighting in editors.
> > 
> > Out of curiosity, I'm wondering if it's acceptable to commit a "ln -s README.md
> > README" change in the repo, at all?
> 
> Yes, that's a good idea.

I'll see if github likes that. :)

Fam
Daniel P. Berrangé July 20, 2016, 8:47 a.m. UTC | #10
On Tue, Jul 19, 2016 at 09:53:30PM +0100, Peter Maydell wrote:
> On 19 July 2016 at 18:47, Stefan Weil <sw@weilnetz.de> wrote:
> > Commit e5dfc5e8e715c572aea44ac4d96c43941d4741c7 renamed README but did not
> > update scripts/checkpatch.pl. Fix this.
> >
> > An update was also needed for qemu.nsi (Windows).
> >
> > Reported-by: Daniel P. Berrange <berrange@redhat.com>
> > Signed-off-by: Stefan Weil <sw@weilnetz.de>
> > ---
> >  qemu.nsi              | 4 ++--
> >  scripts/checkpatch.pl | 2 +-
> >  2 files changed, 3 insertions(+), 3 deletions(-)
> >
> > diff --git a/qemu.nsi b/qemu.nsi
> > index a20f6ef..c037017 100644
> > --- a/qemu.nsi
> > +++ b/qemu.nsi
> > @@ -116,7 +116,7 @@ Section "${PRODUCT} (required)"
> >      File "${SRCDIR}\Changelog"
> >      File "${SRCDIR}\COPYING"
> >      File "${SRCDIR}\COPYING.LIB"
> > -    File "${SRCDIR}\README"
> > +    File "${SRCDIR}\README.md"
> >      File "${SRCDIR}\VERSION"
> >
> >      File "${BINDIR}\*.bmp"
> > @@ -206,7 +206,7 @@ Section "Uninstall"
> >      Delete "$INSTDIR\Changelog"
> >      Delete "$INSTDIR\COPYING"
> >      Delete "$INSTDIR\COPYING.LIB"
> > -    Delete "$INSTDIR\README"
> > +    Delete "$INSTDIR\README.md"
> >      Delete "$INSTDIR\VERSION"
> >      Delete "$INSTDIR\*.bmp"
> >      Delete "$INSTDIR\*.bin"
> > diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
> > index afa7f79..8247305 100755
> > --- a/scripts/checkpatch.pl
> > +++ b/scripts/checkpatch.pl
> > @@ -286,7 +286,7 @@ sub top_of_kernel_tree {
> >
> >         my @tree_check = (
> >                 "COPYING", "MAINTAINERS", "Makefile",
> > -               "README", "docs", "VERSION",
> > +               "README.md", "docs", "VERSION",
> >                 "vl.c"
> >         );
> 
> Does raise the question of whether we should be renaming
> the file in the first place. README is the traditional
> name and fits with all our other basically-plain-text
> document names like COPYING, MAINTAINERS, HACKING.

As mentioned in my original review, we could just create a symlink
between READE & README.md or vica-verca if we wish to preserve the
"normal" filename, but still have github do its magic.

Regards,
Daniel
diff mbox

Patch

diff --git a/qemu.nsi b/qemu.nsi
index a20f6ef..c037017 100644
--- a/qemu.nsi
+++ b/qemu.nsi
@@ -116,7 +116,7 @@  Section "${PRODUCT} (required)"
     File "${SRCDIR}\Changelog"
     File "${SRCDIR}\COPYING"
     File "${SRCDIR}\COPYING.LIB"
-    File "${SRCDIR}\README"
+    File "${SRCDIR}\README.md"
     File "${SRCDIR}\VERSION"
 
     File "${BINDIR}\*.bmp"
@@ -206,7 +206,7 @@  Section "Uninstall"
     Delete "$INSTDIR\Changelog"
     Delete "$INSTDIR\COPYING"
     Delete "$INSTDIR\COPYING.LIB"
-    Delete "$INSTDIR\README"
+    Delete "$INSTDIR\README.md"
     Delete "$INSTDIR\VERSION"
     Delete "$INSTDIR\*.bmp"
     Delete "$INSTDIR\*.bin"
diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index afa7f79..8247305 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -286,7 +286,7 @@  sub top_of_kernel_tree {
 
 	my @tree_check = (
 		"COPYING", "MAINTAINERS", "Makefile",
-		"README", "docs", "VERSION",
+		"README.md", "docs", "VERSION",
 		"vl.c"
 	);