Both .mailmap and .travis.yml only make sense in the context of a
git checkout, so we should not include them in releases.

Signed-off-by: Andrea Bolognani <abologna@redhat.com>
---
 meson.build     | 9 +++++++++
 tools/nodist.sh | 5 +++++
 2 files changed, 14 insertions(+)
 create mode 100755 tools/nodist.sh

diff --git a/meson.build b/meson.build
index e765ed6..341abe9 100644
--- a/meson.build
+++ b/meson.build
@@ -263,6 +263,15 @@ if git
     foreach file : [ 'libvirt-dbus.spec', 'AUTHORS.rst' ]
         meson.add_dist_script('tools/dist.sh', meson.build_root(), file)
     endforeach
+
+    nodist_files = [
+        '.mailmap',
+        '.travis.yml',
+    ]
+
+    foreach file : nodist_files
+        meson.add_dist_script('tools/nodist.sh', file)
+    endforeach
 endif
 
 
diff --git a/tools/nodist.sh b/tools/nodist.sh
new file mode 100755
index 0000000..f820147
--- /dev/null
+++ b/tools/nodist.sh
@@ -0,0 +1,5 @@
+#!/bin/sh
+
+FILE=$1
+
+rm -f $MESON_DIST_ROOT/$FILE
-- 
2.25.4

On Wed, May 06, 2020 at 12:16:24PM +0200, Andrea Bolognani wrote:
> Both .mailmap and .travis.yml only make sense in the context of a
> git checkout, so we should not include them in releases.

Does it really matter if we include them though ? AFAIK they don't cause
any harm, and it feels like we're just creating extra work for ourselves
by deciding to exclude them.  The same can be said of the .gitlab-ci.yml
file, and supporting things we'll be introducing like the dockerfiles.

One of the things I like about meson is that we can essentially stop
worrying about what goes into the dist tarball, and simply let it have
the entire of git contents (aside from the couple of generated files
we added). So my gut feeling says to not bother trying to exclude files.

> 
> Signed-off-by: Andrea Bolognani <abologna@redhat.com>
> ---
>  meson.build     | 9 +++++++++
>  tools/nodist.sh | 5 +++++
>  2 files changed, 14 insertions(+)
>  create mode 100755 tools/nodist.sh
> 
> diff --git a/meson.build b/meson.build
> index e765ed6..341abe9 100644
> --- a/meson.build
> +++ b/meson.build
> @@ -263,6 +263,15 @@ if git
>      foreach file : [ 'libvirt-dbus.spec', 'AUTHORS.rst' ]
>          meson.add_dist_script('tools/dist.sh', meson.build_root(), file)
>      endforeach
> +
> +    nodist_files = [
> +        '.mailmap',
> +        '.travis.yml',
> +    ]
> +
> +    foreach file : nodist_files
> +        meson.add_dist_script('tools/nodist.sh', file)
> +    endforeach
>  endif
>  
>  
> diff --git a/tools/nodist.sh b/tools/nodist.sh
> new file mode 100755
> index 0000000..f820147
> --- /dev/null
> +++ b/tools/nodist.sh
> @@ -0,0 +1,5 @@
> +#!/bin/sh
> +
> +FILE=$1
> +
> +rm -f $MESON_DIST_ROOT/$FILE
> -- 
> 2.25.4
> 

Regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|

On Wed, 2020-05-06 at 11:29 +0100, Daniel P. Berrangé wrote:
> On Wed, May 06, 2020 at 12:16:24PM +0200, Andrea Bolognani wrote:
> > Both .mailmap and .travis.yml only make sense in the context of a
> > git checkout, so we should not include them in releases.
> 
> Does it really matter if we include them though ? AFAIK they don't cause
> any harm, and it feels like we're just creating extra work for ourselves
> by deciding to exclude them.  The same can be said of the .gitlab-ci.yml
> file, and supporting things we'll be introducing like the dockerfiles.
> 
> One of the things I like about meson is that we can essentially stop
> worrying about what goes into the dist tarball, and simply let it have
> the entire of git contents (aside from the couple of generated files
> we added). So my gut feeling says to not bother trying to exclude files.

It's certainly not a deal breaker to have a couple extra files in the
release archives, but I think it's much tidier not to have them
there. It's really not that much extra work anyway... But, if you and
Pavel share the opinion that we shouldn't bother, I will not try to
push this further :)

-- 
Andrea Bolognani / Red Hat / Virtualization