Similar to commit c3c2cc6, use a literal newsline instead of \n
inside the brackets.
---
Pushed as a build fix.

 docs/schemas/news.rng | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/docs/schemas/news.rng b/docs/schemas/news.rng
index ec4313f..a633e22 100644
--- a/docs/schemas/news.rng
+++ b/docs/schemas/news.rng
@@ -45,7 +45,9 @@
     <element name="change">
       <element name="summary">
         <data type="string">
-          <param name="pattern">\n[^\n]+\n +</param>
+          <!-- Use literal newline instead of \n for bug in libxml2 2.7.6 -->
+          <param name="pattern">\n[^
+]+\n +</param>
         </data>
       </element>
       <optional>
-- 
2.10.2

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

On Tue, Apr 04, 2017 at 09:05:58 +0200, Ján Tomko wrote:
> Similar to commit c3c2cc6, use a literal newsline instead of \n
> inside the brackets.
> ---
> Pushed as a build fix.

This is ugly. How about dropping support for platforms which can't use
new enough libraries rather than band-aiding the code?

Nobody in their right mind would use upstream software with very old
distros.
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

On Tue, Apr 04, 2017 at 09:12:11AM +0200, Peter Krempa wrote:
> On Tue, Apr 04, 2017 at 09:05:58 +0200, Ján Tomko wrote:
> > Similar to commit c3c2cc6, use a literal newsline instead of \n
> > inside the brackets.
> > ---
> > Pushed as a build fix.
> 
> This is ugly. How about dropping support for platforms which can't use
> new enough libraries rather than band-aiding the code?
> 
> Nobody in their right mind would use upstream software with very old
> distros.

We still target RHEL6 as a platform for libvirt to build on and as such
people reasonably expect tests to pass when built there. It is mildly
ugly but not a serious maint burden to us long term.

Regards,
Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://entangle-photo.org       -o-    http://search.cpan.org/~danberr/ :|

--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list

On Tue, Apr 04, 2017 at 10:52:39 +0100, Daniel Berrange wrote:
> On Tue, Apr 04, 2017 at 09:12:11AM +0200, Peter Krempa wrote:
> > On Tue, Apr 04, 2017 at 09:05:58 +0200, Ján Tomko wrote:
> > > Similar to commit c3c2cc6, use a literal newsline instead of \n
> > > inside the brackets.
> > > ---
> > > Pushed as a build fix.
> > 
> > This is ugly. How about dropping support for platforms which can't use
> > new enough libraries rather than band-aiding the code?
> > 
> > Nobody in their right mind would use upstream software with very old
> > distros.
> 
> We still target RHEL6 as a platform for libvirt to build on and as such
> people reasonably expect tests to pass when built there. It is mildly
> ugly but not a serious maint burden to us long term.

Yes I know we still target it. That's why I said we should drop it. I
don't expect it to happen, but the constant breakages due to symbol
shadowing (since gcc is too old) or this bother me. And I actually don't
see much use in keeping it. Using upstream libvirt on rhel6 would be
insane for a production environment. Using rhel6 for a development
environment is insane by itself.
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list