You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@apr.apache.org by Rainer Jung <ra...@kippdata.de> on 2013/10/24 01:39:09 UTC

Re: svn commit: r1535157 - /apr/apr/trunk/configure.in

On 23.10.2013 22:22, jorton@apache.org wrote:
> Author: jorton
> Date: Wed Oct 23 20:22:57 2013
> New Revision: 1535157
> 
> URL: http://svn.apache.org/r1535157
> Log:
> * configure.in: Fix r1535027 for Linux kernel versions x.0.y.
> 
> Modified:
>     apr/apr/trunk/configure.in
> 
> Modified: apr/apr/trunk/configure.in
> URL: http://svn.apache.org/viewvc/apr/apr/trunk/configure.in?rev=1535157&r1=1535156&r2=1535157&view=diff
> ==============================================================================
> --- apr/apr/trunk/configure.in (original)
> +++ apr/apr/trunk/configure.in Wed Oct 23 20:22:57 2013
> @@ -681,7 +681,7 @@ case $host in
>          ;;
>      *linux*)
>          os_major=[`uname -r | sed -e 's/\([1-9][0-9]*\)\..*/\1/'`]
> -        os_minor=[`uname -r | sed -e 's/[1-9][0-9]*\.\([1-9][0-9]*\)\..*/\1/'`]
> +        os_minor=[`uname -r | sed -e 's/[1-9][0-9]*\.\([0-9]+\)\..*/\1/'`]

Solaris sed only supports basic regexp, which knows about "*" but not "+".

So instead of "[0-9]+" it could be "[0-9][0-9]*" or "[0-9]\{1,\}",
whichever you find less annoying.

Regards,

Rainer

Re: svn commit: r1535157 - /apr/apr/trunk/configure.in

Posted by Rainer Jung <ra...@kippdata.de>.
On 25.10.2013 12:27, Joe Orton wrote:
> On Thu, Oct 24, 2013 at 01:39:09AM +0200, Rainer Jung wrote:
>> On 23.10.2013 22:22, jorton@apache.org wrote:
>>> --- apr/apr/trunk/configure.in (original)
>>> +++ apr/apr/trunk/configure.in Wed Oct 23 20:22:57 2013
>>> @@ -681,7 +681,7 @@ case $host in
>>>          ;;
>>>      *linux*)
>>>          os_major=[`uname -r | sed -e 's/\([1-9][0-9]*\)\..*/\1/'`]
>>> -        os_minor=[`uname -r | sed -e 's/[1-9][0-9]*\.\([1-9][0-9]*\)\..*/\1/'`]
>>> +        os_minor=[`uname -r | sed -e 's/[1-9][0-9]*\.\([0-9]+\)\..*/\1/'`]
>>
>> Solaris sed only supports basic regexp, which knows about "*" but not "+".
>>
>> So instead of "[0-9]+" it could be "[0-9][0-9]*" or "[0-9]\{1,\}",
>> whichever you find less annoying.
> 
> Surely we don't have to care about what works with Solaris sed on a 
> *linux* host?

Hah, caught me ;)

Regards,

Rainer


Re: svn commit: r1535157 - /apr/apr/trunk/configure.in

Posted by Joe Orton <jo...@redhat.com>.
On Thu, Oct 24, 2013 at 01:39:09AM +0200, Rainer Jung wrote:
> On 23.10.2013 22:22, jorton@apache.org wrote:
> > --- apr/apr/trunk/configure.in (original)
> > +++ apr/apr/trunk/configure.in Wed Oct 23 20:22:57 2013
> > @@ -681,7 +681,7 @@ case $host in
> >          ;;
> >      *linux*)
> >          os_major=[`uname -r | sed -e 's/\([1-9][0-9]*\)\..*/\1/'`]
> > -        os_minor=[`uname -r | sed -e 's/[1-9][0-9]*\.\([1-9][0-9]*\)\..*/\1/'`]
> > +        os_minor=[`uname -r | sed -e 's/[1-9][0-9]*\.\([0-9]+\)\..*/\1/'`]
> 
> Solaris sed only supports basic regexp, which knows about "*" but not "+".
> 
> So instead of "[0-9]+" it could be "[0-9][0-9]*" or "[0-9]\{1,\}",
> whichever you find less annoying.

Surely we don't have to care about what works with Solaris sed on a 
*linux* host?

Regards, Joe