[lfs-dev] chapter05/expect.xml

Milton Ott milton at mettech.me
Fri Jul 10 19:59:49 PDT 2015

Hi Bruce Dubbs, Armin K and Nathan.

Thanks for replying.  I allowed my email client to reformat the patch
I sent and only
noticed it when checking the mailing list archive. In hindsight I
should of just introduced
myself . My one and only chance at making a good first impression is
gone now. I guess
I tried too hard to and ended up doing the opposite. Such is life I guess.

For posterity if nothing else here is the patch proper.

chapter05/expect.xml | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)

Index: LFS-20150628-systemd/chapter05/expect.xml
--- LFS-20150628-systemd.orig/chapter05/expect.xml
+++ LFS-20150628-systemd/chapter05/expect.xml
@@ -46,8 +46,7 @@
    This will ensure that our test suite tools remain sane for the
final builds of our

-<screen><userinput remap="pre">cp -v configure{,.orig}
-sed 's:/usr/local/bin:/bin:' configure.orig > configure</userinput></screen>
+<screen><userinput remap="pre">sed -i.orig 's:/usr/local/bin:/bin:'

    <para>Now prepare Expect for compilation:</para>


I've sent a small paypal donation using the donation button. Tx ID
Bye for now.

On Sat, Jul 11, 2015 at 4:41 AM, Armin K. <krejzi at email.com> wrote:
> On 10.07.2015 19:16, Bruce Dubbs wrote:
>> Armin K. wrote:
>>> On 10.07.2015 17:01, Nathan Coulson wrote:
>>>> On 10 July 2015 at 07:13, Milton Ott <milton at mettech.me> wrote:
>>>>> All apologies lfs-dev, this a great start isn't it. The patch is no
>>>>> good. I forgot to double check
>>>>> before sending and its come back to bite me.
>>> Patience is the key. Main developer is from USA, which means you're
>>> almost a day ahead of him. Having no response doesn't mean it's not
>>> good, but rather it's not read at the moment.
>>>> Originally chapter 5 was written that way due to older versions of sed
>>>> not supporting -i (allowing it to work on older hosts).  Not sure if
>>>> this is still a concern or not.
>>> Minimum listed sed version is 4.1.5 and that version supports -i.orig
>>> just fine. I just downloaded the version and built it to test. Note
>>> that the requirement is GNU sed, not any other, so the change is fine
>>> by me.
>> The change is fine, but I don't think necessary.  One of the objectives if LFS is to show how commands are used so users can become familiar with them.  You still see the two line format occasionally and we certainly use the -i form in most places.
>>   -- Bruce
> That's exactly a good reason to use it. There's no example that I've yet
> seen using such command but there are several ones with renaming the files
> and utilizing a sed after that.
> --
> Note: My last name is not Krejzi.
> --
> http://lists.linuxfromscratch.org/listinfo/lfs-dev
> FAQ: http://www.linuxfromscratch.org/faq/
> Unsubscribe: See the above information page

More information about the lfs-dev mailing list