Archaic wrote:
On Sun, Dec 25, 2005 at 11:41:21AM +, Andrew Benton wrote:
The sooner it's merged, the more testing it will get before LFS-6.2
Agreed.
And thirded. Any of the editors should feel free to do the merge
whenever they have the time and inclination to do so. If it's not
me
On Mon, Dec 26, 2005 at 01:04:13PM +, Matthew Burgess wrote:
>
> And thirded. Any of the editors should feel free to do the merge
> whenever they have the time and inclination to do so. If it's not
> merged by the time I get back from my break I'll merge it myself.
Done.
--
Archaic
Wan
On Fri, 23 Dec 2005, Dan Nicholson wrote:
I haven't done any research yet, but I'm attaching the ICA report for
1v2. With the exception of farce-extras (too big to move around), you
can see the results in http://students.washington.edu/dbnichol/lfs/ .
I'm going out of town in the morning, so
On 12/23/05, Dan Nicholson <[EMAIL PROTECTED]> wrote:
> I haven't done any research yet, but I'm attaching the ICA report for
> 1v2. With the exception of farce-extras (too big to move around), you
> can see the results in http://students.washington.edu/dbnichol/lfs/ .
One thing that looks strang
On Mon, Dec 26, 2005 at 12:00:46PM -0700, Archaic wrote:
>
> Done.
And reverted. :/
Patrakov, here's a list of things outstanding (might not be all
inclusive):
1) All the new patches need to be properly formatted to the LFS style.
You can use
http://www.linuxfromscratch.org/patches/downloads/MA
Archaic wrote these words on 12/26/05 14:03 CST:
> I noticed Randy had pointed out some things, but didn't look closely,
> so if there might be overlap in what he saw and what I saw.
I'm just trying to point out typos/spelling/grammar fixes. There
is probably no conflict in anything you've pointe
On Mon, Dec 26, 2005 at 01:03:47PM -0700, Archaic wrote:
>
> 3) GDBM. Can Man-DB use Berkeley? If so, would that be preferable?
> (Would also allow us to remove the arpd sed in IPRoute2.
Assuming GDBM stays, should we install the compatability library? There
are other instances where LFS does not
On Mon, Dec 26, 2005 at 02:06:53PM -0600, Randy McMurchy wrote:
>
> I'm just trying to point out typos/spelling/grammar fixes. There
> is probably no conflict in anything you've pointed out.
Not yet. :) But don't worry, I'll leave the grammar stuff to you. It's
not my strong suit. ;)
--
Archaic
There are some policy decisions being made in the UTF-8 patch that
should be commented on here first before implementing the patch.
Likewise, some commentary needs to be rethought.
I have a rendered copy of the (now reverted) patched book here:
http://linuxfromscratch.org/~archaic/utf8-book
From
Archaic wrote:
There are some policy decisions being made in the UTF-8 patch that
should be commented on here first before implementing the patch.
Likewise, some commentary needs to be rethought.
I have a rendered copy of the (now reverted) patched book here:
http://linuxfromscratch.org/~archa
Archaic wrote:
On Mon, Dec 26, 2005 at 01:03:47PM -0700, Archaic wrote:
3) GDBM. Can Man-DB use Berkeley? If so, would that be preferable?
(Would also allow us to remove the arpd sed in IPRoute2.
Assuming GDBM stays, should we install the compatability library? There
are other instanc
On Tue, Dec 27, 2005 at 07:50:24AM +0500, Alexander E. Patrakov wrote:
>
> Do this as you wish. Either GDBM (with or without the compatibility
> library) or Berkeley DB is good for Man-DB. I chose GDBM because of
> smaller package size.
Okay, I will start another thread about just this topic as
In order to make LFS usable in UTF-8 locales, and different man program
was chosen, man-DB. That program requires a database backend. It can
support GDBM or Berkeley DB. I'll list what I think are the pros of
each.
GDBM:
1) Small
2) Easier to install
3) No external dependencies
BDB:
1) More widel
Archaic wrote these words on 12/26/05 21:48 CST:
> Please comment on this thread with your choice and the reasoning for
> your choice.
To me, it is a no-brainer. Berkeley-DB should be installed. We
already perform an abortion in one LFS package so that BDB doesn't
have to be installed, and then t
On Mon, Dec 26, 2005 at 10:06:59PM -0600, Randy McMurchy wrote:
>
> It can be argued that neither of the above is a valid reason to
> choose GDBM. What exactly is easier?
cd build_unix &&
../dist/configure --prefix=/usr \
--enable-compat185 \
--enable-cxx &&
make LIBSO_LIBS="-lpthread" LI
Randy McMurchy wrote these words on 12/26/05 22:06 CST:
> Archaic wrote these words on 12/26/05 21:48 CST:
>>Please comment on this thread with your choice and the reasoning for
>>your choice.
>
> To me, it is a no-brainer. Berkeley-DB should be installed.
I must admit that I didn't lay out all t
On Mon, Dec 26, 2005 at 10:22:45PM -0600, Randy McMurchy wrote:
>
> 1. GDBM is an unmaintained package that won't change. This is good
> and bad. Good because we can count on it being predictable, bad in
> that if there is ever something in man-db that needs updates, GDBM
> will not work any longe
Archaic wrote these words on 12/26/05 22:20 CST:
> On Mon, Dec 26, 2005 at 10:06:59PM -0600, Randy McMurchy wrote:
>
>>It can be argued that neither of the above is a valid reason to
>>choose GDBM. What exactly is easier?
>
> cd build_unix &&
> ../dist/configure --prefix=/usr \
> --enable-com
Archaic wrote these words on 12/26/05 22:31 CST:
> The LFS Book is *NOT* producing a UTF-8 only or UTF-8 default system
> (ala Fedora Core). The ability to use UTF-8 is being implemented, but
> the choice is still on the end user.
I saw *many* changes and patches that affect core LFS packages in
On Mon, Dec 26, 2005 at 11:05:03PM -0600, Randy McMurchy wrote:
>
> I saw *many* changes and patches that affect core LFS packages in
> the UTF patch applied (for only a short while) today. Many of which
> looked to me as though it changed core functionality.
The patches add UTF support. There ar
Randy McMurchy wrote:
Archaic wrote these words on 12/26/05 22:31 CST:
The LFS Book is *NOT* producing a UTF-8 only or UTF-8 default system
(ala Fedora Core). The ability to use UTF-8 is being implemented, but
the choice is still on the end user.
I saw *many* changes and patches that
On 12/26/05, Archaic <[EMAIL PROTECTED]> wrote:
> Please comment on this thread with your choice and the reasoning for
> your choice.
I prefer gdbm coz the bdb api keeps changing with every new release.
An added benefit of gdbm is the small number of required dependencies.
--
Tushar Teredesai
On Tue, Dec 27, 2005 at 12:44:02AM -0500, Tushar Teredesai wrote:
>
> I prefer gdbm coz the bdb api keeps changing with every new release.
> An added benefit of gdbm is the small number of required dependencies.
I guess we really should find out if the man-db devs have a preference.
Not all apps
Archaic wrote:
BDB:
1) More widely used as a DB backend
BDB surely has many techincal advantages. But there is one problem for
the hybrid of the UTF-8 and alphabetical books: Perl looks for GDBM and
BDB as for optional dependencies. If we build the database library after
Perl, differences
On Tue, Dec 27, 2005 at 10:54:40AM +0500, Alexander E. Patrakov wrote:
>
> BDB surely has many techincal advantages. But there is one problem for
> the hybrid of the UTF-8 and alphabetical books: Perl looks for GDBM and
> BDB as for optional dependencies. If we build the database library after
Archaic wrote:
On Tue, Dec 27, 2005 at 12:44:02AM -0500, Tushar Teredesai wrote:
I prefer gdbm coz the bdb api keeps changing with every new release.
An added benefit of gdbm is the small number of required dependencies.
I guess we really should find out if the man-db devs have a pref
Randy McMurchy wrote:
Archaic wrote these words on 12/26/05 23:29 CST:
Is BLFS going to adopt /usr/X11R7? If not, then it will obviously break
your build method. And since /usr/X11R7 will suffer the same problems as
/usr WRT to assumptions made by other packages, would now be the time to
defa
On Tue, Dec 27, 2005 at 10:56:16AM +0500, Alexander E. Patrakov wrote:
> >
> Debian builds man-db with GDBM. The packager and the author of man-db is
> the same person, so you can count this as a dev's preference.
There could be many reasons for his choice, one likely one could be it
was easier t
Archaic wrote these words on 12/26/05 23:41 CST:
> Forgive my pedantic need, but are you hoping /usr doesn't get used or
> /usr/X11R7 doesnt' get used?
Trying to answer both yours and Chris' questions at once.
I like large packages installed in their own neat little place. This
is how *I* like i
Hello,
I agree with Randy that the concept of dead keys should be explained in
the UTF-8 book. Even MSDN does this. The question is what is the proper
form of such explanation. Available alternatives:
1) Add a short entry on the "Acronyms and Terms" page.
2) Add a section on the "Configuring
Randy McMurchy wrote:
Keep in mind, that LFS has dropped perfectly good packages in favor
of another (dropped net-tools for ...hmm...can't even remember as I
don't use it) because it is unmaintained.
This alone should be a reason to absolutely discard GDBM as a choice
for support of man-db.
Archaic wrote:
In order to make LFS usable in UTF-8 locales, and different man program
was chosen, man-DB. That program requires a database backend. It can
support GDBM or Berkeley DB.
Yet another alternative would be to leave the usual Man package in the
book, configure it with "+lang none" (
I wrote:
Yet another alternative would be to leave the usual Man package in the
book, configure it with "+lang none" (see explanation in the man-i18n
hint) and import the whole "Hacks" section from the man-i18n hint. But
I think that the instructions would contain too many "ifs" then (i.e.:
t
Alexander E. Patrakov wrote these words on 12/27/05 00:15 CST:
> Forgot to say: if this variant is chosen, the LiveCD will continue using
> Man-DB, because users can't be expected to do any configuration on the
> LiveCD.
At this point, I think a new thread should be started, one where
the crux
Randy McMurchy wrote:
Alexander E. Patrakov wrote these words on 12/27/05 00:15 CST:
Forgot to say: if this variant is chosen, the LiveCD will continue using
Man-DB, because users can't be expected to do any configuration on the
LiveCD.
At this point, I think a new thread should be
35 matches
Mail list logo
