Re: teTeX on HURD (Was: My Hurd speaks PostScript)
>>>>> Marcus Brinkmann writes:
MB> On Sat, Nov 20, 1999 at 09:52:11AM +0530, Kapil H. Paranjape
MB> wrote:
>> I didn't incorporate libtool since: (a) I wanted to do the build
>> with minimal changes. (sheer laziness ...) (b) The klibtool
>> script said it was a modification of the libtool scripts so I
>> didn't want to go into those modifs.
>>
>> But I suppose I should have tried.
Please... never try to convert a package to libtool unless you have
discussed it with the package's maintainer, and they say they want
your help. As with any Debian package, there is no point in making
major changes unless you know for sure that the upstream maintainer
wants it done the same way that you plan on doing it.
MB> Not necessarily. It's a good idea to keep changes minimal by all
MB> means. I too was too lazy to investigate the diff between
MB> klibtool and libtool, maybe it's a trivia, maybe not. It only
MB> matters if the produced libraries are incorrect.
The differences are probably pretty big, but I suspect they make
compatible sonames.
For those interested in history, Karl Berry was one of the people who
was involved with the early development stages of libtool. He
intended to use it for a TeX release, but libtool-0.5 (summer, 1996)
was a bloated, shakey mess. I began a total rewrite of libtool,
but I didn't think it would be ready in time for Karl to use.
And so, I recommended that he scavenge code from libtool 0.5, and hack
together his own script. That's what he did, and that's what klibtool
is. It's simpler and more effective than his old shared library
system, but it would still be nice to move TeX to pure libtool one
day.
Just make sure you do it with the maintainer's blessing!
--
Gordon Matzigkeit <gord@fig.org> //\ I'm a FIG (http://www.fig.org/)
Committed to freedom and diversity \// I use GNU (http://www.gnu.org/)
Reply to: