Uups, missed the list... > -----Original Message----- > From: Daniel Veillard [mailto:veillard redhat com] > Sent: Thursday, March 28, 2002 4:43 PM > To: Henke, Markus > Cc: 'xml gnome org' > Subject: Re: [xml] Current version of config.guess > > > On Thu, Mar 28, 2002 at 04:48:24PM +0100, Henke, Markus wrote: > > Hello Daniel, > > > > the "./configure" process of libxml2-2.4.19 doesn't > > recognize the new MPE platform (MPE/iX 7.0) and breaks > > with an appropriate message. Well, that's propably no > > problem for anybody else but me 8), but i've tested > > that it works with the current version of "config.guess" > > (ftp://ftp.gnu.org/pub/gnu/config/config.guess), so it's > > maybe a good idea to update that file in libxml2!? > > Could be that this will prevent problems with other > > "exotic" platforms... > > The auto* tools drive me nuts each time I try to change > anything in them. So I stick with the version provided > with my OS, currently Red Hat 7.2 . I you want to update them > go ahead, you're more courageous than me. Good joke, i havn't got the foggiest idea about auto* tools :] I've just replaced the file "config.guess" in libxml2 with the version provided on the GNU FTP server. > > Another, but related thing: > > It's been a while that you've asked me for the platform > > information for MPE/iX with the prospect that it's maybe > > possible to force the enabling of trio in "configure" for > > that platform (which is necessary) > > (http://mail.gnome.org/archives/xml/2002-January/msg00184.html). > > Is there a chance that this will be added to forthcoming > > versions of libxml2 (It's "hppa1.0-hp-mpeix")? > > For the moment i still hack "configure" with a static > > NEED_TRIO=1, which is ugly but works... > > Is the enclosed patch working (you will need to rebuild configure) ? Surprise: I havn't auto* tools on this platform... :) However, i've applied the patch to the already generated "configure" (the position is obvious) but it doesn't work since WITH_TRIO is overwritten later in "configure" where NEED_TRIO is evaluated. I've fixed the patch (WITH_TRIO -> NEED_TRIO), i guess it was just a typo. The modified patch is attached. With NEED_TRIO it works perfect. > Daniel Many thanx, Markus
Attachment:
mpe.patch
Description: Binary data