Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 19 Dec 2007 06:45:11 -0800
From:      Jeremy Chadwick <koitsu@FreeBSD.org>
To:        Barry Byrne <barry.byrne@wbtsystems.com>
Cc:        ports@FreeBSD.org, hiromi@tac.tsukuba.ac.jp, beech@FreeBSD.org, perky@freebsd.org
Subject:   Re: FreeBSD Port: horde-base-3.1.5_1
Message-ID:  <20071219144511.GA15338@eos.sc1.parodius.com>
In-Reply-To: <006501c8424b$975d8c90$c5010c0a@SUNYA>
References:  <000f01c8422b$3515d7b0$c5010c0a@SUNYA> <20071219113418.GA11378@eos.sc1.parodius.com> <006501c8424b$975d8c90$c5010c0a@SUNYA>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Dec 19, 2007 at 02:29:40PM -0000, Barry Byrne wrote:
> > -----Original Message-----
> > From: Jeremy Chadwick [mailto:koitsu@FreeBSD.org] 
> > Sent: 19 December 2007 11:34
> > To: Barry Byrne
> > Cc: beech@FreeBSD.org; ports@FreeBSD.org; 
> > hiromi@tac.tsukuba.ac.jp; ache@freebsd.org; perky@freebsd.org
> 
> > > Checked the patched file (NLS.php) and I had something like this;
> > > 
> > >  \xa0 \xa0 \xa0 \xa0/* avoid FreeBSD issapce(3) bug */
> > >  \xa0 \xa0 \xa0 \xa0if(NLS::getCharset() == "UTF-8"){
> > >  \xa0 \xa0 \xa0 \xa0 \xa0 \xa0setlocale(LC_CTYPE,"C");
> > >  \xa0 \xa0 \xa0 \xa0}
> > > 
> > > When I replaced the \xa0 with tabs, all was fine again.
> > 
> > The parse error is probably because PHP does not have Unicode or
> > multibyte character support enabled.  The patch file in question
> > contains actual Unicode characters.  You might try rebuilding PHP with
> > Zend support for multibyte characters (it's an option in 
> > "make config")
> > and see if that improves things.
> 
> Jeremy,
> 
> Thanks for the quick response.
> 
> You're correct in assuming I don't have Multibyte enabled.
> 
> However, I just rebuilt php on another machine and enabled multibyte,
> then built horde-base on that, but still had the same characters
> in the NLS.php file on that, so not sure what the problem is.

Okay, so it's probably a Unicode vs. UTF-8 parser thing.

My comment about the code needing to remain in ASCII stands, though
(this isn't your fault, of course).  The patch should really be fixed to
use literal whitespace (ASCII 0x20) or tabs (ASCII 0x09).

I have commit access to fix this, but I want to clear it + discuss with
those involved with the PR first.

-- 
| Jeremy Chadwick                                    jdc at parodius.com |
| Parodius Networking                           http://www.parodius.com/ |
| UNIX Systems Administrator                      Mountain View, CA, USA |
| Making life hard for others since 1977.                  PGP: 4BD6C0CB |




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20071219144511.GA15338>