Re: Re: 4.2.3 and the new PEAR package
| From: | Adam Maccabee Trachtenberg | Date: | Sun, 25 Aug 2002 15:52:09 +0000 |
| Subject: | Re: Re: 4.2.3 and the new PEAR package | ||
| References: | 1 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-8652@lists.php.net to get a copy of this message | ||
On Sun, 25 Aug 2002, Pierre-Alain Joye wrote:
> On Sun, 25 Aug 2002 14:32:13 +0200
> "Tomas V.V.Cox" <cox@idecnet.com> wrote:
>
> > Pierre-Alain Joye wrote:
> > >
> > I've only received from you problems with go-pear. Is there any problem
> > with the pear cmd under Windows? Umm, perhaps the branch does not
> > include the lastest versions.
>
> There is problem with env variables, I solved them using getenv and
> tests before. Especially with system_root and tmp.
FWIW, output buffering for go-pear still doesn't work correctly for me
under 4.3.0-dev. (This is Redhat 7.3.) If I explictly set
output_buffering=off in my php.ini file, everything is fine; however,
with output_buffering=4096 (default for php.ini-recommended) I can't
get it to work correctly even if I add "-d output_buffering=off" on
the command line.
On a related note, I've been playing with "pear build" on Unix and
I've stumbled across some similar $_ENV bugs. There's an $_ENV['HOME']
that fails; I'm currently working on fixing up an opendir('modules')
call that seems to be an include_path issue. Since php.ini-recommended
removes E from variables_order, PEAR authors should never assume they
can use $_ENV, IMHO.
I am working on a patch to pear once I track down all the issues.
-adam
--
adam maccabee trachtenberg
adam@trachtenberg.com