Re: RFC: CLI behave like SH or PERL/RUBY/PYTHON?
| From: | Yasuo Ohgaki | Date: | Wed, 23 Oct 2002 08:22:23 +0000 |
| Subject: | Re: RFC: CLI behave like SH or PERL/RUBY/PYTHON? | ||
| References: | 1 2 | Groups: | php.dev |
| Request: | Send a blank email to php-dev+get-89969@lists.php.net to get a copy of this message | ||
Edin Kadribasic wrote:
> OTOH, having implicit_flush turned on makes writing interactive command line
> programs easier. Some programs (like pear installer) might even depend on it.
It may, but the current implementation is broken as I
mentioned in other mail.
I don't argue with usefulness of implicit flushing. It
should be optional and should behave more consistently.
> The way cli overwrites some ini settings is a whole different issue and I
> don't thinks we should start mixing the two on this thread.
Ok. Then we can narrow down a bit.
PG(implicit_flush)=1 in php_cli.c was bogus as described
in other mail in this thread.
It's better to have more generic/consistent way of
setting the implicit flushing. Any objection about
this?
Whether I like it or not, ob_implicit_flush() and/or
OG(implicit_flush) may be used to turn on/off
SAPI level flushing. This should be be enough.
(current ob_implicit_flush() is just enabling SAPI
level flushing)
To be honest, I prefer to have implicit_flush() and use
PG(implicit_flush) instead of using ob_implicit_flush()
for SAPI flushing.
PS: Almost all script should not be affected by this
change, since php.ini setting is overwritten and implicit
flushing was disabled by php.ini.
i.e. php.ini-dist/recommended users were not affected by
PG(implicit_flush)=1 in php_cli.c.
--
Yasuo Ohgaki