syntax error at (eval 289) line 1, near "require nothing:::"
{
'status' => 500,
'content' => {},
'route' => undef,
'mojo.started' => [
1279789227,
662131
],
...
...
I'm affraid, this can be somehow exploitable..
Regards,
Tomas Hruby
What I suggest is to use WebSockets + Pure Perl Ping implementation,
just not to bother with forks, output capturing and so on.
2010/7/22 Sebastian Riedel <kra...@googlemail.com>:
>> http://community.livejournal.com/ru_perl/366844.html
>
> That sentence doesn't make any sense and i don't speak russian...
http://en.wikipedia.org/wiki/HTTP_pipelining
You received this message because you are subscribed to the Google Groups "Mojolicious" group.
To post to this group, send email to mojol...@googlegroups.com.
To unsubscribe from this group, send email to mojolicious...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/mojolicious?hl=en.
What about it?
Non buffered output is a myth, there is pretty much always a buffer involved on some layer, especially portability is a huge problem.
The closest you will get in Mojo/Mojolicious is the response callback.
$self->res->body(sub {...});
Which means you don't actually process the request when it hits your Mojolicious controller, but later when the server binding invokes the callback.
$self->res->code(200);
$self->res->headers->content_length(...);
$self->res->body(sub {
# Do stuff and return response body chunk wise
});
Note that you need to know the Content-Length in advance or use chunked Transfer-Encoding.
> Same feature exists in CGI – nph scripts.
Also a myth, the server can still do whatever it wants, there are no guarantees.
To make it absolutely clear, what you are proposing here would mean that whenever one of your handlers writes a chunk of information, the whole server process would have to block and wait for the data to travel all across the internet and back, before being able to do anything else again.
Doesn't sound like such a good idea, does it? :)