RE: Opera's SSE Tests on GitHub [Was: <different topic>]

On Mon, 26 Nov 2012, SULLIVAN, BRYAN L wrote:

>> As far as I know there is nothing special compared to the default
>> apache+PHP setup on w3c-test.org. The relevant PHP files are in the
>> resources/ directory of the testsuite. Is there something that doesn't
>> seem to be working correctly?
>
> To implement SSE using PHP for server scripting, I have found a number of SSE dependencies that require configuration of PHP, Apache (or other server), or the intersection of PHP and the underlying web server. These include:
> - turn output compression off (Apache)
> - turn output buffering off (Apache?)
> - set implicit flush (PHP)
> - PHP 5.4 (I have not been able to get SSE working with PHP 5.3)
>
> I'm not sure of the interaction of these, but I have found very little 
> useful information about server configuration for SSE, and had to spend 
> a long time hacking on the server config to get it to work. Not being a 
> real web developer ;-) my experience may be anecdotal but I think it 
> represents an important data point and indicates a real need for 
> documentation. But I'll be glad to kick the tires of the Opera tests 
> once they are on the W3C test server, to see if and how it works there.

It is possible that there's something magic in our setup, but I don't see 
anything obvious. I haven't dug through the PHP config or anything yet 
though. But this is all very hypothetical. If the tests work on the W3C 
server maybe we don't need to have this conversation at all. Is there some 
evidence that they don't?

> And PHP is only one of the server options (and a poor man's one at that, 
> AFAICT). Shouldn't we be testing this with different server 
> environments?

Not obviously. I mean as long as we have one server that correctly 
fulfills its side of the spec bargin that is sufficient for our purposes; 
the goal is to ensure that the clients implement the spec correctly, not 
determine the relative merits of PHP vs node.js vs twisted vs whatever for 
the server side.

As it happens I think we are going to need a custom server for testing 
anyway, and am currently trying to gather requirements. Ability to test 
server sent events in a natural way seems like a reasonable one.

Received on Monday, 26 November 2012 22:24:09 UTC