Hi Sergey,

/publish page isn't a part of the protocol, it's an extra feature
implemented by http://pubsubhubbub.appspot.com.

The new version of the spec we are working on (
https://superfeedr-misc.s3.amazonaws.com/pubsubhubbub-core-0.4.html)
doesn't specify how the publishing happens at all. Hubs are free to
implement their own policies, and of course it's a good practice to
document them.

I added a TODO for myself to document the presence of the /publish handler
on http://pubsubhubbub.appspot.com.

Roman.

On Sun, Aug 19, 2012 at 11:26 AM, Sergey Lukin <[email protected]>wrote:

> I just tried sending a POST request with query string parameters and empty
> body to http://pubsubhubbub.appspot.com and received 204 No Content
> response.
>
> Here is the Request:
>
> POST 
> /publish?hub.mode=publish&hub.url=http%3A%2F%2Fsergeylukin.com%2Ffeed&something=else
>  HTTP/1.1
> User-Agent: Fiddler
> Host: pubsubhubbub.appspot.com
> Content-length: 0
>
>
> and here is the Response:
>
> HTTP/1.1 204 No Content
> Cache-Control: no-cache
> Content-Type: text/plain
> Expires: Fri, 01 Jan 1990 00:00:00 GMT
> Date: Sat, 18 Aug 2012 16:15:35 GMT
> Server: Google Frontend
> Content-Length: 0
>
>
> What do you think?
> Does it mean that pubsubhubbub.appspot.com is running a different version
> (not 0.3) of PSHB specification?
>
>
> On Sunday, October 18, 2009 11:20:38 PM UTC+2, Tim Bray wrote:
>>
>> Since all the publisher is sending is a couple of fields, why does it
>> have to send a request body?  For example, to hook up my own blog, all
>> I have to send is
>>
>> hub.mode=publish&hub.url=http:**//www.tbray.org/ongoing/**ongoing.atom<http://www.tbray.org/ongoing/ongoing.atom>
>>
>> So I'm wondering why I can't just POST an empty body to
>>
>> http://pubsubhubbub.appspot.**com/?hub.mode=publish&hub.url=**
>> http://www.tbray.org/ongoing/**ongoing.atom<http://pubsubhubbub.appspot.com/?hub.mode=publish&hub.url=http://www.tbray.org/ongoing/ongoing.atom>
>>
>> No content-type, no content-length, less work all around?
>>
>>  -T
>>
>>

Reply via email to