hc-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Roland Weber <ROLWE...@de.ibm.com>
Subject Re: 4.0 Requirements: Proxy chains
Date Wed, 12 Jan 2005 14:44:48 GMT
Hi Oleg,

what's wrong with having an InMemoryCookieStore in
http-cookie and a StdCredentialsProvider in http-auth?

When I first started using HttpClient (2.0 API), my
biggest problem was getting rid of cookie handling.
I know it can be switched off by now, but I still don't
like the idea of having to pass in an object that might
get modified, if I don't want it to be modified.

cheers,
  Roland




Oleg Kalnichevski <olegk@apache.org> 
12.01.2005 15:38
Please respond to
"HttpClient Project"


To
HttpClient Project <httpclient-dev@jakarta.apache.org>
cc

Subject
Re: 4.0 Requirements: Proxy chains






Odi, Roland

I wholeheartedly agree HttpState in its current form must go. I really
like tie idea to replace it with CredentialsProvider and CookieStore
interfaces and encourage the users to provide their own application
specific implementations. This said we still have provide a simple in
memory cookie store and a simple credentials store and I see no big deal
combining those two stores for the sake of simplicity. 

class InMemoryHttpState implements CookieStore, CredentialsProvider

Oleg

On Wed, Jan 12, 2005 at 10:08:41AM +0100, Roland Weber wrote:
> Hi Ortwin,
> 
> > Motivated by Roland Weber, I find the HttpState class bad. Especially 
> > the inheritance hierarchy seems weird. Why would HttpState (which 
sounds 
> 
> > very general) extend a CookieStore (which sounds very specific)?
> > 
> It was supposed to implement CookieStore and CredentialProvider,
> thereby pulling both functionalities together into one class.
> 
> cheers,
>   Roland

---------------------------------------------------------------------
To unsubscribe, e-mail: httpclient-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: httpclient-dev-help@jakarta.apache.org



Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message