Article delegate-en/986 of [1-5169] on the server localhost:119
  upper oldest olders older1 this newer1 newers latest
search
[Top/Up] [oldest] - [Older+chunk] - [Newer+chunk] - [newest + Check]
[Reference:<_A985@delegate-en.ML_>]
Newsgroups: mail-lists.delegate-en

[DeleGate-En] Re: Changing user instead of server, again
22 Jan 2001 09:12:35 GMT Christian Couder <pyebqbdyi-fa3r2xotsrfr.ml@ml.delegate.org>


On Mon, Jan 22, 2001 at 02:57:00PM +0900, Yutaka Sato wrote:
> On 01/19/01(19:25) you Christian Couder <pyebqbdyi-fa3r2xotsrfr.ml@ml.delegate.org> wrote
>  |'USER toto@titi..' is rewritten as 'pop://titi.com/toto'
>  |
>  |and that's a problem because I would like
>  |   
>  |'USER toto@titi..' rewritten as 'pop://localhost/toto@titi..'
> 
> Then add one more MOUNT like this:
> 
>   MOUNT="//* pop://localhost/*[%1@%0]"
> 
> The input from client like "USER foo@bar" will be rewritten into
> canonical form "//bar/foo" then MOUNTED to "pop://localhost/foo@bar"
> by the above MOUNT parameter.

I see. I suppose this will work fine indeed.

>  |Would it be possible to not have a user name with "@host" rewritten 
>  |to a canonical form like "//host/user" before MOUNT is applied ?
> 
> No, at least in the current specification.
> 
>  |But it seems that MOUNT are not applied if the host is specified in 
>  |the SERVER parameter...
> 
> Since I intend to control any rewriting by the generic "mount"
> mechanism, any exceptional jumper is not desired.

I understand.
 
>  |And I also would like to be able to do this kind of rewrites:
>  |
>  |'USER toto_stuff'           ---> 'pop://localhost/toto@titi.com'
>  |'USER toto_stuff@titi..'  ---> 'pop://localhost/toto@titi.com'
>  |
>  |the 'stuff' part can change so I have to match the '_' character.
> 
> A pair of MOUNT parameters:
> 
>   MOUNT="toto_*            pop://localhost/toto@titi.com"
>   MOUNT="//titi.com/toto_* pop://localhost/toto@titi.com"
> 
> will do.

The problem here is that the 'toto' part can change too. In fact I would like the above rewrites to work for any 'toto' and any 'stuff'.

> Any problems should be solved by extending or generalizing the
> "mount", and generalizing matching and rewriting of MOUNT has
> been one of what must be done in future since the beginning of
> DeleGate.  I realized a part of it in the last weekend and it
> will be available in the next release.

Could you send me your new 'mount.c' file or your changes, or tell me how I can get them ?

Do you already know what will be the general rules for matching and rewriting MOUNT ? Perhaps I can help you implement or test them.

Regards,
Christian.


  admin search upper oldest olders older1 this newer1 newers latest
[Top/Up] [oldest] - [Older+chunk] - [Newer+chunk] - [newest + Check]
@_@V