Is James appropriate as a development testbed for a proxy service?

[Alain Ravet <al...@wanadoo.be>]

Hi,


I'm about to start developing a proxy based service, the TDD/Test First 
way.

Whether or not I use James mailets,  I'll need a super fast (low 
overhead) testbed (a pop "server, and a smtp "server").

I foresee that there will be hundreds of small tests, talking to a 
pop/smtp server setup in a clean and given state (users, waiting 
messages, etc..).
Obviously, the "servers" 1°/setup phase, and 2°/ responsiveness has to 
be ULTRA FAST, with the lowest possible overhead.

I previously wrote a mock pop server. It's suitably superfast, but it's 
too simple, and it's only POP.


Questions :
----------
 - is James appropriate (fast enough) for this usage?
 - is there an alternative?

Alain Ravet


---------------------------------------------------------------------
To unsubscribe, e-mail: server-user-unsubscribe@james.apache.org
For additional commands, e-mail: server-user-help@james.apache.org

Re: Is James appropriate as a development testbed for a proxy service?

[bill parducci <bi...@parducci.net>]

Alain Ravet wrote:

 > Are you using it for acceptance tests, or did you use for unit tests,
 > while developing?
 >
 > Could you give a few numbers (rough estimates) : number of tests, time
 > taken, setup time, ..
 > Did you consider, at some point, using/writing mock servers to speed
 > up some tests.
 >

since james is the platform for our application (a mailet based tool), i 
am not sure if our testing methodologies are pertinent to what you are 
doing. unit testing is done internally with the mailet, but also how it 
  interacts with james (so, to answer the question about using it for 
development, the answer is most assuredly yes)

the number of formal tests are in the hundreds, but again since this 
james is core to the overall application architecture this may not 
answer question, especially since we have a 'honey pot' setup for long 
running tests and how the system reacts to a wide variety of inputs; it 
is ever so kind of the spammer community to provide us with copious 
quantities of wildly variable e-mail content just for signing up for a 
few 'free' contests :o)

b



---------------------------------------------------------------------
To unsubscribe, e-mail: server-user-unsubscribe@james.apache.org
For additional commands, e-mail: server-user-help@james.apache.org

Re: Is James appropriate as a development testbed for a proxy service?

[Alain Ravet <al...@wanadoo.be>]

Bill ,

Just to be sure :

>we have been using james for that for over a year now
>(generic platform for proxying e-mail for analysis). it has worked quite
>well. on the other hand 'fast' and 'small' are kinda relative ;o)
>
>  
>

Are you using it for acceptance tests, or did you use for unit tests, 
while developing?

Could you give a few numbers (rough estimates) : number of tests, time 
taken, setup time, ..
Did you consider, at some point, using/writing mock servers to speed up 
some tests.

Alain

---------------------------------------------------------------------
To unsubscribe, e-mail: server-user-unsubscribe@james.apache.org
For additional commands, e-mail: server-user-help@james.apache.org

Re: Is James appropriate as a development testbed for a proxy service?

[bill parducci <bi...@parducci.net>]

anecdotally, we have been using james for that for over a year now
(generic platform for proxying e-mail for analysis). it has worked quite
well. on the other hand 'fast' and 'small' are kinda relative ;o)

b

On Wed, 2003-12-10 at 02:12, Alain Ravet wrote:
> Hi,
> 
> 
> I'm about to start developing a proxy based service, the TDD/Test First 
> way.
> 
> Whether or not I use James mailets,  I'll need a super fast (low 
> overhead) testbed (a pop "server, and a smtp "server").
> 
> I foresee that there will be hundreds of small tests, talking to a 
> pop/smtp server setup in a clean and given state (users, waiting 
> messages, etc..).
> Obviously, the "servers" 1°/setup phase, and 2°/ responsiveness has to 
> be ULTRA FAST, with the lowest possible overhead.
> 
> I previously wrote a mock pop server. It's suitably superfast, but it's 
> too simple, and it's only POP.
> 
> 
> Questions :
> ----------
>  - is James appropriate (fast enough) for this usage?
>  - is there an alternative?
> 
> Alain Ravet
> 
> 
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: server-user-unsubscribe@james.apache.org
> For additional commands, e-mail: server-user-help@james.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: server-user-unsubscribe@james.apache.org
For additional commands, e-mail: server-user-help@james.apache.org