RE: [OpenSTA-devel] Re: OpenSTA cookie handling
Brought to you by:
dansut
From: Goldberg, C. <cor...@th...> - 2004-11-29 15:39:04
|
>This is not that different from what is currently created for the >user in SCL though - the created SCL relies on the cookie names not >changing from the time of recording but it effectively does >something similar. Do you agree? yes, but this is often useless. Cookies usually contain sessions ID's or some sort of dynamically generated value that is set at login time. So it will not be the same as the value when recorded. >What you are talking about here I think is something that happens >entirely "behind the scenes" though - something that requires no >scripting at all and doesn't care what the names of the cookies are. >You would effectively give the VU it's own browser cookie store that >the replay engine knows about and will use. Yes? exactly >To be flexible >there would have to be some way of initializing the cookie store >before any HTTP requests at script start. Otherwise you couldn't >simulate users who had previously "logged in", etc. >Do they handle it invisibly? Or is it something that the user is >alerted to and/or can see in the "scripts"? How much control does >the user have? i think it should work like: if the user does not do anything in the scripts, cookies are handled (stored/resent) automatically for you. To simulate returning users (persistant cookies), you can override this by providing the cookie in your code. |