On 20/06/06, Willingham Geoff <[EMAIL PROTECTED]> wrote:
Sorry - should have been a bit clearer...
Quote: "There is no "data store" element."
By 'data store' element, I had meant a new Post Processor that I will
write. Basically, an element that takes the result, and sticks it in a
hashtable (or a string, or some other reference). The Link Parser could
be modified to take an optional parameter: the name of the 'data store'
it should retrieve the result from. If the name is not specified, it
should just use the most recent result (current behaviour).
OK - though the response data could be quite large, so it might be
best to store just the object reference. This can be done using a
thread variable.
The only problem I currently have is working out how to get the Link
Parser to walk the element tree, looking for that named element (and
deciding on the behaviour should the named element not be an instance of
the 'data store' post processor class).
No need - just read the thread variable.
Sound reasonable ?
Geoff
-----Original Message-----
From: sebb [mailto:[EMAIL PROTECTED]
Sent: 20 June 2006 12:33
To: JMeter Developers List
Subject: Re: How to access a response other than the most recent
On 20/06/06, Willingham Geoff <[EMAIL PROTECTED]> wrote:
> Unfortunately, the file download appears in the middle of a very long
> page sequence (40+), each of which uses a Link Parser to pass hidden
> fields on to the next request. No matter where the file requests are
> made, they create a break in the chain of forms :/
>
> Do you think it would be possible to make use of the View Results Tree
> to access previous results, given that it can be set to cache all
> results ?
Don't know without looking carefully.
I probably won't have time for that until late July.
> Maybe a simple Post Processor that holds the results for later parsing
> would also work. Due to time constraints, I cannot currently take the
> time to split the Link Parser into two entities, but I may be able to
> knock up a simple 'data store' type post processor. If I do this, how
> easy would it be to crawl the Test Element tree, to find the relevant
> 'data store' element ?
There is no "data store" element.
Samples are passed directly to Listeners.
Some listeners store results, but these may be in a different form -
e.g. the Aggregate Report stores just the timings, not the data.
> Thanks for the help and ideas ;)
I suppose it might be possible to have an option to ignore certain
sample results (or types of results) for the purpose of resetting the
"last sample" value. This would get round the problem for you. But I'm
not sure how useful that would be to anyone else - it would need to be
optional (via a property, or an extra option on the sampler or
something).
The other possibility might be for you to use the RE post-processor to
extract the links. This might be hard work - or it might be easy.
> Geoff
>
> -----Original Message-----
> From: sebb [mailto:[EMAIL PROTECTED]
> Sent: 20 June 2006 11:32
> To: JMeter Developers List
> Subject: Re: How to access a response other than the most recent
>
> I'm not sure you can do that - the previous sample is saved locally,
> and is not in any tree. Although one could additionally save the next
> last sample, this would not be a generic solution.
>
> Can you download the file first?
>
> The Link Parser is a bit odd, as it is a Pre-Processor and a
> Post-Processor in one.
> Perhaps you can create a version that splits it in two - one part
> extracts the links, the other applies them.
>
> S.
> On 20/06/06, Willingham Geoff <[EMAIL PROTECTED]> wrote:
> > Greetings
> >
> > I'm currently using the Http Link Parser to fill forms as part of my
> > JMeter script. However, im running into a problem caused by the
> > design of the site being tested. In several locations, we have a
> > link
>
> > that performs two actions.
> > a) transitions to the next page, and
> > b) uses javascript to open a new window, and download a file.
> >
> > Whilst both actions have been modelled in my script, the call to
> > fetch
>
> > the file means that the following request cannot use a Http Link
> > Parser, because the previous response (the downloaded file) does not
> > contain any form or link data.
> >
> > I am hoping to modify the Http Link Parser to allow the user to
> > specify the name of a HttpRequest object, so that the Link Parser
> > can then use the response to that object.
> >
> > However, having looked at the code, I am uncertain as to how I can
> > access the tree of test elements, how to find the specified Http
> > Request, and how to access the response to that request.
> >
> > Can anyone point me in the right direction ?
> >
> > cheers
> >
> > Geoff
> > p.s. Cheers for the help on finding the classes for the Link Parser
> > last week :)
> >
> > This e-mail is confidential and privileged. If you are not the
> intended recipient please accept our apologies; please do not
> disclose, copy or distribute information in this e-mail or take any
> action in reliance on its contents: to do so is strictly prohibited
> and may be unlawful. Please inform us that this message has gone
> astray before deleting it. Thank you for your co-operation.
> >
> >
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
