Mike, These sound really useful. I'll update and try them out in the next few days.
Cheers, Scott -- Scott Eade Backstage Technologies Pty. Ltd. Web: http://www.backstagetech.com.au > From: "Mike Stover" <[EMAIL PROTECTED]> > Reply-To: "JMeter Developers List" <[EMAIL PROTECTED]> > Date: Fri, 12 Jul 2002 23:31:08 -0400 > To: [EMAIL PROTECTED] > Subject: built-in functions for jmeter > > So, I have implemented enough that the first built-in function is now > functional. There is no > GUI help for writing it yet, so using it will be a complete mystery to ya'll. > > Briefly, it's a function that let's you write a regular expression to grab a > string from > the previous response text and make a new string with it. The arguments are: > > 1. The regular expression to apply to the response text > 2. How to generate a new string. This is like a template string that includes > groups from the > regular expression. You'll need to learn something about Perl regular > expressions to > understand this part. You can refer to any particular group in your regular > expression with a > syntax like $1$ (gets replaced with the first group from your regex). > 3. Which match to use. Your regex might find multiple matches. You can pick > a number (1 is > the first match, 2 the second, etc), you can use RAND to indicate JMeter > should pick one at > random, ALL means to use all matches (this is kinda complicated, so nevermind > for now). > 4. The between text. Say what? Has to do with that ALL option mentioned > above. > 5. Default text. If your regex doesn't match, you can set the default string > to use instead. > > 3,4, and 5 are optional and can be entirely absent. > > Example: > ${__regexFunction(<a href="(/[^"]*)",$1$,RAND,,/index.html)} > > This finds links that start with '/' and groups the path to be used in the > template. The template > is simply the path parsed from the regex. A random match will be chosen each > time the > function is run, no between text is given, and the default is string if no > match is found is > '/index.html'. > > If you need to use a comma in any of these values, you'll need to use the HTTP > encoding for > a comma, which you'll have to look up yourself. > > I was surprisingly easy to use. The hardest part is writing the regular > expression, which > shouldn't be much of a surprise, I guess. > > -- > Michael Stover > [EMAIL PROTECTED] > Yahoo IM: mstover_ya > ICQ: 152975688 > > -- > To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> > For additional commands, e-mail: <mailto:[EMAIL PROTECTED]> > -- To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]> For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
