This is my first mail to the list.
I've working with jmeter and I noticed a small problem when loading
jtl files back into a listener. The problem relates to empty tokens in
the jtl
I'm working with JMeter 2.2, JDBC Samplers and Summary report. The jtl
file looks this way
1160833236328,62,Prepared update,,,Grupo de Hilos 1-1,text,true
1160833236406,0,Prepared Select,,,Grupo de Hilos 1-1,text,true
1160833236406,31,Callable Query,,,Grupo de Hilos 1-1,text,true
1160833236437,16,Callable update,,,Grupo de Hilos 1-1,text,true
1160833236453,0,Prepared update,,,Grupo de Hilos 1-1,text,true
Notice the ",,,", well this confused the StringTokeniker used in
OldSaveService.java at line 142. StringTokenizer silently ignores
those empy tokens and the result is that the JTL is not parsed right.
I've made a patch that allows jmeter to parse those files. I'm willing to
contribute this code back, but I'm not sure how to do it. Should I fill a
bug report and put it there?
I also have also detected another problem when loading xml jtl into a
listener. Due to the way that SampleResultConverter.java is recreating
SampleResult from the xml. The startTime of all SampleResults are set
to "now" (System.currentTimeMillis) and that brokes the Summary Report
listener that uses starttimes to figure out the throughput. I also
have a patch for it that I could send if someone tells me about the
right way to contribute it. Should I open a bug report? or this
behavior is known and accepted.
If this is not the right place to post these issues please let me know
and point me in the right direction. Thanks in advance
--
/Rubén
--
/Rubén
--------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
