Both my threadgroup and HTTP sampler have Umlauts (like ä=a_umlaut) in their names. These do not appear properly in any of the .xml log files (from either view results tree, assertion log, aggregate report or table results). More precisely, an ä appears as a sequence of 5 Bytes, which looks like a character to UTF-8 conversion took place twice on top of each other. The 5 bytes are 0xc3 83 e2 82 ac. The XML files contents (the 5 above bytes replaced with äääää): <?xml version="1.0" encoding="UTF-8"?> <testResults> <sampleResult ... threadName="UserSammelanträääääge1-1" label="HTTP Sammelanträääääge" time="220" ... It should have been "Sammelanträge" (with 2 bytes for ä as per UTF-8). The Umlauts appear fine in the GUI. Regards, Jörg Höhle On Suse Linux 9.1, jdk 1.4.2-b2
Umlauts did not appear OK for me in all of the GUI - there were problems with the text display in the tree visualiser. I've checked in some changes to this visualiser and httpsampler which should hopefully solve the problem of displaying accented characters in the response data pane. Not yet solved the problem of output to the log file
Seems to have been fixed by the move to using XML output based on Xstream. Please re-open if there are still problems in the nightlies or the forthcoming version 2.1
This issue has been migrated to GitHub: https://github.com/apache/jmeter/issues/1313